Networks and Religion: Ties That Bind

1. Religious groups recruit primarily through social networks. People are much more likely to join religious groups where they know people than where they do not. Take, for instance, John Lofland and Rodney Stark’s (1963) study of people converting to the Unification Church (aka, the Moonies). Stark later remarked that when he and Lofland first began watching the group, it “had never succeeded in attracting a stranger" (Stark 1996). Everyone who joined already knew someone who was a member. Moreover, Stark and Lofland witnessed a number of people who were sympathetic with the group’s doctrines, but in the end did not join because they had numerous ties with people who disapproved of the Moonies. This led them to conclude that the people who ultimately joined tended to be those whose ties to group members exceeded their ties to nonmembers. Or to put it differently, conversion typically involved aligning one's beliefs and practices with those of one's friends and family.

In another study, Rodney Stark and Bill Bainbridge (1980) looked at the role that social ties play in recruiting people to the Mormon Church. They found that when Mormon missionaries go door to door, their success rate is only 0.1 percent. Referrals provide a somewhat higher rate of success (7 percent for covert referrals and 8 percent for overt referrals), but their highest rates of success occurred when Mormons invited non-Mormon friends and relatives into their homes to meet Mormon missionaries. In those instances, they enjoyed a success rate close to 50 percent. This suggests that the best strategy for conversion is not cold-calling but forming friendships with non-Mormons. Stark and Bainbridge noted that an article in the Mormon Church’s official magazine provided detailed instructions on how to recruit new members, and a recurring theme was the importance of building close personal ties with non-Mormons. It also explicitly instructed its readers that they should downplay or avoid discussing religion while forming these ties. Only later were they to bring up that they were Mormons

Another way of looking at these findings is that missionaries do not serve as the primary instrument of recruitment to the Mormon faith. Instead, recruitment is accomplished primarily by the rank and file of the church as they construct intimate interpersonal ties with non-Mormons and thus link them into a group network (Stark and Bainbridge 1980:1386-87).

At about the same time the Stark and Bainbridge study appeared, David Snow and his colleagues (Snow, Zurcher, and Ekland-Olson 1980) published an article that highlighted the same dynamic: Successful social movements, religious or otherwise, recruit primarily through social networks of friends and families. All of the groups they studied, except the Hare Krishna, recruited over 50 percent of their members through either kinship or friendship networks with several recruiting over 90 percent of their members through such networks. The Hare Krishnas were the lone exception because they demand exclusive participation from their members and require them to sever all extra-movement ties. Thus, they have virtually no social ties outside of the group through which they can recruit, which forces them to recruit from public places. That's why they are so small. The moral of the story? Successful religious groups must maintain open social networks in order to grow.

2. People of faith cluster just like everybody else: with similar others. The phenomenon of homophily, "birds of a feather, flock together," appears ubiquitous to human nature. We tend to befriend similar others, and this is even true between and within religious groups.

For example, in the 1960s Samuel Sampson (1968) spent a year in a Roman Catholic monastery observing the social interactions among a group of monks. During his stay a “crisis in the cloister” occurred that was in reaction to some of the changes introduced by the Second Vatican Council, known informally as Vatican II. These changes were an attempt to modernize the Roman Catholic Church and included changes, such as saying the Mass in the vernacular (rather than in Latin) and granting nuns the freedom not to wear habits (à la, Sally Field as "The Flying Nun").

This conflict ultimately resulted in the expulsion of four monks and the voluntary departure of several others. In the end, only four monks remained. While he was there, Sampson coded four types of relational data that he further subdivided into positive and negative ties. He had each monk rank his top three choices for each type of relation, which included esteem and disesteem, liking and disliking, positive and negative influence, and praise and blame. Some of these relations he coded at different points in time, which are captured by the three graphs below (Time 1, Time 2, and Time 3). What they illustrate is how, over time, the monks increasingly clustered into groups, based largely on their opinion regarding the changes from Vatican II.

Sampson's Monks: Time 1

Sampson's Monks: Time 2

Sampson's Monks: Time 3

Sampson's monks are not unusual. Subnetworks form in most large religious groups, and the same is true of religious scholars. They tend to cluster together into like-minded subgroups. One the face of it, this isn't surprising, but it can have profound effects on the conclusions they draw. As we will see below, when like-minded people come together to debate a topic or an issue, the group tends to gravitate to the most radical position held by a member in the group. Thus, we shouldn't be surprised of the conclusions of groups such as the Jesus Seminar, which fall outside the mainstream of even liberal orthodoxy.

3. People on the periphery of a religious group are more likely to leave than those who are located in the center. In the same study in which Rodney Stark and Bill Bainbridge (1980) studied Mormon recruitment, they also examined a "doomsday" group, which had formed primarily along kinship ties, and Stark and Bainbridge found that members who had direct kinship ties to the group's leaders were less likely to defect (i.e., leave the group) than were others. In particular,

Members who were direct kin of the leaders, only 14% quit. Of those who were related to kin of the leaders, but not directly to the leaders (e.g., in-laws), 25% defected. But of those who had no relatives in the group, two-thirds left prematurely. For those who had to abandon their families as well as their faith, defection was rare. But for those without familial ties to the group, defection was the rule! (p. 1383).

