Testimony before the Select Committee to Investigate the January 6th Attack on the United States Capitol

Chairman Thompson and Members of the Select Committee, thank you for your work on this critical threat to our democracy and to the security of the American people. It is an honor to present this testimony to your committee. My name is Dr. Rachel Kleinfeld, and I am a Senior Fellow in the Democracy, Conflict, and Governance Program at the Carnegie Endowment for International Peace. I serve on the bipartisan Boards of Directors of the National Endowment for Democracy and Freedom House, and am also a member of the bipartisan National Task Force on Election Crises in the United States. My last two books focused on political violence and the rule of law. I will speak to three issues in this testimony:

  • Trends in acceptance of political violence in the U.S. and how former President Trump’s claims of election fraud have enhanced the acceptability of such violence;
  • Why the structure of today’s political violence is inflicting long-term damage on our democracy;
  • What these trends mean for addressing the problem, and what Congress can do.

Overall, I hope to drive home this point. Acceptance of political violence has been rising sharply over the past five years. The acceleration in violent incidents from 2016 to 2020 – and their probable fall in 2021 – does not mean the threat has receded. Instead, it tells us something about the nature of the threat and its causes. The trends we see in the spring of 2022 are indicative of violent behavior in the U.S. that can be motivated spontaneously by social events and can be directed purposefully for political goals. Politically-catalyzed violence should be expected to rise with the election calendar and to fall between campaigns. The damage that this violence itself, and the conspiracies driving it, are causing to our democracy are already substantial and are likely to produce significant democratic decline if not arrested soon.

1. Trends in Acceptance of Political Violence

According to nearly every measure of both beliefs and actual incidents, political violence is considered more acceptable than it was five years ago before then-President Trump took office.


The most consistent survey data comes from scholars Lilliana Mason and Nathan Kalmoe, who have been tracking public opinion on political violence using identical questions and methods for the last five years.1 From 2017 when their tracking begins, support for political violence rises across several measures prior to the midterm elections and declines after the elections. It also spikes (especially for Republicans) around then-President Trump’s first impeachment, and again drops afterward. Support for violence from 2017 through the summer of 2020 is generally quite close across parties but somewhat higher for Democrats, though as I’ll show later, actual incidents of violence are far higher for Republicans.

Rachel Kleinfeld
Rachel Kleinfeld is a senior fellow in Carnegie’s Democracy, Conflict, and Governance Program, where she focuses on issues of rule of law, security, and governance in post-conflict countries, fragile states, and states in transition.
More >

The 2020 election season was an inflection point that led to a step-change in acceptance of violence as a political tool, particularly among Republicans. During the month of the election, Republican support for violence leaps across each of Kalmoe and Mason’s questions. Democratic justifications rise in response to some questions but fall in others, and their support moves after Republican opinion and grows less quickly.

By February 2021, 25% of Republicans and 17% of Democrats felt threats against the other party’s leaders were justifiable, and 19% of Republicans and 10% of Democrats believed it was justified to harass ordinary members of the other party. One in five Republicans (20%) and 13% of Democrats claimed that political violence was justified “these days”.

In each case, support for political violence has doubled for Republicans since 2017 and has grown for Democrats.

To put this level of support into context: In 1973 during the most violent period of Northern Ireland’s Troubles, 25% of Catholics and 16% of Protestants agreed that “violence is a legitimate way to achieve one’s goals.”2 The U.S. is fast approaching these numbers.

Other polls trace the same leap in acceptance of violence following the election, particularly among Republicans. For instance, similar questions were posed a year apart by scholar Larry Bartels and the AEI Survey Center on American Life. In January 2020, half of Republicans and Republican-leaning independents agreed that “The traditional American way of life is disappearing so fast that we may have to use force to save it.” By January 2021, this number had risen to 56%, and by June 2021 another poll found 68% support across both parties.3

The perception that an election was stolen is known to be a major instigator of political unrest worldwide. Researchers have found that incumbents play an outsized role in shaping whether elections that are perceived as fraudulent (whether that perception is real or imaginary) lead to violent riots or street protests. Political parties can also reduce the risk of violence by channeling anger towards non-violent demonstrations. This is one reason it is so disturbing to note that by late November, 2020: “a significant portion of the online infrastructure” that had been built to support Trump’s reelection had been retooled to promote election misinformation and upcoming protests, according to the Atlantic Council’s DFR Lab, which has done a forensic analysis of that cyberinfrastructure.4

The consciously propagated false narrative regarding election theft is directly linked to the growing support for violence on the right. Those who believed the election was fraudulent were twice as likely in Kalmoe and Mason’s survey data to endorse a military coup and were more likely to justify armed citizen rebellion.5 Of the 9% of Americans who agreed that “force is justified to restore Trump to the Presidency” in a separate University of Chicago survey in June 2021, 90% believed Biden’s presidency was illegitimate.6 Throughout 2021, a time during which the Bright Line Watch organization found that 73-74% of Republicans felt that President Biden was not the rightful winner of the election, a separate 22,900 person poll found that almost 1 in 5 among Republican men claimed that violence was justifiable “right now”.7

Some experts believe that these numbers are emotional or philosophical statements and thus overstate the acceptance of actual violence. Abstract questions with no definition of violence opens surveys to that claim, and no doubt a percentage of respondents are not serious. Yet there is reason to believe a significant number of respondents mean what they say.

For instance, Bright Line Watch found that when survey respondents were asked about concrete examples and were forced to pay more attention before answering, support for violence reduced – but even with these safeguards, they found in November 2021 that among Republicans who most strongly identified with the party, 17% were willing to endorse violence to usher Trump back into the presidency, 18% endorsed threats, and 9% were willing to justify violence if the other party wins the 2024 election.

Similarly, a month after January 6, 2021, Kalmoe and Mason polled respondents on what they meant when they spoke of violence. When presented with concrete actions, a third of those who initially endorsed political violence picked “none of these options”, suggesting some posturing had been occurring. Of the remaining two-thirds, however, 7% justified “widespread violence by armed groups that kill lots of people”, 17% endorsed “violence by armed individuals that might kill a few people,” and 22% endorsed “fistfights and beating people up”, while 24% intended property crime. Respondents on the right and left justified these forms of violence. When asked: “Do you believe the use of violence to take over state government buildings to advance your political goals is justified?” a quarter of Republicans and 13% of Democrats agreed, while 12% of Republicans and 11% of Democrats agreed that “assassination of opposing political leaders is justified to advance political goals” was justified.8

Incidents of Threat, Harassment and Violence

Another way of measuring the growing acceptance of violence is the rise in actual incidents, which demonstrate the mainstreaming of threats, intimidation, and violence in American political life.

Threats against members of Congress

Threats against members of Congress are more than ten times as high as just five years ago. From 902 threats investigated by Capitol Police in 2016, they leapt to 3,939 in the first year of the Trump presidency, 5,206 by 2018, 6,955 in 2019, 8,613 in 2020, and hit 9,600 in 2021.9

Armed Demonstrations

In the 11 weeks between the election and Inauguration Day, armed actors at protests grew by 47% compared to the 11 weeks prior to the elections, and organized paramilitary groups grew by 96%. Stop the Steal demonstrations were more than four times as likely to feature armed actors or unlawful paramilitary compared to other demonstrations. More than a fifth of all Stop the Steal rallies nationally featured armed actors or unlawful paramilitary.10

Armed demonstrations are legal in many states; however, they are 6.5 times more likely to lead to violence than demonstrations without the presence of firearms. In 2020, 6.2% of all pro-Trump demonstrations were armed, numbers that climbed to 8.8% in 2021. These rates are not only increasing but are far higher than protests that were not supportive of Trump: from January 2020 to November 2021, only 1.5% of all other demonstrations across the United States involved armed protestors.11

Signifying their political intent, armed pro-Trump demonstrations were far more likely to occur at legislative buildings: 47.3% of all armed pro-Trump protests occurred on legislative grounds compared to 12.2% of all other armed demonstrations. Armed protests on legislative grounds increased by nearly 20% between 2020 and 2021 and were also more likely to turn violent.12 For example, on January 6, in addition to events in Washington D.C., 12 states were forced to disrupt activities and evacuate their capitol buildings and/or faced crowds threatening or breaching state capitol buildings.13

Prior to the 2020 election, armed demonstrations occurred for a variety of reasons around the country, from Second Amendment rallies to militias countering Black Lives Matter protests, and in a few instances, armed Black paramilitary groups. Following the election, however, each peak of armed protest coincided with election-related activities that had previously been purely procedural, such as election tabulation, safe harbor deadlines, electoral college results and Inauguration Day.14 Thus, violence was taking on the pattern seen in other countries where popular anger is coalesced by politicians and put to the service of political goals.

