THE RELATION AMONG INTRINSIC RELIGIOSITY, RELIGIOUS FUNDAMENTALISM AND ATTITUDES TOWARD OUT-GROUPS IN MUSLIMS FROM BOSNIA AND HERZEGOVINA

The relation between religion and prejudice has been shown in American Christians and Western Europeans, but it is currently unknown whether this effect can be generalized to other religions and cultures. To address this issue we conducted a study in which we assessed the personal religiosity of Muslim students from Bosnia and Herzegovina, who had to report their attitudes toward their in-group, as well as different out-groups. Consistent with prior findings, participants showed explicit preferences toward their own group relative to other religious and non-religious out-groups. As expected, we also found a relation between religiosity and out-group attitudes. Taken together, our results indicate that religiosity is negatively related to tolerance toward specific value-violating out-groups. Thus, we have extended previous findings in a different cultural and religious context.

In all major Abrahamic religions one could find explicit examples of tolerance teachings.For instance, central to the Judeo-Christian tradition is the teaching to "love your neighbor as yourself" (Leviticus 19:18) and to "do to others what you would have them do to you" (Matthew 7:12).Similarly, Mohammed is quoted as saying, "None of you really has faith unless he desires for his neighbor what he desires for himself" (Lutfiyya & Churchill, 1970, p. 58).Outside the Abrahamic tradition, Buddhism stresses compassion toward all of humanity and all of life, and within Hinduism compassion is one of the virtues needed to follow the path of righteousness (Rye et al., 2000).Similarly, across cultural contexts religiosity is related to highly valuing benevolence (i.e. concern for welfare of others) among Christians, Jews, and Muslims (Saroglou, Delpierre, & Dernelle, 2004), as well as Buddhists (Saroglou & Dupuis, 2006).Theoretically, as incorporation of religious teachings into everyday life increases, prejudice and discriminatory behavior decreases.Nevertheless, previous research do not show a clear consensus on whether religious adherents follow the implications of the Golden Rule to a greater degree than nonreligious people.In fact, early studies on religion and prejudice show quite the opposite.For instance, Allport and Kramer (1946) found that church members exhibited more racial prejudice than church nonmembers.Stouffer (1955) similarly obtained that frequent religious attendance predicted more intolerance for groups holding different ideologies (e.g.socialism).In line with these early studies, more recent research have found that measures of different dimensions of religiosity, including self-ratings of religiosity, religious fundamentalism, Christian orthodoxy, and intrinsic religiosity, correlate positively with prejudice toward out-groups perceived to violate religious worldviews (for meta-analysis, see Hall, Matz, & Wood, 2010;Whitley, 2009).
However, despite this converging evidence, it is worth noting that most of the previous studies on religion and prejudice have been conducted using predominantly Christian samples in North America or Western Europe (although there are some notable exceptions, e.g.Griffin Gorsuch, & Davis, 1987;Hunsberger, Owusu, & Duck, 1999;Karpov, 2002).This cultural uniformity in past research raises the following question; should we expect the same relation between religiosity and prejudice to generalize and surface in other cultures and other religious beliefs?In other words, is there something universal about religion in its capacity to influence out-group attitudes?Given that cultural variability (e.g.Triandis, 1989) may moderate the link between religion and social behavior, it is important that findings derived from a particular religious sample are replicated in other cultures, if generalized conclusions are to be drawn regarding the religion in question, as well as religion in general.Following this, another limitation of the previous research is associated with being generally interested in the relation between religion and prejudice toward unique specific groups that are not directly comparable across different cultures.Indeed, guided by the political reality in North American and West European countries, the studies have mostly been focused on attitudes toward black people, Jewish people, homosexuals (for a review, see Hunsberger & Jackson, 2005), and most recently, after the 9/11 attacks, on attitudes toward Muslims (e.g.Rowatt, Franklin, & Cotton, 2005).Very little research has questioned the way religiousness influence attitudes toward other religious and ethnic out-groups.Examination of these relations could illuminate the way some facets of religion relate to a wider range of social groups, and generate potential explanations for the observed group dynamics.

