A Study Among Turkish-Dutch Muslims Maykel Verkuyten

Religious Group Identification and Inter-Religious Relations: A Study Among Turkish-Dutch Muslims Maykel Verkuyten

To cite this version:

Maykel Verkuyten. Religious Group Identification and Inter-Religious Relations: A Study Among Turkish-Dutch Muslims. Group Processes and Intergroup Relations, SAGE Publications, 2007, 10 (3), pp.341-357. 10.1177/1368430207078695. hal-00571655

HAL Id: hal-00571655 https://hal.archives-ouvertes.fr/hal-00571655 Submitted on 1 Mar 2011

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Group Processes & Intergroup Relations 2007 Vol 10(3) 341–357

Religious Group Identiﬁ cation and Inter-Religious Relations: A Study Among Turkish-Dutch Muslims

Maykel Verkuyten Utrecht University

Following social identity theory, this research examines the relationship between group identifi cation and intergroup relations by focusing on religion. Religious as well as Dutch national group identifi cation was examined among Turkish-Dutch Muslims. Identifi cation was studied in relation to general affective ratings of multiple religious groups (Muslims, Christians, Hindustanis, Jews and non-believers) and the endorsement of Islamic group rights. The results show that Muslim identifi cation was more like a nominal (high or ‘total’ identifi cation) than a continuous variable, and that many participants showed low identifi cation with the national group. The affective ratings of religious out-groups were quite negative, particularly of the Jews and non-believers. Muslim identifi cation was positively and strongly related to feelings toward the religious in-group and to the endorsement of Islamic group rights. National identifi cation was positively related to feelings toward the religious out-groups, but only for ‘total’ Muslim identifi ers, supporting the mutual intergroup differentiation model. The fi ndings are discussed in relation to social psychological thinking about group identifi cation and the importance of religion for intergroup relations. keywords minority groups, national identity, religious group relations, religious identifi cation

A Central assumption of social identity theory There is considerable empirical evidence that (SIT; Tajfel & Turner, 1986) is that a positive in an intergroup situation those with high in- social identity is typically based on favorable group identifi cation are more likely to show a intergroup comparisons. Brown (2000) argues variety of group level responses relative to those that a plausible inference to draw from this assumption is that identifi cation with a group is Author’s note positively related to the tendency to favor that Address correspondence to Maykel group over other groups. People with high and Verkuyten, Faculty of Social Sciences, low psychological commitment to their group Utrecht University, Heidelberglaan 2, 3584 (high and low identifi ers) can be expected CS Utrecht, The Netherlands to differ in their reactions and evaluations. [email: [email protected]]

Copyright © 2007 SAGE Publications (Los Angeles, London, New Delhi and Singapore) 10:3; 341–357; DOI: 10.1177/1368430207078695 Group Processes & Intergroup Relations 10(3) shown by low identifi ers (see Ellemers, Spears, the associations of these group identifi cations & Doosje, 1999). with feelings toward Muslims, Hindustanis, However, not all studies support the link Christians, Jews, and non-believers. Third, both between in-group identifi cation and in-group group identifi cations were examined in relation favoritism and the assumption of this link has to the endorsement of Islamic group rights in also been criticized theoretically (e.g. Turner, the Netherlands. With this study, I hope to make 1999). A positive correlation between indi- a contribution to a further understanding of the vidual differences in group identifi cation and important processes of group identifi cation. In individual differences in the degree to which particular, I will examine the assumed continuous the in-group is favoured over an out-group is not nature of group identifi cation and will discuss always found and can be quite weak (see Hinkle the role of dual identifi cation. In addition, & Brown, 1990; Mullen, Brown, & Smith, 1992). although religion is an important dimension for In addition, Turner (1999; see also McGarty, defi ning a positive social identity in comparison 2001; Turner & Reynolds, 2001) has criticized the to dissenters and non-believers, social psychology idea of a simple positive correlation between has not paid much attention to religious identity identifi cation and in-group favouritism because (but see Argyle & Beit-Hallahmi, 1975; Batson & SIT argues that in-group favouritism is a func- Burris, 1994; Blaine & Crocker, 1995). Religion tion of, for example, status positions and unifi es the community of believers around a con- beliefs about the nature of the groups and the sensus of values and truths, and in doing so makes intergroup context. Furthermore, group iden- meaningful in-group and out-group distinctions tifi cation is a multifaceted construct that can be which contribute to social divisions and current examined in various ways (see Ashmore, Deaux, confl icts in many parts of the world. & McLaughlin-Volpe, 2004), and there is also the possibility of dual or hybrid identities (Verkuyten, Religious identifi cation 2005). Turner (1999, p. 22) further argues that it is important to focus on identities that ‘cor- There is a great deal of variation and inconsist- respond to the subjective division of the social ency in the way that group identifi cation is con- world by the subjects in relation to the intergroup ceptualized and measured (see Ashmore et al, attitudes obtained’. The identities should have 2004; Cameron, 2004; Jackson & Smith, 1999). real meaning and the evaluative dimensions For example, a distinction among various di- used for assessing intergroup relations should be mensions or elements of identifi cation is made, relevant in relation to various out-groups. Most such as evaluation, importance and attachment. social psychological studies have examined Group identifi cation is further interpreted in identifi cation processes among student groups terms of situational sensitive self-categorizations, and the in-group favouritism found is typically but also in terms of more stable differences in due to a more positive evaluation of the in-group the degree to which psychologically central compared to a less positive or neutral evaluation and valued group memberships have developed. of an out-group. Few studies have focused on The former interpretation focuses on the real-world situations where group identifi cation changing ways that people defi ne themselves can be very strong and clear negative out-group and the extent to which they feel attached to a reactions and evaluations may exist. specifi c group in a specifi c context (e.g. Ellemers The present study focuses on religious and et al., 1999). The latter one argues that some Dutch national group identifi cation among people, for whatever reason, are more inclined Turkish-Dutch Muslims. Identifi cation is studied than others to see themselves as a group member in relation to general affective ratings of multiple and to value their group membership (e.g. religious out-groups and the endorsement of Phinney, 1992). Islamic group rights. I examined, fi rst, the nature In both interpretations, group identifi cation of Muslim and Dutch group identifi cation and can be considered as an individual difference their interrelationship. Second, the focus was on variable which affects various group level