FBI hate crimes

Reported hate crimes peaked in the year of the September 11, 2001 attacks at 9,730 incidents. They then declined, hitting a low of 5,479 in 2014. They started growing after that by up to a few hundred annually – but the trend line rose sharply during the former President’s administration. From 2016 to 2017 hate crimes rose by over a thousand cases, and they rose by nearly the same amount again in 2020. In that year (the last numbers the FBI have reported), 8,263 hate crimes were recorded by the FBI, the highest number since 9/11. The rise is especially notable because 452 agencies that had previously reported statistics failed to report in 2020, meaning the actual increase is greater.15

White Supremacist Propaganda and Organizing

White supremacist propaganda and organizing have also grown dramatically. As measured by the Anti-Defamation League, the presence of public, openly white supremacist activity rose more than twelve-fold from 421 incidents in 2017 to 5,125 in 2020. While 2021 saw a slight drop, there were still around 13 incidents of white supremacist propaganda occurring every day – compared to just over one incident a day five years ago. White supremacists have also moved from the shadows to the mainstream: in 2021 there were at least 183 incidents where white supremacists hung banners on highway overpasses or other highly visible locations, a 40 percent increase from the year before. Organizing is also no longer clandestine: in 2021 white supremacists held 108 public events, more than double the year before and the most recorded in the last five years.16

2. Why the structure of political violence today is so dangerous to democracy

The growing acceptance of political violence and the volume of incidents is deeply worrying. However, as problematic is the structure of today’s political violence.

On the right, violence previously the purview of radical, fringe groups is now mainstream: From the 1970s until the mid-2010s, violence remained an activity of extremist groups on the fringes of society. In the late 1970s, for instance, only 6% of Americans justified violence for political goals.17

Throughout past decades, political violence in the U.S. operated in the classic, clandestine terrorist cell model. Intensely ideological organizations such as the Weathermen or Operation Rescue pulled recruits away from family and friends to deepen their connection to the violent group. In the late 1960s and 1970s, most political violence was committed by organizations on the far left acting in the name of perceived social justice, environmental, and animal rights causes, generally against property. From the 1980s through the 2010s, violence shifted towards the right with the rise of white supremacist, anti-abortion, and militia organizations, though environmental and animal rights attacks continued. Violent events diminished greatly from the 1970s peak, but lethality grew as targets shifted from property to people: minorities, abortion providers, and federal agents.

Today, the majority of individuals committing spontaneous or organized political violence do not formally belong to any radical group. Instead, violent beliefs and activities have moved into the mainstream. Ideas once confined to hardcore white supremacists – such as the Great Replacement theory – appear on Fox News and even in the halls of Congress.18 White supremacists have become commonplace on popular gaming sites – a 2019 study by the Anti-Defamation League found that 23% of respondents who played online multiplayer games had been exposed to white supremacist ideology and one in ten had encountered Holocaust denial.19 Women stumble into the violent Q-Anon conspiracy via a host of mothering blogs and Instagram yoga wellness pages. Once people begin to consider these violent ideas, a slippery language of memes, slang, and jokes blurs the line between posturing and supporting violence, slipping past mental brakes and normalizing activities and ideologies long viewed as beyond the pale of normal politics.

While desire is bipartisan, violence is overwhelmingly on the right, suggesting a role played by political leaders: Despite similar sentiments regarding violence among Democrats and Republicans from 2017 to pre-election 2020, incidents of violence are overwhelmingly on the right. The Global Terrorist Database (GTD), managed by National Consortium for the Study of Terrorism and Responses to Terrorism (START), a Department of Homeland Security Center of Excellence, is considered the most comprehensive global dataset for domestic, transnational, and international terrorist incidents since the 1970s. It charts a rise in violent far-right terrorist activity that begins in 2013 (possibly in reaction to the reelection of America’s first Black president) but skyrockets in 2016.

As START director William Braniff testified to the Congressional Homeland Security and Governmental Affairs Committee in 2019, “Among domestic terrorists, violent far-right terrorists are by far the most numerous, lethal and criminally active. Over the last several decades, they are responsible for more: failed plots; successful plots; pursuits of chemical or biological weapons; homicide events; and illicit financial schemes than international terrorists, including HVEs.”

Source: Global Terrorism Database, classification extended by author<sup>20</sup>

Similar trends are found in FBI statistics (the FBI’s investigations into white supremacists have tripled since 2017), the Joint Regional Intelligence Center of DHS, and independent counts from organizations such as ACLED. While the Global Terrorism Database is still compiling data from 2020 and 2021, and left-wing violence has risen, the difference appears to remain: in 2020, the Center for Strategic and International Studies found that violence from both sides reached the highest recorded levels since they began collecting data in 1994 – but right-wing incidents were still nearly three times as high (73) as left-wing (25).21

This discrepancy points to the role of political leaders. In both Israel and Germany, research has found that domestic terrorists are emboldened when they believe that politicians encourage violence or that authorities will tolerate it from their side of the political spectrum. Trends in U.S. political violence show similar patterns.22 The Oath Keepers, for example, issued a call to action as early as 2016 to “protect” the polls after then-candidate Trump’s claims of potential fraud. Militia groups mobilized in response to Trump’s statements surrounding his inauguration and with his rhetoric on border security.23 At regular intervals they provided security to Trump campaign rallies and events from 2016 to 2020, and in all of these instances there was no negative governmental reaction. At least one Oath Keeper has claimed that she was providing official security at the rally on the mall on January 6, 2020 for VIPs and legislators; Trump supporter Roger Stone appeared to use Oath Keeper militia members as part of his personal security outfit that morning.24 Given that history, it is reasonable that they would assume they were secure from prosecution for their violence at the Capitol on January 6, should Trump prevail in the 2020 election. Indeed, Stewart Rhodes’ statements suggest that he expected lenience directly from then-President Trump – specifically pardons for those arrested for the January 6 riots and money for their legal defense.25

The left is growing in anger and violence is inching up: While violent events to date have been overwhelmingly concentrated on the right, early evidence suggests that they rose for the left in 2020 and that partisans on the left are growing increasingly angry. The most recent December 2021 poll by Lilliana Mason and Nathan Kalmoe found increased justifications of violence from the left: 21% of Democrats (and 16% of Republicans) thought punching members of the other party was justified, while 13% of Democrats (and 9% of Republicans) justified killing at least some members of the other party. This emotion is not yet appearing in many actual incidents, which global research suggests would require that potential perpetrators feel a greater sense of impunity from prosecution. But these numbers indicate ominous trends were left-wing leaders to encourage violence, given the well-established effects on followers’ actions.

Extremist groups are using mainstream causes to recruit, expanding their membership: In the aftermath of the January 6 insurrection, daily internet monitoring showed right-wing violent extremists were encouraging members to use mainstream conservative causes and local rallies to increase recruitment while flying under the radar of national news. For instance, the Proud Boys, a European chauvinist group known for politically violent protest, held a number of rallies using anti-critical race theory to recruit new members. In Oregon, the Proud Boys, People’s Rights (founded by anti-government activist Ammon Bundy), and Patriot Prayer used Telegram to call on supporters to appear at a local school to oppose COVID mask requirements, triggering the school district into lockdown.26 Extremist organizations – particularly the Proud Boys, Three Percenters, and Boogaloo Bois, and their respective affiliates, were present at nearly half of all the armed demonstrations carried out for any cause in 2021, an increase from their presence at 1/3 of assorted rallies in the year prior.27 As these examples also demonstrate, violent organizations that are often in competition and which previously focused on different targets and goals are now beginning to collude under a more encompassing, melded ideology.28 Their recruiting tactics build on efforts over the last five years to mainstream violent extremism: the militia movement, which previously focused its violence on the federal government, for instance, mobilized against the Women’s March in 2016 and myriad Black Lives Matter marches in 2020.

As extremist organizations capitalize on mainstream concerns, more people enter a radical world that once required joining secretive fringe groups in-person. These trends are expanding the pool of potentially violent individuals, as well as those who might not be violent normally, but who will act violently when excited by the crowd dynamics of an emotional rally.

Targets are becoming partisan, not just ideological: In past decades, violent actors may have been ideologically left (such as many environmental and animal rights terrorists) or right (such as most militia members and abortion clinic attackers), but they were not overtly partisan. In the previous graph of GTD data showing political violence by ideology, many anti-Semitic, anti-government, pro-gun, and other types of attack were classified in past years as “single-issue”, rather than “right- or left-wing”, because while the ideology may be associated with a political side, the perpetrator showed no partisan affiliation or motivation, or seemed to have a mix of motivations from different sides of the political spectrum. Such violence also did not track the electoral calendar and was not used to affect the prospects of either political party.

Now, individuals committing violence on the right tend to be closely aligned with a political party and can be mobilized for partisan purposes. For instance, In November 2021, Bright Line Watch found that right-wing individuals who felt a stronger connection to the Republican Party endorsed concrete violent actions at significantly higher rates.29 Concerns that were once the purview of very separate extremist groups, or were ideological but not partisan – from policies concerning masks, vaccines, election rules, and school curricula, and encompassing dislike of Jews, Muslims, people of different races, immigrants, and other causes – are all triggers that now reinforce a shared identity that is also connected to a political party. For instance, several individuals who took part in armed counter-protests to BLM events have since been charged in the January 6 attacks. In contrast, Bright Line Watch found that on the left, violent sentiment was higher among respondents who felt more distant from the Democratic Party.30 While violence for causes associated with the left still appear disconnected from the Party and political cycle, post-election violence and intimidation on the right began to align with the political calendar in 2016 and 2018, and in 2020 became highly connected to election dates and procedures.31

Globally, democracies facing this pattern see intimidation, violence, and harassment rise and fall around each election cycle when elections are close or contentious. Election day itself is often not violent, though intimidation may take place. Instead, violence tends to ramp up in the months prior to election as politicians try to solidify bases of support within more homogenous political parties by creating distinctions with other racial, ethnic, or religious groups, often using dehumanizing rhetoric that inflames their supporters. Violence also occurs after an election but before the final settlement of power (Inauguration Day, in the United States), as politicians whip up anger and intimidation among their supporters to affect vote tallies or as a tool to jockey for the fruits of power. Violence then subsides until the next close election. What may appear to be spontaneous or mob-led is, upon inspection, closely linked to the electoral calendar and serves a political function.32 This a pattern that previously occurred in America during the rise of the anti-immigrant “Know-Nothing” party and again following Reconstruction among Southern Democrats. For more on this pattern, see my attached article from the Journal of Democracy.

While data for 2021 is not generally available (FBI data, for instance, is generally published in November or December of the following year); incidents appear to have lessened. Given the evidence suggesting that extremist organizing, armed protest, and violent beliefs are all high or increasing, this pattern should not lull anyone into complacency. It is instead a feature to be expected with violence that is encouraged by politicians for a political purpose.

Organized violence on the right is being committed by established community members: Most criminal violence globally is committed by young, unmarried, childless, unemployed men with low levels of education. This is also the demographic profile of the majority of Americans who commit spontaneous hate crimes, who also generally have prior criminal records.33

This is not the demographic of those involved in violence and intimidation at right-wing political events and armed demonstrations such as Stop the Steal rallies. These individuals are more likely to be middle-aged, married with children, middle class, and have jobs. The majority of those arrested for the January 6 insurrection belong to this more established demographic.34 They also gained most of their news from the mainstream media, not far-right or social media, and 25% have college degrees.35 Respondents who told the December 2021 Washington Post-University of Maryland poll that violence was justified were also more likely to have college degrees.36

The individuals who are taking part in violent right-wing or pro-Trump events and espousing violent political beliefs do not have the profile of extremists, criminals, or hate crime offenders. Indeed, many people arrested on January 6, like the majority of violent extremists today, have no history of violence and aren’t part of a violent group. Instead, America is facing a mainstreaming of violence among people who are well-established in their communities and who seem to view their violence not as a criminal act but as an extension of political behavior. That behavior is clearly influenced by the conspiracy that the 2020 election was “stolen”. In fact, far from being on the fringe, greater community involvement on the right has been found by AEI’s Survey Center on American Life to correlate with greater belief in conspiracies such as Stop the Steal and Q-Anon, as does church membership among white Evangelicals.37 Problematically, this was also the demographic picture as Nazi extremism mainstreamed among regular Germans in the 1930s – in fact, German towns with more civic associations saw a faster rise in Nazi Party membership.38

Damage to Our Democracy from these Political Violence Trends

These trends pose at least seven significant risks to our democracy:

  1. Violence, intimidation, and harassment can be used for political gain

    The melding of social causes with extremist groups and their connection to partisan politics have created a large population of Americans willing to consider engaging in violence and intimidation. Violence may be catalyzed by anger at a variety of social events such as mask mandates or gun policies that provoke a sense of threat to a shared identity. However, violence and intimidation can also be intentionally wielded as a partisan tool to affect elections and democracy itself. Politicians, media leaders, and organizations can amplify charged rhetoric in particular campaigns and geographies in order to increase harassment, intimidation, and the threat of violence against particular targets and thereby deter potential opponents or pressure officials to resign. For instance:
    • Adam Gonzalez and Anthony Kinzinger, 2 of 10 Republicans who voted to impeach Trump after January 6, are not running for reelection after threats of violence. In Gonzalez’s case, these included threats against his wife and young children. Immediately after his announcement, Trump tweeted “1 down, 9 to go!” suggesting that violent intimidation of opponents could be used to chase them from office.39
    • Brad Raffensperger, the Republican Secretary of State of Georgia who refused to “find 11,780” votes requested by then-President Trump, faced death threats against himself and his family that increased in late November, 2020 after Trump declared him an “enemy of the people." The family went into hiding following manifold FBI-investigated threats, a break-in intended to intimidate their widowed daughter-in-law, and the appearance of out-of-state Oath Keepers at their home.40 On January 6, 2021 Raffensperger had to be evacuated from his office by the Georgia Capitol Police as the head of Patriots USA, a former Ku Klux Klan member, searched for him through the capitol halls while armed protestors rallied outside.41 Raffsenperger interpreted the threats as an effort to compel his resignation, but when that did not occur, former-president Trump endorsed a serious primary challenger who supports false claims that the 2020 election was illegitimate.
    • The 13 House Republicans who supported greater funding for highways, bridges, and other infrastructure had their phone numbers publicized by their colleague Marjorie Taylor Greene. A number admitted to receiving threats afterward – providing a new set of tactics for “whipping” future votes and compelling Members of Congress to vote against the wishes of their districts.42

    While violence can be used by politicians, we know from other countries with similar dynamics that it cannot be controlled by them. Mobs can turn against the individuals and factions using them, just as quickly as they can turn against others whom they conceive of as enemies.
  2. Americans may be intimidated from political service, particularly women, parents, and minorities

    Threats and violence, such as the illustrative cases above, are being made against Republican men who upheld the results of a valid election in 2020. However, research by the Institute for Strategic Dialogue found that during the 2020 election season, female Democratic politicians received ten times the number of abusive Facebook comments as male Democrats and female Republican politicians were subject to twice as much abuse as male Republicans. Being a visible minority made abuse even more likely.43 Similarly, a 2019 study of mayors found that 13% had faced physical violence, with female mayors facing nearly three times the rate of violence as male counterparts.44

    Intimidation and actual violence have already caused a number of politicians to retire. While some politicians may choose to downplay the effects of threats against themselves as “part of the job”, harassment may particularly dissuade parents who fear threats directed against their young children, such as Al Schmidt, the Republican election board member in Philadelphia who resigned after being singled out by then-President Trump and targeted with threats so significant that his children were given Philadelphia police protection, and Lauren McLean, the mayor of Boise, Idaho and a mother of two who now has a security detail after two years of threats including armed mobs with torches and pitchforks who have come to her home. A poll by the National League of Cities found that 81% of local leaders have experienced threats or violence in the last few years.45
  3. Elections risk becoming less free and fair and more prone to mistakes

    Threats of violence against election workers, formerly almost nonexistent, are now frequent – a 2022 Brennan Center poll found that 1 in 6 election officials had experienced threats. Half were threatened in person – not just online or on the phone. The DOJ has received reports of more than 850 threats against election officials since the 2020 elections, but the Brennan Center poll found that more than half the cases of threat had not been reported to law enforcement, suggesting actual intimidation is far higher.46

    A national 2020 survey found that the “political environment” was the reason given by nearly a quarter of election officials planning to retire – and that survey occurred prior to the death threats and harassment that skyrocketed since the election.47 In California, one of the few states for which we have numbers, 15% of election officials retired between the 2020 election and June 2021.48 By 2022, the Brennan Center found that nationally, 1 out of 3 election officials knew at least one official who had left the job because of fear or threats.

    Mass retirements threaten to inflict unintended errors into U.S. elections. America’s elections are the most decentralized in the world. They rely on large numbers of workers and volunteers who must master highly localized information and systems to ensure smooth elections across approximately 10,000 jurisdictions. They also increasingly involve cybersecurity and computer savvy, with hacking, mis- and dis-information, computerized ballots, and the designation of elections as Critical Infrastructure. The median election official has been working in her role for over 12 years – meaning that half have served even longer; in the largest districts, most have served 15 to 20 years. As professionals are threatened out of their jobs, new hires will be challenged to master the arcana of local requirements, opening elections to more human errors in an environment of partisan recrimination, which will further erode trust in our electoral system.

    Election officials also risk being replaced by more partisan officials whose false beliefs in fraud may lead them to engage in purposeful malfeasance – such as Colorado’s Mesa County Clerk, who has been indicted for tampering with voting machines. As of this writing, individuals who deny that the 2020 election was legitimate are running for Secretary of State in 18 states in 2022, representing 2/3 of all Secretary of State elections that year. In seven states, election deniers are running for Governor, Attorney General, and Secretary of State.49

    The growth in MAGA-affiliated election deniers seeking these jobs may be intentional: former-President Trump’s communications infrastructure has particularly focused followers on doing something to counter his claims of election fraud, and former advisor Steve Bannon has used his popular podcast to encourage listeners to run for formerly arcane local election roles. Such actions suggest a political intent behind intimidating highly qualified election officials from both parties into premature retirement.50
  4. America risks losing public health workers and other critical functions

    While violence is being used politically, it is also being mobilized around a variety of social causes by mainstream and extremist groups. In the last two years, overlapping and expanding groups of Americans have been organized around vaccine and mask mandates, school curricula, and other cultural flashpoints. As already discussed, violent extremist groups such as the Proud Boys have used these social causes to recruit greater numbers of people to their extremist movement. In states that had stay-at-home orders for more than ten days during the spring 2020 coronavirus period, Moonshot, an anti-extremist internet monitoring company, found a 21% increase in engagement with white supremacist content.51

    Public health officials were the first to feel the brunt of this overlap between mainstream concerns and extremist propaganda. A CDC survey from June 2021 of over 25,000 state and local public health workers found that 11.8 percent had received work-related threats since March 2020.52 These numbers hide intense geographic pockets: for instance, 80% of the public health directors polled by the Colorado Association of Local Public Health Officials reported threats to themselves or their private property in the first three months of the pandemic.53 As of October 2021, over 500 state and local public health officials had left their positions – the greatest sudden loss of health officials in American history.54

    Election officials were targeted next, then school boards. However, it’s important to see these different targets as part of the same strategy: mobilizing against an ever-increasing set of enemies that are also being singled-out by mainstream conservative outlets has proven to be a useful recruitment tactic for extremist groups, suggesting that other professions will be targeted for harassment and intimidation in the future. The result is likely to harm one institution after another as different parts of our governmental and community fabric find it increasingly difficult to recruit and retain candidates and staff.
  5. Violence is likely to increase over time

    Violence tends to lead to more violence. While some are disgusted by violent acts, others come to see violence as heroic or simply more possible. Thus, in the week after the 2017 white supremacist Unite the Right rally in Charlottesville, Moonshot’s internet monitoring found a 40% increase in searches suggesting a desire to kill Black Americans, a 200% increase in searches indicating a desire to kill ethnic minorities, and an 1,800% increase in searches indicating a desire to kill Jews. They also uncovered 22,571 searches suggesting a desire to take part in a violent group, a 400% increase. These feelings do not remain online: that year also witnessed an actual 17% increase in FBI-reported hate crimes. Similarly, Nathan Kalmoe and Lilliana Mason happened to be conducting one of their regular surveys during the 2018 midterm election season over a period that encompassed a series of mail bombs sent by a Trump supporter to a variety of Democratic targets and the Pittsburgh synagogue shooting. Two days after the synagogue shooting, the scholars recorded the highest rates of approval for violence to that date.55 Moonshot found a sudden rise in extremist search traffic after the Pittsburgh shooting similar to the one that occurred after Charlottesville’s rally. While these repeated spikes subside quickly, Moonshots’ internet monitoring data suggests they remain elevated above prior levels, growing slightly with each event.56
  6. It is becoming less safe to exercise our freedoms of speech and assembly

    The increase in political violence has made protests increasingly dangerous, damaging Americans’ rights to free speech and assembly. Plowing cars into civilians used to be a tactic favored by overseas terrorists. It had been recorded just twice in the United States before James Alex Fields Jr. murdered Heather Heyer by driving into a crowd of counter-protestors at the Charlottesville Unite the Right rally. Yet from George Floyd’s murder on May 25, 2020 through September 30, 2021, at least 139 drivers drove their cars into protests across America, injuring 100 – sometimes severely – and killing four.57 One reason the tactic may be increasing is a lack of accountability – many drivers have not been charged, and three states – Florida, Iowa, and Oklahoma – have passed laws since 2017 that provide drivers with some legal immunity for hitting protestors “unintentionally”. In Oklahoma, drivers are immune from civil and criminal penalties.

    The vast majority of attacks have been from the right against protests for racial justice and other left-aligned causes, though police themselves have driven into protests at least 11 times during this time period. However, there are instances of use by the left, and in addition to the 139 cases targeting protestors, five police have also been struck by private vehicles at protests during the same time period.58
  7. Paramilitary groups may become a part of political life

    In weaker democracies, armed militias that lean towards a political party but whose actions can be denied and distanced by politicians are common – from Sinn Fein’s role in Northern Ireland during the Troubles to India today. Globally, politicians often outsource coercion and intimidation to professional instigators who specialize in the use of violence, similar to outsourcing robocalls and direct mail.59 Repeated security provision from the Oath Keepers militia to Trump campaign rallies – such as those mentioned earlier in Texas, Minnesota, and Washington D.C. – suggest that such alignment is possible even in the United States.60 In India, whose once-vibrant democracy provides the closest similar case, the trend has been for the violence specialists to eventually dispose of their political middlemen and run for office themselves.61 To understand where this can lead: 11 of India’s current national legislators face open cases for murder, 30 have attempted murder charges and 10 serving legislators have been convicted of such serious crimes – a doubling from ten years ago.62

    In America, the growing acceptability of paramilitary groups in political activity is not limited to the Oath Keepers. The 1st Amendment Praetorian (1AP), has been repeatedly linked to the Trump campaign, including providing security for former Trump National Security Advisor Michael Flynn at Stop the Steal rallies.63 Moreover, the normalization of this phenomenon has grown beyond the Trump campaign: in 2017, the Portland branch of the Republican Party voted to allow militia groups, including members of the Oath Keepers and Three Percenters, to act as security at their public events.64 In Idaho, the National Guard supported a legislative bill to repeal the state’s anti-militia law.65 Symbolic acceptance is also growing. In Texas, Allen West, then-Chair of the Republican Party, posed with armed militia members just days after the January 6 riot, and appeared at a rally with the leader of the Oath Keepers militia in March after the latter was under investigation for his involvement in the January 6 attack.66

3. Addressing Political Violence and Safeguarding Democracy

Today, violence and intimidation are being used to frighten people from exercising their full rights as citizens and elected officials. These tactics are increasingly practiced by people who don’t fit the previously understood behavioral characteristics of terrorists or criminals, are only occasionally members of extremist groups, and are not isolated or particularly radicalized – in fact, they are better seen as aggressive individuals swimming within a sea of similarly-minded, otherwise mainstream community members. This has three important implications that make fighting this phenomenon more difficult and argues for rapid intervention before it metastasizes further:

  • Strategies to help individuals leave violent groups are still needed, but are less likely to succeed at scale

    Globally, established methods to pull people out of extremist groups involve strengthening radicalized individuals’ bonds to mainstream society. Typical terrorists, guerrillas, and gang members are attracted to these groups at moments when their identities are destabilized – such as during adolescence or after job loss. They seek to bolster a sense of self within an encompassing community. Many young offenders – who are the majority of violent criminals – simply age out of violence. Meanwhile, becoming educated, holding a job, being married, and having children are all factors that protect against violent group membership, and thus are common tactics used intentionally by programs working to pull individuals out of violent gangs, terrorist, and other extremist groups.

