Exclusive Religious Beliefs and Social Capital: Unpacking Nuances in the Relationship between Religion and Social Capital Formation

Daniel Auguste

ABSTRACT

Religion has long been associated with the promotion of social capital and social well-being. It has also been implicated, however, in the promotion of social division and exclusion. Using data on indicators of exclusive religious beliefs and social capital collected by the World Values Survey across forty-seven countries, this study sheds light on this apparent paradox. Results from mixed-effects ordered and binomial logistic regressions show that, on the one hand, the greater the extent to which people believe in the authenticity of their own religion and the inauthenticity of others, the less likely they are to accept immigrants, people of a different race, or those who speak a different language as neighbors. They are also less likely to trust people of another religion and nationality. On the other hand, results show that religious individuals are more likely than their nonreligious counterparts to trust people of another nationality and to accept immigrants, people of a different race, and those who speak a different language as neighbors. These findings reveal an important distinction between individuals’ overall religiosity and their exclusive religious beliefs. The findings help explain why religion has been implicated simultaneously in the promotion of both social cohesion and exclusion. Finally, this analysis suggests that the connection between religion and social exclusion may be a consequence of religious intolerance rather than religiosity.

Keywords: social capital, social inclusion, social exclusion, religion

We live at a time in which polarization and exclusion are burgeoning and considered the only way to resolve conflicts. We see, for example, how quickly those among us with the status of a stranger, an immigrant or a refugee, become a threat, take on the status of an enemy. An enemy because they come from a distant country or have different customs. An enemy because of the color of their skin, their language or their social class. An enemy because they think differently or even have a different faith.

—Pope Francis, in his homily for the Ordinary Public Consistory for the Creation of New Cardinals (November 2016)

We are observing an increasing use of religion as justification for social exclusion and anti-multiculturalism in many countries around the world. Fifty-five years ago, Martin Luther King Jr. brought similar phenomena to public attention in the United States when he stated that “the church is the most segregated major institution in America…eleven o’clock on Sunday morning…[is] the most segregated hour of America” (1964). Recent research using data on religious affiliation across six countries (Canada, China, Jordan, Turkey, United States, and South Africa) found that children from religious households demonstrated a low level of altruistic behaviors and a high level of punitive behaviors compared to those from secular households (Decety et al. 2015). However, research also finds that religion fosters civic engagement, philanthropy (Lam 2006; Lim and MacGregor 2012), community empowerment and development (Barnes 2005; Pattillo-McCoy 1998; Reese and Shields 2000), and altruism and social cohesion (Putnam and Campbell 2010:456–61). How do we reconcile these seemingly paradoxical narratives about religion as a source of social cohesion and its common use to justify social differentiation and exclusion in contemporary societies? This study sheds light on this apparent paradox by examining the link between individuals’ exclusive religious beliefs and the tendency to form bonding social capital (i.e., the propensity to accept and trust familiar or like-minded others) and bridging social capital (i.e., the propensity to accept and trust unfamiliar or unlike-minded others). Exclusive religious belief is here conceptualized as the degree to which individuals believe in the authenticity of their own religion and the inauthenticity of others. Thus, the analysis examines the impact of the extent to which people believe in the authenticity of their own religion and the inauthenticity of others on their propensity to form bonding and bridging social capital. By emphasizing individual exclusive religious beliefs, as opposed to disproportionately focusing on religious affiliations (as in most previous research), this study improves our understanding of the multifaceted ways that religion may influence social capital.

REVIEW OF RELEVANT RESEARCH AND THEORETICAL FRAMEWORK

Conceptualization of Social Capital: Bonding and Bridging Social Capital

Social capital is understood as social relationships and social norms, such as social network interaction and trust (Bourdieu 1985; Coleman 1988). Building on this conceptualization of social capital, research has identified two primary forms of social capital: bonding and bridging (Putnam 2000:22). Bonding social capital refers to the tendency of people to engage in social interactions and form trust with people like them or like-minded others. Bridging social capital pertains to the tendency of people to form trust and social networks with unlike-minded and unfamiliar individuals, such as people of different races, religions, nationalities, or belief systems. Bonding social capital is characterized by network closure and tends to reinforce group homogeneity, whereas bridging social capital tends to foster network diversity and social inclusion (Storm 2015). I use this typology of social capital to examine the relationship between exclusive religious beliefs (measured as the extent to which people believe that their religion is the only acceptable religion) and the tendency to form bonding and bridging social capital.

Review of Relevant Research on Religion and Social Capital

Understanding the link between religion and social capital formation has long been a focus of sociological inquiry. Scholars like Alexis de Tocqueville ([1835] 1954) and Emile Durkheim ([1912] 1995) pioneered the study of religion and social capital development, and this branch of research has gained increasing popularity among social scientists with the new conceptualization of social capital offered by the political scientist Robert Putnam in his famous work Bowling Alone (2000) and by the work of James Coleman (1988). In the United States, some argue that religion constitutes a key source of community cohesion and is one of the most important forms of social capital endowment (Putnam and Campbell 2010). Research also suggests that religious Americans are significantly more likely to develop friendships and social interactions with people who are socially and racially different from them (Putnam and Campbell 2010). Other research goes further to conclude that Protestant values and norms foster social trust within religious communities that spill over to the broader community (Fukuyama 1995). The basis of this argument is that religious communities provide a safe environment for social interactions and the development of interpersonal relationships necessary for the formation of trust that stretches across social and racial divides (Fukuyama 1995).