342 Verkuyten religious group identification and inter-religious relations responses. Group identifi cation, for example, is found that 74% of Muslim participants indicated an important factor affecting a person’s readiness that their religion was ‘very important’ in living to use a social category for self-description, their lives and 21% said ‘fairly important’. and people who feel highly committed to their Furthermore, around 80% indicated that they group are inclined to act in terms of their group visited a mosque once a week or more. In membership. Group identifi cation implies that comparison, the highest percentage for ‘very collective beliefs and values that characterize important’ for the Hindu, Sikh, Catholic and the in-group become normative and part of the Church of England participants was 46% for psychological self. People start to think, feel the Sikhs. and act in terms of the way that the in-group is Among a representative sample from the city understood. These group understandings can of Rotterdam, Phalet and Güngör (2004) found be relatively stable or enduring refl ecting, for that Islam was considered ‘very meaningful and example, cultural features, but they are also important’ in one’s life by 87% of the Turkish responsive to social events and current threats and 96% of the Moroccan population. These (Tajfel & Turner, 1986). percentages were similar for younger (18–30 Triandis (1992) points out that in some col- years of age) and older participants (> 30 years). lectivist cultures one either is or is not a mem- In addition, around two-thirds of the Turks and ber of the in-group. In these cultures, group Moroccans had a very strong Muslim identity. identifi cation is not so much a matter of degree For the great majority of Muslims, Muslim and one cannot be more or less identifi ed with identity was a given and not being a Muslim was a group: group identification is more of a not a real option. The same has been found in nominal rather than a continuous variable. The surveys in Brussels, Belgium (Phalet, 2004), and orientation and commitment to the in-group in other European countries (Haddad & Smith, is normative and total, rather than optional 2001; Vertovec & Rogers, 1999). and differing in strength. Religious group The data for the Rotterdam study were col- identifi cation can be similar. Religion is often lected in 1999 but it is highly unlikely that these of profound importance to people’s lives and percentages have dropped. In the Netherlands religious groups are among the more salient a policy of multiculturalism was adopted in the buttresses of identity. The lives of observant 1980s in response to the increased infl ux of believers are organized around their religi- ‘foreigners’. The recognition that many ‘guest- ous beliefs, values and practices. These ideas and worker’ migrants would remain in the country values involve religious truth-claims and absolute led to a policy for ‘integration with retention moral principles that defi ne what it means to be of the own identity’ (Entzinger, 2003, p. 63). a believer of a particular religion and that lead Dutch policies saw immigrants according to to dogmatic thinking (Altemeyer, 2002). Islam their group membership and not primarily is a religion that presents guidelines, referred as individuals. The ‘pillarization’ tradition of to in the Quran (1:6) as the ‘straight way’, for institutionalized pluralism provided a wide living in accordance with the will of Allah. These range of cultural opportunities and group rights, fundamental principles are known as the Five such as local voting rights for non-nationals and Pillars of Islam. The fi rst pillar is the Shahada or public funding of Islamic schools. However, declaration of faith and has a central place in the much has changed since the 1980s (see Joppke, lives of Muslims.1 A person becomes a Muslim 2004). The previous ‘ethnic minorities policy’ with the declaration of the Shahada in front of has gradually been replaced by a policy of civic two witnesses, and one either is a Muslim who integration (Entzinger, 2003) and in public is committed to Islam or one is not. debates multiculturalism has been described Research has shown that for Muslims living in as a ‘drama’ and a ‘failure’, and assimilation Western Europe, religion has great importance has been proposed as the only viable option in the way they live their lives. In a nation-wide (e.g. Schnabel, 2000). In the last fi ve years in study in Great Britain, Modood et al. (1997) the Netherlands, Islam has increasingly become

343 Group Processes & Intergroup Relations 10(3) the ‘negative other’. In the Dutch media, Islam participants with a forced-choice test to measure has become symbolic for problems related to ethnic identifi cation, such as in studies using ethnic minorities and immigration (see Ter black and white dolls or questions such as ‘do Wal, 2004) and some politicians have defi ned you feel Dutch or Turkish’. As a result, identi- Islam as a backward religion and Muslims as a fi cation was studied as an either/or phenomenon ‘fi fth column’, and have argued that ‘a cold war and the possibility of a hyphenated or dual against Islam is unavoidable’ (see Verkuyten & identity was disregarded. The result was, for Zaremba, 2005). example, the assumption that the more strongly As a result, the current public discussion people identifi ed with their own minority group, strongly focuses on the need to compel Islamic the more they would distance themselves from groups to assimilate because they are a threat to others. Dutch values and identity. Thus, the emphasis on However, in a situation where people are free Islamic groups has become more prominent but to describe themselves, they do not always use it is not new. In the Netherlands, these groups this dichotomous scheme of group identifi - are at the bottom of the ethnic hierarchy and ca ti on (e.g. Hutnik, 1991; Sanchez & Fernandez, they have been the prototypes or paradigmatic 1993). Fin dings from cross-cultural studies examples of minority groups for quite some time suggest that group identi fi cation is not necessar- (Hagendoorn, 1995). Hence, Islamic groups ily a singu lar given but may be constitu ted of clearly face high levels of threat to the value of their hyphenated identities that indicate varying religious identity and the public condemnation degrees of identifi cation with both the ethnic of Islam and the plea for assimilation can lead minority group and the majority group simul- to strong in-group identifi cation among these taneously. For example, for many Turks living groups (Verkuyten & Zaremba, 2005). Thus, it in the Netherlands it is often not a question of was expected that Muslim identity would be very being Turkish or Dutch but a question of the important for most Turkish-Dutch participants, extent to which they feel Turkish as well as the making Islamic group identifi cation more like a degree to which they feel Dutch (Verkuyten, nominal rather than a continuous variable. 2005). They can consider themselves members of their ethnic in-group but also as having to Dutch identifi cation relate to, if not quite be members of, the majority group. In acculturation research, Berry’s (1984) two- These identities do not have to contradict each dimensional model has become one of the most other or get in each other’s way because they are used frameworks. The model has been supported of different kinds: they are differently defi ned or empirically in various studies (e.g. Ryder, Alden, situated on different levels of abstraction. One & Paulhus, 2000; Ward & Rana-Deuba, 1999) and can be a member of an ethnic group as well as proposes a combination of culture maintenance a superordinate national category. There is and mainstream contacts, resulting in four dif- little problem as long as these identities are not ferent acculturation positions: assimilation, defi ned on the same level of abstraction and in separation, integration and marginalization. contrasting or competing terms. Depending on By analogy with the two-dimensional model the situation the one or the other is relevant of acculturation, ethnic identification and and becomes salient. identifi cation as a member of the new society However, different identities do sometimes can be thought of as two dimensions that vary get in each other’s way. For Polish Tatars, for independently. Hutnik (1991) developed such example, being simultaneously a Tatar, a Muslim a model, independently of Berry’s work. and a Pole is central to their self-understanding Traditionally, identifi cation with one’s own (see Cieslik & Verkuyten, 2006). Islam as practiced ethnic group was considered to be inversely by the Polish Tatars is the main characteristic related to identifi cation with other groups, in par- distinguishing them from the Catholic Poles. ticular the majority group. Research presented The Tatars perceive themselves as real Muslims