    This makes the growth of violence by individuals who are middle-aged, married, with jobs, education, and children particularly worrisome. Their demographics defy the well-tested strategies used to help groups of violent people return to society. The standard paths to delink violent individuals from radical fringes and reconnect them to their communities do not work if the fringe has become mainstream such that the broader community supports many of their beliefs. And while different tactics exist to pull individuals out of cults and conspiracies, these methods are individualistic and cannot easily be scaled.
  • Violence is likely to be more common

    The majority of political violence is now committed by people who do not belong to violent organizations. This is likely to make it more virulent. Violent groups tend to fall apart of their own accord through infighting, corruption, sexually divisive behaviors, or criminal activity; movements are not subject to these forces. Moreover, in-person cells of violent extremists frequently hold back their own violence. Because members of extremist cells are often family, close friends, or community members, they do not wish to attract law enforcement attention that could lead to arrest. While the tendency of such cells to inhibit actual violence often leads to splintering into more violent factions, it can also suppress violence. The smaller splinter groups that do move forward with violent plans involve fewer individuals who can be targeted with surgical interventions. We know that right-wing extremist organizations with weak or decentralized leadership are more likely to use violence, and that followers are more violent than leaders – logically, ungrouped movements are likely to be more violent still.67 Meanwhile, law enforcement interventions are more possible because splintering often creates informants. Large numbers of separate individuals united only by common cause or a sense of camaraderie on the internet have no such natural brakes.
  • Law enforcement strategies are necessary, but not at all sufficient

    Law enforcement strategies were crucial to reducing a wave of violent white supremacists in the 1980s and the growth of militias in the 1990s – however, these strategies are likely to be less effective today unless combined with other interventions. Previous law enforcement successes relied on infiltrating violent groups and gaining informants. This is still possible - indeed, among the subset of plots perpetrated by extremists in the military, those that involved groups of people were more frequently foiled.68 But most violent incidents today are perpetrated by individuals who are not part of any organized group, making informants harder to find, and plea-bargain-based prosecutorial methods (used to amass evidence against ringleaders) less possible. Meanwhile, law enforcement often relies on Racketeer Influenced and Corrupt Organizations (RICO) laws that were crafted to fight organized crime in order to prosecute violent groups. This is still feasible against organized groups like the Oath Keepers, as evidenced by a number of cases involving January 6 conspirators. However, movements and ideologies don’t fit easily into RICO categorizations. Those who instigate violence but do not run a formal organization or commit violence themselves may particularly be more difficult to prosecute.

What Congress Can Do

I urge this Congress to take seriously the danger that increased acceptance and acts of political violence pose to the American public and to our democratic system. Congressional action should focus on three goals:

  • Reduce the threats from violent individuals and those on the path towards violence;
  • Reduce the normalization of violence within mainstream society;
  • Reduce the incentives for political actors to engender politically-motivated violence for fundraising and electoral purposes.

Actions that Congress could take to meet these goals include:

  1. Increase accountability via:
    • Federal legal penalties for individuals who threaten or use violence against election officials and elected officials, their families, and their property, including political party officials and party headquarters. Not all violence is equal. While most violence is directed towards a unique victim, some is intended to intimidate a group of Americans, which is why hate crime laws delineate these acts differently from similar violence that lacks animus against a group. Intimidation and violence against elected officials, election officials, their families, and political parties are not only crimes against these individuals, but are violence that affects our democratic system. They should be recognized as a special class that is particularly heinous and penalties should be federal and meted out accordingly.

      Such a law should make it a federal crime to intimidate, threaten, or coerce federal, state, and local elected officials, election officials, and their families, including volunteers and people processing or scanning ballots, tabulating, canvassing, certifying, or officially auditing election results. It should carry federal penalties for violence and threats of violence against party officials and offices and should also ban the publishing of personal information for malicious purposes of all of these classes of individuals and immediate family members to prevent third-party threats via doxing. Finally, it should include federal penalties for intentionally damaging or threatening to damage physical property being used as a polling place, tabulation center, storage facility, auditing location, or other election infrastructure involved in voting, the handling of voter registration, or ballot information. It could include party offices, legislative, and judicial buildings because of their symbolic and real role in our democracy. These crimes should include criminal penalties as well as civil enforcement mechanisms to enable preventative action and to provide harmed individuals with legal standing.

      Because legislation without enforcement is toothless, Congress should also legislate reporting requirements compelling the Department of Justice (DOJ) to publicly report on actions taken against these threats, with data broken down by elected officials, candidates, election officials, and political parties.
    • Federal legislation banning private militias. Paramilitary activity now frequently involves travel by individuals across state lines. While 29 states have anti-militia statutes, 25 have regulations on paramilitary activity, and all states have some form of prohibition on private militia activity, these prohibitions are often unknown or poorly understood by the state officials who must implement them. They are also frequently hard for many law enforcement agencies that lack complex investigatory capabilities to enforce, particularly in states where implementing legislation doesn’t exist or attorneys general lack general criminal law enforcement authority. In addition to these enforcement difficulties, the lack of such a federal law allows the public misunderstanding to continue that private militias are lawful. A federal statute should involve criminal penalties as well as civil enforcement mechanisms.69
  2. Dramatically increase funding for:
    • Elections, which require a significant increase in dedicated federal funding along the lines of the current White House budget to meet needs that have become vastly more complex in the face of massive mis-and dis-information and violent threats to state and local staff.
    • Programs that strengthen individual and community resilience to violence and protect targeted groups and their communities. This would include significantly expanding funding pools such as the Targeted Violence and Terrorism Prevention Grant Program, though legislation should also explore administering these grants through non-security agencies such as the Department of Education or Health and Human Services, while ensuring through safeguards and oversight that communities do not feel stigmatized.
    • Programs that help individuals and communities avoid recruitment by violent extremists, such as those that provide inoculation online and in university and other group settings where violent extremist recruitment may be high; Veterans Administration programs to particularly assist that community in resisting extremism and recruitment into violent groups; mental health interventions for those at risk of violence who have greater than average propensity for mental health needs; and programs that foster rehabilitation and reintegration of formerly violent individuals who have renounced that path.
    • Support for refugee resettlement and immigrant integration. The counties showing the highest rates of spontaneous hate crimes and sending violent insurrectionists to Washington D.C. on January 6 are places with declining white populations and higher rates of immigrant settlement.70 This demographic change appears to be a particular trigger for violence. Integration programs could help.
  3. Craft a political pact to reduce violent rhetoric and imagery among candidates and Members of Congress

    Study after study shows that political leaders’ rhetoric is particularly influential in normalizing violence among their followers, inflaming already angry people, and focusing those inclined to violence on particular targets.71 Candidates are among the major offenders in mainstreaming violent rhetoric and actions. Thus, one of the most important things Congress could do to reduce the growing acceptance and undertaking of political violence is to stop Members and candidates for Congress as well as down-ballot candidates in their parties from encouraging it themselves. Prominent members of each party should make regular statements denouncing violence from their own side – shown to have far more impact than denouncements focused on their opponents. Meanwhile, Congressional and political party leadership should agree to a cross-partisan pact, with teeth, for Members of Congress and ideally party candidates that defines rhetoric that 1) encourages violence, 2) enhances the likelihood of violence, or 3) dehumanizes fellow Americans; censures candidates and party officials who use such rhetoric or imagery, and includes disincentives to do so, such as banning access to pooled or party funds, refusing to offer endorsements in primary campaigns or even the general election, stripping committee assignments, refusing co-sponsorship of legislation, and so on. A bipartisan adjudication panel of former Members or an independent bipartisan panel could be created to determine whether rhetoric, ads, imagery, or actions reach the level of censure.
  4. Reduce political incentives for violence through ECA Reform and majority winner laws

    Winner-take-all election systems such as America’s experience far more violence globally than other election systems – one reason that Northern Ireland altered its election system as part of the peace treaty that ended its Troubles. It is obvious that systems in which winning one extra vote, even with just a plurality of voters, allows a candidate to have a total win of the district incentivize violence and other efforts to game the system. To reduce systemic incentives for violence and political extremism, Congress should:
    • Pass a law that
      • Requires states to elect members of Congress by majority rather than plurality winners; and
      • Provides funding to encourage states to experiment with various forms of majority-winner systems, from conventional to ranked-choice and other forms of ballot and election procedures.
      • Clarify the Electoral Count Act, to reduce incentives for using political violence in order to achieve victory after votes have been cast.
  5. Reduce the appearance of extremist leanings or bias among law enforcement

