Attendees at Donald Trump's July rally in Greenville, North CarolinaKevin Lamarque / Reuters

Over the past month, President Donald Trump has embarked on a concerted push to place race at the heart of the 2020 election, first by saying that a group of four progressive congresswomen of color should “go back [to] the totally broken and crime infested places from which they came” and then with a sustained campaign against Representative Elijah Cummings, an African American Democrat. Trump has been using race as a political wedge for nearly a decade, dating back to his campaign against an Islamic community center in lower Manhattan; these moves are, as I have argued, a more explicit version of that long-standing strategy.

Commentators seeking to contextualize this political strategy have sometimes labeled it “white identity politics,” a phrase that mirrors the label (often pejorative) given to politicians who have emphasized race and gender issues. Calling it “white identity politics” also emphasizes the way in which whiteness, though commonly treated as a default or an absence of race, is very much an identity of its own.

But simply labeling Trump’s strategy as white identity politics doesn’t differentiate it from other race-based approaches to politics, much less explain why it works, what its limitations might be, or to whom it appeals.

These are all questions that the political scientist Ashley Jardina explores in a book published earlier this year, aptly titled White Identity Politics. (She is a professor at Duke, where I sometimes teach journalism.) Jardina’s research finds that it isn’t just pundits and political scientists who have zeroed in on whiteness as an affirmative political identity: Many white Americans are identifying themselves with their racial group as well. That’s a departure from recent years, though it has likely happened at other times in American history as well, and it has important political ramifications. White identity was an important predictor of voting for Trump.

But Jardina finds some surprising things about white identity politics. For one thing, there seems to be a real psychological divide between whites who hold animus to other racial groups and those who show little sign of typical racial prejudice but are concerned about protecting their own group—though in practice, they often end up supporting politicians and policies that do hurt minority groups, as with Trump. Meanwhile, despite common oversimplifications about who these voters are, Jardina finds little evidence to suggest they are largely members of an economically fragile working class.

Trump’s political success has been built in part on his ability to appeal to both whites who are prejudiced and those who are not, using the same policy ideas. But moves like his attacks on the “squad” or Cummings test the outer limits of this two-pronged strategy, threatening to turn off whites who don’t think of themselves as prejudiced. There’s been a 10-percentage-point drop in white identifiers—whites who indicated their racial identity is really important to them—since the 2016 election. Trump’s recent moves toward cutting budgets for entitlement programs popular among white identifiers also risk alienating the voters who helped put him in office. Jardina walked me through her research, and discussed how her findings might apply to the president’s recent racist outbursts. This interview has been edited and condensed for clarity and concision.


David A. Graham: I see a growing number of people using the phrase white identity politics, sometimes to mean different things. What do you mean when you refer to white identity politics?

Ashley Jardina: The term refers to the psychological attachment to their racial group that many whites in the United States possess. Whites with a racial identity feel a sense of racial solidarity with their group and see whites as having similar interests. White identity politics refers to the way in which this sense of racial solidarity influences whites’ view of the political world. Generally what that looks like is whites with a sense of racial identity prefer political candidates and policies that protect their group’s interests. In the U.S., protecting these interests often means attempting to preserve privileges and advantages that whites, on average, have relative to other racial and ethnic groups.

Graham: Is this a recent phenomenon?

Jardina: I think of it as an episodic phenomenon. Until recently, it might seem like we haven’t seen white identity influencing whites’ political preferences in any serious way. But if we had historical political polling data, we might. For example, it isn’t hard to imagine that the white backlash to the civil-rights movement was not just about racial animus, but also about whites feeling like their political power was going to shift markedly as African Americans achieved more political rights and opportunities. Another moment white identity politics was likely at play was in the U.S. in the 1920s, just before the passage of the 1924 Immigration Act. Coming on the heels of a large influx of immigrants from eastern and southern Europe, many Americans were worried about the changing racial composition of the United States. If you read the Congressional Record at the time, politicians were having very similar conversations to the ones we’re having today. They expressed opinions about what the racial composition of the country should look like and considered from where we should limit immigration. Most of these conversations were about maintaining the image of the U.S. as a “white” nation. Members of Congress even talked about preserving the “Nordic stock” of the nation.