344 Verkuyten religious group identification and inter-religious relations even though theirs has developed as a specialized circumstances, people will try to maintain or local variant of Islam. Furthermore, the social restore a positive and distinct collective identity, and cultural aspects of their religious life are for example by increased in-group favouritism. central for maintaining Tatar ethnic solidarity. The public condemnation of Islam and the plea In contrast, Catholicism is strongly connected for assimilation implies that Islamic groups with the idea of the Polish nation and a central clearly face threats to the value of their religious aspect of ‘being Polish’ (Warminska, 1997). identity. Other religious groups present a threat Polish national identity is largely based on being to the integrity of the Islamic in-group strug- Catholic and includes negative stereotypes of gling to maintain a valuable and distinctive other religions, particularly Islam. Hence, the identity. To enhance the value and distinctiveness Tatars’ claim to Polishness can be problematic of their religious in-group, group members can when their religious background is emphasized derogate other religious groups (Rothgerber & too much, making it diffi cult to fi nd the right Worchel, 1997). Hence, we expected that the balance between being Polish and Muslim at feelings towards religious out-groups would be the same time. rather negative. In the Netherlands, being simultaneously However, there may also be differences in feel- a Muslim and a Dutch national is often not ings towards various religious out-groups living easy. In public debate the secular nature and in the Netherlands: e.g. Christians, Hindustanis, Christian heritage of Dutch society is increas- Jews and non-believers. Different out-groups can ingly emphasized. Among Turkish-Dutch people, enjoy varying degrees of social acceptability. In a close association between Turkish and Muslim relation to ethnicity, there is evidence for this in identifi cation has been found (r > .60), and countries such as Canada, the United States, the several studies in the Netherlands have found former Soviet Union and the Netherlands (for that around a third of the participants of different reviews see Hagendoorn,1995; Owen, Eiser, & ethnic minority and age groups indicate a dual McFaul, 1981), and among ethnic majority and or hyphenated identity (e.g. Turkish-Dutch), minority group members (Berry & Kalin, 1979; whereas about half of them identify only with Verkuyten, Hagendoorn, & Masson, 1996). their own group (e.g. only Turkish) and not There can be many reasons for the different with the Dutch (see Verkuyten, 2005). Thus, evaluation of out-groups. For example, research it was expected that two groups of participants has shown (see Hagendoorn, 1995; Verkuyten could be distinguished: low and high Dutch et al., 1996) that cultural differences, negative identifi ers. stereotypes and the degree to which out-groups are perceived as threatening the status and Multiple religious out-groups interests of the in-group play a role. In addition, concerns about beliefs, values and norms that In general, there is a close relationship between defi ne the collective identity are important. religion and prejudice. The more religious an Religious belief is not about personal preferences individual is, the more prejudiced he or she is or social conventions, but about convictions. likely to be (see Batson & Burris, 1994; Scheepers, It is concerned with moral good and divine Gijsberts, & Hello, 2002). Particularly, people truth that is diffi cult to reconcile with moral whose religious beliefs provide clear-cut moral and epistemic diversity. The observant believer truths tend to be negative towards out-groups. believes that he or she is right and will fi nd it Feelings towards out-groups are also related, diffi cult to have positive feelings towards non- however, to the intergroup situation (Tajfel & believers that implicitly challenge his or her Turner, 1986). life. Atheism cannot do without believers, and In the Netherlands, questions of religious and vice versa. In the Rotterdam study, around 45% cultural differences and confl icts have become of the Islamic participants indicated that they core issues in public and political debates. had ‘completely no sympathy’ for non-believers According to SIT, under identity threatening (Phalet & Güngör, 2004). Hence, we expected

345 Group Processes & Intergroup Relations 10(3) that the feelings towards non-believers would identifi cation can be expected to be associated be more negative than those towards Christians with less negative religious out-group feelings. and Hindustanis. People of the latter two groups Hence, an interaction effect between Dutch and do at least consider religion important and Islamic group identifi cation was expected. Christianity has, of course, many similarities with Islam. Islamic group rights In addition, we expected the feelings towards Jews to be quite negative. In many European Berry and Kalin (1995) argued that groups are countries, including the Netherlands, there more in favour of cultural diversity and group is a growing concern about increased anti- rights when they see advantages for themselves. Semitism, which certainly is not only limited to Several theories have emphasized the role of Islamic groups but which has also been manifest group interests in the dynamics of intergroup in, for example, mosques across Europe and relations (e.g. Blumer, 1958; Bobo, 1999; among Muslim youth (Antisemitism Research, Tajfel & Turner, 1986). For religious minority 2002). One source for this antisemitism is the groups, minority rights offer the possibility of confl ict between Palestinians and Israelis, and maintaining and expressing their own distinct- between observant Muslims and observant ive religious identity, and obtaining more equal Jews in particular. The concern about anti- social status in society. Hence, we can expect that Jewish tendencies among Muslims is further the participants will support Islamic group rights. illustrated by the many books and websites that In addition, the more strongly Muslims identify fi nd it necessary to argue that Islam actually with their religious in-group, the more likely denounces antisemitism. they are to consider it important to express and The affective ratings towards the various preserve their own religion and to participate as religious out-groups living in the Netherlands religious group members in social and political were expected to be related to Muslim and Dutch life. In contrast, from the perspective of common national identifi cation. Participants with a very group identity (Gaertner et al., 1993), it can strong Muslim identifi cation were expected be expected that Dutch identifi cation will be to be more negative than participants with a negatively related to the attitude towards Islamic less strong Muslim identifi cation. In addition, group rights. This is particularly likely in the higher Dutch identifi cation was expected to context of a very strong Muslim identity that be associated with less negative religious out- minimizes feelings of identity threat (Hornsey group feelings. The reason for this expectation & Hogg, 2000). is that a common national identity implies that religious group distinctions can be subsumed Method into a superordinate category. This shared category can reduce negative feelings because the Participants religious out-group members become fellow A questionnaire was used and in total there national in-group members (Gaertner, Dovidio, were 217 Turkish-Dutch participants. On an Anastasio, Bachman, & Rust, 1993; Gaertner, open-ended question, 206 of these participants Mann, Murrell, & Dovidio, 1989). Hornsey and described themselves as Muslim (Sunni), and Hogg (2000) have shown that such a reduction these participants have been included in the is particularly likely when the superordinate analyses. Of these participants 68.8% were (national) category membership is combined males and 31.2% females. The participants with a strong (religious) subgroup identity (see were between 16 and 62 years of age and their also Hewtsone & Brown, 1986). Such a combin- mean age was 30.18 (SD = 10.63). The participants ation helps to reduce threats to a valued identity were recruited in the Utrecht region and were that may result from assimilation to the national asked to participate in research on contemporary category. This means that for strong Islamic social issues. In the introduction it was explained group identifi ers in particular, higher Dutch that the study was on people’s opinions about