Stark and Bainbridge's findings have been replicated in other studies. For example, Pamela Popielarz and J. Miller McPherson (1995) studied voluntary organizations in general (not just religious groups), and found that they lose members located at the edge (i.e., the periphery) of their group faster than they do core members, primarily because members at the edge have fewer ties within the organization and more without than do core members. However, members on the periphery are also more likely to leave because they are more likely to have other groups competing for their time and money.

Michael Emerson and Christian Smith (2000) found that this phenomenon helped explain why a newly-formed Seattle church's attempt to establish an interracial church was unsuccessful. The church was founded by an African-American pastor, who prior to founding the church, visited white and black churches in the area where he told them of his vision of an interracial church and asked for volunteers to be charter members. After a year of preparation, the church held its first public service, and the congregation was almost evenly split between blacks and whites (p. 147). However, as the church grew, the congregation began losing its white members, and within three years of its founding, fewer than 10 whites remained within the congregation. The first few white members (who were charter members) who left said they left because they felt like outsiders, that their voices were not being heard. Regardless of whether this indeed the case, it had a spiraling effect on the church's composition. As Emerson and Smith put it,

there were fewer within-church social network ties to keep white there and recruit new white members. Whites leaving also meant an increasingly great number of social ties outside the church for the remaining whites, making the church less central for them, and making them feel increasingly like outsiders, and that there needs were not being met (p. 149).

4. Religious ties sometimes coerce people to attend church. Our social ties cannot only entice us to join a particular congregation, in certain social contexts they can also coerce us into attending worship services. Chris Ellison and Darren Sherkat (1995) documented this phenomenon in their study of the rural, southern Black church. Before getting to their study, however, a little background is in order.

According to W. E. B. DuBois, after the Civil War, the church became the central social institution in the African American community. It was its primary vehicle of communication, entertainment, and education. Black churches also functioned as mutual-aid societies that helped members survive financial crises, especially when loved ones died or became extremely ill. And for males it became a vehicle of power, upward mobility, and economic success, which is why black churches were often overflowing with aspiring ministers and preachers who literally waited in the wings for a chance to preach and hopefully gather a following for themselves.

The central role of the Black Church was especially pronounced in the rural south, which led some scholars to argue that the southern rural Black church is a semi-involuntary institution. By this they meant that in the rural south, church participation serves as a sign of social legitimacy. In order to be respectable, one needs to go to church, and those who don’t are sanctioned. This differs for Black churches in urban settings because urban settings offer more secular opportunities for achieving personal status and prestige. Thus, there is less pressure for people to attend church.

Ellison and Sherkat tested this thesis using data on church attendance of African-Americans. They expected to find that not only will the rural, southern black church will have the highest overall rate of attendance, but it will also have the highest percentage of people who attend at intermittent levels. Why? Because the highest percentage of people who don't want to attend but feel compelled to do so will be those living in the rural South. However, they won't attend weekly. Only often enough so that they're not sanctioned by the wider community.

Ellison and Sherkat found support for their hypothesis. Moreover, once they broke the analysis down by region and rural differences, they found that the frequency of contact with friends is a statistically significant predictor of church participation among rural southerners but not among non-southerners. This suggests that rural southerners do, in fact, participate in church activities for reasons other than religious ones while non-southerners do not. To be sure, this is not a direct test of the semi-involuntary thesis, but it's results are consistent with it. They suggest that in the rural south peoples’ social networks sometimes force them to attend church when they otherwise would not.

References

• Ellison, Christopher G., and Darren E. Sherkat. 1995. "The 'Semi-involuntary Institution' Revisited: Regional Variations in Church Participation Among Black Americans." Social Forces 73(4):1415-37.
• Emerson, Michael O., and Christian S. Smith. 2000. Divided by Faith: Evangelical Religion and the Problem of Race in America. Oxford: Oxford University Press.
• Lofland, John, and Rodney Stark. 1965. "Becoming a World-Saver: A Theory of Conversion to a Deviant Perspective." American Sociological Review 30:862-75.
• Popielarz, Pamela A., and J. Miller McPherson. 1995. "On the Edge or in Between: Niche Position, Niche Overlap, and the Duration of Voluntary Association Memberships." American Journal of Sociology 101(3):698-720.
• Sampson, Samuel F. 1968. "A Novitiate in a Period of Change: An Experimental and Case Study of Relationships." Unpublished Dissertation. Sociology Department: Cornell University.
• Snow, David A., Louis A. Zurcher, and Sheldon Ekland-Olson. 1980. "Social Networks and Social Movements: A Microstructural Approach to Differential Recruitment." American Sociological Review 45:787-801.
• Stark, Rodney. 1996. The Rise of Christianity: A Sociologist Reconsiders History. Princeton, NJ: Princeton University Press.
• Stark, Rodney, and William Sims Bainbridge. 1980. "Networks of Faith: Interpersonal Bonds and Recruitment to Cults and Sects." American Journal of Sociology 85(6):1376-95.