Our Own Worst Enemies

In under a decade, violence has become a shockingly regular feature of American political life. In 2017, a left-wing extremist shot and nearly killed Republican House Majority Leader Steve Scalise and four other people. In 2021, a mass of right-wing insurrectionists stormed the U.S. Capitol to try to stop the Democratic president-elect, Joe Biden, from taking office. And in this year’s presidential campaign, there have, as of this writing, been two thwarted assassination attempts against Republican nominee Donald Trump, along with a torrent of threats directed at political figures of all stripes. Indeed, the election in November could well be not only the most consequential in modern U.S. history but also the most dangerous.

But for all the warranted dismay, the mounting frequency of such events should not have come as a surprise, for Americans or for observers around the world. As analysts have pointed out, there are many possible reasons for the surge in violence. Some experts have cited the steady weakening of critical democratic institutions and, relatedly, the antidemocratic tendencies of destitute and isolated white conservatives. Others have pointed to the radicalizing effects of partisan gerrymandering and polarization. Still more have highlighted social media and militias. Many analysts have blamed Trump.

Each of these factors is indeed helping foster contentious U.S. politics. But all this commentary overlooks the predominant structural dynamic driving the new era of violence. The principal danger to the United States is not any out-of-control technology or fringe militia group. It is not economic grievances run amok. It is not even Trump, who is as much a symptom of what ails the United States as he is a cause. Instead, the greatest source of danger comes from a cultural clash over the nature of the United States’ identity—one with profound implications for who gets to be a citizen. Its key actors are not isolated radicals but large numbers of ordinary Americans. According to new research carried out by my team at the University of Chicago, tens of millions of Democrats, Republicans, and independents believe that political violence is acceptable. Many of them hail from the middle and upper class, with nice homes and college educations.

The country’s fight over its national identity has multiple dimensions. But the most serious is demographic change. In 1990, 76 percent of the U.S. population identified as white. In 2023, the U.S. Census Bureau put that figure at a little over 58 percent. By 2035, the share is set to fall to 54 percent; a decade later, it will dip below 50 percent. These changes have led to rising anger among conservatives, many of whom see increased ethnic diversity as an existential threat to their way of life. These voters have embraced Trump and his nationalist movement, which promise to stop such change in its tracks. Trump’s exclusionary policies and rhetoric have, in turn, prompted a ferocious backlash from liberals, who embrace demographic change—or who at least fear that conservative success will cost Americans hard-won freedoms.

The anger on both sides is in keeping with historical precedents. Scholars have long understood that social change and demographic shifts are a potent catalyst for violence. And as elsewhere, the turn toward force in the United States is fundamentally populist in nature. The millions of Americans who support political violence have concluded that their country’s elites are so thoroughly corrupt and that their democracy is so completely broken that riots, political assassinations, and coercive attacks are acceptable and even necessary to bring about the supposedly genuine democracy that people deserve. This kind of thinking is endemic to all kinds of populist movements, in which people angrily latch on to a political leader, party, or movement to overcome the so-called establishment.

Unfortunately, violent populism is likely to grow more pronounced in the years ahead. Throughout history, societies in which large numbers of people support political violence are much more likely to experience unrest. There is no way to stop the United States’ demographic shift, and even if there were, doing so would be a mistake: the country’s diversity makes it stronger. The United States may not be on the precipice of a full-scale civil war, as some have predicted. But the country is entering an era of intense deadly conflict—one replete with politically motivated riots, attacks against minorities, and even assassinations.

DANGER ZONE

Throughout U.S. history, Americans have experienced several waves of violent populism. In the early 1920s, following a massive wave of Catholic immigration to the United States, millions of people signed up to join the nativist and white supremacist Ku Klux Klan. The KKK and its allies then carried out repeated attacks against Black people, Jews, and Catholics. During the 1960s and 1970s, the United States had to contend with major political assassinations and large urban riots, many of them conducted by right-wing extremists and left-wing terrorist groups such as the Weather Underground. The violence of this era was also spurred on by social issues, including the fight to offer Black Americans equal rights, and by growing dissatisfaction with the war in Vietnam.