Research conducted in other countries echoes this conclusion from U.S.-based research about the positive impact of religious community on social cohesion and social capital (Dingemans and Van Ingen 2015). For example, using data across German regions and generalized trust as a measure of social capital (i.e., trust in most people), Richard Traunmüller (2011) finds that people who are affiliated with a religious denomination demonstrate a greater propensity to trust in most people than their nonreligious counterparts. In the United Kingdom, research also finds that people who identify as Protestants are more trusting of others than those who are nonreligious (Storm 2015). Using data from fifteen Western European countries, other research finds a similar association between generalized trust and religion (Paxton, Reith, and Glanville 2014). The common conclusion in these previous studies is that religiosity (commonly measured as affiliation with a religious denomination) constitutes a key determining factor of social capital and cohesion.

Building on the argument about religion as a source of social capital endowment, some research tries to disentangle the potential relative effects of different religious traditions on social capital production. In doing so, this research theorizes that some religious ideologies tend to promote stronger symbolic boundaries between insiders and outsiders than others (Welch et al. 2004). Scholars argue that religious traditions promoting clear boundaries between insiders and outsiders tend to foster strong in-group trust and networks but also promote out-group distrust, intolerance, and distance between unlike-minded others (Traunmüller 2011). Although empirical evidence lags behind theoretical claims about differences in religious traditions, a common conclusion emerging from this theoretical approach to religion and social capital stipulates that Protestant religions tend to instill in people religious values that create a culture of trust that extends to the broader society. On the other hand, research claims that non-Christian religious values, such as Buddhist, Islamic, and Confucian values, tend to promote strong in-group trust but outgroup distrust and social exclusion (Fukuyama 1995).

Within this sub-branch of research, some studies also emphasize the relative importance of different subsets of the Christian religious traditions for the promotion of social capital. Some scholars argue that Catholic and conservative Protestants have a greater tendency to form trust in people they know and socialize more with like-minded others compared to mainstream Protestants (Blanchard 2007; Welch, Sikkink, and Loveland 2007). In the United States, research shows that social distance, measured as residential racial segregation, is greater in counties where the conservative Protestant population is high (Blanchard 2007). Other scholars find that, in the United States, White Protestant religious organizations tend to be characterized by network

closure and strong in-group social connection (Altemeyer 2003; Emerson and Smith 2000). These studies, overall, emphasize a greater importance of mainstream Protestantism in fostering social capital as compared to non-Christian, Catholic, and conservative religious traditions.

Although these findings seem to dominate the narrative about the relationship between religion and social capital, recent research shows them to be inconclusive. For example, research

suggests a greater and positive impact of non-Christian religious tradition (compared with their mainstream Protestant counterparts) on bridging social capital formation. Research conducted

on U.S. college campuses finds that students who identify as Muslim, Hindu, or Buddhist are more likely than their Protestant counterparts to engage in cross-racial interaction (Park and Bowman 2015), have roommates of different racial groups, and participate in cross-racial/cultural awareness activities training (Cole and Ahmadi 2010). Within this line of inquiry, some go further to question the argument about the positive association between religion and social capital. Hence, some scholars argue that because religion tends to create trust through in-group socialization and networks, religious diversity, for example, may lower social integration; consequently, religion may adversely affect social capital formation (Delhey and Newton 2005).

In the same vein, testing the argument that religion facilitates the formation of social trust that bridges across social groups, in an experiment that paired individuals within unknown social networks, research shows that being religious has no impact on the likelihood that individuals will trust unfamiliar network members (Paciotti et al. 2011). This finding echoes Eric Uslaner’s (2002) argument that religion may present a barrier to trust, as it may foster strong bonds between individuals within a given religious organization at the expense of outsiders. Using survey data from a range of U.S. states and 105 countries, other results show that respondents who acknowledge that religion is important in their lives have a low tendency to place trust in most people (i.e., generalized trust) (Berggren and Bjørnskov 2009). Similarly, studies conducted on U.S. college campuses show that being religious and a member of a student religious organization are adversely related to cross-racial interaction (Park 2012; Park and Kim 2013). In summary, these studies highlight the inconsistencies in empirical research about the relationship between religion and social capital formation. This variability indicates that existing empirical analyses do not fully capture the complex and multifaceted ways in which religion influences social capital and may facilitate social cohesion. Thus, in addition to denominational affiliation, a closer examination of personal beliefs may improve our understanding of the multidimensional nature of the relationship between religion and social capital.

