The ‘Trump Effect’ and Evangelicals

Secret support for Trump and the crisis of whiteness.

Donald Trump campaign rally in Mobile, Alabama. Photo by Mark Wallheiser/Getty Images

By Eddie S. Glaude Jr.

I find myself in many interesting environments. I’m the president of the American Academy of Religion, I appear on Morning Joe, I write for Time magazine, I teach at Princeton. It can be confusing at times. But I hope these various contexts mean that I have something to say to this audience of journalists and academics, and perhaps it will be something that will spark debate.

It is certainly the case that white evangelicals, and, I suspect, some evangelicals of color, played a central role in the election of Donald Trump. What do we know? We know from the exit poll data that about 81 percent of white evangelicals voted for Trump, 60 percent of white Catholics voted for him, and even the majority of Mormons—61 percent—voted for him. This came as a surprise to many people in the media, myself included. Throughout the primary and general election there seemed to be a consensus that the so-called value voter found the character question surrounding Donald Trump insurmountable. His personal flaws definitely disqualified him from consideration, at least for some. This was certainly said to be true during the controversy with the Khans, the Gold Star family, and after the release of the audio and video footage with Trump’s horrific comments about his behavior with regard to women. We see the early news coverage reflecting this assumption.

Now, postelection, we know otherwise. We might even call this the “Trump effect,” which is something akin to the Bradley effect but with a slight variation. Of course, the Bradley effect involved discrepancies between voter opinion polls and election outcomes when you have a white candidate and a nonwhite candidate running against each other.1 In that situation, white voters dissemble when asked about who they support, for fear that, in stating their true preferences, they might be seen as holding racist views or having racist motivations when voting for the white candidate. In the sociological literature and the political science literature, this has been called the phenomenon of “social desirability bias.” The worst thing you can be called in the United States is a racist. Remember when George W. Bush said that the lowest moment in his administration was when Kanye West said, “George W. Bush doesn’t like black people, he hates black people.” (That’s the lowest moment for him? Really?)

In this case, we didn’t have a person of color running for office, but instead we had a white candidate running for president who embraced views that were widely considered racist—from building a wall, to his comments about the Indiana judge, to the appointment of Steve Bannon, and so on. So we had a number of white evangelicals behaving like Rose Aller said she had in a November 15 Washington Post article: they kept their support of Donald Trump secret.2 I remember sitting around the table on Morning Joe where folks said, “Is the silent Trump voter out there?” And they just kept saying, “No, no, no, no. They’re not.” But by not expressing their commitments early on, by not revealing what they actually felt, this obviously complicates how we might have read the motivation of these voters for supporting Trump. It also impedes our attempt to understand the contradictions of commitments that we all hold.

This is an edited version of a talk delivered during the “Religious Literacy and Journalism Symposium” held December 8–9, 2016, at HDS. The symposium was sponsored by the Religious Literacy Project, in collaboration with Boston University.

This was the first election in which white Christians were clearly a demographic minority, according to PRRI data—43 percent today, down from 54 percent in 2008.

To my mind, what happened in the 2016 election is not simply racial, although it is definitely that. Here, I agree with Robert P. Jones that the evangelical support of Trump reflects a convergence of economic insecurity, anxiety over demographic shifts, and in some ways a moral panic over the changing nature of the culture of the United States.3 I think a lot of this is rooted in the crisis of whiteness. This was the first election in which white Christians were clearly a demographic minority, according to the Public Religion Research Institute (PRRI) data—43 percent today, down from 54 percent in 2008. In 2008, only 40 percent of the country supported same-sex marriage. Today, it’s legal in all fifty states, and six out of ten Americans support it—a cultural shift. I think the writing is on the wall. White evangelical Protestants made up 22 percent of the population in 1988 and still commanded 21 percent of the population in 2008, but today only 17 percent of Americans claim the moniker. I think the data is showing—and I’m just echoing the PRRI data—that young adults ages 18 to 29 are less than half as likely to be white Christians as seniors ages 65 and older. Today, only three in ten young white adults are Christian.

Though we can challenge some of this data, it shows that this election may be the white Christian America’s last gasp. In my other writing, I have said that the election of Donald Trump reflects white America’s last gasp. James Baldwin wrote in his 1961 essay “The Dangerous Road before Martin Luther King”: “White America is dead; the question is how long and how expensive the funeral will be.”

The second brief point I want to make involves our use of language. I noticed that in the early reporting on the election, the term “evangelical” was used to represent an undifferentiated grouping of people who supposedly hold similar religious and political commitments. But by the end of the 2016 election cycle, the adjective “white” was added. So early on it’s just “evangelical,” and in postelection articles it’s “white evangelical.” I think this was a significant development, the result of a concerted effort among some evangelicals to force the media to treat them with a bit more nuance. Think about Jim Wallis and his op-ed in USA Today, in which he tried to insist that there were “tens of millions of Americans who fit the theological definition of evangelical, but who do not support such a narrow definition of ‘moral issues’ and clearly do not support Trump or his bigotry.”4

I find it interesting that the problem has been repeated in the very literature that props up our discussion today. Might we think of the word “evangelical” and its function as a variant of a dog whistle? If the “Trump effect” is a variant of the “Bradley effect,” our use of “evangelical” can be read as a variant of a dog whistle. It isn’t a racist dog whistle, certainly, but it calls out a particular grouping of white people for special attention and orients the listener accordingly. When we use “evangelical” in our writing, it orients the reader or the listener in a particular sort of way, much like the expressions “working class,” and “middle America” do—as if black people and brown people don’t live in “middle America,” as if black people and brown people aren’t “working class.” What is the adjective “white” doing when it describes the noun “evangelical”? I’d like to explore this more in thoughtful conversation.


  1. This phenomenon was named after Tom Bradley, the Los Angeles mayor who lost the 1982 California governor’s race, though he was ahead in the polls going into the election.
  2. Julie Zauzmer, “Hopeful and Relieved, Conservative White Evangelicals See Trump’s Win as Their Own,” The Washington Post, November 15, 2016.
  3. Jones is the author of The End of White Christian America (Simon & Schuster, 2016).
  4. Jim Wallis, “Evangelicals Aren’t Who You Think,” USA Today, October 23, 2016.

Eddie S. Glaude Jr. is the William S. Tod Professor of Religion and African American Studies at Princeton University. This is an edited version of a presentation he delivered in the “Donald Trump and Evangelicals” panel during the “Religious Literacy and Journalism Symposium” held December 8–9, 2016, at HDS.

Please follow our Commentary Guidelines when engaging in discussion on this site.