Donald Trump has unified different groups of white voters around his Presidency, the political scientist Ashley Jardina writes.

In her new book, “White Identity Politics,” the Duke political scientist Ashley Jardina examines the increasing relevance of white identity in America. Drawing on data from American National Election Studies surveys and her own research, Jardina finds that about thirty to forty per cent of white Americans say that white identity is important to them, and she adds an interesting twist—that this group only partly overlaps with the group of white Americans who hold racist views. According to Jardina’s analysis, about thirty-eight per cent of white people who highly value their white identity are at or below the mean level of racial resentment, while forty-four per cent of white people who say their racial identity is less important are at or above that level. “For those invested in racial equality, this outcome should be of little comfort,” Jardina writes, of white Americans asserting their identity, with or without explicit racial resentment. In the past, she notes, when white people have been asked to share resources and power, they have not responded “by leveling the field; instead, they have expanded the scope of who is considered white, allowing the racial hierarchy to remain more firmly in place.”

I recently spoke with Jardina by phone. In our conversation, which has been edited and condensed for clarity, we discussed which voters exhibit the strongest feelings of white identity, how Donald Trump has unified different groups of white voters around his Presidency, and why some Americans who score low on tests of racism still identify strongly as white.

How has white identity changed over the past several decades?

One thing that’s different is how salient and politically relevant it is. We don’t have good public-opinion data going back in time to indicate that levels of white identity in the population have changed, or that now more people are identifying with their racial group than in the past. But what’s certainly clear is the extent to which white identity, or racial identity for some whites in the United States, matters for how they view the political and social world.

Think about white identity as being episodic and contextual. It’s politically relevant when something happens in the environment that makes it relevant, or when élites try to activate it, but it’s not always a force in politics in the way that we’re observing it to be today. If we could go back to the nineteen-twenties, in the wake of massive immigration to the United States, or if we could go back to the civil-rights movement, there are periods when there was a challenge to the dominant status of whites. There’s a possibility that the United States was no longer going to be defined by whiteness. These are places in time in which we might have seen white identity matter just as we’re seeing it matter today.

How do you come to the conclusion that white identity tends to experience spikes when there is a change in the country, or when politicians wield ideas of whiteness or racism, without all the survey data we would like?

We do have some data from the past. I’m not the first political scientist to interrogate the idea of white identity. There were people—mainly people who were interested in racial prejudice and racial resentment, some of whom were actually my advisers—who looked to see if white identity mattered in the mid-nineteen-nineties and in the early two-thousands, and this is a period in time in which the country was really different. We hadn’t seen the election of the nation’s first black President. And so, when political scientists looked to see whether white identity mattered for political attitudes and behavior, they really found nothing. There wasn’t much to uncover.

I think we social scientists were looking for white identity in the wrong places. One of the distinctions I make in the book is between the ideas of identity and prejudice. This is the distinction between in-group attitudes and out-group attitudes. For good reason, a lot of work on whites’ racial attitudes in the U.S. has focussed on racial prejudice. We want to understand where discrimination and bigotry and all these things come from, and what the consequences are.

But there was a tendency to think about white identity as just another manifestation of prejudice. So what people did is they went out and they said, “O.K., white identity ought to project things like opposition to welfare, opposition to things that benefit blacks.” Maybe, if we’d gone back earlier in time, and looked to see if white identity affected opinions on immigration or support for policies that benefit whites, we would have seen different effects and different results. We probably would have seen that white identity did matter politically to some extent, even well before we saw the big political things in the environment that I argue activate white identity.

How much of a connection is there between strongly identifying with whiteness and racist attitudes?

Not as strong as you might think. It’s certainly the case that there are some people who identify as white and who are also racist. But it’s not a one-to-one relationship. The connection is fairly weak, and that’s for two reasons. One is that there are a lot of white people who are more racially prejudiced who do not identify as being white, and the converse is true. Then there are a number of white people who feel strongly attached to their group but who aren’t particularly prejudiced.

How do you understand the first of those? Because that’s not intuitive.

The first one? I actually think the first one is the more intuitive.

Tell me why.

One reason that we haven’t talked a lot about whiteness in the past is because whites don’t have to confront their racial identity the way that people of color in the United States traditionally have. So we…