Theoretical Link between Personal Religious Beliefs and Social Capital Formation

As stated above, the heavy emphasis on religious affiliation combined with inconsistent findings in existing research highlight the need for a greater emphasis on personal religious beliefs (as opposed to grouping individuals based on their religious affiliations) that represent more directly the theoretical link between religion and the process of social network development (i.e., relationship-building, social trust, and attitudes toward others). I focus on personal beliefs because denominational affiliation provides a limited understanding of the extent to which individuals adhere to the precepts of their denominations (Lam 2002:408), hence how religion influences individual social interactions and behavior. For this reason, I contend that personal religious beliefs, and particularly those pertaining to inclusionary or exclusionary attitudes, are important to fully capture the extent to which religion may influence individual social interactions. Therefore, I will examine the extent to which an individual’s exclusionary religious beliefs influence the likelihood of forming bonding and bridging social capital.

To this end, I situate my theoretical framework in the symbolic boundary literature, detailing sources of symbolic distinctions people make between themselves and others (Lamont and Molnár 2002). Symbolic boundaries generate group identity by excluding people perceived as outsiders while including those considered insiders (Alexander 2006; C. Taylor 2002). The symbolic boundary literature also demonstrates that religion (often conceptualized as denominational affiliation) is an instrument that people use to draw symbolic differentiations (Edgell and Tranby 2010) and form “subcultural identities” (Smith 1998). However, although religious affiliation reveals important aspects of the processes underlying symbolic boundary creation, empirical analysis of how the process of boundary creating operates on the personal level to produce differential propensities to form bonding and bridging social capital is insufficient. This is unfortunate because, net of religious denomination, one may expect perceptions about social differentiations between people and social groups to be more pronounced among people who hold strong views about symbolic differences between their religious traditions and other religions. That is partly because beliefs (both conscious and subconscious) that people hold tend to shape their life philosophy or values, which in turn often influence their behavior (Smalley 2011; Swidler 1986), including their social interactions and tendency to develop trust in others. Thus, because religion tends to influence people’s moral standards, personal religious beliefs that embody inclusive or exclusive values provide a clearer

understanding of the mechanism by which religion influences socialization patterns (i.e., people’s decision to socialize with and trust people of different ethnic, racial groups, and cultural background or with those who are like them) than measures of religion that group individuals into religious affiliations.

In this regard, someone who believes that there are strong differences regarding the authenticity of their religion and other religions may also hold views about social reality and human nature that are marked by strong symbolic and social differentiation between social groups and individuals. In the same vein, because birds of a feather flock together (McPherson, Smith-Lovin, and Cook 2001), people who believe that there are strong differences in the authenticity of their religion (compared to other religions) may be less likely to form social networks with individuals who are different from them. These people may also be more likely to form inclusive social networks and trust with people of similar worldviews or racial, ethnic, and cultural backgrounds. In other words, personal religious beliefs that embody inclusive religious values may be more strongly associated with bridging social capital, whereas exclusive religious values may be more strongly associated with bonding social capital. That is, the worldviews of people in this study who believe that their religion is the only acceptable religion may be characterized by social exclusion values and differentiation. Based on this argument, I propose the following hypotheses (see Figure 1 for a summary).

Hypothesis 1: The greater the extent to which people believe that their religion is the only acceptable religion, the greater the tendency they have toward forming bonding social capital as defined here.

Hypothesis 2: The greater the extent to which people believe that their religion is the only acceptable religion, the lower the tendency they have toward forming bridging social capital as defined here.

Figure 1. The link between personal and exclusive religious beliefs and bonding and bridging social capital.

DATA, MEASUREMENT, AND METHODS

Data

The World Values Survey (WVS) examines changes in values and beliefs across the world and how those changes relate to transformations in other social phenomena. The WVS is a country-representative survey administered to individuals aged eighteen years and older in approximately one hundred countries, representing roughly 90 percent of the global population. The data were collected using face-to-face interviews and interviews conducted by telephone in remote areas. The survey now contains six waves of data spanning 1981 to 2014. This analysis is conducted using data from the most recent wave (i.e., Wave 6, 2010–2014), which enables me to capture the current state of attitudes about social trust and acceptance and the potential influence of religious beliefs on these attitudes. After conducting listwise deletions, the sample size includes 50,371 observations distributed across forty-seven countries. The countries under study here are Algeria, Armenia, Australia, Azerbaijan, Belarus, Chile, China, Colombia, Cyprus, Ecuador, Estonia, Germany, Ghana, Iraq, Japan, Jordan, Kazakhstan, Kyrgyzstan, Lebanon, Libya, Malaysia, Mexico, Morocco, the Netherlands, Nigeria, Pakistan, Peru, the Philippines, Poland, Romania, Russia, Rwanda, Singapore, Slovenia, South Korea, Spain, Sweden, Taiwan, Trinidad and Tobago, Tunisia, Turkey, Ukraine, United States, Uruguay, Uzbekistan, Yemen, and Zimbabwe.

Measurement of Exclusive Religious Beliefs and Bonding and Bridging Social Capital

Eight dependent variables are examined in this study. Exclusive religious beliefs were measured by asking respondents how much they agreed with the following statement: “The only acceptable

religion is my religion.” Possible responses were treated as a dichotomous variable where “agree” and “strongly agree” are coded “1” and “disagree” and “strongly disagree” are coded “0” (see Table 2 for detailed description). Answer choices for this question were originally strongly agree (1), agree (2), disagree (3), and strongly disagree (4).