346 Verkuyten religious group identification and inter-religious relations the ‘Netherlands and Dutch society’ and that Under each target group a scale was presented we wanted to know what people thought about running from 0 to 100. the societal situation in the country. The study For assessing the attitude towards Islamic was presented as focusing on opinions of people group rights in the Netherlands, eight items in general, but the participants were recruited that are relevant in the Dutch context were used. by a Turkish-Dutch assistant. The participants These items were partly taken from Verkuyten received €5 for their cooperation. and Yildiz (2006) and focus on public rights. The items were: ‘The right to establish own Measures Islamic schools should always exist in the Muslim group identifi cation was assessed by Netherlands’; ‘some Islamic holy days should six items using 7-point scales (with 1 as ‘dis- become offi cial Dutch holidays’; ‘Dutch TV agree strongly’, 4 as ‘neutral’, and 7 as ‘agree should broadcast more programes by and for strongly’). The items are similar to questions Muslims’; ‘in the Netherlands the wearing of a used in various social psychological studies and headscarf should not be forbidden’; ‘ Muslims in Dutch studies on ethnic identifi cation (see should establish an Islamic political party’; ‘The Verkuyten, 2005). The items were: ‘My Muslim Dutch government does not really listen to what identity is an important part of my self’; ‘I Muslims want’; and ‘in the Netherlands, more identify strongly with Muslims’; ‘I feel a strong measures against discrimination of Muslims attachment to Muslims’; ‘Being a Muslim is a are needed’. Items were measured on scales very important part of how I see myself’; ‘I am ranging from 1 (disagree strongly) to 7 (agree proud of my Islamic background’; and ‘I feel a strongly), and the eight-item scale was internally strong sense of belonging to Islam’. Cronbach’s consistent with Cronbach’s alpha which was .88. alpha was .94. A higher score indicated a stronger endorsement Dutch identifi cation was assessed by four items of Islamic group rights. using the same 7-point scale. The items were: ‘I identify with Dutch people’; ‘I feel that I am Results2 Dutch’; ‘I feel connected to the Netherlands’; and ‘Being Dutch is an important part of how Group identifi cations I see myself’. Cronbach’s alpha was .85. On the basis of a 7-point scale, the mean score In order to measure religious group feelings, indicated strong Muslim group identifi cation the participants were given the well-known (M = 5.81, SD = 1.79). The distribution was ‘feeling thermometer’. This thermometer has negatively skewed (–1.45) and the mode was 7.0. been successfully used in different studies of In total, 50.5% of the participants had a score both ethnic majority and ethnic minority group of 7 indicating that for half of the sample their participants, including in the Netherlands Muslim identity was an integral or inextricable (e.g. Dijker, 1987; Verkuyten & De Wolf, 2002). part of how they saw themselves. Because of the It is intended as a global measure of in-group skewness of the distribution, a median split was and out-group feelings. The exact wor ding of used for making a distinction between high the instructions was: (M = 4.59, SD = 1.88) and ‘total’ (M = 7.0) Muslim Use the ‘feeling thermometer’ to indicate whether group identifi cation. you have positive or negative feelings about different The mean score for Dutch identifi cation was religious groups living in the Netherlands. You may 3.80 (SD = 1.68, median 4.0), which is around use any degree between 0 and 100, but you have to the neutral midpoint of the scale. Of the partici- choose one. 100 degrees indicates very positive or pants, 46.6% had a score below the midpoint warm feelings and zero degrees indicates very cold indicating low identifi cation with the Dutch, or negative feelings. and 45.7% had a score above the midpoint Following the instruction, fi ve religious groups indicating high Dutch identification (7.7% were listed in the following order: Hindustanis, at the midpoint). The high and total Muslim Christians, Muslims, Jews and non-believers. identifiers differed significantly for Dutch

347 Group Processes & Intergroup Relations 10(3) identifi cation, t(206) = 4.73, p < .001. Dutch feelings towards non-believers (–.41, p < .001), identifi cation was stronger in the former group but not signifi cantly towards the other religious compared to the latter (M = 4.33, SD = 1.53, and out-groups. The affective ratings of the four out- M = 3.28, SD = 1.68, respectively). groups were signifi cantly and positively related

Classification of the participants in four (ps < .001; range between .37 and .58). identity positions can be achieved by using the A repeated measures multivariate analysis bipartite split of the two identifi cation scales. For of variance was conducted with the fi ve religious Dutch identifi cation, the scores at the neutral groups’ ratings as a repeated measures factor. midpoint of the scale were disregarded. The Muslim identifi cation (high versus total) and results are shown in Table 1. All four combin- Dutch identifi cation (low versus high) were ations of group identifi cation are evident, but the independent factors. The analysis yielded two combinations were predominant: (1) total a significant main effect for group ratings, Muslim identifi cation and low Dutch identi- F(4, 206) = 139.28, p < .001. As shown in Table 2, fi cation; and (2) high Muslim identifi cation participants clearly had the most positive feel- and high Dutch identifi cation. Both measures ings towards Muslims. For the religious out- were signifi cantly and negatively related but not groups, the mean score for the Christians was strongly (Phi = –.24, p < .01). around the neutral midpoint (50) of the scale, followed by the Hindustanis, the non-believers Thermometer ratings and then the Jews. The feelings towards the For the thermometer questions the focus was latter two groups were quite negative. Around on the affective ratings of Muslims, Hindustanis, a third of the participants indicated extremely Christians, Jews and non-believers. The rating negative feelings (score zero) towards these two of the Muslims was negatively related to the groups, and more than half scored below the neutral midpoint of the scale. In contrast, more than half of the participants indicated extremely Table 1. Participant Identifi cation (%) with the Muslim positive feelings (score 100) towards Muslims. group (high versus total) and with the Dutch Except for the difference in feelings towards (de-identifi cation versus identifi cation) Jews and non-believers, all pair-wise compari- Muslim identifi cation sons between the fi ve groups were signifi cant

(p < .01). In addition, the scores for the Muslims High Total s and Christians were significantly above the Dutch identiﬁ cation neutral midpoint of the scale (t = 20.66, p < Low 19.3% 31.3% .001, and t = 2.21, p < .05, respectively), whereas (N = 37) (N = 60) the scores for the Hindustanis, Jews and non- High 30.7% 18.8% believers were below the midpoint (t = 1.76, (N = 59) (N = 36) p = .08, t = 6.38, p < .001 and t = 5.49, p < .001, χ2(1, 192) = 11.02, p < .001. respectively).

Table 2. Percentages, mean scores, and standard deviations for the thermometer ratings of the ﬁ ve religious groups Thermometer ratings

0 10–40 50 60–90 100 Mean SD Muslims 2.4 3.9 6.7 33.3 51.7 83.8 23.7 Christians 9.1 13.8 30.6 41.2 5.3 53.9 25.7 Hindustanis 12.4 23.0 31.1 28.7 4.8 46.7 27.1 Non-believers 31.4 19.5 18.7 23.7 6.7 37.3 33.3 Jews 33.5 21.9 18.2 20.7 5.7 35.3 33.4

348 Verkuyten religious group identification and inter-religious relations

The analysis showed a signifi cant interaction participants, Dutch high identifi ers had less effect between group ratings and Muslim iden- negative feelings towards religious out-groups tifi cation, F(4, 206) = 27.67, p < .001. Simple than Dutch low identifi ers. The participants main effect analyses indicated an effect for with total Muslim identifi cation who also had identifi cation on the feelings towards Muslims low identifi cation with the Dutch had the most and towards non-believers. As shown in Table 3, negative or cold feelings towards the religious compared to high identifi ers the participants out-groups. with total Muslim identifi cation indicated more positive feelings towards Muslims (explaining Islamic group rights 28% of the variance), and more negative feelings The mean score for Islamic group rights indicated towards non-believers. Muslim identifi cation was that the participants were in favour of special not related to the feelings towards Christians, rights for Muslims (M = 5.52, SD = 1.34, median Hindustanis and Jews. = 5.87). Regression analysis was used to predict There was also a signifi cant interaction effect the endorsement of Islamic group rights. Muslim between group ratings and Dutch identifi cation identifi cation and Dutch identifi cation as well as (F(4, 206) = 3.30, p < .01). Simple main effects their interaction were used as predictors. In the analyses showed signifi cant effects (ps < .05) for regression equation, the high Muslim identifi ers the four religious out-groups, but not for the were compared with the total identifi ers using Muslim in-group (see Table 3). Compared to contrast coding (Aiken & West, 1991). In this low identifi ers, Dutch identifi ers reported less procedure the criterion measure is regressed on negative feelings towards all four out-groups. the predictors and their interactions with the As expected, these effects for out-group feelings contrast between the two groups. The contrast were qualifi ed by a three-way interaction effect be- is coded –1 and +1 respectively for the high and tween group ratings, Muslim identifi cation and total Muslim identifi ers. Dutch identifi cation, F(4, 206) = 2.97, p < .05. The regression model explains no less than Further analyses indicated signifi cant effects 48% of the variance in the endorsement of for the four religious out-groups (ps < .05, Islamic group rights, F(3,189) = 57.25, p < .001. for the Jews, p = .054). As shown in Figure 1, Muslim identification was a significant and Dutch identifi cation turned out to be related strong positive predictor (β = .73, t = 9.51, to the out-group ratings for the participants p < .001). The total Muslim identifi ers were more with total Muslim identifi cation only. For these in favour of Islamic group rights than high Muslim

