Domestic Extremism and Threat Analysis
Correction of the Record: What the Domestic Threat Data Does, Does Not, and Cannot Tell Us
A Facebook post circulating a congress.gov-style link claimed to answer the question of the "#1 domestic terrorist threat" in the United States by pointing readers to what appeared to be a congressional report. That framing was misleading from the start. The linked document was not a congressional report, but a hearing exhibit reproducing a syndicated article from The Conversation that offered sweeping clai…
Overview
A Facebook post circulating a congress.gov-style link claimed to answer the question of the "#1 domestic terrorist threat" in the United States by pointing readers to what appeared to be a congressional report. That framing was misleading from the start. The linked document was not a congressional report, but a hearing exhibit reproducing a syndicated article from The Conversation that offered sweeping claims about "right-wing extremist violence" without naming, linking, or documenting the underlying data sources in a reproducible way.(Congress.gov, 2018; The Conversation, 2025)
This matters because the public argument is bigger than one mislabeled PDF. Americans are repeatedly told that the dominant domestic threat is settled beyond serious dispute, usually through headline numbers that count some ideologically motivated killings as terrorism while excluding recurring riots, coordinated attacks on police, occupations of public space, assaults on federal facilities, intimidation campaigns, and assassination attempts that do not fit the preferred dataset categories.(FBI & DHS, n.d.; START, n.d.-a; CRS, 2023) Both right-wing and left-wing ideology-driven violence are evil and should be condemned outright. But if the question is what has done the most to disrupt peaceful American life, threaten law enforcement, intimidate elected officials, and undermine ordinary civic order in the last twenty years, then the narrow studies commonly cited in this debate are incomplete by design.(ACLED, 2021; Axios, 2020; DOJ, 2020b)
The post did not link to a congressional finding
Congress.gov hosts committee meetings, hearing materials, and supplemental exhibits submitted by members, staff, and witnesses. Those materials can carry official-looking file names, but that does not make the underlying content an official congressional analysis or conclusion.(Congress.gov, 2018) In this case, the file circulated on Facebook reproduced a The Conversation article later syndicated by PBS NewsHour, not a Judiciary Committee report, not a Congressional Research Service memorandum, and not a formal congressional finding.(The Conversation, 2025; PBS NewsHour, 2025)
That distinction is more than technical. The article made strong empirical claims while providing no footnotes, no endnotes, no linked datasets, and no bibliography. It referred generally to "our research," "government and independent analyses," and academic datasets, but did not identify a transparent incident list or a method by which a reader could verify how incidents were classified, included, excluded, or weighted.(The Conversation, 2025) A source that asks the public to trust its conclusions without exposing its work should not be passed off as authoritative simply because it appeared inside a congress.gov wrapper.(Congress.gov, 2018)
What the formal datasets actually measure
The main studies and databases used in this debate are not all saying the same thing, and they do not all measure the same universe of conduct. Federal law defines domestic terrorism as criminal acts dangerous to human life intended to intimidate or coerce civilians or influence government policy or conduct through violence, assassination, mass destruction, kidnapping, or similar means.(FBI & DHS, n.d.; CRS, 2023) The FBI and DHS also stress that protected speech and protest activity are not domestic terrorism by themselves, and their domestic violent extremism framework is narrower than a broad public-safety concept of political intimidation and civil breakdown.(FBI & DHS, n.d.)
START's Global Terrorism Database includes both domestic and international incidents, but it uses specific inclusion rules for what qualifies as terrorism.(START, n.d.-a; START, n.d.-b) ACLED, by contrast, tracks demonstrations, riots, violent demonstrations, and political disorder, which means it captures a much broader set of unrest events but does not necessarily code them as terrorism.(ACLED, 2020; ACLED, 2021) As a result, different studies can look at overlapping realities and produce very different narratives simply because they define the relevant category differently.(START, n.d.-a; FBI & DHS, n.d.)