Bonding social capital is represented by two items measuring trust in familiar others. Respondents were asked how much they trusted (1) “their neighbors” (2) and “people they know personally.” Choices were treated as ordinal outcomes, which rank from lowest to highest levels of trust. Bridging social capital variables were measured using two sets of questions. First, respondents were asked how much they trusted (1) “most people,” (2) “people of another religion,” (3) and “people of another nationality.” Possible responses were also treated as ordinal outcomes, ranking from lowest to highest levels of trust. Second, respondents were asked to indicate whether they would like to have as neighbors (1) “people of a different race,” (2) “immigrants or foreign workers,” and (3) “people who speak a different language.” Possible

responses are treated as dichotomous outcomes, with yes = 1 and no = 0 (see Table 1).

Table 1. Descriptive statistics for personal and exclusive religious belief, bonding and bridging social capital variables

Individual and Country-Level Control Variables

Other individual-level factors, independent of one’s religious beliefs, may influence the type of people with whom one socializes and the tendency that one has to trust familiar or unfamiliar people. For example, employment, education, marital and parental status, health condition, age, and income may be important influences. For instance, an employed person may be more likely to encounter people outside his or her close network of family and friends than an unemployed individual. Moreover, research shows that schools tend to provide opportunities for people to meet other individuals from different social backgrounds (Borgonovi 2012). Thus, employment and education may affect differences in socialization patterns and the propensity to form trust outside of one’s personal religious circle. Men and women tend to exhibit different socialization patterns (Benenson 1990; Rose and Rudolph 2006) and propensities for trusting others (Haselhuhn et al. 2015). Some scholars suggest that men are more trusting than women (Alesina and La Ferrara 2000). Being parents, particularly of young children, can also alter the structure and size of social networks (Cronenwett 1985). Moreover, older individuals tend to be more trusting than their younger counterparts (Welch, Sinkink, and Loveland 2007). Given the theoretical importance of these individual-level factors for social interaction and trust, I account for them in this analysis.

Given the importance of religious community for social capital formation (Putnam and Campbell 2010), an individual’s religious denomination is controlled for in this analysis based on whether respondents belonged to a religious denomination. Answers vary significantly across individuals and countries, but are classified into five categories: (1) Catholics, (2) Protestants, (3) non-Christians, (4) Muslims, and (5) Nonreligious. Table 2 presents the description of these variables. The countries differ greatly in their level of economic development (measured as gross domestic product per capita [GDP/capita] in U.S. dollars). For instance, in the years the survey was conducted (2011–2014), the U.S. GDP per capita was $46,568.60 (2011), Sweden’s was $40,890.70 (2011), Mexico’s was $13,430 (2012), and Nigeria’s was $16,29.50 (2011), whereas Zimbabwe had a GDP per capita of only $369.15 (2012). Research implicates differences in economic development level in cross-country differences in social capital development. Some scholars argue that the radius of trust in others tends to be greater in developed societies than in their less-developed counterparts (Delhey, Newton, and Welzel 2011). To account for individual-level differences in bonding and bridging social capital that may be due to economic disparities, this analysis controls for differences in economic development levels across countries. Twenty-one control variables are examined (see Table 2 for a detailed description of these variables).

Methods

This study assumes cross-national differences in exclusive religious beliefs. For this reason, multilevel modeling techniques are used, permitting the estimation of cross-context dynamics that may influence how personal religious beliefs shape social relations and attitudes pertaining to the formation of bonding and bridging social capital. Furthermore, the social capital variables used here are measured on both binary and ordinal scales (see Table 1 for a detailed description). As a result, mixed-effects logistic regressions for ordered and binary outcome variables are employed using Stata 14. The analysis is conducted in two stages. The first stage tests the initial hypothesis that the greater the degree to which people believe that their religion is the only acceptable religion, the greater the tendency to form bonding social capital. To this end, the analysis estimates the effect of the independent variable the only acceptable religion is my religion on the two dependent variables that measure bonding social capital: (1) trust in neighbors, and (2) trust in people one knows personally. The next stage tests the second hypothesis that the greater the extent to which people believe that their religion is the only acceptable religion, the lower the tendency toward forming bridging social capital. The analysis estimates the effect of the independent variable, the only acceptable religion is my religion, on the six dependent variables measuring bridging social capital: (1) most people can be trusted, (2) trust in people of another religion, (3) trust in people of another nationality, (4) accepting people of a different race, (5) accepting immigrants or foreign workers, and (6) accepting people who speak a different language as neighbors.

Table 2. Descriptive statistics for the control variables.

RESULTS

The Relationship between Exclusive Religious Beliefs and Bonding Social Capital

Models 1 and 2, in Table 3, test the impact of religious exclusivity attitudes on the bonding social capital measures while controlling for individual-level factors and countries’ level of economic development. However, they do not control for individual religious affiliation. In Model 1, the coefficient that gauges whether respondents believe their religion is the only acceptable religion is positive and significant (b=0.21, p < 0.001; odds=1.23). This result suggests that respondents who believe that their religion is the only acceptable religion are 1.23 times more likely to place trust in their neighbors compared to those who do not hold this belief. Model 2 assesses whether these same beliefs affect trust in people they know personally. However, this indicator is insignificant, suggesting that religiously exclusive attitudes are no more or less likely to influence such trust.