Table 3. Mean thermometer ratings (and standard deviations) for participants with high and total Muslim identifi cation, and for participants with low and high Dutch identifi cation Muslim identifi cation Dutch identifi cation

High Total F-value Low High F-value Muslims 71.3 96.0 74.34*** 86.6 82.4 0.36 (27.1) (9.4) (24.1) (23.2) Christians 56.2 51.7 0.16 46.6 61.6 16.17*** (21.9) (18.9) (26.9) (23.2) Hindustanis 49.8 43.7 1.31 41.5 51.1 4.88* (24.8) (29.0) (28.0) (25.8) Non-believers 49.2 25.8 23.53*** 29.1 44.3 4.95* (31.7) (30.8) (32.3) (33.0) Jews 38.2 32.5 0.15 26.3 43.2 15.31*** (32.0) (34.6) (30.3) (34.1) *p < .05; ***p< .001.

349 Group Processes & Intergroup Relations 10(3)

Figure 1. Mean thermometer ratings by Muslim and Dutch identifi cation identifi ers (M = 6.46, SD = 0.61, and M = 4.56, high Muslim identifi ers, Dutch identifi cation SD = 1.19, respectively). Dutch identifi cation was not related to Islamic group rights. did not make a signifi cant contribution to the prediction of the endorsement of Islamic group Discussion rights. The interaction between Muslim and Dutch identifi cation was also not signifi cant, Identifi cation establishes a psychological link but had a marginal effect (β = –.21, t = 1.62, between the individual and the group. As p = .098). soon as people identify with their group, that In an additional regression analysis, the con- group becomes the basis for thinking, feeling tinuous centered score of Dutch identifi cation and acting. Hence, for social psychologists was used as a predictor (Aiken & West, 1991). working within the tradition of social identity In this analysis, the interaction effect between theory, essential questions are, ‘how do people Muslim identifi cation and Dutch identifi ca- identify with a group, and precisely what are the tion was signifi cant (β = –.27, t = 2.07, p < .05). consequences of such identifi cation’ (Hogg & Separate regression analyses for high Muslim Abrams, 1988, p. 2, their italics). Group iden- identifiers and for total Muslim identifiers tifi cation is conceptualized in various ways (e.g. showed that Dutch identification was not Ashmore et al., 2004; Cameron, 2004; Jackson signifi cantly associated with Islamic group rights & Smith, 1999), and social psychological studies for the former group (β = .12, t = 1.17, p > .10). have examined various groups and different For the latter group this association was nega- intergroup situations. However, few studies (e.g. tive (β = –.22, t = 2.32, p < .05). Thus, Dutch Argyle & Beit-Hallahmi, 1975; Batson & Burris, identifi cation was related to lesser endorsement 1994; Blaine & Crocker, 1995) have focused on of Islamic group rights for participants with religious identifi cation and feelings towards total Muslim identifi cation, whereas for the multiple religious out-groups, and few studies

350 Verkuyten religious group identification and inter-religious relations have examined dual or hyphenated identities and non-negotiable, making the idea of religious the attitude towards religious group rights. This changes or adaptations an oxymoron. is unfortunate because religious distinctions are The results further indicate the importance among the most important and most problematic of examining dual or hyphenated identities. In ones in many present-day societies and religion social psychology, there is increasing interest is an important dimension for defi ning a positive in the fact that most individuals are simultan- identity. eously members of multiple social groups. This study focused on religious and Dutch There is, for example, research on the effects of national identifi cation among Islamic (Sunni) cross-categorization on in-group bias (e.g. Crisp, Turkish-Dutch participants. The results show Hewstone, & Rubin, 2001; Deschamps, 1977, that for most participants, Muslim identity is very and on social identity complexity (e.g. Brewer & important. No less than half of the participants Pierce, 2005; Roccas & Brewer, 2002). However, had the highest possible score on the six-item for studying dual and hyphenated identities the Muslim identifi cation measure indicating ‘total’ nature of the groups is also important. Language group identifi cation. Hence, for many partici- and culture, for example, differ from religion pants, Muslim identity does not seem to be in that one can master more than one language optional or a matter of strength of identifi cation. and internalize more than one culture, whereas Other Dutch studies among Muslims have found religious identity is exclusive (Zolberg & Long, similar results (e.g. Phalet & Güngör, 2004). 1999). Cross-cutting memberships and dual This total Muslim identifi cation is probably identities are possible as among bilinguals related to global and national developments. and biculturals, but it is something else to be a For example, the increased global tensions and ‘Christian Muslim’ or a ‘Hindu Jew’. divergences between the Western and Islamic This study examined religious and national world may also force Turkish-Dutch Muslims to identity. The results indicate that it is important a position of having to defend and stress their to examine religious identifi cation in relation to religion. In addition, in the Netherlands, the national group identifi cation (see also Hutnik, public condemnation of Islam and the plea 1991). In the Netherlands, it is often thought for assimilation has increased the salience and that Muslim identifi cation implies low identifi - importance of Muslim identity (Verkuyten & cation, or even de-identifi cation, with the Dutch. Zaremba, 2005). Islamic immigrant groups face Muslim identity is thought to be contradictory high levels of threat to the value of their religious to Dutch national identifi cation because of identity which can lead to increased in-group incompatible values, norms and beliefs. Our identifi cation among these groups. results showed a negative association between However, the total religious identifi cation Muslim and Dutch identifi cation, but they also found is probably also related to the nature of indicated that a total Muslim identifi cation does monotheistic religions in general, and Islam in not necessarily imply low Dutch identifi cation. particular. Very strong Muslim identifi cation Furthermore, around a third of the participants among West European immigrants was also found indicated high Muslim identifi cation together in the 1990s when the intergroup tensions were with Dutch national identifi cation. Hence, the much less strong (e.g. Modood et al., 1997). fi nding that a large number of participants have In addition, being a Muslim seems to imply a a very strong Muslim identity does not imply normative group commitment that is related that all of these immigrant people would not be to Islamic religion. For many Muslims, the de- interested in developing a sense of commitment claration of faith symbolizes one’s belief and to the nation. commitment to Islam: one either is a believer The importance of examining dual identities or one is not. Religion is about convictions is also evident from the results for the affective and divine truths, and for most observant be- ratings of the various religious out-groups. Social lievers, the core of their religious identity is identity theory is a theory about intergroup