The present research
In order to address these issues, we conducted a study in which Muslim students from Bosnia and Herzegovina had to complete different measures of personal religiosity and attitudes toward their own group, as well as toward different out-groups.Serbs (Orthodox) and Croats (Catholic), in particular, were selected as target out-groups, due to the nature of the past conflicts between these ethnic groups during the 1992-1995 war in Bosnia and Herzegovina (BH), where the belligerents were strictly divided under these monikers i.e. they were considered combatants in a vicious war.Besides the conflict ridden past (the Balkan wars of the 90s), the post-war period with the transition from a communist regime in Yugoslavia to a democratic society and an independent state of BH, resulted in numerous changes in the social strata.The population estimations, as well as the recent preliminary census data (in progress) show that the majority of people identify with a certain religion, and that their religions are usually related to their ethnic identities, which means that they are inseparable (Velikonja, 2003).Thus, a strict division exists within Bosnia and Herzegovina across the lines of Croats (Catholics -where the majority feels more connected to the neighboring Croatia than to Bosnia and Herzegovina), Serbs (Orthodox -where the majority feels more connected to the neighboring Serbia) and Bosniaks (Muslims -the majority, who perhaps feel a deeper connection to Bosnia and Herzegovina).To explore the relation between religiousness and attitudes toward out-groups on a wider level, we have also assessed the feelings toward typical value violating groups (i.e.homosexuals and atheists), and groups with whom our sample does not have a direct contact, but that is still considered to be different (i.e.Jewish people).We have also included a neutral group (i.e. the unemployed) that is not in any way associated with a religious or between groups dynamic within the Bosnian political context.Higher religiosity should not be related to prejudice toward this group, because it is not in any way related to in-group valuations.In choosing the measures of personal religiosity we have closely followed previous work.In order to encompass a wider range of religious beliefs, and because religiousness could contain a wide variety of dimensions and characteristics, we have chosen to employ two measures related to two core facets of religion -religious fundamentalism and intrinsic religiosity.Religious fundamentalism (RF; Altemeyer & Hunsberger, 1992), as cited, is "the belief that there is a set of religious teachings that clearly contains the fundamental, basic, inerrant truth about humanity and deity, and that this truth must be followed today according to the fundamental unchangeable practices of the past."(Altemeyer, 1996, p. 157).RF is often considered the religious manifestation of the right-wing authoritarianism, reflecting obedience to authority, aggression toward outgroups, and conventionalism (Altemeyer & Hunsberger, 2005).Empirically, RF has often been linked to various types of prejudice, including racial prejudice (e.g.Altemeyer & Hunsberger, 1992;Lythe, Finkel, & Kirkpatrick, 2001;Smith, Stones, Peck, & Naidoo, 2007), prejudice against Muslims (e.g.Rowatt et al., 2005) and anti-gay prejudice (e.g.Altemeyer, 2003;Fulton, Gorsuch, & Maynard, 1999;Schwartz & Lindley, 2005).Intrinsic religiosity (IR; Allport & Ross, 1967) pertains to motivation for religious behavior rather than the actual behavior itself, and it is descriptive of people who fully embrace and internalize a belief, as well as those people who try to live their lives according to religious teachings.An evidence regarding the relation of IR to prejudice has been mixed, with reports of both positive and negative relations to prejudice (see Fulton et al., 1999, for a review).Nevertheless, IR has been associated with prejudice towards homosexuals (Fisher, Derison, Polley, Cadman, & Johnston., 1994;Herek, 1987;Wilkinson, 2004), racial prejudice (Batson, Naifeh, & Pate, 1978) and religious prejudice as well (Rowatt et al., 2005).
Finally, we have also included a measure of group identification that relates to the level of affective identification one expresses toward one's own group, independent of any external cohesive factors, such as religion or statehood.Given that religion can serve as a common denominator that differentiates groups and the nature of our sample (i.e. the inter-ethnic conflict that has facilitated the divisions between groups), we have found it necessary to include a measure that could possibly control or moderate for prejudice expression not depending on religion per se.Based on the patterns observed in the previous research, we have hypothesized that higher self-reported religiousness, measured on two scales that pertain to different facets of religious experience, in a Muslim sample should be correlated with less positive attitudes toward different out-groups.