Models 3 and 4 (Table 3) test the robustness of the results from Models 1 and 2 by controlling for religious denominations using the four dummy variables. Results in Model 3 show that denominational affiliation does not eliminate the effects of exclusionary attitudes on trust in neighbors (b=0.122, p<0.05). Yet the influence of such attitudes is diminished compared to the baseline model. This significant reduction in the effect of exclusive religious beliefs after controlling for religion affiliation indicates that religious community matters for social capital formation. Yet personal religious beliefs are still important in influencing trust in neighbors. Model 4 considers trust in people whom respondents know personally. However, as in Model 2, denominational ties do not affect the insignificant influence of exclusionary attitudes on personal trust levels. Results in Models 3 and 4 highlight nuances in the relationship between personal religious beliefs, religious denomination, and social capital formation. They suggest that the effect of personal religious beliefs on social capital varies by type of bonding social capital.

Several additional observations: Although controlling for religious denomination in Model 3 does not eliminate the effect of personal and exclusive religious beliefs on bonding social capital, the religious affiliation variables have significant effects on bonding social capital. This finding is consistent with existing research that emphasizes the importance of religious organizations for bonding social capital (Putnam 2000; Putnam and Campbell 2010). In Models 3 and 4, each of the religious indicators are influential, which suggests that respondents who are affiliated with a religious denomination are more likely than nonreligious individuals to exhibit bonding social capital. For instance, in Model 3 (Table 3), people who belong to the Muslim, Christian, non-Christian, and Catholic religious categories are, respectively, 1.86 (odds =1.86), 1.20 (odds =1.20), 1.31 (odds =1.31), and 1.29 (odds =1.29) times more likely to trust in neighbors compared to their nonreligious counterparts. Together, these results show that both personal religious beliefs and religious affiliation may foster bonding social capital. The following analysis considers bridging social capital formation.

Table 3. Mixed-effects ordered logistic regressions of bonding social capital.

The Relationship between Exclusive Religious Beliefs and Bridging Social Capital

Models 1–6 in Table 4 examine the effects of exclusionary religious beliefs on the six measures of bridging social capital and control for economic development, religious denomination, and the individual-level controls. Model 1 tests the effect of the exclusionary attitudes on the likelihood that one will trust in most people. This relationship is not significant. Next, Model 2 tests the effect of the exclusionary attitudinal indicator on the likelihood that one would trust in people of another religion. This relationship is negative and significant (b=–0.502, p< 0.001), meaning that respondents who believe that their religion is the only acceptable religion (compared to those who do not hold this belief) are less likely to trust people who belong to other religions. Additionally, the exclusionary attitudinal variable undermines trust in people of another nationality (b=–0.467, p< 0.001, Mode 3).

Model 4 estimates the effect of the exclusionary attitudinal variable on the likelihood that one would accept people of a different race as neighbors and suggests negative effects (b=–0.42, p<0.001). This means that people who believe that their religion is the only acceptable religion are less likely to accept people of different races as neighbors compared to people who do not share this religious belief. Model 5 investigates the impact of beliefs that one’s religion is the only acceptable religion on the likelihood that one would accept immigrants or foreign workers as neighbors; this relationship is also predictive and negative (b=–0.262, p<0.001). Finally, the impact of exclusionary views on acceptance of people who speak a different language as neighbors is examined in Model 6. Negative effects are apparent (b=–0.320, p<0.001). In summary, I find negative and significant effects of the variable that assesses exclusionary religious beliefs on five of the six measures of bridging social capital.

The results also demonstrate that religion affects social capital on multiple levels. It can be adversely or positively associated with social capital formation depending on the measures of religion and social capital employed. In Model 3 (Table 4) the negative effect of exclusionary beliefs (b=–0.47) and positive effects of denominational affiliations (b=0.29 for Muslim, b=0.26 for Christians, and b=0.09 for Catholic) on acceptance of people of another nationality evidence this argument. The consistent negative effect of exclusionary religious beliefs and the inconsistent effects of religious denomination on five measures of bridging social capital (Table 4, Models 2–6) support my argument that, in addition to religious affiliation, personal religious belief is important for understanding nuances in the ways that religion influences social capital formation. These results also highlight the importance of religious affiliation for social capital formation, which is consistent with previous research (Putnam 2000; Putnam and Campbell 2010).

These findings improve on previous research; they show that the importance of religious denomination depends on types of social capital. For instance, only being a Muslim has a significant effect on acceptance of people of different races and that effect is negative (b=–0.323, Model 4) and only being a Christian has a significant and positive effect on acceptance of immigrants (b=0.135, Model 5). In sum, these findings support both my argument that existing research has not fully captured the multidimensional ways that religion relates to social capital formation and that personal religious beliefs are crucial for understanding the nuanced relationship between religion and social capital development.

