Why Tribalism Took Over Our Politics
Social science gives an uncomfortable explanation: Our brains were made for conflict
By Aaron Zitner
Ahead of his arrest on Thursday in Georgia, Donald Trump repeatedly told his supporters about the legal peril he faced from charges of election interference. But the danger wasn’t his alone, he said. “In the end, they’re not coming after me. They’re coming after you,” he told a campaign rally.
It was the latest example of the Republican former president employing a potent driver of America’s partisan divide: group identity. Decades of social science research show that our need for collective belonging is forceful enough to reshape how we view facts and affect our voting decisions. When our group is threatened, we rise to its defense.
The research helps explain why Trump has solidified his standing as the front-runner for the Republican presidential nomination despite facing four indictments since April. The former president has been especially adept at building loyalty by asserting that his supporters are threatened by outside forces. His false claims that he was the rightful winner of the 2020 election, which have triggered much of his legal peril, have been adopted by many of his supporters.
Democrats are using the tactic, too, if not as forcefully as Trump. The Biden campaign criticized Republicans in Wednesday’s presidential debate as “extreme candidates” who would undermine democracy, and President Biden himself has accused “MAGA Republicans” of trying to destroy our systems of government.
The split in the electorate has left many Americans fatigued and worried that partisanship is undermining the country’s ability to solve its problems. Calling themselves America’s “exhausted majority,” tens of thousands of people have joined civic groups, with names such as Braver Angels, Listen First and Unify America, and are holding cross-party conversations in search of ways to lower the temperature in political discourse.
Share of those in each party who view the other party very unfavorably
Yet the research on the power of group identity suggests the push for a more respectful political culture faces a disquieting challenge. The human brain in many circumstances is more suited to tribalism and conflict than to civility and reasoned debate.
The differences between the parties are clearer than before. Demographic characteristics are now major indicators of party preference, with noncollege white and more religious Americans increasingly identifying as Republicans, while Democrats now win most nonwhite voters and a majority of white people with a college degree.
“Instead of going into the voting booth and asking, ‘What do I want my elected representatives to do for me,’ they’re thinking, ‘If my party loses, it’s not just that my policy preferences aren’t going to get done,’ ” said Lilliana Mason, a Johns Hopkins University political scientist. “It’s who I think I am, my place in the world, my religion, my race, the many parts of my identity are all wrapped up in that one vote.”
Trump, in responding to his indictment in Georgia for conspiring to overturn his 2020 loss in that state, amplified the sense of threat by telling a party gathering that they were engaged in a “final battle” that he described as “an epic struggle to rescue our country from the sinister forces within who hate it.” The criminal prosecutions, his campaign said in a fundraising email Thursday, were designed “to intimidate you out of voting to save your country.”
More than 60% of Republicans and more than half of Democrats now view the other party “very unfavorably,” about three times the shares when Pew Research Center polled on it in the early 1990s. Several polls find that more than 70% within each party think the other party’s leaders are a danger to democracy or back an agenda that would destroy the country.
Party allegiance can affect our judgment and behavior, many experiments show. When Shanto Iyengar of Stanford University and Sean J. Westwood, then at Princeton University, asked a group of Democrats and Republicans to review the résumés of two fictitious high-school students in a 2015 study, their subjects proved more likely to award a scholarship to the student who matched their own party affiliation. People in the experiment gave political party more weight than the student’s race or even grade-point average.
In a landmark 2013 study, Dan Kahan, a Yale University law professor, and colleagues assessed the math skills of about 1,000 adults, a mix of self-described liberals, conservatives and moderates. Then, the researchers gave them a politically inflected math problem to solve, presenting data that pointed to whether cities that had banned concealed handguns experienced a decrease or increase in crime. In half the tests, solving the problem correctly showed that a concealed-carry ban reduced crime rates. In the other half, the correct solution would suggest that crime had risen.
The result was striking: The more adept the test-takers were at math, the more likely they were to get the correct answer—but only when the right answer matched their political outlook. When the right answer ran contrary to their political stance—that is, when liberals drew a version of the problem suggesting that gun control was ineffective—they tended to give the wrong answer. They were no more likely to solve the problem correctly than were people in the study who were less adept at math.
To explain why the animosity in American politics is greater today than in the past, some researchers have focused on the nation’s political “sorting”—the fact that Americans have shifted their allegiances so that the membership of each party is now far more uniform. In the past, each party had a mix of people who leaned conservative and liberal, rural residents and urbanites, the religiously devout and those less observant.
Today, our partisan identities have come into alignment with the other facets of our identity, which heightens our intolerance of each other even beyond our actual political disagreements, Mason said. Political party has become a “mega-identity,” she said, magnifying a voter’s political allegiances and amplifying the biases that innately come from belonging to a group.
“When you go to cast a ballot, whatever part of your identity is under the most threat is going to influence your choice the most,” Mason said.
Researchers have been trying to use what they have learned about social identity to develop tactics to diminish hostility between the parties.
One recent experiment led by Stanford researchers showed 25 messages to a large set of Americans, and then assessed their views of the opposite political party and other attitudes. The tested “interventions” included videos and quizzes intended to show that many people within each party hold centrist views on policy and exaggerate the presence of hard-line views among the other party.
The strategies that worked best at reducing partisan animosity essentially modeled good behavior, highlighting what Democrats and Republicans have in common as Americans or presenting people making a good-faith effort to understand someone with differing views, said Robb Willer, a Stanford sociologist who led the research. But follow-up work showed that the effects diminished over time.
Researchers tested a TV ad that featured both Utah Gov. Spencer Cox, a Republican, and his Democratic opponent for governor in 2020, Chris Peterson, in which they committed to honoring the presidential election results. Cox last month began a one-year term as chairman of the National Governors Association and said he would devote much of the year to promoting civility in politics through a “Disagree Better” program that draws on the Stanford research.
Willer said the most effective messages might be broadcast at an important time in the election calendar, or used in school civics classes.
“It’s a bit of a David and Goliath situation, to be sure,” Willer said. “All the more reason to invent a slingshot.”