Love Thy Neighbor: Exploring Religious and Social Openness among Prospective Theologians in Germany and Turkey

Abstract

Amidst increasing globalization and religious diversity, acknowledging and embracing openness towards religious and/or cultural others has become crucial for societal cohesion and international relations. Theological scholars, holding significant potential in mitigating inter-religious and intercultural prejudices, can play a pivotal role in addressing this challenge. However, it is acknowledged that theologians themselves may harbor such biases. This study, conducted within the framework of the Religious Openness Hypothesis, employed an online questionnaire among theology students, seen as future multipliers of religiosity, in Germany and Turkey (N = 513) using convenience sampling. The results reveal the consistent relation of religiosity to all forms of prejudice among German Christians, with a linked defense against secularism potentially leading to self-isolation and the protection of their own worldview against religious or cultural outgroups. In contrast, the (generally high) prejudice among Turkish Muslims appears to be rooted not primarily in religiosity or defense against secularism but in fundamentalism and, most likely, in other socio-cultural factors such as politics and education. For both subsamples, religiosity was positively linked with xenosophia, particularly when accounting for fundamentalism. The article concludes by proposing curriculum implications for universities and schools in both cultural contexts.

1. Introduction

The commandment ‘Love thy neighbor’ is a well-known principle found in all Abrahamic religions, including Judaism (e.g., Leviticus 19:18), Christianity (e.g., Mark 12:31), and Islam (e.g., Quran 5:32). Despite the prominence of this teaching, one cannot help but question why it appears to be inadequately practiced in light of historical and contemporary conflicts on societal, economic, territorial, or religious fronts involving these religions. This paradox has spurred extensive research efforts, dating back to the 1940s (Allport and Kramer 1946). Researchers in this field have undertaken various approaches to address this paradox by exploring different aspects of religiosity (Allport and Ross 1967), examining various cultural contexts, and considering diverse samples (for an overview see Utsch and Demmrich 2023).

In this paper, our focus is directed towards comparing theology students of two different religious traditions. Studying prospective theologians is of utmost importance since it offers insights into the perceptions and biases towards outgroups of soon-to-be multipliers of religion in society. Our study centers on Muslims in Turkey and Christians in Germany, exploring their potential prejudice not only against each other but also against shared religious and/or cultural outgroups, such as Jews and refugees. Regarding this, a central study aim is to approach the religiosity–prejudice paradox within the context of the Religious Openness Hypothesis. This hypothesis posits that religious traditions, even in their conservative forms, harbor the potential for openness (e.g., Watson et al. 2015a), including intellectual but also social openness, i.e., a willingness to embrace others and strangers (Watson et al. 2019).

Based on our findings, our objective is to derive practical implications for the theology curriculum and religious education at schools. This is of specific interest since prejudice against each other (Muslims against the West, Christians against Muslims), against Jews, and against refugees is not just widespread in both societies and hotly debated in the wider public (Decker et al. 2022; Evangelical Church in Germany 2022; Öztürk and Pickel 2022; Ünal 2014) but also contains explosive potential for social cohesion in both societies. Although previous studies found that theologians and theology students might be highly prejudiced (Kayıklık 2001; Yapıcı 2004b; cf. Erdiç 2022)—a bias that can be attributed, at least partially, to their higher religiosity, conservatism, and fundamentalism—we nevertheless regard a huge potential for (prospective) theologians as they have a significant capacity as multipliers of religion and religiosity in their respective societies (e.g., as clerics, teachers of religion, spiritual counselors).

2. Classical Research on Religiosity and Prejudice

Prejudice is defined as the negative evaluation of another social group or individual (Crandall and Eshleman 2003) or the derogation of those who are socially different (Loewenthal 2004). It can manifest directed towards individuals based on various factors, such as social class, gender, sexual orientation, political beliefs, religious affiliations, disabilities, or ethnic backgrounds. The exploration of the relationship between religiosity and prejudice can be traced back to the foundational work of Gordon Allport (Allport [1954] 1979). In his seminal studies, Allport noted, “religion makes and unmakes prejudice” (Allport [1954] 1979, p. 444). This widely cited quote highlights the paradoxical interplay between religiosity and prejudice, a conundrum often considered one of the most intricate aspects of the psychology of religion (Loewenthal 2004).

A nuanced exploration of various forms of religiosity can offer valuable insights into unraveling this paradox (e.g., Streib and Klein 2014; Yapıcı 2004b). Within Christian samples, extrinsic religiosity tends to show a positive association with prejudice, while intrinsic religiosity displays a less pronounced or negative relationship (Allport and Ross 1967; Hunsberger and Jackson 2005). Similarly, the more central religiosity is within an individual’s personality (resembling intrinsic religiosity, see Huber and Huber 2012), the less likely that person is to harbor prejudice against strangers, foreigners, and Muslims (Yendell and Huber 2020). In contrast, fundamentalism—a form of religiosity that encompasses exclusivity, superiority, universality, and restoration of a ‘Golden Age’ (Pollack et al. 2023)—emerges with the strongest correlations with outgroup biases (for a review, see Williamson and Demmrich 2024).

Interestingly, investigations into the connection between religiosity and prejudice among Muslims in Turkey, though still in its early stages (Ulu 2019), have produced results that challenge our prior understanding. Contrary to expectations, intrinsic religiosity has frequently demonstrated a positive correlation with various forms of prejudice, whereas extrinsic religiosity has shown a negative relationship with outgroup biases (Cirhinlioğlu 2010; Gürses 2001; Kayıklık 2001; Yapıcı and Kayıklık 2005).

Another crucial aspect that could contribute to the resolution of this paradox is the specific target group of the prejudice (Argyle 2000; Küpper 2010). Within Christian samples, prejudices against ethnic minorities or disabled individuals tend to show a negative correlation, at least with personal religiosity versus institutionalized religiosity (Maftei et al. 2023; Scheepers et al. 2002). However, positive correlations emerge when examining prejudices against groups such as homosexuals, feminists, atheists, and individuals of different religious affiliations, including immigrants/refugees (Clobert et al. 2017; Deslandes and Anderson 2019 for Christian and Muslims; Hoffarth et al. 2018). This differentiation helps clarify why certain forms of prejudice against groups that challenge or diverge from one’s religious moral beliefs are often considered acceptable, even by those who are intrinsically religious (Batson et al. 1993; Hunsberger and Jackson 2005; Whitley 2009; Yapıcı 2004a, 2004b).