351 Group Processes & Intergroup Relations 10(3) differentiation rather than out-group derogation, out-groups and in relation to political and ideo- dislike and other forms of negative affect (Brown, logical circumstances (Verkuyten & Zaremba, 2000). The majority of social psychological 2005). studies focus on situations in which the in-group Dutch national identifi cation was found to is evaluated a little more favourably than an out- be positively associated with out-group feelings group. Few studies have examined situations in towards religious groups in the Netherlands. which there are clear negative feelings towards Compared to Dutch low identifi ers, high national out-groups and in which these feelings differ identifi ers had a positive mean score towards for various out-groups (Brewer, 1999). The Christians, a neutral score towards Hindustanis, present results show that the affective ratings and less negative scores towards Jews and non- towards Christians, Hindustanis, Jews and non- believers. These results seem consistent with the believers were quite low, indicating neutral to common-group identity model for the reduc- negative or cold feelings. The mean scores for tion of in-group bias (Gaertner et al., 1989, the Christians and Hindustanis as target groups 1993). A common national identity implies were around the neutral midpoint of the scale, that religious group distinctions are subsumed and the mean scores for Jews and non-believers into a superordinate category that can reduce were clearly negative, with around a third of the negative feelings because the religious out-group participants indicating the most negative score members are fellow national in-group members. on the thermometer question (zero degrees). However, the results show a further signifi cant The results support the idea that religion can be interaction effect between Muslim and Dutch an important dimension for making meaningful identifi cation. The positive effects for Dutch and strong in-group and out-group distinctions. national identifi cation were only found for total Religion unifi es a community of believers around Muslim identifi ers. The most negative or cold a consensus of moral values and divine truths. feelings towards the religious out-groups were The observant believer will feel that he or she is found among the total Muslim identifi ers who ‘right’ leading to dogmatic thinking (Altemeyer, identifi ed to a low degree with the Dutch. In con- 2002) and making it diffi cult to have positive feel- trast, participants with total Muslim identifi ca- ings towards other religions, and to dissenters tion and high Dutch national identifi cation had and non-believers in particular. The feelings the most positive feelings (except towards the towards non-believers were very negative. In non-believers). Hence, a simultaneous awareness addition, the affective rating towards the Islamic of both a common national identity and a in-group was signifi cantly and negatively related strong emphasis on one’s Muslim identity—dual to the feelings towards non-believers only. identity—seems to be the best condition for However, not only the nature of the group is relatively favourable feelings towards religious important to understand out-group dislike and out-groups. This result is in agreement with negative affect, but also the nature of the inter- Berry’s (1984, 2005) acculturation model and group situation (Brewer, 1999; Tajfel & Turner, with the mutual intergroup differentiation 1986). The feelings towards Christians and model (Hewstone & Brown, 1986; Hornsey & Hindustanis were less negative than towards Hogg, 2000). The former model argues that the Jews. The negative feelings towards the latter integration strategy results in the least intergroup are most likely related to the growing group confl ict. The latter one argues that em- anti-Jewish sentiments in present-day Islam, phasizing a superordinate category is effective which is exacerbated by the Israeli–Palestinian in promoting positive intergroup relations, confl ict. In recent years, antisemitism has in- provided that the integrity of the original sub- creased, particularly among Muslim populations groups is maintained. This latter model might around the world, including Western Europe be particularly adequate or relevant for minority (Antisemitism Research, 2002). Hence, the groups that face high levels of identity threat and results indicate the importance of studying show strong group identifi cation, such as Islamic negative group relations in relation to different minority groups in the western world.

352 Verkuyten religious group identification and inter-religious relations

Muslim identifi cation was also related to the This does not mean that the nature and con- affective rating of Muslims and the endorsement tent of the group identity is not important. of Islamic minority group rights. Compared to Religious groups differ in terms of their identity high identifi ers, total Muslim identifi ers had defi ning beliefs, values and norms. Religion more positive feelings towards Muslims and is about convictions and divine truths, and were more in favour of group rights. The differ- for most observant believers, the core of the ences were quite substantial because Muslim religious identity is considered absolute and identifi cation accounted for not less than 28% non-negotiable. But there are also differences of the variance in the affect rating and for 48% between and within religious groups, such of the variance in the endorsement of Islamic as between Hindustanis and Christians, and group rights. These results further show that the between Sunni and Shiite Muslims. These groups dichotomous distinction between high and differ in various ways and the acceptance and total Muslim identifi cation is an important one. tolerance of ‘others’ can be seen as more or less Individuals who feel totally and normatively self-defi ning. Muslim identity and what repre- committed to their religious group did show sents the core of the religion is not a fi xed given very strong in-group favouritism and were also but is disputed and constructed in different ways much more in favour of political demands for and in the context of negotiating intergroup group rights. For the total Muslim identifi ers, for relations and organizing collective action (see example, the mean score for the endorsement Kahani-Hopkins & Hopkins, 2002). of Islamic group rights was very high (6.46 on In evaluating the present results, some re- a 7-point scale). strictions should be considered. For example, The fi ndings do not only have theoretical the study was correlational making it impossible implications but they also raise practical issues. to determine the causal direction of the effects. For one thing, there is the question of how group The arguments and analyses presented were rights can be reconciled with the emphasis on based on social identity theory, but future individualism and meritocracy that is central research should examine the causal impact of in most western liberal states (Barry, 2001). religious identifi cation. Furthermore, the order Furthermore, an emphasis on groups and group of the two group identifi cation measures was rights can lead to reifi ed group distinctions that not counterbalanced. endanger social unity and cohesion (Brewer, In addition, religious identification was 1997). In addition, there is the question of how measured with items that are commonly used in a very strong Muslim identity can go together social psychological research (see Ashmore et al., with an accepting or tolerant attitude towards 2004; Jackson & Smith, 1999). However, the way religious out-groups. The present results do not that the items were phrased might have affected allow for much optimism but should not be the results and a more ‘extreme’ phrasing could taken to show that negative religious out-group lead to a different distribution of scores.3 For feelings are intrinsic to Islam. Following social example, instead of presenting participants with a identity theory, it has been argued that group statement like ‘Being a Muslim is a very important identifi cation and intergroup relations should part of how I see myself’ one could phrase the be examined in their wider social and political statement as ‘Being a Muslim is the only important context. For example, the results also indicate part of how I see myself’. The latter statement is that for total Muslim identifi ers, Dutch national more concerned with the possible exclusionary identifi cation is related to more positive feelings nature of Muslim identifi cation and therefore towards religious out-groups and a less stronger can shed additional light on the question of total endorsement of Islamic group rights. Hence, religious identifi cation. In addition, it seems the meaning and implications of Muslim important to examine in future studies different identifi cation should be considered in relation dimensions of religious identifi cation, such as to other group memberships and the intergroup religious behaviors and practices (see Ashmore situation. et al., 2004; Verkuyten, 2005). It is possible that