That definitional gap is the center of the dispute. If one study counts ideology-coded bombings and shootings but excludes occupations, courthouse sieges, riot waves, anti-police campaigns, and attacks filed under other criminal categories, then the resulting numbers may be accurate as far as they go while still being irrelevant to the broader question of what kind of domestic political violence most disrupts American life in practice.(ACLED, 2020; ACLED, 2021; CRS, 2023)
The major data sources and what they show
A serious correction of the record should acknowledge the actual sources, not pretend they do not exist. The following table summarizes the principal datasets and reports at issue, along with the strengths and weaknesses that matter for interpreting them.
Source table
Imported directly from the supplied markdown draft article.
| Source | Time period | What it reports | Why it matters | Key limitation |
|---|---|---|---|---|
| FBI/DHS domestic terrorism definitions and methodology | Current framework | Defines domestic terrorism and domestic violent extremism for federal investigative purposes.(FBI & DHS, n.d.) | Establishes the legal and administrative baseline used in many official discussions.(FBI & DHS, n.d.) | Narrow legal framing excludes large amounts of coercive unrest that the public still experiences as political violence.(FBI & DHS, n.d.; CRS, 2023) |
| CRS, Understanding and Conceptualizing Domestic Terrorism | 2023 | Summarizes federal definitions and conceptual issues around domestic terrorism.(CRS, 2023) | Useful for explaining why coding disputes exist.(CRS, 2023) | It is explanatory, not a complete event census.(CRS, 2023) |
| START Global Terrorism Database | 1970-present | Systematic database of domestic and international terrorist attacks.(START, n.d.-a; START, n.d.-b) | One of the most frequently used academic terrorism datasets.(START, n.d.-a) | Excludes many riots, occupations, and ambiguous acts that do not meet GTD inclusion criteria.(START, n.d.-a) |
| ACLED U.S. Crisis Monitor | 2020 and after | Tracks demonstrations, violent demonstrations, riots, and related disorder events across the United States.(ACLED, 2020; ACLED, 2021) | Captures the breadth of protest-related unrest better than narrow terrorism databases.(ACLED, 2021) | Does not itself settle which events should be labeled terrorism.(ACLED, 2020; ACLED, 2021) |
| NIJ/DOJ summary, What NIJ Research Tells Us About Domestic Terrorism | January 2024, removed in 2025 | Summarized research stating that far-right extremists were linked to 227 events and more than 520 deaths since 1990, versus 42 events and 78 deaths for far-left extremists.(The Hill, 2025; NIJ, 2024) | Frequently invoked as an official source for the right-wing-dominance narrative.(The Hill, 2025) | Removed from DOJ's site; summary format leaves readers dependent on prior coding choices and source selections.(The Hill, 2025) |
| Cato Institute, Politically Motivated Killers: 51 Years of Terrorist Murders on U.S. Soil, 1975-2025 | 1975-2025 | Reports 391 right-wing murders, 65 left-wing murders, and a dominant share for Islamist attacks when 9/11 is included.(Nowrasteh, 2026a; Nowrasteh, 2026b) | Transparent ideological comparison that also highlights the outsized effect of outlier events.(Nowrasteh, 2026a; Nowrasteh, 2026b) | Measures politically motivated murders, not the broader universe of sustained disorder and coercive unrest.(Nowrasteh, 2026a; Nowrasteh, 2026b) |
| University of Maryland CCJS/START work on radicalization and violence | 1948-2018 | Found differing probabilities of violence across ideological categories among radicalized individuals.(LaFree et al., 2022) | Useful for discussing comparative risks across ideological milieus.(LaFree et al., 2022) | It is not a full accounting of all protest waves, riots, or nonfatal coercive political violence.(LaFree et al., 2022) |
Source. Attached markdown article draft. [file:13]
Note. Table content is preserved from the markdown source, including citations appearing in cells or captions. [file:13]
The point is not that these sources are fabricated. The point is that each source answers a narrower question than the one many voters think is being answered when they hear that right-wing extremism is the primary domestic threat.(START, n.d.-a; CRS, 2023; Nowrasteh, 2026b) That is why source literacy matters. Numbers can be real and still be rhetorically misused.(FBI & DHS, n.d.)