Concerning Muslims, similar results have been shown. For example, a positive association between religiosity and inter-religious prejudice has been identified among Muslims (Hasnain 2007; Tausch et al. 2009; for general prejudice: Karaca 2023). This tendency is particularly pronounced when they belong to a religious majority and assess a religious minority (for Turkey: Yapıcı 2004a; globally: Guiso et al. 2003). Conversely, religiosity shows a negative correlation with anti-refugee prejudice (Çiftçi 2022). This pattern is reversed when considering extrinsic religiosity (Tepe et al. 2019).

Given these puzzling outcomes related to specific forms of religiosity and target groups of prejudice, we aim to approach the religiosity–prejudice paradox from a novel perspective. Specifically, we intend to test the Religious Openness Hypothesis, extending its application to social openness, using xenosophia as a positive indicator and prejudices as negative indicators.

3. A New Perspective: Integrating Prejudice and Xenosophia into the Religious Openness Hypothesis (ROH) as Indicators of Social Openness

Xenophobia, characterized as “the anxiety of the other and strange” (Streib 2022, p. 1), finds its opposite in xenosophia, defined as an appreciation of the wisdom available in foreign religious traditions and worldviews. This appreciation extends to being open to re-examining one’s current worldview (Streib and Klein 2018). Xenosophia served in previous studies as a valid negative indicator for both explicit and implicit prejudices, particularly those of an inter-religious nature, such as anti-Muslim prejudice (Streib and Klein 2014, 2018). Xenosophia is usually negatively correlated to fundamentalism (Watson et al. 2019; Streib 2010) but general religiosity seems to exert a preventive effect against xenophobic attitudes among German adolescents (Streib 2010).

In the present study, we aim to incorporate xenosophia into the Religious Openness Hypothesis (ROH) as an indicator of social openness. The ROH posits that traditional religions harbor potential for expressing intellectual openness (i.e., importance of open intellectual engagement), psychological openness (i.e., receptivity, curiosity, and non-defensiveness towards one’s inner and outer world, such as integrative self-knowledge or openness to new experience), and social openness (i.e., a willingness to embrace others and strangers) (e.g., Ghorbani et al. 2013; Watson et al. 2014, 2015b, 2015c). Essentially, the ROH contends that religions, even in conservative forms, possess resources and potentials to integrate faith with various aspects of openness. Empirical support for the ROH has been demonstrated concerning intellectual openness in diverse cultural–religious contexts, including US-American Christians (Watson et al. 2011, 2014), Iranian Muslims and Christians (Ghorbani et al. 2014, 2018; Watson et al. 2015a), Malay Muslims (Tekke et al. 2015), Pakistani Muslims (Khan et al. 2018), Muslims with a Turkish migration background in Germany (Demmrich and Akçe 2022), and Indian Hindus (Kamble et al. 2014). These studies have yielded intriguing results. In the case of US-American Christians and Muslims of Turkish origin in Germany, negative or no bivariate relations were observed between religiosity and intellectual openness. Nevertheless, when this relationship was controlled for confounding variables such as fundamentalism, positive relations emerged. However, samples from Muslim, Hindu, and Christian populations outside the West exhibited positive bivariate relations between religiosity and intellectual openness without the need to control for other variables. Additionally, these samples demonstrated positive correlations between religiosity and various indicators of psychological openness (for summaries, see Demmrich and Akçe 2022; Watson 2019).

To date, only one study has explored the Religious Openness Hypothesis (ROH) in the context of social openness, specifically examining xenophilia (the potential for a love of the stranger) among a sample of N = 279 Christian undergraduate students from the USA (Watson et al. 2019). Positive correlations between different forms of religiosity and xenophilia revealed conservative religious potentials for social openness. An even stronger xenophilic surround emerged when fundamentalism was controlled for. Whether a similar relationship can be identified in diverse cultural contexts remains an open empirical question.

According to Watson and colleagues (Watson et al. 2014; Watson 2019), contrasting findings can be clarified by considering the sociological context. They propose that it is not a specific religious tradition itself that fosters openness. Instead, they argue that “secular standards drive a wedge” between religiosity and various facets of openness, encompassing intellectual, psychological, and social openness:

Beliefs that conservative religious commitments are under attack by secularism may lead to anxieties associated with culture war. Defense mechanisms reduce anxiety, and the defense mechanism of splitting reduces anxieties by minimizing ambiguities in perceptions of self and others as all good or all bad.

(Watson et al. 2019, p. 253)

This concept, labeled as the ‘defense against secularism’, has been established as a mediator in the connection between religiosity and (intellectual) openness, particularly among Christian students in the US (Watson et al. 2015b). Its incorporation into cross-cultural studies on the Religious Openness Hypothesis (ROH) has been recognized as a noteworthy research desideratum (Demmrich and Akçe 2022; Kamble et al. 2014; Watson et al. 2019).

Drawing on previous research findings, it becomes intriguing to investigate whether a similar polarization between religiosity and openness, observed among Christians in the US, also exists among Christians in Germany. This inquiry can provide insights into whether such polarization is rooted in a defensive response to secularism. Parallel to previous findings (Watson 2019), such polarization between religiosity and intellectual openness could also lead to contrasting relationships with social openness. Social openness may positively correlate with intellectual openness but negatively with religiosity, the latter aligning with classical prejudice research among Christians.

Even more captivating is the exploration of the dynamics in the relationship between religiosity and openness in Turkey. Despite undergoing significant secularization efforts, Turkey notably stands as a Western society where a substantial majority of the population remains unwavering in its commitment to faith and religion (Köse 2021; Yapıcı and Ağılkaya-Şahin 2021). The Turkish population has demonstrated resilience against secularism across various facets and has endured “persistent tensions between traditionalism and modernism, religiosity and secularism, until recent times” (Yapıcı and Ağılkaya-Şahin 2021, p. 203). Conversely, some scholars, such as Köse (2021), contend that in contemporary Turkey, a coexistence or even a robust interaction can be observed between religion and secularity or modernity. Notably, neither the Religious Openness Hypothesis nor the application of the Religious Reflection Scale, used to test the ROH (Dover et al. 2007), has been explored among Muslims in Turkey (or Christians in Germany) thus far.

4. The Present Study

In line with previous research on the religious openness hypothesis (ROH), we predict that there is a positive relationship between religiosity and intellectual openness among Muslims in Turkey (short: Turkish Muslims) but that such a positive relationship among Christians in Germany (short: German Christians) appears only when defense against secularism and fundamentalism are controlled for (Hypothesis 1).

Furthermore, we expect positive bivariate relationships between social openness on the one hand and religiosity and intellectual openness on the other hand for Turkish Muslims. In contrast, we expect negative bivariate relationships between social openness and religiosity and positive bivariate relationships between social and intellectual openness for German Christians (Hypothesis 2).

Finally, and similar to previous studies on intellectual openness, we assume that the relationship between religiosity and social openness is mediated by defense against secularism in both subsamples while simultaneously controlling for fundamentalism and sociodemographic variables (Hypothesis 3).

5. Method

5.1. Sample and Procedure

All procedures undertaken in this study involving human participants adhered to the ethical standards outlined in the 1975 Helsinki Declaration and the ethics guidelines established by the institutions of all authors. The survey was administered through an online questionnaire, conducted in German for the German subsample and in Turkish for the Turkish subsample, utilizing the SoSci Survey application between February and August 2023.

Convenient sampling was employed in both countries, involving outreach to theology departments through university teachers, study coordinators, or student councils. These contacts were requested to disseminate our invitation letter, along with the survey link, introducing the study as an examination of the religious life of theology students in Germany and Turkey. Participation in the study was voluntary. To encourage participation, Turkish participants were offered bonus points for exams, while German participants were entered into a EUR 20 book voucher raffle (due to the absence of a point system in Germany). Personally identifiable information required for the incentives was collected separately from the online questionnaire using SoSci Survey, in compliance with German data protection laws.

The questionnaire implemented the reCAPTCHA plugin to prevent multiple unauthorized entries by bots. Rigorous screening questions ensured the inclusion of participants studying theology of religious education, residing in Germany or Turkey, and self-identifying as Christian or Muslim, respectively.

A prior power analysis using G*POWER revealed that n = 105 participants per group would be sufficient to detect a medium effect size (d = 0.50) of mean differences (two-tailed) with a power of 0.95. To detect correlations of the same effect size (r = 0.30, one-tailed) with the same power, N = 115 participants would be necessary. Participants who completed less than half of the questionnaire or finished in less than half of the median time (median time = 701.50 s) were excluded. The final sample consisted of a total of N = 513 participants (M_age = 24.08; SD_Age = 6.20, range: 19–64 years), including n = 345 Turkish Muslims (M_age = 23.35; SD_Age = 4.75, range: 19–50 years) and n = 168 German Christians (M_age = 25.71, SD_Age = 8.07, range: 19–64 years, significantly older than the Turkish Muslim subsample, see

Table 1. Descriptive statistics and ANOVAs between both groups of all metric and ordinal variables included.

	Turkish Muslims (n = 345)		German Christians (n = 168)		F
	M	SD	M	SD
Faith-oriented reflection (1–5)	4.55	0.44	3.47	0.72	445.72 ***
Intellect-oriented reflection (1–5)	4.48	0.51	3.96	0.64	99.75 ***
Fundamentalism (1–4)	3.38	0.54	2.00	0.79	541.13 ***
Defense against secularism (1–5)	2.67	0.70	1.53	0.67	282.00 ***
Xenosophia (1–5)	3.74	0.76	3.74	0.72	0.00
Anti-Muslim/anti-Western attitudes (1–6)	4.27	0.69	1.71	0.81	1328.52 ***
Anti-Semitism (1–6)	4.59	0.92	1.65	0.60	1373.83 ***
Anti-refugee (1–6)	3.87	1.13	1.92	0.91	366.92 ***
Age (in years)	23.35	4.75	25.71	8.07	17.19 ***

Notes. *** p ≤ 0.001, rest ns. numbers in brackets behind variable names indicate their scale range.

).

Of the participants, 66.5% were female and 33.3% were male (rest: non-binary), which is also reflected in the subsamples (Turkish Muslims: 67.0% female, rest: male; German Christians: 65.5% female, 33.9% male, rest: non-binary). Among Turkish Muslims, the vast majority (85.5%) indicated a Sunni affiliation, with the rest consisting of Shia (0.3%), and 10.7% having other denominational affiliations, mostly indicating subgroups of Sunni affiliation. No Alevi participated. Similarly, among German Christians, the majority (83.9%) indicated a Catholic affiliation, while 16.1% indicated a Protestant affiliation. No other denominational affiliation was reported. The data file can be obtained from the corresponding author upon request.

5.2. Measurements

In addition to sociodemographic variables, the study assessed religiosity, intellectual openness, and social openness (with xenosophia as a positive indicator and prejudices as negative indicators), as well as defense against secularism and fundamentalism. All used scales, including their items, are displayed in Table S1 in the supplementary material.

Sociodemographic variables included age (in years), gender (male, female, or non-binary), and denominational affiliation (for Turkish Muslims: Sunni, Shia, Alevi, other; for German Christians: Protestant, Catholic, other).

Religiosity: Following the tradition of the ROH, we utilized the same measure of religiosity as employed in most previous studies on the same hypothesis—the faith-oriented reflection subscale of the Religious Reflection Scale (Watson et al. 2011). This seven-item subscale measures individuals’ reflection on their own religious norms, thoughts, and practices (e.g., “I believe that through science and religion one can really understand the meaning of life”), with responses recorded on a five-point scale ranging from 1 = strongly disagree to 5 = strongly agree. The Muslim version was administered to Turkish Muslims, while the Christian version was given to German Christians. The scale was internally translated into Turkish and German through independent back and forth translations until satisfactory concordance was achieved. Internal consistencies were acceptable to good (α = 0.69 for Turkish Muslims, α = 0.78 for German Christians).

Intellectual Openness: Consistent with prior ROH studies, intellectual openness was assessed using the intellect-oriented reflection subscale of the Religious Reflection Scale (Watson et al. 2011). This five-item subscale explores the integration of compatible insights and the rejection of incompatible insights from other social realities (e.g., “I believe as humans we should use our minds to explore all fields of thought from science to metaphysics”). The answer format and translation procedure mirrored those of the faith-oriented reflection subscale. Internal consistencies were found to be acceptable (α = 0.65 for Turkish Muslims, α = 0.64 for German Christians).

Xenosophia: This six-item scale was developed by Streib et al. 2010; e.g., “We can learn from each other what ultimate truth each religion contains”), with responses recorded on a five-point scale from 1 = does not apply to 5 = applies. The original German scale was translated into Turkish by independent back and forth translation. Internal consistencies were acceptable and good (α = 0.63 for Turkish Muslims, α = 0.73 for German Christians).

Prejudices: A central aim of this study is to investigate the prejudice of Turkish Muslims and German Christians towards each other. To achieve this, anti-Muslim prejudice was assessed among the German Christian subsample by a four-item scale (Evangelical Church in Germany 2022; e.g., “Muslims should be banned from immigrating to Germany”) with a six-point answer format from 1 = strongly disagree to 6 = strongly agree. The scale was originally in German, and internal consistency was good with α = 0.69. As no established concept exists in the social sciences with regard to anti-Christian attitudes, an anti-Western prejudice scale, designed for Muslims (Albaghli and Carlucci 2021), was used as a pendant for the Turkish Muslim subsample. It comprises 11 items, answered on a six-point scale from 1 = strongly disagree to 6 = strongly agree. The original English scale for Arabs was translated by independent back and forth translation into Turkish and culturally adapted by replacing ‘Arab/ic’ with ‘Islam/ic’ or ‘Muslims’ (e.g., “Western governments are responsible for all the suffering in the Islamic [previously: Arabic] world”). Internal consistency was good with α = 0.71.

Another aim of the study is to examine additional prejudice towards shared religious and/or cultural outgroups. To achieve this, anti-Semitism and anti-refugee prejudice were also assessed. Anti-Semitism was measured using an eight-item scale (Evangelical Church in Germany 2022), covering primary anti-Semitism (three items, e.g., “The influence of the Jews is still too great today”), secondary anti-Semitism (two items, reflecting guilt defense in the German context, e.g., “Claims for reparations against Germany often no longer benefit the victims, but a Holocaust industry of resourceful lawyers”; and Holocaust denial/perpetrator–victim turn in the Turkish context, see Heitmeyer (2005), e.g., “The Jews are exploiting the memory of the Holocaust for their own benefit today”), and Israel-related anti-Semitism (three items, e.g., “Other nations may also have their dark sides, but Israel’s crimes are the most serious”). Participants responded on a six-point scale from 1 = strongly disagree to 6 = strongly agree. The original German scale was translated into Turkish using independent back and forth translation. Internal consistency was good to excellent (α = 0.90 for Turkish Muslims, α = 0.77 for German Christians). Anti-refugee prejudice was measured with two items (Evangelical Church in Germany 2022, e.g., “Refugees who live here threaten my personal way of life”). Answer format and translation process mirrored those of the anti-Semitism scale. Despite the scale’s brevity with two items, reliability was acceptable (α = 0.56 for Turkish Muslims, α = 0.66 for German Christians).

Defense against secularism: This 17-item scale (Watson et al. 2015c, e.g., “The theory of evolution is an example of how science and reason are dedicated to eliminating faith”) utilized a five-point answer format from 1 = strongly disagree to 5 = strongly agree. The original English scale for Christians was translated into German by independent forth and backtranslation. Moreover, it was translated into Turkish and adapted to Muslim terms using the same procedure. Internal consistency was excellent (α = 0.89 for Turkish Muslims, α = 0.95 for German Christians).

Religious fundamentalism: The fundamentalism scale measures a form of religiosity that encompasses exclusivity, superiority, universality, and a restoration of a ‘Golden Age’ (Pollack et al. 2023) and was developed by Pollack et al. (forthcoming). The Muslim version in Turkish was used for Turkish Muslims and the Christian version in German for German Christians. Participants responded to four items on a four-point-answer scale from 1 = strongly disagree to 4 = strongly agree (e.g., “There is only one true religion”). Internal consistency was acceptable to good (α = 0.68 for Turkish Muslims, α = 0.84 for German Christians).

6. Results

6.1. Descriptive Statistics, ANOVAs, and Zero-Order Correlations

Table 1 displays the means and standard deviations of all metric and ordinal variables separately for both subsamples, along with significance tests between these groups (ANOVAs). Despite the previously mentioned significant difference in age between the German Christian and the Turkish Muslim subsamples, the Turkish Muslim subsample scored significantly higher on both faith-oriented reflection and intellect-oriented reflection, as well as higher on fundamentalism and defense against secularism. Furthermore, the average score of Turkish Muslims on the anti-Western scale was notably higher than the average score of German Christians on the anti-Muslim scale. Additionally, within the Turkish Muslim subsample, a significantly higher mean score in both anti-Semitic and anti-refugee prejudice was observed compared to the German Christian subsample. Interestingly, both subsamples did not differ significantly in xenosophia, and both scored relatively high on average.

With regard to Hypotheses 1 and 2,

Table 2. Zero-order correlations for all included variables for Turkish Muslims (above the diagonal) and for German Christians (below the diagonal).

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
Faith-oriented reflection (1)		0.55 ***	0.53 ***	0.13 *	0.25 ***	0.21 ***	0.31 ***	−0.08	0.02	0.13
Intellect-oriented reflection (2)	0.24 **		0.26 ***	−0.02	0.37 ***	0.08	0.22 ***	0.09	−0.04	0.05
Fundamentalism (3)	0.61 ***	0.01		0.33 ***	0.01	0.37 ***	0.28 ***	−0.29 ***	0.04	−0.06
Defense against secularism (4)	0.43 ***	−0.12	0.50 ***		0.03	0.15 **	0.15 **	−0.25 ***	0.00	0.08
Xenosophia (5)	−0.12	0.29 ***	−0.42 ***	−0.23 **		−0.16 **	0.00	0.08	−0.17 **	0.06
Anti-Muslim/anti-Western attitudes (6)	0.21 **	−0.17 *	0.35 ***	0.59 ***	−0.23 **		0.47 ***	−0.14 *	−0.02	−0.06
Anti-Semitism (7)	0.18 *	−0.08	0.07	0.52 ***	0.00	0.52 ***		0.02	0.00	−0.05
Anti-refugee (8)	0.21 **	−0.09	0.31 ***	0.46 ***	−0.23 ***	0.69 ***	0.41 ***		−0.10	−0.03
Age (9)	0.07	0.10	0.06	−0.10	0.13	0.02	−0.02	−0.03		−0.14 *
Gender (10)	−0.23 **	0.03	−0.30 ***	−0.16	0.19 *	−0.16 *	−0.04	−0.19 *	−0.11