Table 4. Mixed-effects ordered binomial regressions of bridging social capital.*

DISCUSSION

Since the works of Emile Durkheim ([1912] 1995) and Max Weber ([1905] 1930), religion has been associated with the promotion of social welfare, cohesion, and development. In contrast, religion has also been implicated in the nurturing of division and oppression, such as racial segregation, discrimination, and White supremacy in general, and as justification for social exclusion, such as anti-immigrant and nationalist rhetoric, in particular (Ben-Nun Bloom, Arikan and Courtemanche 2015; M. Taylor 2016). By examining the relationship between individuals’ exclusive religious beliefs and bonding and bridging social capital, this analysis provides empirical tools to reconcile this apparent paradox of religion as both an agent of social cohesion and division.

A key finding here suggests looking beyond religious affiliation and considering personal religious beliefs to better understand some of the benefits and challenges associated with religious beliefs. By simultaneously examining the influence of individuals’ exclusive religious beliefs and denomination on bonding and bridging social capital, this analysis shows that both sentiments of social exclusion and inclusion may emerge. Overall, my findings show that people who believe that their religion is the only acceptable religion tend to be less trusting and accepting of diverse people, while at the same time religious individuals (compared with their nonreligious counterparts) are more trusting and accepting of unfamiliar others. These findings

indicate that the relationship between religion and social capital is complex. Different dimensions of religion may have differential impacts on social capital formation.

The results of this analysis also show that what many respondents believe about the authenticity of their religion (compared to other religions), independent of individuals’ religious affiliation, reveals why religion may promote both social inclusion and exclusion. Furthermore, these findings demonstrate that the way social capital is conceptualized is also crucial for understanding the complexity through which religion influences social inclusion and exclusion. Depending on how social capital is measured, research may conclude that religion serves as a bridge uniting a society’s social strata, whereas others may find religion to be a source of division and exclusion. This argument is supported here by the finding that the effect of religion on social capital varies by different measures of bonding and bridging social capital.

Moreover, my findings suggest that research that uses only generalized trust (i.e., trust in most people) as a measure of social capital could conclude that religion has no effect on social capital formation. However, using both generalized trust and other measures of social capital as was done in this endeavor provides a more systematic examination of the different ways that religion affects social capital formation. The importance of how social capital is conceptualized is also demonstrated, given that religious affiliation had a significant effect on some measures of bridging social capital here and no significant effect on others. For instance, the results show no significant difference between those affiliated with a religious denomination (except for Muslims and Christians) and the nonreligious in the likelihood of accepting people of different races, immigrants, and people who speak a different language as neighbors. Thus, the inconsistent previous research may be partly because religion relates differently to various aspects of social capital and because prior studies often used a narrow range of measures of social capital.

CONCLUSION

The findings of the present analysis suggest that the effects of religion may vary based on its conceptualization (e.g., personal religious beliefs and religious affiliation) and how social capital

is measured. Conceptualizing religion both as denominational affiliation and individuals’ exclusive religious beliefs and considering multiple measures of bonding and bridging social capital, this analysis illustrates ways to reconcile the seemingly contradictory findings in previous research about the relationship between religion and social capital. These results show that religion can both foster and hinder social capital development. My finding that respondents who believe that their religion is the only acceptable religion express diminished acceptance of people of different races, immigrants, and people who speak a different language than those who do not hold such a belief lends support to this argument. Individuals who hold such religious beliefs may express superiority that fosters exclusion of diverse groups.

Yet that religious individuals, overall, have a high propensity (as compared to their non-religious counterparts) to accept immigrants, people of different races, and those who speak a different language, and to trust people of another nationality, demonstrates that religion may also promote social inclusion and acceptance of multiculturalism. These findings suggest caution

when reducing individuals to religious categories or affiliations. Thus, this study demonstrates that understanding individuals’ personal religious beliefs is also crucial. The empirical analysis supports my argument that denominational affiliation provides a limited understanding of the extent to which religion influences one’s social interactions. The findings also justify Pui-Yan Lam’s (2002) concern about using solely religious affiliation for under standing the ways that religions influence individual behavior. In sum, the findings support my theoretical framework (summarized in Figure 1) that exclusionary religious beliefs will increase the tendency toward forming bonding social capital but decrease the propensity toward building bridging social capital.

Although this analysis shows that individuals’ exclusionary beliefs undermine the formation of bridging social capital, due to data limitations this analysis did not investigate other potential motives such as economic, ethnic, class, or political identity-based motives for creating symbolic boundaries that include some people while excluding others. In-depth interviews across

groups may glean additional motives and motivations for both inclusive and exclusionary behavior. Furthermore, given the findings of this study and the multilevel nature of the relationship between religion and social capital illustrated in this study, future research may extend this analysis to the impact of societal-level religious beliefs on potential cross-country differences in social capital. For example, research suggests that the radius of trust in others (i.e., the width of the circle of others that people imagine when they say “they trust in most people”) varies across cultures and by level of economic prosperity (Delhey, Newton, and Welzel 2011). Future research might examine the potential influence of societal-level religious beliefs that embody the values of religious superiority and exclusion on the radius of trust in a society.