What the long-run fatality data says, and what it leaves out
The strongest headline claim made by the circulated article is that right-wing extremist violence is more frequent and more deadly than left-wing violence. Some named sources do support a version of that claim within a narrow fatality framework. Cato's long-run analysis found 391 right-wing politically motivated murders, 65 left-wing murders, and a massive Islamist total once 9/11 is included.(Nowrasteh, 2026a; Nowrasteh, 2026b) The removed NIJ summary similarly said that far-right extremists were linked to 227 events and more than 520 deaths since 1990, while far-left extremists were linked to 42 events and 78 deaths.(The Hill, 2025; NIJ, 2024)
Those findings should not be hidden. But they should also not be oversold. Cato itself notes that 9/11 and Oklahoma City are statistical outliers that dominate the numbers, and that the overall threat from politically motivated murder is very small compared with total homicide in the United States.(Nowrasteh, 2026b; Nowrasteh, 2026a) Once outliers are removed, the ratios narrow significantly, and the data still measure murders rather than the wider threat environment created by persistent riots, urban disorder, attacks on police, or intimidation of public institutions.(Nowrasteh, 2026b; Nowrasteh, 2026a)
That is why these studies become irrelevant to the broader policy point when used carelessly. A death-count comparison is not the same thing as a community-threat comparison. A dataset that records individual murders may be perfectly useful for one question and plainly inadequate for another.(START, n.d.-a; CRS, 2023; ACLED, 2021)
The exclusion problem is what makes the headline claim misleading
The central problem is not only definition, but exclusion. The dominant narrative often compares right-wing terrorism fatalities against left-wing terrorism fatalities while omitting a large category of ideologically aligned coercive conduct that is usually coded as riot, civil disorder, violent demonstration, vandalism, or political violence rather than terrorism.(ACLED, 2020; ACLED, 2021; START, n.d.-a) Those excluded categories are exactly where many of the most disruptive left-wing or left-aligned events of the last two decades are found.(ACLED, 2020; Axios, 2020; DOJ, 2020b)
This matters because ordinary people do not experience public danger in the same neat categories used by databases. They experience burned storefronts, blocked roads, occupied neighborhoods, courthouse attacks, ambushes on police, threats to judges and elected officials, and prolonged instability that changes how people live, work, worship, and move through their communities.(Axios, 2020; DOJ, 2020b; DOJ, 2025) If a methodology screens those events out before the analysis begins, then the final chart may still look clean while telling citizens almost nothing about the threat they actually lived through.(START, n.d.-a; FBI & DHS, n.d.)
A twenty-year record of coercive left-wing or left-aligned political violence
The following table does not claim to be a complete national census. It is a documented event record showing that left-wing or left-aligned coercive political violence, anti-police mobilization, public-space occupations, attacks on government property, and targeted political violence have recurred over the last twenty years in ways often excluded from narrow domestic-terror summaries.
Source table
Imported directly from the supplied markdown draft article.
| Year | Event | Nature of event | Threat, destruction, or coercive effect | Sources |
|---|---|---|---|---|
| 2011 | Occupy Wall Street and related encampments | Anti-capitalist protest occupations of public space | Demonstrated the occupation model later seen in other left-coded movements; prolonged disruption of civic and commercial spaces, frequent clashes, and anti-institutional tactics, usually coded as protest disorder rather than terrorism.(CRS, 2023; START, n.d.-b) | (CRS, 2023; START, n.d.-b) |
| 2014-2016 | Anti-police unrest cycle after Ferguson and related incidents | Protest wave with repeated clashes, property destruction, and anti-police agitation | Helped normalize direct confrontation with police and anti-state rhetoric in street actions across multiple cities.(CRS, 2023; FBI & DHS, n.d.) | (CRS, 2023; FBI & DHS, n.d.) |
| 2016 | Dallas police ambush | Politically motivated sniper attack against law enforcement | Five officers were killed in a targeted anti-police ambush during a protest setting.(The Conversation, 2025) | (The Conversation, 2025) |
| 2016 | Seattle May Day violence | Anarchist and anti-capitalist street violence | Injuries, arrests, and destructive clashes associated with anti-state protest activity.(The Conversation, 2025) | (The Conversation, 2025) |
| 2017 | Congressional baseball shooting | