By Paul A. Djupe, Denison University

The zombie thesis is back – non-church attending Trumpers are to blame for all the nation’s dysfunctional politics. I can’t believe people are still claiming this. How many times do I need to write this post and present evidence? In a recent article in The Atlantic, Daniel K. Williams decided to take another run at it. But it was way back in 2022 when he first wrote a piece for the Patheos blog suggesting that low attending White Protestants in the South are the ones who actually take the repugnant stands on public policy, not the attenders. Here’s the argument: “[W]hen white southerners quit attending church, they don’t lose the church’s political conservatism, moralism, or individualism. Instead, they become hyper-individualistic, strongly devoted to law and order, and overwhelmingly politically conservative (if they vote at all). But they’re also cynical and distrustful of others.” It sounds as if it’s based on evidence.

Our thorough investigation in “The Supposed Threat of Low Attending White Protestants” came up empty. No surprise: attenders are more conservative and non/low attenders take more liberal positions. As we conclude: “Instead, low attenders in 2020 appear to be more supportive of gun control, abortion rights, removing Trump, and other liberal positions. They are less religious and appear to be less moralistic. There certainly is not evidence to suggest that they have lost any willingness to care for other people. Quite the opposite. In fact, they look close to the religious nones, for whom one reason why they tend to leave congregations is because of political disagreement. There’s certainly a lot of grist for that mill here.”

Just a few weeks ago, David French let loose the zombie thesis and my response with data from 2020-2023 was, “I’ve addressed this several times before in several different ways and I’m surely missing some posts, but let me say it again: church attendance is linked to Trump support.”

Well the zombie argument didn’t stay down; it’s back and, apparently, The Atlantic doesn’t fact check anything. So, here’s some more evidence. I’m using a sample of 2,323 American adults with weights applied to return a sample that looks like Census figures for the adult population.

First, are there Christian nationalists (CN) who don’t attend? Not really. Looking at the CN scale (using the Whitehead and Perry measures) above .75 (the top quarter) and we find about 25 percent who attend seldom or never. There are zero never attenders in the second to top category and a few more than average in the very top. But it’s consistent that there are few never attending people with high CN scores.

But maybe those low attending Christian nationalists happen to be the ones more in love with Trump. (Even if they are the “always-Trumpers,” there simply aren’t enough of them to make a difference, but let’s check anyway.) The following comes from a simple model that interacts the CN scale with an attendance measure split into two – low attenders (a few times a year or less) versus high attenders (few times a month or more often) and controls for a few demographics (age, race, gender, and education). It’s clear from the evidence that Christian nationalism is linked to greater warmth toward Trump and attendance doesn’t make a bit of difference. CN is associated with a climb of just over 60 points (out of 100) and attendance changes that result by perhaps a point or two – an insignificant difference.

Williams suggests that attendance is essential for learning how to live with difference. “But without a church community, in many cases, the nation’s political system becomes their church—and the results are polarizing. They bring whatever moral and social values they acquired from their church experience and then apply those values in the political sphere with an evangelical zeal.”

Yeah, so, about that. Here’s a look at affective partisan polarization measured (in a standard way) as the absolute value of the difference between a feeling thermometer score for Democrats minus one for Republicans (0-100 range). Again, this is from a sparse model that controls for some demographics. The figure below shows how Christian nationalism is linked to polarization, allowing that relationship to differ between low and high attenders. It’s plain to see that there’s no difference. Maybe high attending anti-Christian nationalists might have more polarization than low attenders, but there is almost zero difference among those with high Christian nationalism scores.

The crux of Williams’ argument is that church attenders will temper their attitudes because of the important relationships they have with diverse sets of people. Outside of that context and they will treat their attitudes as God – politics as the new church – and damn the consequences for others. So that’s why it’s interesting that church attending Christian nationalists are more likely to agree with two “need for chaos” statements: “When I think about our political and social institutions, I cannot help thinking ‘just let them all burn.’” and “I fantasize about a natural disaster wiping out most of humanity so a small group of people can start all over.” Low attenders show only half the increase in support for chaos as their Christian nationalism climbs (~.06 points vs .11 points on a 0-1 scale). Put another way, church attendance boosts the relationship between Christian nationalism and chaos. Almost needless to say that it is the antithesis of concern for the community when you are ready to burn it down.

Yet again, the evidence doesn’t support the zombie thesis that the church attenders are ones who want to maintain society and social relations. It is not at all surprising these days that what matters is the ideas, like Christian nationalism, which are reinforced by sources that transcend congregations. Just the other day, Mike Huckabee argued on a sizable Christian network (TBN) that if Trump loses in 2024, elections will be decided by “bullets rather than ballots.” And he’s not alone in that provocation (see Mike Flynn). As we show in several datasets, those sorts of views – that “we” need to turn to violence to regain control – are linked to Christian nationalist beliefs and church attendance. And as I argued just a few weeks ago, “We Should Probably Stop Thinking Religion is a Solution to MAGA.” The bigger concern that seems to flow from these investigations is the decline of religious influence that could mitigate these anti-social and anti-democratic orientations. For one, it’s not clear that many clergy are trying and, two, I doubt they would have much influence if they did. We should study that directly instead of drawing out arguments based on dubious assumptions about a pro-social church and without systematic evidence.

Professor Paul A. Djupe directs the Data for Political Research program at Denison University, is an affiliated scholar with PRRI, the series editor of Religious Engagement in Democratic Politics (Temple), and co-creator of religioninpublic.blog. Further information about his work can be found on his website and on Twitter and Threads.