Method Participants
A total of 251 undergraduates from the University of Sarajevo (143 women; M age = 20.0 years, SD age = 1.5) participated in the exchange for partial course credit.We excluded from the data analysis 31 participants who selfidentified as Catholics, 11 participants who self-identified as Orthodox, 13 who self-identified as atheists, and 15 participants who did not indicate their reliprimenjena psihologija, str.379-394 gious affiliation.The remaining 181 participants were all Muslims, all believed in God and were Bosniaks (151 women; M age = 20.6 years, SD age = 3.8).

Instruments and procedure
Upon arrival, each participant was individually placed in a lab room with a desk and a personal computer.Participants first completed a series of thermometer items that assessed feelings toward different out-groups (e.g.Croats, Serbs, Jewish people, atheists, homosexuals and the unemployed), as well as toward their own group (e.g."Your feelings toward Bosniaks are"; 0 = very cold, 10 = very warm).In order to assess the attitudes toward different out-groups, we calculated the scores of affective distance (for a similar method, see Cottrell & Neuberg, 2005;Johnson, Rowatt, & LaBouff, 2012) by subtracting the thermometer item rating of the in-group (i.e.Bosniaks) from the thermometer item ratings of out-group (e.g.Serbs).Lower scores indicated higher affective distance with the out-group (i.e.colder feelings toward the out-groups relative to the in-group).Once participants finished with the thermometer items, they completed the rest of the instruments.
Intrinsic religiosity scale (Hoge, 1972).Scale consisted of ten items (e.g."My faith involves all of my life"; 1 = does not relate to me; 5 = completely relates to me).Cronbach alpha values for this scale was .83.
Religious Fundamentalism Scale (Altemeyer & Hunsberger, 1992).Scale consisted of twelve items (e.g."God has given humanity a complete, unfailing guide to happiness and salvation, which must be totally followed''; 1 = very strongly disagree, 9 = very strongly agree).Cronbach alpha values for this scale was .74.
In-group identification scale (Meeus, Duriez, Vanbeselaere, Phalet, & Kuppens, 2009).This is a six-item scale, which measured the affective identification with the ethnic in-group (e.g."I am proud to be a member of my group.","I feel connected with other people from my group."; 1 = not at all; 5 = completely).Cronbach alpha values for this scale was .75.

Results
The means, standard deviations and correlations among all measures are presented in Table 1.In accordance with the previous research, participants consistently reported more positive feelings toward their ethnic group (i.e.Bosniaks) than toward the out-groups.As we could see in Table 1, this intergroup bias was stronger for atheists and homosexuals than it was either for Croats, Serbs Jews, or the unemployed.To test our main hypothesis, correlations were computed among different dimensions of religiosity and attitudes toward each out-group (Table 1).As hypothesized, different measures of religiosity were generally related to the intensity of intergroup bias.Precisely, the measure of IR was significantly related to affective distance for all groups except for the unemployed.The only non-significant correlations for the GI and RF measures were towards Croats.As expected, none of the measures correlated with the unemployed group, as this group was not related to any type of in-group markers.
Furthermore, as we could see in Table 1, the different dimensions of religiosity and group identification measure were strongly inter-correlated, and thus confounded with one another.Therefore, a multiple regression analysis was conducted to simultaneously control for the contribution of IR, RF, and GI on the relative measures of affective distance (Table 2).When entered into a regression analysis, only IR emerged as negatively correlated with affective distance towards all groups, except for Jewish people (even though the trend was suggestive of significance with a p value of .07)and the unemployed.The RF and GI measures turned out to be significant only for the value violating groups (homosexuals and atheists).This is not surprising, as research has shown that once again, these two groups are regarded with the highest amount of prejudice and correlate negatively with multiple measures of religiosity.

Discussion
The relation between religion and prejudice has been shown in West Europeans and American Christians, but it is currently unknown whether this effect could be generalized to other religions and cultures.To address this issue, we assessed the personal religiosity in Muslim students from Bosnia and Herzegovina who reported their feelings toward different out-groups.In line with previous research, the results revealed the existence of classic intergroup bias with participants showing explicit preferences toward their own group relative to other religious and non-religious out-groups (Brewer, 1999;Mummendey, Klink, & Brown, 2001;Rowatt et al., 2005).This affective distance between Bosniaks and out-groups was stronger for atheists and homosexuals than it was either for Croats and Serbs, which might be surprising given that these ethnic groups were associated with the previous war atrocities, committed in the 90s Balkans conflict.Therefore, our results have confirmed once again that atheists (Edgell, Gerteis, & Hartmann, 2006) and homosexuals (Falomir-Pichastor & Mugny, 2009;Rowatt, LaBouff, Johnson, Froese, & Tsang, 2009) are the least likely groups to be socially accepted among a variety of other religious and ethnic minority groups.Regarding our main hypothesis, we have found evidence supporting the existence of a relationship between different facets of religiousness and intergroup bias.Consistent with previous findings obtained in Christian populations, religiosity in a Muslim sample correlated negatively with attitudes toward atheists (Duckitt & Sibley, 2007;Jackson & Hunsberger, 1999), homosexuals (Whitley, 2009), and Jewish people (Duriez & Hutsebaut, 2000;Rudman, Greenwald, Mellott, & Schwartz, 1999).This relationship was not restricted to these specific groups.Indeed, similar results were observed for the ethnic out-groups from Bosnia and Herzegovina, where higher levels of religious fundamentalism and intrinsic religiosity correlated with higher levels of affective distance toward Serbs and Croats.Furthermore, when we simultaneously entered measures of religiosity and the group identification measure into a regression analysis, IR emerged as a significant predictor for all the groups (except for Jewish people and the unemployed), while RF was a significant predictor only for the value violating groups.Taken together, our results indicate that religiousness is positively related to relative derogation of different out-groups.On a theoretical level, this represents an important contribution for the psychology of religion since the previous evidence was mostly restricted to the North American and West European Christian populations.Thus, we extended previous findings in a different cultural and religious context, as we found similar patterns with the Muslim sample as well.
However, it is important to note that some of our results are at odds with past research.Although we replicated the findings of IR and RF literature on prejudice toward atheists and homosexuals (e.g.Altemeyer, 2003;Wilkinson, 2004), our results were more ambivalent for the ethnic out-groups (Serbs and Croats in relation to Bosniaks).Precisely, upon analysis, we found that only the measure of intrinsic religiosity predicted affective distance towards these groups, subsuming the measure of group identification and religious fundamentalism.Given that previous research conducted in North American and West European samples showed a negative relation between IR and prejudice (e.g.Batson, Eidelman, Higley, & Russell, 2001;Donahue, 1985) one might wonder why we found the results which went in opposite direction?To answer this question, we should take into consideration differences in normative rules which regulated active proscription or encouragement of prejudice by religious institutions.For example, Herek (1987) found that an intrinsic religious orientation was positively linked with prejudice against gay men and lesbians.However, consistent with some previous research, it was negatively linked with racism.As explained by the author, this different relationship between I and racism, compared to negative attitudes toward gay men and lesbians, might be attributable to differences in church teachings.Consistent with such reasoning, Duck and Hunsberger (1999) have found evidence confirming the tendency for people to report that racial prejudice is proscribed by their religious group, and that gay/lesbian intolerance is non-proscribed, on average.However, a different side of the coin is present as well.Although it seems that the mainstream contemporary religions generally preach against racial prejudice (Duck & Hunsberger, 1999;Rowatt et al., 2009), the research also suggests that these tendencies may be culturally and geographically specific.For example, in South Africa (which is known for the past apartheid politics), the racial prejudice has been religiously non-proscribed (Lafferty, 1990).If we look closer at Bosnia and Herzegovina, the 1992-1995 war created animosity in between-group dynamics, which is still present in every segment of social life.The state atheism promoted by the past communist regime was replaced by a nationalist ideology in which religion was present in every sphere of social life (West, 2012).In such a political context, religious institutions entertain close partnerships with political parties which frequently use religion as one of the main factors serving as a separator of ethnic groups.For that purpose, religious teachings have often been adapted to highlight the difference among groups, and in some situations, they have even been adapted to a non-proscriptive stance (Velikonja, 2003).Given that intrinsic religiousness is related to greater involvement with religious institutions and the importance of religion in life (Wulff, 1997), we can speculate that individuals high in intrinsic religiousness have internalized any teachings by their religious representatives that may derogate other ethnic groups (see also Duck & Hunsberger, 1999).Therefore, the higher affective distance with Serbs and Croats, among those high on the intrinsic orientation, may reflect conformity to perceived mosque positions regarding out-groups, since such conformity reinforces one's image as a good group member.
Curiously enough, Jewish people, a group that is not represented so much in the social strata of Bosnia and Herzegovina, correlates with higher affective distance on all three measures.Even when controlling for the measures in a regression analysis, we could see that there is a marginal significance on the measure of intrinsic religiosity.This seems to suggest that no actual physical contact or social closeness is needed for out-group derogation to establish itself.Nevertheless, the neutral group (unemployed), while carrying a certain negative connotation, do not correlate in any way with the used measures.This gives a dose of specificity to religious influence on out-group attitudes, as this group is not in any way related to certain markers that could be used as prejudice signals.