My argument is that the reason white identity politics matters today in a way it didn’t matter in the 1980s or 1990s or even 2000s is because of a confluence of things happening in the political and social environment. The country is changing demographically because of immigration that took place in the 1990s and early 2000s, and because of differences in birth rates across racial and ethnic groups. The U.S. is becoming far more racially and ethnically diverse at a rapid pace, and perhaps most symbolic of these changes was the election of Barack Obama, the nation’s first black president. These are all factors in the political environment that some whites, I argue, are interpreting as a threat to their group’s power and their status.

Graham: You draw some conclusions about what the typical white identifier looks like. Can you sketch that?

Jardina: It’s not who you might expect. I think the term white identity politics often conjures up this image of a working-class white man who maybe lost his manufacturing job and feels he’s being left behind. There’s not a lot of evidence that such a person is the typical white identifier. People high on white identity tend to be older and without college degrees. Women are actually slightly more likely to identify as white than men. And white identifiers are not exclusively found among those in the working class. White identifiers have similar incomes, are no less likely to be unemployed, and are just as likely to own their own home as whites who do not have a strong sense of racial identity.

Graham: There’s an idea in circulation that white identity politics is driven less by poverty than by a sense of fragility—a sense that one’s economic status is endangered. Is there evidence for that?

Jardina: I find very little evidence that a sense of economic vulnerability leads whites to adopt a racial identity. I asked people whether they think their families are better or worse off than they were a few years ago and find that their answers to this question are not at all predictive of whether they identify as white. I also looked to see whether people who were worried about losing their job might be more likely to identify as white, and I don’t find a relationship there, either. There really isn’t a strong relationship between subjective or objective economic circumstances and the propensity to adopt this identity.

In addition to whether or not someone went to college, I find that higher levels of white identity tend to be associated with certain personality traits. Whites who are more authoritarian or who score higher on this metric we call social-dominance orientation—a psychological predisposition that leads individuals to prefer hierarchy, or to believe that society should be organized hierarchically—these are the white people more inclined to identify with their racial group.

Graham: You say about 38 percent of whites score high on racial identity but low on racial resentment. What does that look like in practice?

Jardina: This is a really important distinction for a number of reasons. When social scientists think about how people act as groups in society, we make this distinction between the negative attitudes that people hold toward out-group members and the attitudes people have toward members of their in-groups. Traditionally, we have often focused on the negative out-group attitudes that white people have toward people of color. We call these attitudes racial prejudice.

White identity is an in-group attitude. There isn’t necessarily a strong relationship between feeling favorable toward one’s own racial group and strongly disliking members of other racial groups. Many white identifiers aren’t especially racially prejudiced in the classical sense. Nevertheless, they do want to do things that benefit their group, and while they’re not necessarily motivated to do so at the expense of other racial or ethnic groups, it turns out that the policies and candidates white identifiers support in the name of their group’s interests can hurt other groups. In a world in which whites have a disproportionate share of power and resources, having a preference for protecting your group inherently preserves a system of racism and racial inequality.

This distinction between whites who have a sense of racial identity and whites who are racially prejudiced matters a lot for today’s politics. Many of Trump’s racist or racially charged remarks likely appeal to two distinct sets of white voters. Take a recent example, when Trump told several members of Congress to “go back to your countries.” We might think that this remark, which is racist because it suggests that these women of color are not truly American, would only seem acceptable to the most racist of whites. But this sentiment might also appeal to white identifiers who look around a more racially and ethnically diverse nation and worry that they are no longer seen as prototypical members of the United States.

Graham: I suspect a lot of people will view this skeptically: Is there really a difference here? The example that persuaded me in your book concerned “racialized” programs like welfare and Medicaid.

Jardina: I find that white identity isn’t at all associated with views on a lot of policies that we know traditionally are overwhelmingly associated with racial prejudice. People with high levels of racial animus are far less supportive of welfare and Medicaid, social-welfare policies that have been associated with erroneous and disparaging stereotypes about African Americans. White identity is unrelated to attitudes on these policies. Whites with high levels of identity are not any more supportive of reductions to these policies than whites with no sense of racial identity. There are some social-welfare policies that I and other scholars have argued are traditionally associated with whiteness or with disproportionately benefiting white people: Social Security, Medicare. These are especially popular among white identifiers.