    Law enforcement agencies that are believed to lean towards one side of a partisan, ideological, or extremist divide are a major global factor known to drive political violence, because these leanings indicate to violent individuals that their actions are more likely to go unpunished. While the vast majority of America’s 800,000 law enforcement officers uphold the rule of law, the U.S. nevertheless faces this dilemma. A 2006 FBI intelligence assessment titled “White Supremacist Infiltration of Law Enforcement” raised the problem, and in 2015 a second report identified “active links” between some law enforcement officers and subjects of investigation who were white supremacists, militias, and other extremists. The problem has been written about by multiple police chiefs and was discussed at the Congressional hearings on “Confronting Violent White Supremacy” by the Committee for Oversight and Reform on May 26, 2021.72

    Officers need not be extremists themselves to augment political violence. An atmosphere of friendliness towards militias or partisan bias also has consequences: a lawsuit in Texas alleges with telephone evidence that a local police department intentionally failed to come to the aid of a Biden campaign bus that was boxed in and at risk of being forced off the highway by a caravan of cars known as a “Trump train”.73 Despite evidence that right-wing protests are twice as likely as left-leaning protests to turn violent and despite the rise in far-right political violence and hate crimes, U.S. police intervened two-thirds more frequently in left-leaning than right-leaning protests in 2020, and in 2021, intervention towards right-wing protests declined even further.74

    However, America’s policing system, like its elections, is the most decentralized in the world. Thus, to address extremists in their ranks or perceptions of partisan bias within law enforcement agencies, cash-strapped municipalities must allocate scarce funds away from schools, parks, or daily police work toward police internal investigations. The Department of Justice can review local police agencies, but there are around 18,000 jurisdictions – its nationwide ambit and necessarily limited resources mean that many cases will be overlooked. Moreover, federal action can generate backlash. Congress could:
    • Adapt the 2021 DOD directive on extremism and legislate its applicability to the 26,000 federal law enforcement officers who currently operate in over 65 different agencies.75 The legislation should particularly encompass a constitutionally and personnel appropriate adaptation for law enforcement of DOD Instruction 1325.06 “Handling Protest, Extremist, and Criminal Gang Activities Among Members of the Armed Forces” and should include sections on screening, regular training on prohibited extremist activity, and procedures for reporting extremist behaviors.
    • Make federal grants to local police, primarily provided through the COPS and Byrne grant programs of the DOJ, contingent on the adoption of similar guidelines for local forces, with constitutionally appropriate language that can help local departments adopt such rules.76
    • Pass legislation granting state attorneys general the authority to initiate pattern or practice suits against local law enforcement agencies in their states using the federal courts, while augmenting DOJ funding for pattern or practice lawsuits to allow for greater investigation of departments that may be aligned or perceived to be aligned with private violence.77

To ensure that these measures are having the intended effect, and to encourage momentum within the agencies intended to carry out this work, Congress should also enhance the requirements on public reporting of domestic terrorist violence. Congress should require the regular publishing of publicly available statistics on violent terrorist acts in the United States such as those required in the FT2020 NDAA, broken down by ideology such as that used by the Global Terrorism Database, as well as reporting on threats against public officials, election officials, and political parties. They should also require the aforementioned DOJ reporting on action taken against threats.


The former President’s refusal to accept the results of a free and fair election – foreshadowed in 2016 before he was certain of his victory, and brought to fruition by the Stop the Steal movement in 2020, have caused great harm to American democracy. Conspiracy theories, (defined as explanatory beliefs about a group of people who collude in secret for malevolent goals), are deeply detrimental to democracy. Their tendency to find malevolence in governing elites alienates believers from democratic politics. They also tend to blame minority groups for societal problems, which can pave the way for marginalization of and violence against these groups. These two tendencies together mean that conspiracy theories are correlated internationally with the formation of political extremist groups.

One of the most consistent findings in the psychology of conspiracy theories is that a person who believes in one is more likely to believe in more. Over time, governing authorities begin to seem fundamentally deceptive, various minorities appear threatening, and future conspiracies become more and more plausible in this light. Moreover, once conspiracy theories take hold, they are extremely difficult to change.78

The deliberate sowing of a conspiracy such as Stop the Steal has thereby opened the body politic to further conspiracies, such as the violent Q-Anon movement and lies perpetrated by Russia, Iran, and other foreign powers. It has decreased trust in democratic institutions and increased harm to minorities. It has also increased the normalization of political violence in America and caused political violence and extremism to burgeon.

America faces a political violence problem that, though violence is currently low, is already causing serious damage to our democracy. Thanks in large part to conspiracies perpetrated by the former President and those close to him, the sheer number of Americans now willing to consider violence has grown precipitously. Even if only a small percentage are likely to act, they constitute tens of millions of individuals. Numbers this large put the country at risk for stochastic terrorism – a situation in which it cannot be predicted who will respond with violence when a political leader calls for it – but it is nearly statistically certain that someone will.

The time to tackle this risk is between election cycles when violence recedes and emotions are cooler. That time is now.


1 The precise questions were “When, if ever, is it OK for [own party] to send threatening and intimidating messages to [opposing party] leaders?”; “When, if ever, is it OK for an ordinary [own party] in the public to harass an ordinary [opposing party] on the internet in a way that makes the target feel unsafe” and “How much do you feel it is justified for [own party] to use political violence in advancing their political goals these days?” Kalmoe, N. P., & Mason, L. Radical American Partisanship: Mapping Violent Hostility, Its Causes, and the Consequences for Democracy (Chicago: University of Chicago Press, 2022), 88.

2 Hayes, B.C., and McAllister, I. “Sowing Dragon’s Teeth: Public Support for Political Violence and Paramilitarism in Northern Ireland.” Political Studies, 49, (2001): 901-922.

3 Bartels, L. M. “Ethnic antagonism erodes Republicans’ commitment to democracy.” Proceedings of the National Academy of Sciences of the United States of America, 117(37), (2020): 22752–22759. https://bit.ly/3tXzr3W; Cox, D. “After the ballots are counted: Conspiracies, political violence, and American exceptionalism.” The Survey Center on American Life. Feb. 11, 2021. https://bit.ly/3qTqSoB “Understanding the American Insurrection Movement: A Nationally Representative Survey.” 2021. NORC. University of Chicago. https://bit.ly/3IVNPO8

4 “#StopTheSteal: Timeline of Social Media and Extremist Activities Leading to 1/6 Insurrection.” Feb. 10, 2021. Just Security. https://bit.ly/36OUEnA

5 Kalmoe, N. P., & Mason, L. Radical American Partisanship: Mapping Violent Hostility, Its Causes, and the Consequences for Democracy (Chicago: University of Chicago Press, 2022), 163-165.

6 “Understanding the American Insurrection Movement.” NORC University of Chicago. 2021. bit.ly/3IVNPO8

7 Because men commit the vast majority of violence, percentages for men are likely to be higher than for Republicans of both genders or Americans as a whole. They are also, for the same reason, more indicative of actual violent trends. “Tempered expectations and hardened divisions a year into the Biden presidency,” Bright Line Watch, Nov. 2021 Survey, https://bit.ly/3NCZwgs; “Americans’ views on violence against the government,” Report #80. The Covid States Project, Jan. 2022. https://osf.io/753cb/.

8 Kalmoe, N. P., & Mason, L. Radical American Partisanship: Mapping Violent Hostility, Its Causes, and the Consequences for Democracy (Chicago: University of Chicago Press, 2022).

9 Wire, S. D. “Threats against members of Congress are skyrocketing. It’s changing the job.” Los Angeles Times. Sept. 20, 2021. https://lat.ms/3wTYI0r; Edmondson, C., & Walker, M. “One menacing call after another: Threats against lawmakers surge.” The New York Times. Feb. 9, 2020. https://nyti.ms/3uJPiSHl

10 Report: Election 2020 “Political Violence Data and Trends.” Bridging Divides Initiative. 2021. https://bit.ly/3qQjPgk

11 “Fact Sheet: Updated Armed Demonstration Data Released A Year After the 6 January Insurrection Show New Trends.” Jan. 5, 2022. ACLED. bit.ly/35rsVc0. The Armed Conflict Locations and Event Data is a research organization supported by the U.S. State Department in many other countries to track armed conflict and which is supported by private donations for its work in the United States.