353 Group Processes & Intergroup Relations 10(3) for other dimensions, Muslim identifi cation is confl icts in many societies around the world. more like a continuous variable. Furthermore, In addition, a study of religious identifi cation future studies could investigate religious can make a contribution to our thinking about identifi cation among different Islamic groups, the important processes of group identifi cation. such as Alevite and Shiite, in different countries For example, such a study can question the stand- and among Islamic minority and majority groups, ard practice of assuming that group identifi cation as well as among different religious groups. is a continuous variable or a matter of degree. An additional point in relation to group identi- Furthermore, by also examining national group fi cation concerns the interpretation of the result identifi cation it was possible to focus on the for Dutch identifi cation. This study used the role of dual identity. Future studies on both standard social psychological practice of making the origins and consequences of religious and a distinction between low and high identifi ca- dual identifi cations, and studies among various tion. In doing so, participants scoring below the religious groups, should contribute to a further neutral midpoint of the scale (‘disagree’) were understanding of identifi cation processes in considered to indicate low Dutch identifi cation. relation to the nature of the groups and the inter- However, strongly disagreeing with an item such group context. as ‘I identify with Dutch people’ could also indicate de-identifi cation. Group identifi cation Notes can be resisted or rejected. Rather than low identifi cation with one’s group, someone may not 1. The other four are, the establishment of daily want to belong to that group. He or she may prayers, concern for and almsgiving to the want to keep a distance from the expectations needy, self-purifi cation through fasting, and, the pilgrimage to Meccah for those who are able. and demands that follow from the group mem- The fi ve pillars of Islam are the foundation of bership. Thus, de-identifi cation rather than low Muslim life. identifi cation can be the issue. Low identifi cation 2. Although this study does not focus on age and with others or one’s group does not seem the gender differences, preliminary analyses were same as de-identifi cation from others or one’s conducted. For age, no systematic differences group. As weak as it can be, in the former case were found for any of the measures used. there is a sense of belonging and commitment. However, there were three signifi cant differences In the second case, there is a rejection and for gender. Compared to females, males had a distancing from the group and what it means and stronger Muslim identifi cation, had more warm stands for. The consequences of both processes thermometer-like feelings towards Muslims, and endorsed Islamic group rights more strongly. can be different, such as indifference towards the No gender differences were found for Dutch ‘other’ in the former case and an oppositional identifi cation, and for the feelings towards identity and counterculture in the latter (e.g. Christians, Hindustanis, non-believers and Jews. Ogbu, 1993). Hence, it seems important for social 3. I want to thank the acting editor for this psychology to develop criteria and questions for suggestion and example. making a distinction between low identifi cation and de-identifi cation, for example, by using References self-categorization questions. In conclusion, this study examined the rela- Aiken, L. S., & West, S. G. (1991). Multiple regression: tionship between group identification and Testing and interpreting interactions. Newbury Park, intergroup relations by focusing on Muslim CA: Sage. identity among Turkish-Dutch participants. Altemeyer, B. (2002). Dogmatic behavior among students: Testing a new measure of dogmatism. Social psychology has not paid much attention Journal of Social Psychology, 142, 713–721. to religious identity. This is unfortunate Antisemitism Research (2002). Antisemitism because religion is an important dimension for incidents in the Netherlands and provisional report developing a positive social identity and religion for 2002. Amsterdam: Centrum Informatie en is an important factor in social divisions and Documentatie Israel.

354 Verkuyten religious group identification and inter-religious relations

Argyle, M., & Beit-Hallahmi, B. (1975). The social the case of the Polish Tatars. National Identities, psychology of religion. London: Routledge & 8, 77–93. Kegan Paul. Crisp, R. J., Hewstone, M., & Rubin, M. (2001). Ashmore, R. D., Deaux, K., & McLaughlin-Volpe, T. Does multiple categorization reduce intergroup (2004). An organizing framework for collective bias? Personality and Social Psychology Bulletin, 27, identity: Articulation and signifi cance of 76–89. multidimensionality. Psychological Bulletin, Deschamps, J-C. (1977). Effect of crossing category 130, 80–114. memberships on quantitative judgement. Barry, B. (2001). Culture and equality. Cambridge, European Journal of Social Psychology, 7, 517–521. UK: Polity Press. Dijker, A. (1987). Emotional reactions to ethnic Batson, C. D., & Burris, C. T. (1994). Personal minorities. European Journal of Social Psycholology, religion: Depressant or stimulant of prejudice 17, 305–329. and discrimination. In M. P. Zanna & J. M. Olson Ellemers, N., Spears, R., & Doosje, B. (Eds.). (Eds.), The psychology of prejudice: The Ontario (1999). Social identity: Context, commitment, content. symposium (Vol. 7, pp. 149–170). Hillsdale, Oxford, UK: Blackwell. NJ: Erlbaum. Entzinger, H. (2003). The rise and fall of Berry, J. (1984). Multicultural policy in Canada: multiculturalism: The case of the Netherlands. A social psychological analysis. Canadian Journal In C. Joppke & E. Morawska (Eds.), Toward of Behavioural Science, 16, 353–370. assimilation and citizenship: Immigrants in liberal Berry, J. W. (2005). Acculturation: Living nation states (pp. 59–86). London: Palgrave. successfully in two cultures. International Journal Gaertner, S. L., Dovidio, J. F., Anastasio, P. of Intercultural Relations, 29, 697–712. A., Bachman, B. A., & Rust, M. C. (1993). Berry, J. W., & Kalin, R. (1979). Reciprocity of The common ingroup identity model: inter-ethnic attitudes in a multicultural society. Recategorization and the reduction of International Journal of Intercultural Relations, intergroup bias. European Review of Social 3, 99–112. Psychology, 4, 3–24. Blaine, B., & Crocker, J. (1995). Religiousness, race, Gaertner, S. L., Mann, J., Murrell, A., & Dovidio, and psychological well-being: Exploring social J. F. (1989). Reducing ingroup bias: The benefi ts psychological mediators. Personality and Social of recategorization. Journal of Personality and Psychology Bulletin, 21, 1031–1041. Social Psychology, 57, 239–249. Blumer, H. (1958). Race prejudice as a sense of Haddad, Y. Y., & Smith, J. I. (2001). Muslim group position. Pacifi c Sociological Review, 1, 3–7. minorities in the West: Visible and invisible. Walnut Bobo, L. D. (1999). Prejudice as group position: Creek, CA: Altamira Press. Microfoundations of a sociological approach to Hagendoorn, L. (1995). Intergroup biases in racism and race relations. Journal of Social Issues, multiple group systems: The perception of 55, 445–472. ethnic hierarchies. In W. Stroebe & Brewer, M. (1997). The social psychology of M. Hewstone (Eds.), European Review of intergroup relations: Can research inform Social Psychology (Vol. 6, pp. 199–228). practice. Journal of Social Issues 53, 197–211. London: Wiley. Brewer, M. B. (1999). The psychology of prejudice: Hewstone, M., & Brown, R. (1986). Contact is not Ingroup love or outgroup hate? Journal of Social enough: An intergroup perspective. In Issues, 55, 429–444. M. Hewstone & R. Brown (Eds.), Contact and Brewer, M. B., & Pierce, K. P. (2005). Social confl ict in intergroup encounters (pp. 1–44). identity complexity and outgroup tolerance. Oxford, UK: Blackwell. Personality and Social Psychology Bulletin, 31, Hinkle, S., & Brown, R. J. (1990). Intergroup 428–437. comparisons and social identity: Some links and Brown, R. (2000). Social identity theory: Past lacunae. In D. F. Abrams & M. A. Hogg (Eds.), achievements, current problems and future Social identity theory: Constructive and critical challenges. European Journal of Social Psychology, advances (pp. 48–70). New York: 30, 745–778. Harvester Wheatsheaf. Cameron, J. (2004). A three-factor model of social Hogg, M. A., & Abrams, D. (1988). Social identity. Self & Identity, 3, 239–262. identifi cations. Londodn: Routledge. Cieslik, A., & Verkuyten, M. (2006). National, Hornsey, M. J., & Hogg, M. A. (2000). Subgroup ethnic and religious identities: Hybridity and relations: A comparison of mutual intergroup

