The people most likely to believe in political violence may surprise you

A clash between Proud Boys and counterprotesters breaks out, in November 2020. Researchers say that the most socially connected people are more willing to say they’d use political violence.

Even before two people attempted to assassinate former President Donald Trump, national security experts and law enforcement were warning that the United States needed a plan to contend with the possibility of political violence on Election Day. 

Now, in the final weeks of the campaign, researchers have just published new findings about the social lives of people who are likely to endorse using political violence, and be willing to use it themselves. 

The results, based on a nationally representative survey of more than 8,000 Americans, may seem counterintuitive. Basically, people open to the idea that political violence is justified tend to exist at opposite ends of the social spectrum. Those who report having no strong personal or work connections were 2.4 times more likely to say political violence is justifiable than people who said they have 1–4 close relationships. That’s not necessarily surprising, given the recent history of mass shooters and politically motivated assailants who’ve been described by their broader networks as loners.

What’s weird, though, is that people with lots of close connections were also a little more likely to endorse political violence. People who said they had 50 or more strong relationships were 1.2 times more likely to endorse political violence. 

And here’s where it gets a little concerning: The people with no social connections weren’t on average any more likely to say they would be willing to personally commit political violence, even if they believed it was justified. But the people with lots of close relationships? They were 1.5 times more likely than the others to say they’d be willing to be violent for a political cause themselves. 

What’s going on with those super socially connected people? Julia Schleimer, the researcher who led the study, told Vox over email that, compared to the people with just a few close connections, the 50+ cohort tended to be white, higher in income, slightly more educated, and older. But that’s also true of the demographic in the middle (which reported 10–49 social connections) and they weren’t especially open to the idea of political violence. 

In other words, there were no demographic factors about the group that jumped out to the researchers. “One limitation of this study is that we don’t have details on the nature or characteristics of people’s social networks, which likely matters a great deal and is an area for future study,” Schleimer said. But prior studies “give us reason to expect that people with very large social networks may be at greater risk for political violence, if those networks are characterized by antisocial norms, outgroup contempt, and extreme views.” That’s particularly true, she said, when the social networks are homogenous. Sometimes those groups form in person, but increasingly, they also develop online, like the Proud Boys and other far-right groups who organized in the days after Donald Trump’s 2020 election loss and stormed the Capitol on January 6. 

This research, done by University of California Davis’s Violence Prevention Research Program (VPRP) builds on data published earlier this year about Americans’ beliefs in political violence. The good news from the earlier work is that by and large, a vast majority of Americans are opposed to political violence under any circumstances. The more worrying news? A small proportion of Americans are open to the idea that political violence is sometimes justified.

“I personally think that large-scale political violence is really, really unlikely. I feel more sanguine about that prediction, given our 2024 data,” Dr. Garen Wintemute, director of the program, told Vox this summer. “But sporadic outbreaks, particularly if the battleground states remain really close — is it possible? Sure. Might there be attempts to intimidate election officials? Absolutely.”

The Violence Prevention Research Program applies a public health approach to issues like gun violence and political violence — meaning they look for interventions that can try to discourage them from happening. The new findings, Schleimer said, suggest that it’s important to develop approaches that target both those who are very lonely and those who are deeply connected, in, for example, extremist ideological groups. For the lonely, social skills training, community centers, cultural activities and more open and accessible cities can all be helpful. And both groups benefit from anti-violence messages from influential public figures and on social media. For the socially connected people, having a trusted figure who can support them as they begin opening up to different perspectives and challenging the beliefs of their ideology can be especially helpful.

The ideas sound really simple, but the researchers’ previous work suggests that getting someone to reject political violence is perhaps easier than you might think. “For the would-be combatants, a big number would switch if their family asked them not to, or friends, or even some media sources,” Wintemute told Vox. “We can create a climate of nonacceptance for political violence. And in doing that, we can expect that it will work.” Their findings are encouraging, in that respect. But reaching every person who might be open to political violence, in a highly divided country, with this many guns? That’s the tricky part.