Notes. Pearson’s correlations are displayed. Pairwise exclusion of missing values. *** p ≤ 0.001, ** p ≤ 0.01, * p ≤ 0.05; rest ns. Gender was here coded as male = 1, female = 2, non-binary = missing.

presents the zero-order correlations of all included variables for the Turkish Muslims subsample above the diagonal and for the German Christian subsample below the diagonal. With regard to Hypothesis 1, faith-oriented reflection and intellect-oriented reflection showed significant positive intercorrelations in both subsamples, even without controlling for fundamentalism or defense against secularism. The correlation was notably high in the Turkish Muslim subsample and of medium size in the German Christian subsample. In the latter, the correlation was notably elevated to r = 0.37 *** when simultaneously controlling for defense against secularism. Introducing an additional control for fundamentalism did not alter the correlation coefficient. Conversely, in the Turkish Muslim subsample, the correlation between faith-oriented reflection and intellect-oriented reflection slightly increased when controlling for defense against secularism (r = 0.59 ***). When further controlling for fundamentalism, the correlation coefficient slightly decreased to r = 0.53 ***.

Notably, among Turkish Muslims, faith-oriented reflection did not exhibit a linear relationship with defense against secularism. However, both faith-oriented reflection and intellect-oriented reflection were positively correlated with xenosophia and, simultaneously, positively correlated with prejudice, especially anti-Semitism among Turkish Muslims. In the German Christian subsample, a distinct pattern emerged where faith-oriented reflection was strongly associated with defense against secularism. While faith-oriented reflection was not related to xenosophia but was positively correlated with all prejudice measures, intellect-oriented reflection was positively correlated to xenosophia and negatively correlated with anti-Muslim prejudice, but not with other forms of prejudice. Despite a medium positive relationship between faith-oriented reflection and intellect-oriented reflection in the German Christian subsample, a more pronounced polarization between these two reflections appeared concerning social openness. Interestingly, this polarization was not observed in the Turkish Muslim subsample. This disparity may be attributed to a more substantial confounding of these central variables with fundamentalism among the German Christian subsample.

Before proceeding to the multivariate analyses, which account for these confounding variables, a brief examination of the interrelationships between xenosophia and the measures of prejudice is given. In the Turkish Muslim subsample, anti-Western and anti-Semitic prejudices exhibited a high intercorrelation. Anti-refugee prejudices demonstrated a negative correlation with anti-Western prejudice and a non-significant correlation with anti-Semitism. Xenosophia was negatively associated solely with anti-Western prejudice and not with the other two forms of measured prejudice. In contrast, within the German Christian subsample, all three prejudice subscales displayed very high intercorrelations. There, medium negative correlations were seen between xenosophia and anti-Muslim, as well as anti-refugee attitudes, with both prejudiced measures showing strong intercorrelations. However, xenosophia did not correlate with anti-Semitism. Finally, the sociodemographic variables, specifically gender in both subsamples and age in the Turkish subsample, appeared to be confounded with our variables of interest. Therefore, the next multivariate also controls for age and gender.

6.2. Multivariate Hierarchical Regressions

With regard to Hypothesis 3 the regressions included faith-oriented reflection as a measure of religiosity in the first step, defense against secularism in the second step, and the control variables fundamentalism, age, and gender in the third step.

Table 3. Multivariate hierarchical regression analysis on xenosophia for both subsamples.

	Turkish Muslims (n = 345)			German Christians (n = 168)
	1	2	3	1	2	3
Faith-oriented reflection	0.25 ***	0.25 ***	0.34 ***	−0.12	−0.03	0.21 *
Defense against secularism		0.00	0.05		−0.21 *	−0.03
Fundamentalism			−0.18 ***			−0.51 ***
Age			−0.17 **			0.15 *
Gender			−0.03			0.09
R2	0.06 ***	0.06 ***	0.10 ***	0.01	0.05 *	0.23 ***
R2change		0.00	0.05 ***		0.04 *	0.18 ***

Notes. Standardized regression weights (betas) are displayed. Pairwise exclusion of missing values. *** p ≤ 0.001, ** p ≤ 0.01, * p ≤ 0.05; rest ns. Gender was here coded as male = 1, female = 2, non-binary = missing. No multicollinearity since VIF < 10.

presents the regressions on xenosophia for both subsamples separately. In the Turkish Muslims subsample, the strong positive relationship between faith-oriented reflection and xenosophia was not mediated by any other included variable, although both fundamentalism and age were negative predictors. Conversely, in the German Christian subsample, the initial non-significant relationship between this religiosity measure and xenosophia became significant once fundamentalism (negative predictor) and age (positive predictor) were included, referring to suppressor effects. Defense against secularism did not act as a mediator here, fully mediated by fundamentalism but not by defense against secularism.

Table 4. Multivariate hierarchical regression analysis on anti-Western/anti-Muslim prejudice for both subsamples.

	Turkish Muslims (n = 345)			German Christians (n = 168)
	1	2	3	1	2	3
Faith-oriented reflection	0.21 ***	0.20 ***	0.04	0.21 **	−0.06	−0.13
Defense against secularism		0.12 *	0.04		0.61 ***	0.59 ***
Fundamentalism			0.33 ***			0.11
Age			−0.04			0.08
Gender			−0.05			−0.06
R2	0.05 ***	0.06 ***	0.14 ***	0.04 **	0.35 ***	0.37 ***
R2change		0.01 *	0.08 ***		0.31 ***	0.01

Notes. Standardized regression weights (betas) are displayed. Pairwise exclusion of missing values. *** p ≤ 0.001, ** p ≤ 0.01, * p ≤ 0.05; rest ns. Gender was here coded as male = 1, female = 2, non-binary = missing. No multicollinearity since VIF < 10.

displays the variables predicting anti-Western prejudice among Turkish Muslims and, in parallel, predicting anti-Muslim prejudice among German Christians. The initial positive prediction of faith-oriented reflection for these prejudices in both subsamples became non-significant when defense against secularism (for German Christians) or fundamentalism (for Turkish Muslims) was included. The variance explanation ranged from low (for Turkish Muslims) to high (for German Christians).

