James Alicie, left, and Richard M. Birchfield, of Delaware, Ohio, at a Trump rally earlier this month. (Jeremy Pelzer/cleveland.com)
Kwame Anthony Appiah, a professor of philosophy and law at New York University, is the author of “The Lies That Bind: Rethinking Identity.”

You remember the photo, taken in early August, of two men at an Ohio Trump rally whose matching T-shirts read, “I’d rather be a Russian than a Democrat.” (Now you can buy them online for $14.) It was a gibe that spoke to our moment. The Republican brand — as with presidential nominees John McCain and Mitt Romney — used to be pointedly anti-Russian; Romney called Moscow our chief global enemy. In the Trump era, though, you can be a Republican Russophile for whom Vladi­mir Putin is a defender of conservative values. American politics, it has become plain, is driven less by ideological commitments than by partisan identities — less by what we think than by what we are. Identity precedes ideology.

“The Democratic Party today is divided over whether it wants to focus on the economy or identity,” the veteran strategist and pollster Stanley B. Greenberg, a man of the economy-first school, has said. But once you come to grips with the potency of partisan-identity politics, the binary falls away. So does the assumption that the great majority of Republicans who support Trump are drawn to his noxious views. (That’s the good news in the bad news.) Among candidates who eventually won the Republican primaries, after all, his percentage of the vote was the lowest in nearly half a century. Identity groups come to rally behind their leaders, and partisan identification wouldn’t be so stable if it didn’t allow for a great deal of ideological flexibility. That’s why rank-and-file Republicans could go from “We need to stand up to Putin!” to “Why wouldn’t we want to get along with Putin?” in the time it takes to say: Rubio’s out, Trump’s in.

What’s true of partisan allegiance is true of ideological allegiance. In research published earlier this year, political scientist Lilliana Mason conducted a national survey that determined where people stood on various hot-button issues: same-sex marriage, abortion, gun control, immigration, the Affordable Care Act, the deficit. Then they were asked how they felt about spending time with liberals or conservatives. About becoming friends with one. About marrying one.

The best predictor of ideological animus, the study found, wasn’t a respondent’s opinions or even how strongly she held them, but what label she embraced, conservative or liberal. Mason calls this “identity-based ideology,” as opposed to “issue-based ideology.” Other researchers in political psychology prefer to speak of “affective polarization.” Either formulation is a polite way of saying that political cleavages are not so much “I disagree with your views” as “I hate your stupid face.” You can be an ideologue without ideology.

“Implicit bias,” and the special tests designed to measure it, come up often in the wake of police shootings and #BlackLivesMatter. They show in-group preferences among whites and among blacks. But experiments suggest that partisan in-group preferences are far more powerful. We’re more polarized by party than by race. Indeed, while few Americans are still bothered by interracial marriage, recent surveys find that between 30 and 60 percent of people who identify as Democrats or Republicans want their kids to marry in the party. It’s hard to avoid the conclusion that identity politics is the only kind of politics we’ve got.

That’s a feature of people, not simply politics. Long before anyone instructs children to group people into categories, research tells us, they’re programmed to do it anyway, and one of our basic ways of making sense of the world is to form generalizations of the sort linguists call “generics” — such as “bears eat people” or “tick bites give you Lyme disease.” Those generalizations count as true, but it’s not easy to say why. Hardly any bears have eaten people , and less than 2 percent of tick bites transmit the Lyme spirochete. But, as the philosopher Sarah-Jane Leslie has argued, we’re more likely to accept a generic if it involves a reason for concern, such as getting eaten or getting sick.

What’s more, generics encourage us to think of the class in question as a kind, a group with a shared essence. To show how this works, Leslie joined with psychologists Marjorie Rhodes and Christina Tworek to design an experiment in which 4-year-olds were shown pictures of a fictional kind of person they called a Zarpie. The people in the pictures were male and female, black, white, Latino, Asian, young and old. With one group of 4-year-olds, the experimenters made lots of generic remarks. (“Zarpies are scared of ladybugs” and the like.) With another group, they made specific statements, not generic ones. (“Look at this Zarpie! He’s afraid of ladybugs!”) A couple of days later, they showed the kids a Zarpie and said he made a buzzing sound. It turned out that the children who’d heard a lot of generics about Zarpies were much more likely to believe that all Zarpies made buzzing sounds. Generic talk encouraged them to think of Zarpies as a category of person.

Generic remarks about people, in short, encourage you to think of them as a kind, and you’re more likely to accept a generic claim about a group if it’s negative or worrying. (Liberals hate America; conservatives are bigots.) As everyone knows on some level, we’re tribal creatures. We not only belong to groups but are easily triggered to take arms against other groups. Evolutionary psychologists think these dispositions helped our ancestors survive by creating groups they could rely on to deal with the perils of prehistoric life, including other groups competing for resources. True, that was before cable news and social media. But those us-and-them instincts remain an indelible part of human nature.

Still, if tribalism is responsible for some of the worst aspects of our politics, it’s also responsible for some of the best. According to the historian David Herbert Donald, the 19th-century abolitionists belonged to a tribe — essentially, an old-line Northern elite displaced by a new commercial and manufacturing class — that sought to regain its position through ethical crusades. The moral math was correct, but social identity was what helped it spread. Another kind of tribalism helped the civil rights movement go mass. We’re always hearing that the Democrats lost the South when — especially after the Civil Rights Act of 1964 — the party of segregation became the party of civil rights. And the red shift was real. But we don’t pause to reflect on how partisan identity politics actually slowed that defection.

Southerners, after World War II, were increasingly out of step with the national Democratic Party. They were more conservative than the leadership, and they were becoming more affluent than they used to be. (Richer Democrats went Republican before poor ones did.) White identity politics played a big and growing role, particularly in the wake of 1954’s Brown v. Board of Education . The region, by the usual reckoning, should have been a GOP stronghold after the civil rights era.

Yet we all belong to multiple tribes, and many Democrats who had once supported or been reconciled to segregation stuck it out when their leaders reversed course. Almost the entire South went in 1976 for Jimmy Carter, who won by wide margins in notably white states like Arkansas and Tennessee. Voters who had supported states-rights candidates got behind the progressive from Plains, Ga., because — well, they were Southern Democrats, and so was Carter. In national elections, the region didn’t become reliably Republican until the late 1990s. A generation of Southern Democrats had to die first.

To wish away identity politics is to wish away gravity. It burdens us, but it also grounds us. A workable politics enlists its force — and broadens its scope. At its very broadest, this gave us the inclusive nationalisms of FDR, Ike and JFK, with all their limitations. Successful politicians know that I’m with you counts for more than I’m for you. To say identity precedes issues doesn’t mean that issues aren’t important. But high-flown ideas — including a moral commitment to equality — must come down to earth to gain power: They matter when they connect to groups that matter to us, when they enter into a collective sense of who and what we are. For better or worse, it’s only through identity that ideas can change the world. Maybe someone could put that on a T-shirt.

Read more from Outlook:

Follow our updates on Facebook and Twitter.