Still, these eras were exceptions, not the rule. For most of the country’s history, political violence has been relegated to the fringes of society. During the 1980s, the 1990s, and the first decade of this century, the country experienced a smattering of domestic terrorist incidents—most famously the 1995 bombing of a federal building in Oklahoma City. People affiliated with the far-left Earth Liberation Front and the Animal Liberation Front also hit farms and car dealerships. But strikes were few and far between. Aside from the Oklahoma City attack, they rarely dealt substantial damage. The real threat was foreign terrorism, as September 11 made painfully clear.

Today, however, domestic political violence is much more frequent. Statistics collected by the Federal Bureau of Investigation and the Department of Homeland Security show that domestic terrorism incidents increased by 357 percent between 2013 and 2021. According to a study by the Chicago Project on Security and Threats, which I direct, more than 250 people have been prosecuted for threatening nearly 200 of the country’s 1,633 federal legislative, executive, and judicial officials from 2001 to 2023. The average number of these threats increased by 400 percent from 2017 to 2023, from four threats a year to just over 20 threats a year.

Domestic terrorism has occurred on both the left and the right. Although antigovernment and white supremacist extremists conducted 49 percent of all attacks and plots in 2021, anarchists, antifascists, and all kinds of left-wing extremists carried out 40 percent of FBI-registered incidents that year (up from 23 percent in 2020). Democratic and Republican members of Congress have been attacked almost equally since 2017.

Public support for political violence is not limited to the fringe.

Violent populism’s bipartisan nature is even more apparent when one examines instances of collective political violence. After the killing of George Floyd at the hands of Minneapolis police officers in 2020, over 15 million people took to the streets to protest racism and police brutality. Between seven and ten percent of these protests devolved into large-scale riots against police and businesses in the downtown areas of Chicago, Minneapolis, New York City, Philadelphia, Portland, Seattle, and over 100 other American cities—the most protracted series of political riots since the 1960s. Six months later came the January 6 ransacking of the Capitol. As part of it, pro-Trump supporters brought a noose to the surrounding grounds and chanted “Hang Mike Pence” (then vice president) and hunted for U.S. House Speaker Nancy Pelosi. And in late 2023 through June 2024, protesters bent on ending Israel’s war in Gaza stormed and seized campus buildings and physically assaulted students. The country also witnessed over 1,000 separate incidents of anti-Semitism and Islamophobia in just nine months.

These numbers are, by themselves, alarming. But what is even more concerning is the broad backing violent actors appear to have. According to a January 2024 survey conducted by my team along with NORC, a prominent polling organization at the University of Chicago, over 15 percent of Americans—12 percent of Democrats, 15 percent of independents, and 19 percent of Republicans—agree that the “use of force is justified to ensure members of Congress and other government officials do the right thing.” In our more recent June survey, ten percent of respondents (a number that extrapolates to 26 million American adults) agreed that “the use of force is justified to prevent Donald Trump from being president.” Over 30 percent of these people own guns. Twenty percent think that when police are violently attacked, it is because they deserve it. Meanwhile, seven percent of respondents (equating to 18 million American adults) support the use of force to restore Trump to the presidency. This group has even more dangerous capabilities: 50 percent own guns, 40 percent think “people who stormed the U.S. Capitol are patriots,” and 25 percent either belong to a militia or know a militia member.

These numbers alone make it clear that public support for political violence is not limited to the fringe. But to test just how mainstream support for violence runs, my team collected data on respondents’ backgrounds. It found that over 80 percent of the people who back using force to either prevent or facilitate Trump’s election live in metropolitan areas. Thirty-nine percent have had at least some kind of college education. Even on the political right, over 80 percent live in metropolitan areas and 38 percent have at least some college experience. In other words, they are broadly representative of the U.S. population. They cannot be derided as a bunch of yokels.