Table 5. Multivariate hierarchical regression analysis on anti-Semitism for both subsamples.

	Turkish Muslims (n = 345)			German Christians (n = 168)
	1	2	3	1	2	3
Faith-oriented reflection	0.31 ***	0.30 ***	0.36 ***	0.18 *	−0.06	0.08
Defense against secularism		0.11 *	0.09		0.55 ***	0.64 ***
Fundamentalism			0.11			−0.30 ***
Age			−0.02			−0.05
Gender			−0.09			0.00
R2	0.10 ***	0.11 ***	0.13 ***	0.03 *	0.28 ***	0.33 ***
R2change		0.01 *	0.02		0.24 ***	0.05 **

Notes. Standardized regression weights (betas) are displayed. Pairwise exclusion of missing values. *** p ≤ 0.001, ** p ≤ 0.01, * p ≤ 0.05; rest ns. Gender was here coded as male = 1, female = 2, non-binary = missing. No multicollinearity since VIF < 10.

demonstrates the same independent variables predicting anti-Semitism as a dependent variable in both subsamples separately. While the positive prediction of faith-oriented reflection was again fully mediated by defense against secularism among the German Christian subsample, this was not the case for Turkish Muslims. Here, the positive prediction of faith-oriented reflection on anti-Semitism remained after the inclusion of all further variables. In contrast, fundamentalism played a negative role in predicting anti-Semitism among German Christians.

Finally,

Table 6. Multivariate hierarchical regression analysis on anti-refugee prejudice for both subsamples.

	Turkish Muslims (n = 345)			German Christians (n = 168)
	1	2	3	1	2	3
Faith-oriented reflection	−0.08	−0.05	0.11	0.21 **	0.02	−0.05
Defense against secularism		−0.24 ***	−0.16 **		0.46 ***	0.42 ***
Fundamentalism			−0.30 ***			0.09
Age			−0.10			0.00
Gender			−0.06			−0.11
R2	0.01	0.06 ***	0.13 ***	0.05 **	0.22 ***	0.23 ***
R2change		0.06 ***	0.07 ***		0.17 ***	0.02

Notes. Standardized regression weights (betas) are displayed. Pairwise exclusion of missing values. *** p ≤ 0.001, ** p ≤ 0.01; rest ns. Gender was here coded as male = 1, female = 2, non-binary = missing. No multicollinearity since VIF < 10.

presents the prediction of anti-refugee prejudice. For the German-Christian subsample, defense against secularism fully mediated the initial positive relationship with faith-oriented reflection. For Turkish Muslims, faith-oriented reflection was not significant at all, while defense against secularism, as well as fundamentalism, were highly significant negative predictors of anti-refugee prejudice.

7. Discussion

In times of increasing globalization and religious pluralism, concepts such as acceptance of diversity or religious openness gain societal importance since their absence can threaten domestic social cohesion and negatively impact foreign affairs. Of particular concern are recent but separate studies in Germany and Turkey that have revealed high levels of inter-religious and intercultural prejudice (e.g., Decker et al. 2022; Evangelical Church in Germany 2022; Öztürk and Pickel 2022; Ünal 2014). This issue could be addressed by (prospective) theologians, who hold significant potential for promoting peace due to their influence in society as clerics, teachers of religion, spiritual counselors, or holding similar positions. Although ‘Love thy neighbor’ is a central commandment in the Abrahamic religions (and beyond), studies in the psychology of religion often find positive relations between (certain forms of) religiosity and prejudice (e.g., Allport [1954] 1979; Yapıcı 2004b), especially towards other religions, including refugees (e.g., Deslandes and Anderson 2019; Hoffarth et al. 2018).

In this study, we aimed to add a new perspective to classical religiosity–prejudice research by framing the topic within the Religious Openness Hypothesis (ROH). The ROH asserts that religions, even in their conservative forms, have the potential for intellectual, psychological, and social openness (such as xenosophia or the absence of prejudice)—at least when some confounding variables, such as fundamentalism and defense against secularism, are controlled for (Watson et al. 2014, 2015b, 2019). Therefore, we conducted a questionnaire study among theology students in Germany and Turkey, who are soon-to-be multipliers of religion and religiosity in their respective societies (e.g., as teachers of religion at schools or as clerics), to investigate how they perceive each other and how they perceive shared religious and/or cultural outgroups (Jews, refugees). The results are eventually incorporated into implications for curricula for universities and schools at the end of this paper.

The aims of the study were threefold: As predicted in Hypothesis 1, positive correlations between religiosity (faith-oriented reflection) and intellectual openness (intellect-oriented reflection) were found in the Turkish Muslim subsample. Contrary to this hypothesis, a similar positive bivariate relationship between religiosity and intellectual openness was found among German Christians, even without controlling for confounding variables. However, this correlation was particularly pronounced within the Turkish Muslim subsample, where it is strong and highly significant. Previously, some researchers have focused on ideological explanations, as Islam values reasoning and intellectual endeavors, encouraging its members to use their minds effectively in religious and worldly matters (e.g., Dover et al. 2007). However, Watson and colleagues (for an overview, see Watson 2019; Demmrich and Akçe 2022) demonstrated in numerous studies that the crucial reason might be the sociological environment shaping the ideological surround. According to their studies, a positive correlation between religiosity and (intellectual) openness is pronounced in less secularized societies. Adherents of religious traditions in secularized countries might show less openness as they encapsulate themselves from others and defend their faith against secularism. As questioned in the theoretical part, defense against secularism appears to be a controversial point in the Turkish society. On the one hand, Turkish politics attempted to implement laicism, which would suggest a strong defense of the religious (conservatives) (e.g., Yapıcı and Ağılkaya-Şahin 2021)—this is also confirmed by the high rates of defense against secularism in Turkey compared to Germany. On the other hand, controlling for defense (and fundamentalism) did not diminish the already strong relationship between religiosity and intellectual openness. Similarly, defense against secularism did not, except in one case, show an effect on prejudice—this will be addressed again when we discuss Hypothesis 3.