Finally, the finding of a significant and systematic influence of exclusionary religious beliefs on bonding and bridging social capital may have implications for understanding sources of community activism and development. This is important because research finds that religiosity (Calhoun-Brown 1999), the theology of church ministers (Reese and Shields 2000), and religious rituals, such as prayer, music, and social justice sermons (Barnes 2005) influence community activism and development. Building on the findings of this analysis and these prior studies, future research may improve our understanding of the process by which religion influences social activism by examining potential links between people’s exclusionary and inclusionary religious beliefs and their propensity for engaging in community activism. Answers to such queries may improve not only our understanding of the multifaceted nature of the processes by which religion influences social capital but also our knowledge of the underlying mechanism of the observed link between religion and community action for social change.

REFERENCES

Alesina, Alberto, and Eliana La Ferrara. 2002. “Who Trusts Others?” Journal of Public Economics 85(2):207–34.

Alexander, Jeffrey. 2006. The Civil Sphere. New York: Oxford University Press.

Altemeyer, Bob. 2003. “Why Do Religious Fundamentalists Tend to Be Prejudiced?” International Journal for the Psychology of Religion 13(1):17–28.

Barnes, Sandra L. 2005. “Black Church Culture and Community Action.” Social Forces 84(2):967–94.

Benenson, Joyce F. 1990. “Gender Differences in Social Networks.” Journal of Early Adolescence 10(4):472–95.

Ben-Nun Bloom, Pazit, Gizem Arikan, and Marie Courtemanche. 2015. “Religious Social Identity, Religious Belief, and Anti-immigration Sentiment.” American Political Science Review 109(2):203–21.

Berggren, Niclas, and Christian Bjørnskov. 2009. “Does Religiosity Promote or Discourage Social Trust? Evidence from Cross-Country and Cross-State Comparisons.” SSRN: https://ssrn.com/abstract=1478445.

Blanchard, Troy C. 2007. “Conservative Protestant Congregations and Racial Residential Segregation: Evaluating the Closed Community Thesis in Metropolitan and Nonmetropolitan Counties.” American Sociological Review 72(3):416–33.

Borgonovi, Francesca. 2012. “The Relationship between Education and Levels of Trust and Tolerance in Europe.” British Journal of Sociology 63(1):146–67.

Bourdieu, Pierre. 1985. “The Forms of Capital.” Pp. 241–58 in Handbook of Theory and Research for the Sociology of Education, edited by J. Richardson. New York: Greenwood.

Calhoun-Brown, Allison. 1999. “The Image of God: Black Theology and Racial Empowerment in the African American Community.” Review of Religious Research 40(3):197–211.

Cole, Darnell, and Shafiqa Ahmadi. 2010. “Reconsidering Campus Diversity: An Examination of Muslim Students’ Experiences.” Journal of Higher Education 81(2):121–39.

Coleman, James S. 1988. “Social Capital in the Creation of Human Capital.” American Journal of Sociology 94:95–120.

Cronenwett, Linda R. 1985. “Parental Network Structure and Perceived Support after Birth of First Child.” Nursing Research 34(6):347–52.

Decety, Jean, Jason M. Cowell, Kang Lee, Randa Mahasneh, Susan Malcolm-Smith, Bilge Selcuk, and Xinyue Zhou. 2015. “The Negative Association between Religiousness and Children’s Altruism across the World.” Current Biology 25(22):2951–55.

Delhey, Jan, and Kenneth Newton. 2005. “Predicting Crossnational Levels of Social Trust: Global Pattern or Nordic Exceptionalism?” European Sociological Review 21(4): 311–27.

Delhey, Jan, Kenneth Newton, and Christian Welzel. 2011. “How General Is Trust in ‘Most People’? Solving the Radius of Trust Problem.” American Sociological Review 76(5):786–807.

Dingemans, Ellen, and Erik Van Ingen. 2015. “Does Religion Breed Trust? A Cross-National Study of the Effects of Religious Involvement, Religious Faith, and Religious Context on Social Trust.” Journal for the Scientific Study of Religion 54(4):739–55.

Durkheim, Emile. [1912] 1995. The Elementary Forms of Religious Life. Translated by Karen E. Fields. New York: Free Press.

Edgell, Penny, and Eric Tranby. 2010. “Shared Visions? Diversity and Cultural Membership in American Life.” Social Problems 57(2):175–204.

Emerson, Michael O., and Christian Smith. 2000. Divided by Faith: Evangelical Religion and the Problem of Race in America. New York: Oxford University Press.

Francis, Pope. “Homily of His Holiness Pope Francis for the Ordinary Public Consistory for the Creation of New Cardinals.” Retrieved November 14, 2019. (http://w2.vatican.va/content/francesco/en/homilies/2016/documents/papa-francesco_20161119_omelia-concistoro-nuovi-cardinali.html).

Fukuyama, Francis. 1995. Trust: The Social Virtues and the Creation of Prosperity. New York: Free Press.

Haselhuhn, Michael P., Jessica A. Kennedy, Laura J. Kray, Alex B. Van Zant, and Maurice E. Schweitzer. 2015. “Gender Differences in Trust Dynamics: Women Trust More Than Men Following a Trust Violation.” Journal of Experimental Social Psychology 56(2015):104–9.

King, Martin Luther Jr. 1964. “Methodist Student Leadership Conference Address.” American Rhetoric: Online Speech Bank. Retrieved November 28, 2016. (http://www.americanrhetoric.com/speeches/mlkmethodistyouthconference.htm).