FEAR AND LOATHING

It is, of course, one thing for people to support political violence and quite another for them to carry out an attack. But they do not need to become violent themselves in order to foster strife. As scholars have long known, public support for political violence encourages volatile people—those who may actually use force—to act on their worst impulses. The political climate may prompt such people to think their attacks are serving some greater good, or even that they will be glorified as warriors.

In fact, popular support for violence is one of the best predictors of bloodshed. Before The Troubles, in the second half of the twentieth century, Catholics and Protestants in Northern Ireland both grew much more supportive of using force to change the region’s political arrangement. In Spain, support for violence went up before the Basque nationalist Euskadi Ta Askatasuna movement began an assassination campaign against the country’s authoritarian government during the same era. And West Germans were increasingly supportive of attacks before the Baader-Meinhof Gang (also known as the Red Army Faction) conducted a series of bombings and assassinations in the 1970s.

Unfortunately, the U.S. population may well become even more tolerant of political violence in the years ahead. According to our June survey, the Americans most opposed to violent populism are those over the age of 59. They are three times less likely to support violence to restore Trump to the presidency than those between the ages of 30 and 59. Their pacifying effect will therefore wane with time, especially if the next generations of young people remain as supportive of violence as their predecessors are. Although it is possible that today’s youth will grow more opposed to violence as they age, it is far from guaranteed. Time does not inherently deradicalize. About ten percent of those who assaulted the Capitol, after all, were 60 years old or older.

But perhaps the main reason to expect more political violence has to do with a different type of demography: race. The United States is set to transition from a white-majority to white-minority society by 2045. That transition will take place in all 50 states, and it will be especially pronounced in the younger portion of the population. It will also be visible in politics. Indeed, it already is. Today, a quarter of House and Senate members identify as nonwhite, making them the most diverse group of representatives in American history.

The United States’ historic transition from a white-majority to a genuinely multiracial democracy is producing social changes with profound political implications. This power shift in politics, media, and major business and community organizations is the taproot of rising cultural backlash among conservatives—epitomized by Trump and his movement. The shift is, therefore, also the basis for counterreactions among liberals both hopeful for change and fearful that conservative success will obstruct progress, reverse economic and social gains, and establish a political system that does not represent everyone. Both sides’ fears do not have to accord with reality to fuel attacks. Among conservatives and liberals alike, the consequences of political change need only exist in peoples’ minds.

The fact that abstract demographic shifts can lead to panic may be jarring, but it should not come as a surprise. Throughout history, social and demographic change have produced grievances (real and imagined), tensions, and political unrest. As the comparative political scientist Donald Horowitz wrote, when “majorities within a country become minorities . . . anxiety flows from a diffuse danger of exaggerated dimensions.” People begin to fear they will come under siege in their own homes and be dominated by strangers. Such concerns drove violence in Brazil, Lebanon, the Balkans, and parts of the former Soviet Union, among numerous other states.

Americans, particularly liberal ones, may fancy themselves as tolerant enough to avoid acting on ethnic biases. But this pattern of thinking afflicts them just as much as it does their peers elsewhere. In separate experimental studies among Americans and Canadians, the psychologists Robert Outten, Jennifer Richeson, and Maureen Craig reported that exposure to information about white demographic decline increased white sympathy for other whites and increased feelings of fear and anger toward minorities. These sentiments were most pronounced among white conservatives, yet they were evident to a small degree among whites who identify as liberal, as well. Research has also shown that the United States’ demographic shift accounts for the rapid rise of Trump in 2015 and 2016. (During the 2016 presidential campaign, both Trump and the Democratic nominee, Hillary Clinton, clashed over issues tied to race, gender, and cultural identity far more than did previous presidential candidates.) Similarly, studies have illustrated that nationalist and multicultural media such as Fox News, Newsmax, and MSNBC have become far more popular as U.S. demographics change. And according to multiple scholars, white American racial prejudice and solidarity have gone up as the share of Americans who are white has gone down.

