By Amy Erica Smith

Several weeks ago, I posted about the huge variation in attitudes toward clergy’s role in politics around the world. What produces this variation?

One obvious possibility is that people base their attitudes about what clergy should do on what they see religious leaders they trust actually doing. If clergy in Country X stay out of politics, people will tend to believe that behavior is proper, this line of thinking goes; but if clergy in Country Y actively get involved in politics, citizens will instead think it’s normal and good for clergy to behave that way. If this is the case, norms about clergy and politics should vary not only across the world, but also over time within each country, depending on how politically active clergy are.

However, as I will argue in this post based on a case study of Brazil, the story is not quite so simple. My current book project shows that evangelical clergy in Brazil often get involved in election campaigns. What do citizens think about their politicking? Results from three studies suggest that evangelicals do become more accepting of clergy campaigning, but the effects are fairly small. Even at the height of election campaigns, most Brazilian evangelicals are still suspicious of politics in church. This limits what clergy are able to say without alienating their flocks.

In March and April of 2014—several months prior to the October 2014 elections—the LAPOP AmericasBarometer asked Brazilians nationwide whether it was appropriate for “a priest or pastor to support or campaign for a certain candidate at election time,” using a scale running from 1 (“strongly disapprove”) to 10 (“strongly approve”). In this survey, a large share of citizens strongly opposed clergy campaigning: 49.8% of respondents gave clergy campaigning the very lowest approval rating, a 1. Moreover, 85.4% gave the scenario a rating of 5 or below. There were no differences between evangelicals and members of other religious groups in their responses.

A few months later, in June, the 2014 Brazilian Electoral Panel Study (BEPS) again asked citizens, “do you approve or disapprove of a priest or pastor supporting or campaigning” for a candidate? This survey was conducted about a month and a half prior to the beginning of the campaign, but some politicians and clergy had already begun informal networking.

The figure below compares BEPS responses to those from the AmericasBarometer respondents. (Those giving clergy politicking a score of 6 through 10 are coded as “approving.”) Responses of Catholics were nearly identical in the two surveys. However, evangelical approval rose ten percentage points. Perhaps some evangelicals were adjusting their norms to match their clergy behavior as candidates and clergy began to form electoral alliances. Still, it is also important to note that more than three quarters of evangelicals still disapproved of clergy campaigning in June.

Do church attendees become more permissive of religion mixing with politics when they observe their clergy take political stances? In my book project, I find that politically engaged clergy mobilize their congregants. In the process, they might change their followers’ attitudes about the relationship between religion and politics. For instance, one focus group respondent in a Baptist church explained that, “It used to be that I didn’t like politics, but the pastor helped us understand that politics is part of our daily life.” Another person from the same congregation opined that, “a lot of people think politics comes from the Devil….but our pastor shows that politics comes from God.”

In September of 2014, just a few weeks before the first-round election, Taylor Boas and I conducted an online, national-level study recruiting participants through Facebook advertisements. We asked respondents the extent to which they agreed, on a seven-point scale, that it is proper “for people to talk about politics and elections in church.” Though this question did not ask directly about clergy, most Brazilians were still suspicious of bringing politics within church walls. Only 20% of Catholics and 22% of evangelicals agreed with this statement (giving it a 5, 6, or 7), as did just 13% of religious non-identifiers (“nones”) and 15% of those with other religious affiliations.

Immediately prior to the questions on mixing religion and politics was a series of experimental questions examining attitudes toward the evangelical presidential candidate Marina Silva. Marina Silva is an interesting candidate to study because of her counter-stereotypical profile. On the one hand, as a born-again member of the Assemblies of God, she was supported by many Pentecostal and evangelical pastors and laypeople in 2014. On the other hand, most non-evangelicals did not see her as the “religious” candidate, since she rose to prominence as an environmentalist on the Brazilian left. In one experimental condition, respondents read simply that “Much has been discussed about the candidate Marina Silva. To what extent do you agree with her policies?” In two other experimental conditions, Marina Silva was alternately described as an “evangelical candidate” and a candidate “who has been endorsed by many pastors.”

As the figure below shows, the way Marina Silva was described affected evangelicals’ attitudes about mixing religion and politics. When only her name was mentioned, 30% of evangelicals agreed with talking about politics in church, as opposed to just 13% of Catholics and “nones” and 12% of “others.” Her religious identity and endorsements were probably salient for evangelicals, but not non-evangelicals. Non-evangelicals became slightly more supportive of bringing politics into church when reminded of Marina’s religious credentials, though differences are not statistically significant. By contrast, evangelical norms were unaffected by the “pastor endorsement” treatment, and evangelicals were much less supportive of political talk in church when Marina was described simply as an “evangelical.” It appears that evangelicals were unwilling to support a co-religionist based only on shared identity. Finally, the right-hand side of the figure shows that how evangelicals felt about Marina strongly influenced the extent to which they said they supported political talk in church.

Thus, evidence from three surveys conducted during Brazil’s 2014 election indicates that exposure to religious politicking can affect citizens’ support for mixing religion and politics. However, there are limits. Even at the height of the 2014 campaign, only 30% of evangelicals thought it was okay to talk about politics in church. As a result, I argue in my book project that even the most politically engaged clergy had to exercise caution in choosing how and when to make their partisan stances known.

A couple of months ago, my colleague Paul Djupe observed that U.S. citizens are ambivalent about the roles clergy play in politics and public life. Though I teased Paul a few weeks ago for not specifying that he was referring to the U.S., it turns out that that Brazilian citizens are similarly ambivalent. Stay tuned for more thoughts on the sources of this ambivalence, and what it means for democracy.

Amy Erica Smith is Visiting Fellow at the Kellogg Institute for International Studies, as well as Assistant Professor of Political Science at Iowa State University. Find more information about her here.