12 Ibid.

13 “Report: Election 2020 Political Violence Data & Trends.” Bridging Divides Initiative. 2021. https://bit.ly/3qQjPgk

14 Ibid.

15 FBI Hate Crime Statistics: Incidents and Offenses. 2000-2020.

16 “White Supremacist Propaganda Remained at Historic Levels in 2021.” Anti-Defamation League. March 3, 2022. https://www.adl.org/news/press-releases/adl-white-supremacist-propaganda-remained-at-historic-levels-in-2021

17 Even this number was in response to a highly generous question that: “using violence to achieve political goals was sometimes the only way to get injustices corrected.” McClosky, H., & Brill, A. The Dimensions of Tolerance: What Americans Believe About Civil Liberties. Russell Sage Foundation, 1983.

18 Vlamis, K. “GOP Rep. Matt Gaetz said Tucker Carlson is correct about white nationalist “replacement” conspiracy theory and called ADL racist.” Business Insider. Sept. 26, 2021. https://bit.ly/36I4is8

19 “Free to Play? Hate, Harassment, and Positive Social Experiences in Online Games.” Anti-Defamation League. July 2019. https://www.adl.org/free-to-play

20 Where the motive was unknown and unidentifiable, I have removed cases from this ideological classification rather than add an “unknown” line. Perpetrators whose attacks appeared to be motivated by anti-abortion or pro-choice ideology, incel/men’s rights, anti-white, anti-Semitic, anti-police, and other goals without a perceptible political party belief set were classified as “single-issue”; some attackers in this classification appear to be unstable, for instance, seem to be equally angry at all political parties, or hold conspiracy beliefs that meld far-left and far-right ideologies. Similar attacks with a discernable partisan leaning were classified as Far Left or Far Right.

21 See William Braniff, “Countering Domestic Terrorism: Examining the Evolving Threat,” Testimony to the Homeland Security and Governmental Affairs Committee, Sept. 25, 2019. https://bit.ly/3NGdqOD. Seth G. Jones, Catrina Doxsee, Nicholas Harrington, Grace Hwang, and James Suber’s research showcased in O’Harrow, R., Tran, A. B., & Hawkins, D. “The rise of domestic extremism in America.” Washington Post. April 12, 2021. https://www.washingtonpost.com/investigations/interactive/2021/domestic-terrorism-data/

22 Perlinger, A. “Terror isn’t always a weapon of the weak—It can also support the powerful.” The Conversation. Oct. 28, 2018. https://bit.ly/3LNBu0z. Perliger, A. American Zealots: Inside Right-Wing Domestic Terrorism. (New York: Columbia University Press, 2020).

23 McCord, M. “Armed militias are taking Trump’s civil war tweets seriously.” Lawfare. Oct. 2, 2019. https://www.lawfareblog.com/armed-militias-are-taking-trumps-civil-war-tweets-seriously

24 Polantz, K. “Oath Keeper claims she was VIP security at Trump rally before riot and says she met with Secret Service agents. “CNN. Feb 21, 2021. https://cnn.it/3LyxiBk; Mosk, M., Rubin, O., Dukakis, A., & Gallagher, F. “Video surfaces showing Trump ally Roger Stone flanked by Oath Keepers on morning of Jan. 6.” ABC News. Feb. 5 2021. https://abcnews.go.com/US/video-surfaces-showing-trump-ally-roger-stone-flanked/story?id=75706765

25 See for instance: Garcia, N. “Militia-style group, Oath Keepers, will be in Dallas for Trump rally.” Dallas News. Oct. 11, 2019. https://bit.ly/3NwdTTH; “Armed “anti-government” militia to provide escort for Trump supporters.” TRT World. Oct. 10, 2019. https://bit.ly/3wNOVcu Rhodes quotes in Hsu, S. S., & Barrett, D. “Oath Keepers founder Stewart Rhodes arrested for Jan 6 riot.” Washington Post. Jan. 13, 2022; https://wapo.st/3qQYvaS Hoft, J. “Dispatch from the Gulag: Oath Keepers Founder Stewart Rhodes Discusses Ray Epps with The Gateway Pundit” (AUDIO). The Gateway Pundit. March 1, 2022. https://bit.ly/3JVRWLj

26 Holt, J., Kann, Alyssa, Rizzuto, M.” After the insurrection: How domestic extremists adapted and evolved after the January 6 US Capitol attack.” Digital Forensic Research Lab, & Atlantic Council of the United States. 2022. https://www.atlanticcouncil.org/wp-content/uploads/2022/01/After-the-Insurrection.pdf

27 “Fact Sheet: Updated Armed Demonstration Data Released A Year After the 6 January Insurrection Show New Trends.” ACLED. Jan. 5, 2022. bit.ly/35rsVc0

28 “Standing by: Right-wing militia groups and the US election.” ACLED. Oct. 21, 2020. https://bit.ly/3tVyxEP

29 Tempered expectations and hardened divisions a year into the Biden presidency,” Bright Line Watch, Nov. 2021 Survey, https://bit.ly/3NCZwgs

30 Ibid.

31 Armed Conflict Location & Event Data Project. (2022). United States of America (18 March 2022) [Data Set]. ACLED, detailed in their report: US Crisis Monitor Releases Full Data for 2020. ACLED. Feb. 5, 2021. https://acleddata.com/2021/02/05/us-crisis-monitor-releases-full-data-for-2020/;

32 Birch, S., Daxecker, U., & Höglund, K. “Electoral violence: An introduction.” Journal of Peace Research, 57(1), 2020: 3–14.

33 Jensen, M., Yates, E., & Kane, S. “Motivations and Characteristics of Hate Crime Offenders.” START. 2020 https://bit.ly/3uLL9Oc; and Profiles of Individual Radicalization in the United States dataset, both maintained by START and detailed in Braniff (2019).

34 Jensen, M., Yates, E., & Kane, S. “Radicalization in the Ranks.” START, 2022, 45. https://bit.ly/3NCCIx1

35 Pape, R. A. “Deep, Divisive, Disturbing, and Continuing: New Survey Shows Mainstream Community Support for Violence to Restore Trump Remains Strong.” University of Chicago. 2022. https://bit.ly/3iSzsQ9

36 Washington Post and University of Maryland. (2021). Dec. 17-19, 2021, Post-UMD crosstabs. https://docs.google.com/spreadsheets/d/1CyE_aWwBrcvS44t9kRa4eK2sF218GEkMdTaFvjIwTU/edit#gid=0

37 Sweeney, M. M., & Perliger, A. “Explaining the Spontaneous Nature of Far-Right Violence in the United States. Perspectives on Terrorism,” 12(6), 2018: 52–71.; Pape, R. A & The Chicago Project on Security and Threats. April 2021. “Understanding American Domestic Terrorism: Mobilization Potential and Risk Factors of a New Threat Trajectory.” University of Chicago. bit.ly/3IVNPO8; Cox, D. A. “Social isolation and community disconnection are not spurring conspiracy theories.” The Survey Center on American Life. March 4, 2021. https://bit.ly/3uJymvz

38 Satyanath, S., Voigtländer, N., & Voth, H.J. “Bowling for Fascism: Social Capital and the Rise of the Nazi Party.” Journal of Political Economy, 125(2), 2017: 478–526.

39 McLaughlin, S. “Trump celebrates Gonzalez’s exit: ‘1 down, 9 to go!’” The Washington Times. Sept. 17, 2021. https://www.washingtontimes.com/news/2021/sep/17/trump-celebrates-gonzalezs-exit-1-down-9-go/

40 So, L. “Special Report: Trump-inspired death threats are terrorizing election workers.” Reuters. June 11, 2021. https://reut.rs/3JWp3Pq

41 Francis, M. . “Raffensperger escorted out of Ga. Capitol after it was surrounded by pro-Trump mob.” Yahoo News. Jan 7, 2021. https://yhoo.it/3Ds51cS

42 Zremski, J. “Reed, other GOP backers of infrastructure bill report getting threats.” Buffalo News. Nov. 15, 2021. https://bit.ly/3Ds59Jo

43 Guerin, C., & Maharasingam-Shah, E. “Public Figures, Public Rage: Candidate abuse on social media.” Institute for Strategic Dialogue. 2020. https://bit.ly/3uLjs7X

44 Thomas, S., Herrick, R., Franklin, L. D., Godwin, M. L., Gnabasik, E., & Schroedel, J. R. “Not for the Faint of Heart: Assessing Physical Violence and Psychological Abuse against U.S. Mayors.” State and Local Government Review, 51(1), 2019, 57–67.