My team’s research shows that anger about diversity also directly predicts support for violence. According to the January 2024 study, Americans who believe that “the Democratic Party is trying to replace the current electorate with new people, more obedient voters from the Third World”—the so-called great replacement theory—are six times more likely to support using force to restore Trump to the presidency. Americans who believe in the great replacement are also five times more likely to think that “people who stormed the Capitol on January 6 were patriots.” They are three times more likely to either belong to a right-wing militia or know someone who does.

There is no perfect parallel to the great replacement on the left. But the January study did ask respondents whether they believe “America is a systemically racist country against nonwhite people and has always been.” People who answered in the affirmative were roughly two times more likely to support using force to stop Trump than were those who did not. These respondents were also four times more likely to believe that “when the police are attacked, it is because they deserve it.” They were one and a half times more likely to think “the use of force is justified to restore the federal right to abortion.”

ROUGH RIDE

These findings do not mean the United States is headed for a classic ethnic conflict, as happened in Northern Ireland and Bosnia. After all, many white people believe the United States suffers from systemic racism and want to end it. There are Asian, Black, and Hispanic Trump supporters. American political violence is unlikely to manifest in the form of civil war, at least understood as two rival armies standing toe to toe on battlefields or as hundreds of thousands of armed insurgents roaming the country. Such wars are more likely when a state’s political, social, economic, and geographic cleavages generally converge so that political parties, economic classes, and geographic areas all broadly align. And although the overlap between them is increasing, the United States’ racial, economic, social, and geographic factors remain largely divergent. There are Democrats and Republicans in pockets throughout the country, in different economic classes, and in different ethnic groups.

To see why convergence matters, compare the circumstances in the United States today with those in Bosnia in the 1990s. The collapse and fragmentation of the Yugoslav state coincided with growing social, economic, and ultimately political cleavages between Albanians, Croats, and Serbs, as well as with major economic problems. Together, these forces led to a surge in nationalist tensions that produced warfare and mass ethnic violence against civilians. The United States, by contrast, is not on the verge of government collapse. Its economy remains strong.

Although the most fantastical forms of violence may not come to pass, Americans must be prepared for an extraordinary period of unrest. Their country will probably experience years of serious political assassination attempts, political riots, and other instances of collective, group, and individual violence. There could be new militia groups, violence over numerous issues in cities and on college campuses, and outbursts related to elections. Such attacks could even break elements of the American political system, or at least yield institutional changes. Political violence, for example, may lead to serious delays in counting and certifying votes in future elections. It could push U.S. politics in an increasingly autocratic direction as Americans become less confident that elections truly reflect the will of the people and become more open to strongman alternatives. It could also pressure Washington to grant states more autonomy over social and cultural matters. The Supreme Court has already devolved questions of abortion rights to states.

The main point of contestation will, naturally, be who gets to be an American and what rights U.S. citizenship confers. The 2024 election has been a stark illustration of this fact—a battle between the strongly nativist Trump and the Democrat Kamala Harris, a progressive, biracial woman. It has featured radical, determined minorities who support violence to get Trump into office and those who support violence to stop it.

Unlike Trump, Democratic Party leaders have shown little willingness to mobilize progressives to embrace violence in response to electoral losses. But the left is still capable of responding virulently to outcomes it dislikes. If Washington undertakes a high-profile effort to arrest, detain, and deport massive numbers of illegal immigrants, radicals could rally to their defense, including by staging mass protests that may turn violent, and then not back down. They may be especially likely to act if the government sends federal or federally deputized armed agents into so-called sanctuary cities—cities that limit cooperation with federal immigration officers. After the Department of Homeland Security sent agents to arrest, detain, and prosecute protesters in Portland, Oregon, in July 2020, demonstrators confronted agents with wooden shields and other objects, breaking through barricades and assaulting police stations.

