A paper by Notre Dame’s David Campbell (PDF link) finds that evangelicals are more likely to vote for a Republican when they live in a community with more people who do not identify with any religion. Building on a tradition of research in race relations which tests whether integrated communities foster greater social acceptance or stronger separation between groups. Studies in Southern communities in the 1960s had found that white voters were more likely to vote for racially conservative candidates as the fraction of African Americans in the community increased. Whether that trend generalizes beyond the South, and whether other factors could explain it, remains subject to debate.
Pointing out that American evangelicalism “is defined by conflict and threat,” he tested whether geographic proximity might enhance evangelical’s group identification and sense of conflict, and whether secularists (voters who don’t identify with any religion) might feel the same way. The figure reproduced here shows that evangelicals are indeed more likely to vote for a Republican when they live near more secularists, while secularists experience no such effect.
The way the study identifies “secularists” complicates the interpretation of the results. Differences between communities may make it easier for people with loose religious affiliations (and who would be less likely to vote Republican) to avoid attending church. In other communities, social pressure may keep those secular leaners in church, keeping them from being identified as secularists in this study. That pattern would account for some if not all of the trend they describe. The only way to resolve this issue would be if there were a clear way to identify such secularists and if standard demographic questions included a question about those topics.
Assuming Dr. Campbell’s interpretation of the data is correct, it also helps guide us in thinking about how to approach arguments about culture wars. If contact with “secularists” really does rapidly drive evangelicals deeper into the arms of conservative policy and politics, it is a powerful argument for putting evangelical (or at least religious) biologists out front in discussing evolution.
Assuming that some of the effect is real, and not a statistical artifact, it has interesting implications. At the American Prospect, Paul Waldman discusses some of the implications for the recent book binge by “new Atheists.” While these results suggest that secularists did not respond like a cohesive group or tribe in 1996 or 2000, Waldman wonders if those books reveal a new tribe of secularists forming:
the question now is whether non-believers will, in large numbers, begin to define themselves as a tribe of their own. In order to do so, they’ll have to feel at least some measure of antagonism toward those on the outside. That’s what makes a tribe a tribe, after all. (What would Red Sox fandom be without the Yankees, or punk rock without the conformist corporate tools?) But one key question for secular people is who, exactly, the Other is. Is it anyone who is religious? Those who want to convert you to their beliefs? Those who want their beliefs to be enshrined in government policy?
Dawkins, Harris and PZ Myers all seem to favor the first option. Groups like Americans United for Separation for Church and State, the Interfaith Alliance and the ACLU tend to focus on the latter groups, to substantial political effect.