45 Claudio, A. “Mayor Details Alarming Public Harassment of Her and Her Family.” Route Fifty. March 4, 2022. https://bit.ly/3IVmmfI; “ New Report: Harassment, Threats and Violence Directed at Local Elected Officials Rising at an Alarming Rate.” National League of Cities. Nov. 10, 2021. https://bit.ly/37300eI; Mekouar, D. “For Women in Politics, Being Terrorized Comes with the Job.” VOA News. Aug. 23, 2019. https://bit.ly/3tV81LR

46So, L., & Szep, J. “Reuters unmasks Trump supporters terrifying U.S. election officials.” Reuters. Nov. 9, 2021. https://reut.rs/3iRkAkX; Local Election Officials Survey. Brennan Center for Justice. March 2022. https://bit.ly/378Uo2U

47 Gronke, P. Manson, P., Lee, J. and Creek, H. “Understanding the Career Journey of Today’s Local Election Officials and Anticipating Tomorrow’s Potential Shortage.” Democracy Fund. April 20, 2021. https://bit.ly/3JVMTe3

48 Gordon, G. “Documenting and Addressing Harassment of Election Officials.” California Voter Foundation. 2021. https://www.calvoter.org/sites/default/files/cvf_addressing_harassment_of_election_officials_report.pdf

49 “Replacing the Refs.” States United Democracy Center. March 1, 2021. https://statesuniteddemocracy.org/ resources/replacingtherefs/

50 Harris, M. “Steve Bannon’s podcast could help Republicans win in 2024.” Slate. Sept. 15, 2021. https://bit.ly/3JYHxyI; Brown, E. Colorado county elections official Tina Peters is indicted in probe of alleged tampering with voting equipment. Washington Post. March 9, 2022. https://wapo.st/3Ln1EXE

51 Moonshot considers searches for insider references and clear intent such as “14 words”, “1488”, “Heil Hitler”, “How to Join KKK”, “Meine Ehre Heibt Treue”, and similar topics as “white supremacist” searches.

52 Bryant-Genevier, J. “Symptoms of Depression, Anxiety, Post-Traumatic Stress Disorder, and Suicidal Ideation Among State, Tribal, Local, and Territorial Public Health Workers During the COVID-19s Pandemic—United States, March–April 2021.” Morbidity and Mortality Weekly Report, 2021. 70.

53 Smith, M. R., Weber, L., & Barry-Jester, A. M. “Public health workers fighting virus face growing threats.” AP News. April 20, 2021. https://bit.ly/37335vg

54 Ivory, D., & Baker, M. “Public health officials in the U.S. need federal protection from abuse and threats, a national group says.” The New York Times. Oct. 19, 2021. https://nyti.ms/35vsdus.

55 Kalmoe, N. P., & Mason, L. Radical American Partisanship. (Chicago: University of Chicago Press. 2022).

56 “Charlottesville: The Aftermath.” Infographic. Moonshot. https://bit.ly/3uI5NhZ

57 Bidgood, J. “Protesters have been injured or killed by car rammings, but justice is rare.” The Boston Globe. Oct. 31, 2021. https://apps.bostonglobe.com/news/nation/2021/10/vehicle-rammings-against-protesters/tulsa/

58 Ibid.

59 Birch, S., Daxecker, U., & Höglund, K. “Electoral violence: An introduction.” Journal of Peace Research, 57(1), 2020, 3–14.

60 Blake, A. “Oath Keepers seeks armed volunteers for Trump rally in D.C.” The Washington Times. Dec. 10, 2020. https://www.washingtontimes.com/news/2020/dec/10/oath-keepers-seeks-armed-volunteers-for-trump-rall/

61 Vaishnav, M. When Crime Pays: Money and Muscle in Indian Politics. (New Haven: Yale University Press. 2017).

62 Nagar, G. “Lok Sabha Elections 2019: Analysis of Criminal Background, Financial, Education, Gender, and other Details of Winners.” Association for Democratic Reforms. https://bit.ly/3wV3qLA

63 Feuer, A. “Another Far-Right Group Is Scrutinized About Its Efforts to Aid Trump.” The New York Times. Jan. 3, 2022. https://www.nytimes.com/2022/01/03/us/politics/first-amendment-praetorian-trump-jan-6.html

64 Brown, D. “Multnomah County Republicans formally allow militia groups to run security.” Portland Mercury. June 30, 2017. https://bit.ly/38btYhf

65 Suppe, R. “This Idaho bill could repeal a law that bans private militias. It has surprising allies.” Idaho Statesman. Feb 10, 2022. https://www.idahostatesman.com/news/politics-government/state-politics/article258192318.html

66 Kleinfeld, R. “What Gang-Backed Government Could Do to America.” The Atlantic. Jan 6, 2022. https://www.the atlantic.com/ideas/archive/2022/01/mobocracy-january-6-riot-guidebook-gang-backed-government/621174/

67 V. Asal, S. M. Chermak, S. Fitzgerald, and J. D. Freilich, “Organizational‐ level characteristics in right‐wing extremist groups in the United States over time.” Criminal Justice Review (2016) 073401681562697. https://doi.org/10. 1177/0734016815626970 Katarzyna Jasko and Gary LaFree, “Who Is More Violent in Extremist Groups? A Comparison of Leaders and Followers. Aggressive Behavior, 2019, 1-10.

68 Jensen, M.A., Yates, E. Kane, S. “Radicalization in the Ranks: An Assessment of the Scope and Nature of Criminal Extremism in the United States Military.” National Consortium for the Study of Terrorism and Responses to Terrorism. Jan. 17, 2022. https://www.start.umd.edu/pubs/Final%20Report%20for%20SAF%20CDM.pdf

69 McCord, M. B. “Congress Can and Should Address the Threat from Unauthorized Paramilitary Activity.” JustSecurity. Jan. 24, 2022. https://bit.ly/3LBd7TH

70 A charged January 6 insurrectionist was six times more likely to hail from a given county for every one-percent decline in its white, non-Hispanic population. Pape, R. A. “Understanding American Domestic Terrorism.” Chicago Project on Security and Threats. April 6, 2021. https://cpost.uchicago.edu/research/domestic_extremism/ A separate dataset of all violent far-right incidents from 1990 to 2018 found that growth in Asian and Hispanic immigration was the most significant predictor of far-right violence. Spontaneous hate crimes are most common in areas where white flight from the 1960s and formerly rural exurban communities are meeting growing immigrant populations. Perliger, A. “Why Do Hate Crimes Proliferate in Progressive Blue States?” Medium. Aug. 202 2020. https://medium.com/3streams/why-hate-crimes-proliferate-in-progressive-blue-state-72483b2d72a7

71 Kalmoe, N. “Yes, Political Rhetoric Can Incite Violence.” Politico. Oct. 30, 2018. https://politi.co/3qQIHEO

72 “Confronting Violent White Supremacy (Part V): Examining the Rise of Militia Extremism,” House of Representatives Committee on Oversight and Reform, May 26, 2021. https://bit.ly/35xthhu

73 “Lawsuit: Texas city refused escort to protect Biden bus.” AP News. October 30, 2021. https://bit.ly/3wUeZTo

74 Data from the ACLED Research Hub: https://acleddata.com/research-hub-united-states/

75 Report on Countering Extremist Activity Within the Department of Defense. (2021). Department of Defense.


76 Federal funding reaches only a minority of police departments and for larger departments is a tiny percentage of their budgets, however, it is a greater funding source for some smaller departments. It also sends a normative message that is important, and constitutionally-appropriate language would itself help local agencies seeking to address these problems and unclear on legal means to do so.

77 Mazzone, J., & Rushin, S. “State Attorneys General as Agents of Police Reform.” Duke Law Journal, 69, 2020, 75.

78 Wood M. J., Douglas, K.M., and Sutton R.M. “Dead and Alive: Beliefs in Contradictory Conspiracy Theories,” Social Psychological and Personality Science 3(6) 2012: 767-773. Van Prooijen, J. M., and Douglas, K.M. “Belief in Conspiracy Theories: Basic Principals of an Emerging Research Domain.” European Journal of Social Psychology. 48. 2018: 897-908.