Lam, Pui-Yan. 2002. “As the Flocks Gather: How Religion Affects Voluntary Association Participation.” Journal for the Scientific Study of Religion 41(3):405–22.

Lam, Pui-Yan. 2006. “Religion and Civic Culture: A Cross-National Study of Voluntary Association Membership.” Journal for the Scientific Study of Religion 45(2):177–93.

Lamont, Michèle, and Virág Molnár. 2002. “The Study of Boundaries in the Social Sciences.” Annual Review of Sociology 28:167–95.

Lim, Chaeyoon, and Carol A. MacGregor. 2012. “Religion and Volunteering in Context: Disentangling the Contextual Effects of Religion on Voluntary Behavior.” American Sociological Review 77(5):747–79.

McPherson, J. Miller, Lynn Smith-Lovin, and James M. Cook. 2001. “Birds of a Feather: Homophily in Social Networks.” Annual Review of Sociology 27:415–44.

Paciotti, Brian, Peter Richerson, Billy Baum, Mark Lubell, Tim Waring, Richard McElreath, Charles Efferson, and Ed Edsten. 2011. “Are Religious Individuals More Generous, Trusting, and Cooperative? An Experimental Test of the Effect of Religion on Prosociality.” Pp. 267–305 in The Economics of Religion: Anthropological Approaches, edited by L. Obadia and D. C. Wood. Bingley, West Yorkshire, UK: Emerald Group Publishing.

Park, Julie J. 2012. “When Race and Religion Collide: The Effect of Religion on Interracial Friendship during College.” Journal of Diversity in Higher Education 5(1):8–21.

Park, Julie J., and Nicholas A. Bowman. 2015. “Religion as Bridging or Bonding Social Capital: Race, Religion, and Cross-Racial Interaction for College Students.” Sociology of Education 88(1):20–37.

Park, Julie J., and Young Kim. 2013. “Interracial Friendship, Structural Diversity, and Peer Groups: Patterns in Greek, Religious, and Ethnic Student Organizations.” Review of Higher Education 37(1): 1–24.

Pattillo-McCoy, Mary. 1998. “Church Culture as a Strategy of Action in the Black Community.” American Sociological Review 63(6): 767–84.

Paxton, Pamela, Nicholas E. Reith, and Jennifer L. Glanville. 2014. “Volunteering and the Dimensions of Religiosity: A Cross-National Analysis.” Review of Religious Research 56(4):597–625.

Putnam, Robert D. 2000. Bowling Alone. The Collapse and Revival of American Community. New York: Simon and Schuster.

Putnam, Robert D., and David E. Campbell. 2010. American Grace: How Religion Divides and Unites Us. New York: Simon and Schuster.

Reese, Laura, and Gary Shields. 2000. “Faith-Based Economic Development.” Policy Studies Review 17(2):84–103.

Rose, Amanda J., and Karen D. Rudolph. 2006. “A Review of Sex Differences in Peer Relationship Processes: Potential Trade-Offs for the Emotional and Behavioral Development of Girls and Boys.” Psychological Bulletin 132(1):98–131.

Smalley, Susan. 2011. “How Our Beliefs Shape Our Actions in the World.” HuffPost. Retrieved November 14, 2019. (https://www.huffingtonpost.com/susan-smalley/beliefs-actions_b_828596.html).

Smith, Christian. 1998. American Evangelicalism: Embattled and Thriving. Chicago: University of Chicago Press.

Storm, Ingrid. 2015. “Civic Engagement in Britain: The Role of Religion and Inclusive Values.” European Sociological Review 31(1):14–29.

Swidler, Ann. 1986. “Culture in Action: Symbols and Strategies.” American Sociological Review 51(2): 273–86.

Taylor, Charles. 2002. “Democracy, Inclusive and Exclusive.” Pp. 181–94 in Meaning and Modernity: Religion, Polity, and the Self, edited by R. Madsen, M. W. Sullivan, A. Swidler, and S. Tipton. Berkeley: University of California Press.

Taylor, Michael. 2016. “British Proslavery Arguments and the Bible, 1823–1833.” Slavery & Abolition 37(1):139–58.

Tocqueville, Alexis de. [1835] 1954. Democracy in America. New York: Oxford University Press.

Traunmüller, Richard. 2011. “Moral Communities? Religious as a Source of Social Trust in a Multilevel Analysis of 97 Germany Regions.” European Sociological Review 27(3):346–63.

Uslaner, Eric M. 2002. The Moral Foundations of Trust. New York: Cambridge University Press.

Weber, Max. [1905] 1930. The Protestant Ethic and the Spirit of Capitalism. New York: Scribner’s.

Welch, Michael R., David Sikkink, Eric Sartain, and Carolyn Bond. 2004. “Trust in God and Trust in Man: The Ambivalent Role of Religion in Shaping Dimensions of Social Trust.” Journal for the Scientific Study of Religion 43(3):317–43.

Welch, Michael R., David Sikkink, and Matthew T. Loveland. 2007. “The Radius of Trust: Religion, Social Embeddedness, and Trust in Strangers.” Social Forces 86(1):23–46.

About the Author

Daniel Auguste, University of North Carolina at Charlotte