355 Group Processes & Intergroup Relations 10(3)

differentiation and common ingroup identity adolescents and young adults from diverse models of prejudice reduction. Personality and groups. Journal of Adolescent Research, 7, 156–176. Social Psychology Bulletin, 26, 242–256. Roccas, S., & Brewer, M. B. (2002). Social identity Hutnik, N. (1991). Ethnic minority identity: A social complexity. Personality and Social Psychology psychological perspective. Oxford, UK: Clarendon Review, 6, 88–106. Press. Rothgerber, H., & Worchel, S. (1997). The view Jackson, J. W., & Smith, E. R. (1999). from below: Intergroup relations from the Conceptualising social identity: A new perspective of the disadvantaged group. Journal framework and evidence for the impact of of Personality and Social Psychology, 73, 1191–1205. different dimensions. Personality and Social Ryder, A. G., Alden, L. E., & Paulhus, D. L. Psychology Bulletin, 25, 120–135. (2000). Is acculturation unidimensional or Joppke, C. (2004). The retreat of multiculturalism bidimensional? A head-to-head comparison in the liberal state: Theory and policy. British in the prediction of personality, self-identity, Journal of Sociology, 55, 237–257. and adjustment. Journal of Personality and Social Kahani-Hopkins, V., & Hopkins, N. (2002). Psychology, 79, 49–65. ‘Representing’ British Muslims: The strategic Sanchez, J. I. & Fernandez, D. M. (1993). dimensions to identity construction. Ethnic and Acculturative stress among Hispanics: A Racial Studies, 25, 288–309. bidimensional model of ethnic identifi cation. McGarty, C. (2001). Commentary: Social identity Journal of Applied Social Psychology, 23, 654–668. theory does not maintain that identifi cation Scheepers, P. M., Gijsberts, M., & Hello, E. produces bias, and self-categorization theory (2002). Religiosity and prejudice against ethnic does not maintain that salience is identifi cation: minorities in Europe. Review of Religious Research, Two comments on Mummendey, Klink and 43, 242–265. Brown. British Journal of Social Psychology, Schnabel, P. (2000). De multiculturele illusie: Een 40, 173–176. pleidooi voor aanpassing en assimilatie [The Modood, T., Berthoud, R., Lakey, J., Nazroo, multicultural illusion: A plea for adaptation J., Smith, P., Bvirdee, S. et al. (1997). Ethnic and assimilation]. Utrecht, The Netherlands: minorities in Britain: Diversity and disadvantage. Forum. London: Policy Studies Institute. Tajfel, H., & Turner, J. C. (1986). The social Mullen, B. N., Brown, R., & Smith, C. (1992). identity theory of intergroup behavior. In Ingroup bias as a function of salience, relevance, S. Worchel & W. Austin (Eds.), Psychology of and status: An integration. European Journal of intergroup relations (pp. 7–24). Chicago, IL: Social Psychology, 22, 103–122. Nelson-Hall. Ogbu, J. (1993). Differences in cultural frame Ter Wal, J. (2004). Moslim in Nederland: Publieke of reference. International Journal of Behavioral Discussie over de Islam in Nederland [Muslim in the Development, 16, 483–506. Netherlands: Public discussion about Islam in Owen, C., Eisner, H., & McFaul, T. (1981). A half- the Netherlands]. The Hague, The Netherlands: century of social distance research: National Sociaal en Cultureel Planbureau. replication of the Bogardus studies. Sociology and Triandis, H. C. (1992). Comments on Hinkle, Social Research, 66, 80–98. Brown and Ely. Revista de Psicología Social, Phalet, K. (2004). Moslim in Nederland: Religie en 3, 113–123. integratie [Muslim in the Netherlands. Religion Turner, J. C. (1999). Some current issues in and integration]. The Hague: Sociaal Cultureel research on social identity and self- Planbureau. categorization theories. In N. Ellemers, R. Phalet, K., & Güngör, D. (2004). Moslim in Spears, & B. Doosje (Eds.), Social identity: Context, Nederland: Religieuze dimensies, etnische relaties en commitment, content (pp. 6–34). Oxford, UK: burgerschap: Turken en Marokkanen in Rotterdam Blackwell. [Muslim in the Netherlands. Religious Turner, J. C., & Reynolds, K. (2001). The social dimensions, ethnic relations and citizenship: identity perspective in intergroup relations: Turks and Moroccans]. The Hague: Sociaal Theories, themes and controversies. In Cultureel Planbureau. R. Brown & S.L. Gaertner (Eds.), Blackwell Phinney, J. S. (1992). The multigroup ethnic handbook of social psychology: Intergroup processes identity measure: A new scale for use with (pp. 133–152). Oxford, UK: Blackwell.

356 Verkuyten religious group identification and inter-religious relations

Verkuyten, M. (2005). The social psychology of ethnic Ward, C., & Rana-Deuba, A. (1999). Acculturation identity. London: Psychology Press. and adaptation revisited. Journal of Cross-Cultural Verkuyten, M., & De Wolf, A. (2002). Ethnic Psychology, 30, 422–442. minority identity and group context: Warminska, K. (1997). Polish Tartars: Ethnic Self-descriptions, acculturation attitudes and ideology and state policy. In C. Govers & group evaluations in an intra- and intergroup H. Vermeulen (Eds.), The politics of ethnic situation. European Journal of Social Psychology, 32, consciousness (pp. 343–366). London: Macmillan. 781–800. Zolberg, A. R., & Long, L. W. (1999). Why Islam is Verkuyten, M., Hagendoorn, L., & Masson, K. like Spanish: Cultural incorporation in Europe (1996). The ethnic hierarchy among minority and the United States. Politics and Society, 27, and majority youth in the Netherlands. 5–38. Journal of Applied Social Psychology, 26, 1104–1118. Paper received 7 March 2005; revised version accepted Verkuyten, M., & Yildiz, A. A. (2006). The 7 December 2005. endorsement of minority rights: The role of group position, national context, and ideological beliefs. Political Psychology, 27, 527–548. Biographical note Verkuyten, M., & Zaremba, K. (2005). Inter-ethnic maykel verkuyten is a professor at the faculty of relations in a changing political context. Social social sciences at Utrecht University. He is also Psychology Quarterly, 68, 375–386. a senior researcher at the European Research Vertovec, S., & Rogers, A. (Eds.). (1999). Muslim Center on Migration and Ethnic Relations European youth: Reproducing ethnicity, religion, and (ERCOMER) at the same university. culture. Aldershot, UK: Ashgate.

357