In contrast, the medium correlation between religiosity and intellectual openness in the German Christian subsample became strong and highly significant once defense of secularism was controlled for (the additional control for fundamentalism did not add any value). This diverges even from previous results among Muslims of Turkish origin in Germany, where no correlation was found between religiosity and openness unless not controlled for fundamentalism (Demmrich and Akçe 2022; however, defense against secularism was not measured in this earlier study). Thus, an ideological explanation comes to the fore, as opposed to a sociological one. Previous studies in Western, secularized countries among Christians were conducted with psychology student samples (e.g., Watson et al. 2011, 2015c, 2019). In contrast, theology students might be more intellectually aware of the potentials for openness within their own faith (for Muslim Iranian students, see Ghorbani et al. 2014, 2018). Furthermore, there might be a denominational difference. While the aforementioned studies focused on Protestants in the US, our study was conducted among a predominantly Catholic sample. Given that Protestantism reinforced the Augustinian doctrine of the depravity of man, nature, and the world in contrast to Catholicism (Troeltsch 1906), our predominantly Catholic subsample could be more intellectually and socially open towards the world and its people than Protestants. The relatively low scores of defense against secularism, in comparison to Turkish Muslims, seem to support this ideological explanation.

A split of the relatively small German Christian subsample with n = 168 participants in order to perform differences tests between a very small Protestant subsample and a larger Catholic subsample seemed statistically unfeasible. Therefore, future studies should explore alternative approaches to recruit theology students as potential participants (e.g., personally advertising the study in university lectures) to enhance their motivation to participate. Additionally, further studies should aim for more balanced representations of other denominations, such as including Protestants in Christian samples and incorporating Alevi and Shia representations in Muslim samples.

Further exploring the potential polarization between religiosity and intellect in secularized contexts, as posited by the ROH, our second hypothesis stated that the relationships between social openness and both religiosity and intellectual openness would be positive for Turkish Muslims. Conversely, we expected negative bivariate relationships between social openness and religiosity, and positive relationships between social and intellectual openness for German Christians (Hypothesis 2). Our results partially support this hypothesis by revealing a notable polarization between religiosity and intellect, manifested in their contrasting associations with social openness among German Christians compared to Turkish Muslims. Among Turkish Muslims, social openness exhibited both positive (xenosophia) and negative (higher prejudice; in line with Hasnain 2007; Karaca 2023; Tausch et al. 2009) relations with religiosity and with intellectual openness. In contrast, among German Christians, social openness was predominantly negatively associated with religiosity (higher prejudice) but positively associated with intellectual openness (higher xenosophia and lower prejudice).

According to our understanding, these results reflect two further aspects for discussion. First, and in line with classical prejudice literature, religiosity tends to be positively related to inter-religious prejudice, including prejudice against refugees (e.g., Cirhinlioğlu 2010; Clobert et al. 2017; Gürses 2001; Hoffarth et al. 2018; Yapıcı and Kayıklık 2005), while intellectual openness is, across both subsamples, at least less related to prejudice.

Second, xenosophia was positively related to intellectual openness and to religiosity, with the latter relationship observed only among the Turkish Muslim subsample. The concept and measurement of xenosophia used here emphasize the appreciation of wisdom available in foreign religious traditions and worldviews, including openness to re-examining one’s current worldview (Streib 2022; Streib and Klein 2018). In support of the validity of the xenosophia scale (Streib et al. 2010), it was negatively related to anti-Western prejudice among Turkish Muslims and, parallelly, to anti-Muslim prejudice among German Christians (replicating findings by Streib and Klein 2014, 2018). Among German Christians, xenosophia was additionally negatively related to anti-refugee prejudice. In both subsamples, xenosophia was unrelated to anti-Semitism. Anti-Semitism may have a stronger political component that extends beyond an appreciation of the wisdom inherent in the Jewish religion or culture. Similarly, prejudice against refugees seemed to be not related to religiosity among Turkish Muslims, as refugees in Turkey, mainly from Syria, Iraq, and Afghanistan, are Muslims as well. For further studies on Turkish samples, especially, scales with subscales indicating political dimensions of such prejudice would be more appropriate (e.g., scales by Eser and Uygur 2019; Sarıkaya and Güner 2021), given that the reliability of the implemented scale consisting of two items only (Evangelical Church in Germany 2022) was lowest. Another limitation in measurement arises from the disparity between the anti-Western prejudice scale, which encompasses various items addressing political dimensions, and the anti-Muslim prejudice scale, which lacks comparable indicators in this regard. In future research, endeavors to measure prejudice against each other among Muslims and Christians should focus on developing scales that allow for a more equitable and comprehensive assessment of social biases.

Similar to the observed relationship between religiosity and intellectual openness (as outlined in Hypothesis 1), a more pronounced polarization between religiosity and social openness indicators was found among German Christians compared to Turkish Muslims. This aligns with our third hypothesis which posited that the relationship between religiosity (faith orientation) and social openness (xenosophia, absence of prejudice) is mediated by defense against secularism in both subsamples, while also controlling for fundamentalism and sociodemographic variables. Hypothesis 3 was partially supported. In line with the ROH (Watson 2019; summarized in Demmrich and Akçe 2022), defense against secularism fully mediated the relationship between faith-oriented reflection as a measure of religiosity and all measured prejudices among German Christians. Specifically, higher defense against secularism was associated with a greater likelihood of developing inter-religious and intercultural prejudice among Christians in Germany. However, the relationship between religiosity and prejudice did not turn negative; instead, it became non-significant. In contrast, among Turkish Muslims—paralleling the positive relationship between religiosity and intellectual openness (as discussed in Hypothesis 1)—defense against secularism did not mediate the relationship between religiosity and social openness. This lack of mediation held true for xenosophia, where the initial positive relationship with religiosity persisted throughout the entire analysis. The substantial level of defense against secularism within the Turkish Muslim subsample confirms the resistance against secularism in Turkish society (Yapıcı and Ağılkaya-Şahin 2021). However, this defense mechanism does not appear to play a significant role in the intellectual or social openness of religious theology students in this context. Together with the finding that defense against secularism is particularly high among the Turkish Muslims subsample, it confirms that the huge religious share of Turkish society has been resisting against secularism (Yapıcı and Ağılkaya-Şahin 2021) but this defense mechanism does not play a role in intellectual or social openness of religious theology students. Vice versa, defense against secularism was quite low for German Christians but played an important role in the formation of all prejudice measured. These contradictions between the level of defense against secularism and the role it plays in psychological processes forming prejudice should be a topic of further studies in the field.