The distinction between white identity and white racial prejudice also matters when we think about political mobilization. We know that both white identity and racial prejudice were powerful predictors of Trump support. Whites high on racial prejudice and whites high on white identity were both likely to vote for Trump. Trump was an unconventional Republican candidate, in that he parted ways with the traditional GOP party platform: He promised to protect Social Security and Medicare, a campaign promise that appealed distinctly to white identifiers.

But Trump was very strategic and very much set out to attract both the racially prejudiced whites and whites who were high on a sense of identity. For instance, Trump has basically hammered over and over again the issue of immigration—an issue very important to whites who feel a sense of prejudice toward Latinos and to whites who are worried about the loss of their race’s numerical majority in the country. All Trump has to do is say, “I’m going to restrict immigration; I’m going to build a border wall.” This message appeals to both types of white people but for different reasons.

Graham: When Trump adopts these racist attacks, does he risk turning off white identifiers who aren’t high on racial animus?

Jardina: It’s a little complicated. What we do know is that when Trump associates himself with extremist groups or white supremacists, many white identifiers are turned off. It’s a little more complicated when you’re talking about remarks that apparently to some white Americans are more ambiguous in terms of whether they’re seen as racist, like “Send her back.”

There may have been a period in American politics where, when politicians made racist or racialized comments, the public would recoil. One of the things I’ve found in my research is that accusations of racism have become politically ineffective. People often see them as “crying wolf.” Think about the “Go back to where you came from” controversy. In reaction to Trump’s racist remarks, Democrats were outraged and called Trump racist. Republicans simply responded by saying, You just want to make everything about race. You just want to play the race card.

Trump does this all the time. He makes racist remarks and then denies that his remarks were racist. It allows Republicans to completely spin the narrative. After the “Send her back” controversy, a lot of the conservative talking points drew attention away from the “Send her back” language. The narrative became that “these four members of Congress were complaining about America so Trump told them to leave.”

Graham: It’s reframing them as classic “Love it or leave it” rhetoric.

Jardina: Right. But there is some evidence that despite his efforts, Trump has turned away some of his initial supporters. I’ve found that after the 2016 election, there was a 10-percentage-point drop in the number of white people who identify as white. I’m working on a study now with some colleagues that tries to understand why we’ve seen this drop. What we’ve found thus far is that the drop has largely been motivated by dislike or disgust toward Donald Trump. There’s also some evidence that Trump is partly responsible for a reduction in levels of racial prejudice among some whites. Since Trump’s election, white Republicans have not become less racially prejudiced, but white Democrats have.

Both these changes are really interesting and surprising, because social scientists often think of these racial identities and racial attitudes as really stable dispositions—ones that people adopt early in their lives. They don’t tend to shift, even when things are going on in the political environment. The fact that Trump is, in part, causing these shifts is really surprising.

Graham: At the same time, we see Trump talking about cutting budgets in the second term, and his most recent budget cuts entitlements. Is that likely to hurt him with white identifiers?

Jardina: Yes, and it’s an opportunity for Democrats to win over these white identifiers, since Democratic candidates tend to be more supportive of protecting these programs than Republicans. But Trump can always play the immigration card. He can talk about cutting these programs and then distract by turning the public’s attention back to immigration and talking about an immigration crisis.

He’s done this before. Think about Trump’s strategy leading up to the midterm elections. Suddenly, we have a caravan of migrants coming to storm the border. The midterm elections happen, and suddenly, no caravan. It’s a very effective strategy, because many people are concerned about immigration, and it is an issue that is especially important to white Republicans, both those who are high on white prejudice and high on white identity.

Graham: Yeah, but the midterm elections were awful for Republicans. Are you saying the results might have been worse if not for that rhetoric?

Jardina: One thing we need to think about is the difference between changing voters’ attitudes and mobilizing voters. Is Trump’s racist rhetoric actually mobilizing white liberals and mobilizing people of color in response? It’s clear his rhetoric isn’t doing much to turn away Republicans. After his “Send her back” remarks, Trump’s approval ratings actually went up with Republicans. So when we think about the results of the midterms, or look toward 2020, the big question, I think, isn’t whether Trump’s racist messaging is going to alienate his supporters. It’s whether whites and people of color are appalled enough by Trump’s racism to show up to the polls and vote.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.