OUT OF MANY

To avert an era of politically motivated riots and attacks, Americans will need to find some common ground on race and immigration. This will be extremely difficult. Race and ethnicity are social constructs, so activists and leaders can try to help immigrants quickly integrate into U.S. society and to persuade white Americans that they have much in common with their nonwhite counterparts. But this process is unlikely to work fast enough to avoid an era of violent populism. Group boundaries and social identities may not be set in stone, but they are hardly putty. It typically requires generations for new immigrant groups to integrate and for white people to see them as no different from themselves. It took more than a century after Irish immigrants began flooding the United States for the country to elect its first Catholic president, John F. Kennedy.

Perhaps the United States can paper over these divisions with strong economic growth. Americans, after all, routinely rank the economy as the most important issue. But if history is any guide, expanding gross national product is also unlikely to be a panacea. The 1920s—when the Ku Klux Klan exploded in membership—is also known as the Roaring Twenties, as the United States economy grew at an average of over four percent each year. Total wealth in the United States more than doubled from 1920 to 1929. Similarly, the violence and instability of the 1960s occurred when U.S. economic growth averaged five percent annually. In both eras, the violence did not stop until questions of identity were decisively resolved. In the 1920s, that meant victory for the nationalists: Congress passed the Immigration Act of 1924, which effectively closed the United States’ borders. Even then, anti-Black violence continued. That did not plummet until federal legislation in the 1960s put a stop to legal segregation and discrimination, handing victory to progressives. The government also repressed organized violent groups, which lost much of their popular support and crumbled rather than resurged. Unrest still proceeded, on and off, until the United States stopped conscripting men to fight in Vietnam.

Today, a hard end to immigration would not resolve America’s challenges. Even closing the U.S. borders entirely would merely slow the process of whites becoming a minority by roughly a decade. Such a solution would also be unacceptable: liberals are right that a truly multiracial democracy would be good for the country. It will most obviously be good for minority groups, who deserve equal treatment. But white Americans have as much to gain as others from a future in which everyone is judged by their character and not their skin color. There is plenty to celebrate about the country becoming a more perfect union.

Americans must be prepared for an extraordinary period of unrest.

Still, less draconian immigration policies could reduce tensions. Policymakers should find bipartisan ways to decrease illegal immigration, aiming to at least return to the levels under the Obama administration. That means dedicating considerable resources to enforcing current laws and keeping the nation’s borders secure. It also means maintaining sensible pathways to citizenship for the vast majority of immigrants. Adopting such policies would put the White House and Congress on better footing by showing that it is possible to effectively balance the country’s economic needs, social responsibilities, safety, and political concerns. Better immigration rules would also build good faith and illustrate that politicians can pursue long-term solutions to the United States’ problems.

Ultimately, Americans should stay hopeful. Most of them, after all, continue to abhor political violence—even if a significant minority now support it. According to the June survey, 70 percent of Republicans oppose political violence and want leaders to condemn its use. So do over 80 percent of Democrats. Elected officials at all levels of government should listen to their constituents and curtail incendiary rhetoric. Trump, of course, shows few signs of doing so. But the broad condemnation of political violence by both Democrats and Republicans in the aftermath of the attempts on his life has set an important precedent that all other leaders can and should emulate.

There are other reasons to believe that the Republican Party’s leaders might, eventually, embrace a less hostile line. The nature of the U.S. political system can sometimes encourage candidates in primaries to take radical positions in order to appeal to the base, but because the United States has just two viable parties, their candidates perform best in general elections when they reach out to multiple groups. In recent years, the Republican Party has been able to win some elections without moderating. Its candidates would surely have more success, however, if they decided to be more inclusive—a lesson that, eventually, its leaders could accept. Ultimately, the two-party system is one of the United States’ great shock absorbers for social change. It may lead to a soft landing as the country transitions to a multiracial democracy.

Yet for now, the country’s fever is unlikely to break. Support for political violence has gone mainstream. The chief reason—demographic change—is not going away. And there is no easy or just way to reconcile conservatives’ and liberals’ visions. Political trends do not move in straight lines, and predicting the future can be a fool’s errand. But it is safe to say that the United States has a rough road ahead.