The Washington PostDemocracy Dies in Darkness

How can the U.S. help prevent more political violence?

Our research found two strategies that may help restore general social disapproval of attacks such as the one against Nancy Pelosi’s husband

Police stand at the top of a closed street outside the home of House Speaker Nancy Pelosi in San Francisco on Friday. (Eric Risberg/AP)

Late last week, an intruder broke into House Speaker Nancy Pelosi’s home and violently assaulted her husband, Paul, with a hammer, fracturing his skull and sending him to the hospital, where he remains. But that’s just the most recent incident in a string of violence and threats aimed at politicians, election officials and federal law enforcement agents since the Jan. 6, 2021, insurrection at the Capitol. Politically motivated violence appears to be rising in the United States.

How can this be stopped — or at least, slowed?

Societies generally deter violence by maintaining and enforcing social norms. These widely agreed-upon standards of acceptable behavior stay in place because of negative sanctions, such as disapproval, judgments and disrespect. Strong social norms can reduce undesirable behaviors, not only through the threat of sanctions, but through socializing individuals, shaping their sense of right and wrong.

So what happens if not everyone shares and enforces those norms against political violence — if, for example, a president pledges “full pardons with an apology to many” following the Jan. 6 attack? Then, political attacks may face fewer social repercussions — or may even be celebrated and rewarded. And if people feel that such attacks are becoming more acceptable, then violence may increase.

How, then, can the U.S. shore up American opposition to political violence? Just in the past year, researchers have made considerable progress toward understanding which strategies work and which don’t.

Don't miss any of TMC's smart analysis! Sign up for our newsletter.

What works? Correcting inaccurate beliefs about how much the other group supports violence

In our recent Strengthening Democracy Challenge (SDC), a massive online survey experiment, our team tested 25 possible interventions aimed at reducing political animosity, democratic backsliding and support for partisan violence. In April and May, we offered our survey to more than 32,000 participants recruited by Bovitz from opt-in panels. The sample was designed to approximate the national U.S. population of Democrats and Republicans with respect to gender, age, race, education and strength of party identification. We randomly assigned participants to one of the 25 approaches, or to be in one of two control groups. We used weighting to protect against bias that could result from participants dropping out of the study at different rates, depending on which group they were assigned to.

Of the five most effective approaches to reducing support for political violence, three involved correcting misperceptions about the “other” party. This approach also appears to help change other dangerous attitudes, such as the dehumanization of rival partisans.

In a recent paper with Northwestern University political scientist James Druckman, we found that both Republicans and Democrats think the rival party’s members are 3 to 5 times more supportive of violence — and more willing to behave violently — than the other party’s members actually are. People who hold the largest misperceptions — themselves a troubling symptom of our divided country — are more likely to support the use of violence against their rivals. Unless those false beliefs are corrected, they could become a self-fulfilling prophecy; misperceptions could encourage actual violence intended to preempt this perceived, false, threat from the other side.

Fortunately, we find that these misperceptions can be corrected. In two survey experiments, we asked people to guess how much the other party’s members support or are willing to engage in violence, then showed them how little the other party’s members actually support violence. Learning this, Americans’ own support for violence dropped by 34 percent to 44 percent, depending on whether we corrected misperceptions about the other party’s support for violence or willingness to engage in violence, respectively. In one study, this reduction lasted at least one month.

Here's what persuades Americans to support democracy over party

What works? Political leaders denouncing violence

Several studies conducted by other researchers suggest that political leaders publicly denouncing violence can strengthen norms against political violence. Political scientists Lilliana Mason and Nathan Kalmoe found that reading that political leaders denounced violence — such as a statement attributed to Donald Trump that said, “I condemn in the strongest possible terms all acts of violence” — significantly reduced ordinary party members’ own support for political violence.

Notably, in the study we mentioned above, two of the top five interventions included cues from political leaders. Other work also shows that reading comparable statements about election legitimacy from political leaders belonging to one’s own party improves trust in elections, which could also reduce violence at contentious times.

Whether this strategy is effective has thus far only been tested in survey experiments, not in practice. But members of both parties have publicly condemned the attack on Paul Pelosi, which could strengthen norms against political violence.

Is the FBI impartial? Over half of Republicans say 'no.'

What doesn’t work: Encouraging Americans to like members of the other party

You might think that getting people to like members of the other party more would reduce support for violence against the other party. However, we’ve found that the link between how much people like the other party and how much they support the use of violence isn’t quite clear.

For example, one of our experimental interventions involved having respondents write about friendships with people who belong to the other party. Another highlighted John McCain and Joe Biden’s friendship. While both reduced the animosity people said they feel toward the other party, neither reduced support for partisan violence.

So, what now?

While support for violence and reported willingness to commit violent acts are correlated, neither is the same as actually assaulting others. Thus, we cannot be sure that either of the strategies presented here would in fact reduce violent acts.

Nor is it clear how best to measure support for violence. Some new research suggests that Americans across the political spectrum are uniformly opposed to severe forms of political violence — such as arson against opponents or driving into a crowd of protesters.

Nevertheless, research suggests it’s worth trying two strategies. First, broadcast that members of both parties generally do not support violence. Second, encourage political leaders to denounce violence, and publicize statements by those who have denounced violence in the past.

Neither will end the threat of violence. But they could help strengthen long-standing U.S. norms against it.

Professors, check out TMC’s new and improved classroom topic guides.

Joseph Mernyk (@jsmernyk) is a PhD student in sociology at Stanford University.

Sophia Pink is a PhD student at the Wharton School at the University of Pennsylvania.

Robb Willer (@RobbWiller) is a professor of sociology, psychology, and organizational behavior at Stanford University.