In contrast to the German Christian subsample, where the link between religiosity and xenosophia was mediated by defense against secularism, the Turkish subsample exhibited mediation through fundamentalism. This pattern aligns with the findings of Watson et al. (2019) among Christians in the US. Across both subsamples, fundamentalism emerged as a consistently negative and moderately strong predictor of xenosophia, a result that replicated the earlier findings reported by Streib et al. (2010).

In summary, prejudice among German Christians appeared to stem predominantly from the impact of religious individuals residing in a secularized environment. In this context, they defend their own religious worldview, leading to the development of prejudice against those with different religious or cultural affiliations. In contrast to this, the results among Turkish Muslims appeared somewhat puzzling. While anti-Western prejudice was attributed to fundamentalism, religiosity emerged as the sole factor predicting anti-Semitism. In stark contrast to this, prejudice against refugees showed no association with religiosity; instead, defense against secularism and fundamentalism serve as buffering factors against this form of prejudice. Given the context of refugees in Turkey, who share a similar religious and cultural background with Turks (e.g., Syrians, Iraqis, Afghanians), it is plausible that Turkish Muslims, especially those with a fundamentalist worldview and a strong defense against secularism, emphasize the concept of a globalized ummah (Roy 2004) more strongly. This concept depicts Muslims as a transcultural worldwide community bound by religious ties. Another relevant concept in this context, prevalent among the conservative segments of the Turkish population, is the ansar–muhājir relationship (Efe 2019; Mezkit 2017), referencing Prophet Muhammad’s migration from Mecca to Medina. This historical event involved the people of Medina (ansar) welcoming the immigrants (muhājir) who had to leave their homeland due to threats to their lives and religious, social, and financial embargoes. Similarly, fundamentalism served as a buffering factor in the context of anti-Semitism among German Christians. This can be attributed, in all likelihood, to the belief system that includes Jews in the narrative of salvation (Heilsgeschichte), even within a fundamentalist worldview.

Overall, there was a relatively low level of variance explanation concerning prejudice among the Turkish Muslim subsample. Religiosity appeared to have a limited impact on predicting anti-Western and anti-refugee prejudice. Subsequent studies could explore the relative influence of religiosity alongside other psychological variables, such as political attitudes and personality traits (e.g., authoritarianism, Altemeyer and Hunsberger 1992). Finally, future research may center on refining the differentiation between criticism motivated by political considerations directed at “Islam” or “the West” and prejudiced perspectives, as suggested by Imhoff and Recker (2012) on Islam.

Implications

Recognizing that attitudes, including prejudices, are learnable and subject to change (Küpper 2010), the importance of religious education in shaping the relationship between religion and prejudice becomes evident. In light of our findings, we propose that theology students may possess heightened intellectual awareness of the potential openness within their own faith. The cultivation of intellectual openness in our sample positively promoted social openness, indicating that future theologians have the groundwork to offer religious education that promotes respect for human dignity, freedom of thought, morality, and cultural heritage (Selçuk 1997).

According to Yapıcı and Kayıklık (2005), social distance to religious or cultural outgroups arises from socio-cultural norms and values, which are nurtured by religious teachings within a specific socio-cultural environment. If social distances toward other religious groups are perceived and lived as religious teachings, formal religious education becomes crucial to rectify such misconceptions. University theology curricula and school-based religious education can consider our findings, highlighting that religiosity tends to be associated positively with inter-religious prejudice, including prejudice against refugees.

Considering our findings, which emphasize the role of religion in shaping prejudice more significantly among Germans than Turks, the imperative need to adapt theology curricula becomes increasingly apparent. Thus, Yapıcı and Yapıcı’s (2017, p. 20) assertion holds particular significance. To combat anti-Muslim prejudice, steering clear of outdated religious education is highly critical. Theologians, often regarded as role models for religiosity, can play a pivotal role in mitigating prejudice. Consequently, there is a pressing need to explore innovative approaches to conceptualizing religious education and theological training in universities. For example, Schweitzer (2016) advocates for a transformative perspective on theology education that acknowledges Europe’s growing multicultural and multi-confessional diversity. He proposes a comprehensive approach that transcends conventional teachings of Christianity or any specific religious tradition. This approach not only involves establishing connections with various religious communities but also tightly integrating religious education with civil society. The overarching goal is to nurture dialogue and a synthesis of diverse perspectives, cultivating religious identities grounded in peace, tolerance, and a universal ethical framework. This is crucial for promoting inclusivity and tolerance in today’s diverse and interconnected societies. In practical terms, this expanded approach can be pedagogically implemented through interreligious learning, extending beyond traditional religious education classes and permeating the entire school environment (Schweitzer 2011).

Moreover, the German-speaking region has already developed interreligious models for school religious education, including dialogical learning (Knauth and Weiße 2020), difference-hermeneutical learning (Tautz 2007), and trialogical learning (Kusch 2007). However, these models are currently applied rather sparingly in practice. In the Turkish context, there is a prevalence of comparatively more homogeneous concepts or unifying models for religious education. Therefore, there should be an increased discourse in Turkey, emphasizing a transition from memorization and catechism to fostering religious maturity, as recommended by Sarıkaya (2017) and Çelikel (2014), This should involve hermeneutic methods for deeper text analysis, rooted in Islam’s rich theological and philosophical traditions, as described by Mutluel (2013), and include Weirer’s (2021) emphasis on intra-Islamic diversity. This approach, promoting critical thinking and contextual understanding, can enhance religious literacy and counter prejudices, including antisemitism, which may arise from doctrinal misinterpretations. Such an approach would better accommodate and reflect the diverse nature of modern societies (Habermas 2001).

To fulfill this role, theology curricula may incorporate subjects that contribute to openness towards other religions and cultures. For example, intellectual openness can be fostered through critical thinking in the religious context, exploration of topics like politics, society, and religion, sacred texts, globalization and religion, inter-religious relations, religious movements, human rights, and democracy—as offered at some Turkish theology faculties.