Limitations and directions for future research
In our research we have tested the relation between religion and outgroup attitudes (as measured by affective distance) in a different cultural, geopolitical, and religious context.Given this intertwining, it is difficult to disentangle which of these factors really explains our findings.Is there something related to the Bosnian society in general, to the Muslim religion, or to the interaction of these factors that leads itself to the explanation of our findings?This issue could be addressed by testing the religion-out-group attitudes link within other religious samples from Bosnia and Herzegovina, such as Catholics and Orthodox.In addition, future studies should also consider potential differences in between-group dynamics, which may vary as a function of past conflicts in specific areas of the country.From this perspective, the inclusion of samples from other parts of Bosnia and Herzegovina, as well as the neighboring countries (i.e.Croatia and Serbia), could provide better insight into this problem.Similarly, as with most other studies of this nature, it is relevant to note that there is a certain cross-categorization.For instance, some individuals could be Bosniaks and homosexuals at the same time.While this delineating problem was interesting, it was out of scope for this study, as we were interested mainly in establishing a relation between different dimensions of religiosity and general out-group attitudes.Future research should take this into account.
As we have already suggested, the expression of stronger affective distance by religious people (and particularly those high in IR orientation) may depend on whether or not different groups are normatively protected against negative judgments (i.e.whether it is unacceptable or acceptable to express negative judgments about specific groups).It means that the influence of religiosity and normative rules could be confounded on the empirical level.In order to control this possibility, the future research should assess the proscription/nonprescription norms relative to attitudes toward different out-groups.For instance, this could be realized by asking participants to what extent a significant person, such as a religious leader or a majority of members of their religious group, would approve or disapprove of prejudice toward different groups.Similarly, the future studies should include self-report measures in tandem with more discrete measures of behavior towards in-groups and outgroups.Not only would this allow assessing attitudes more objectively, but it would also give us the possibility to test whether the relation between religion and discriminative behavior is mediated by prejudice.
Finally, our study provides correlation evidence, which limits conclusions regarding the causal role of religion on the out-group attitudes.To provide a more stringent test of this hypothesis, the future research should use an experimental approach and try to manipulate religious cognitions.For instance, this could be realized by using different paradigms based on priming methodology (e.g.Chartrand & Bargh, 1996;Johnson, Rowatt, & LaBouff, 2010), where the researchers could test the effects of activation of religious concepts on prejudice.

Conclusion
A recent Pew survey has found that, in most of forty countries tested, a majority of respondents agree that believing in God is essential to morality (Pew Research Center, 2014).Rates were highest in Central Asia and West Africa and in the US, where 53% of the people agree that the belief was necessary to be a good person.Clearly people believe that the religious belief is somewhat necessary for a moral disposition, and it is visible across the cultures.Indeed, a multitude of the research exists relating some form of religious belief to prosociality.However, recently there has been a surge in findings that perpetrate a different, more negative side of religion.We have shown that, in our sample, religiosity is associated with intergroup bias thus offering cross cultural support for the findings that have shown the same on a sample from developed, western Christian countries.Despite the fact that religious dogma is filled with messages that propagate altruistic behavior, religion seems to function in multiple ways.Indeed, it can be used as a tool for demarcation that could have disastrous consequences.

Table 2
Multiple regression analysis results