Half of Americans anticipate a U.S. civil war soon, survey finds

Violence can seem to be everywhere in the United States, and political violence is in the spotlight, with the 6 January 2021 insurrection as exhibit A. Now, a large study confirms one in five Americans believes violence motivated by political reasons is—at least sometimes—justified. Nearly half expect a civil war, and many say they would trade democracy for a strong leader, a preprint posted today on medRxiv found.

“This is not a study that’s meant to shock,” says Rachel Kleinfeld, a political violence expert at the Carnegie Endowment for International Peace who was not involved in the research. “But it should be shocking.”

Firearm deaths in the United States grew by nearly 43% between 2010 and 2020, and gun sales surged during the coronavirus pandemic. Garen Wintemute, an emergency medicine physician and longtime gun violence researcher at the University of California, Davis, wondered what those trends portend for civil unrest. “Sometimes being an ER [emergency room] doc is like being the bowman on the Titanic going, ‘Look at that iceberg!’” he says.

He and his colleagues surveyed more than 8600 adults in English and Spanish about their views on democracy in the United States, racial attitudes in U.S. society, and their own attitudes toward political violence. The respondents were part of the Ipsos KnowledgePanel—an online research panel that has been used widely, including by Wintemute for research on violence and firearm ownership. The team then applied statistical methods to extrapolate the survey results to the entire country.

Although almost all respondents thought it’s important for the United States to remain a democracy, about 40% said having a strong leader is more important. Half expect a civil war in the United States in the next few years. (The survey didn’t specify when.) “The fact that basically half the country is expecting a civil war is just chilling,” Wintemute says. And many expect to take part. If found in a situation where they think violence is justified to advance an important political objective, about one in five respondents thinks they will likely be armed with a gun. About 7% of participants—which would correspond to about 18 million U.S. adults—said they would be willing to kill a person in such a situation.

Kleinfeld says the study’s findings are compelling because of the large number of participants and because it asked about specific scenarios in which participants think violence is justified—such as for self-defense or to stop people with different political beliefs from voting. The sample does slightly overrepresent older people, who are not known to commit much violence worldwide, she says. “So the fact that you’re [still] getting these high numbers … is really quite concerning.”

She is less alarmed by the shaky support for democracy, noting that political gridlock—as in U.S. politics today—can often distort attitudes. “What people mean by ‘democracy’ is pretty fuzzy,” she says. Political paralysis, she adds, can quickly lead people who think, “Yeah, I like democracy,” to also say, “Yeah, I want a strong man” in leadership.

“The findings are scary, but not surprising,” Kurt Braddock, who studies the psychology of extremist communication at American University, wrote in an email to Science. In recent years, he says, the United States has seen an increase in individual willingness to engage in violence—homicides in cities increased 44% between 2019 and 2021, for instance—an attitude he says is likely to spill into the political sphere.

Researchers have criticized the sampling and survey methodology of previous studies that found increasing support for political violence. But the new study generally agrees with earlier efforts, Kleinfeld says. A small survey from 2021, for instance, found about 46% of voters thought the United States would have another civil war, and another showed more than one-third of Americans agree that “The traditional American way of life is disappearing so fast that we may have to use force to save it.” Barbara Walter, a political scientist at the University of California, San Diego, who was also not involved in the study, agrees. But she suspects the survey responses overrepresent the number of Americans who would be willing to turn to violence because, she says, surveys tend to overstate what people actually think. “The numbers always tend to be shocking, but in essence, are probably not true.”

Wintemute and colleagues found that conspiracy theories, some rooted in racism, are helping shape views about political violence. They found roughly two in five adults agreed with the white nationalist “great replacement theory,” or the idea that native-born white voters are being replaced by immigrants for electoral gains. And one in five respondents believed the false QAnon conspiracy theory that U.S. institutions are controlled by an elite group of Satan-worshipping pedophiles. Respondents’ belief in conspiracy theories might partially explain their views on democracy and political violence, Walter says, but she wishes the survey went deeper to explain the particular reasons why the participants would choose to engage in violence.

Both Braddock and Kleinfeld wish the new study responses were broken down by partisan affiliation because previous surveys show engaging in violence is much more prevalent in right-leaning individuals. “That is a key oversight,” Braddock says. “If readiness for civil war is coming from one side [and not in both], we need to know that.” The study team did look at affiliation with certain extremist groups such as the Proud Boys, but Wintemute said his group is now working on follow-up analyses of the survey to look at other political affiliations and will also launch a follow-up survey with the same group of respondents by the end of the year to further examine the role of certain group identities and propensity for political violence.

To reduce the threat of political violence, Braddock says, the first step is to call out the disinformation online and in right-wing media, some of which is taken directly from extremist propaganda. “We need to call that out for what it is before we can begin to address the problems it is causing.” Regulating social media to avoid “incendiary” misinformation from spreading could also help, Walter says. Kleinfeld adds that leaders—from politicians and media personalities to church pastors—can also make a difference. Experiments show courageous leaders can deter their communities from engaging in violence. “Now’s the time to take this seriously and not put our heads in the sand,” Kleinfeld says.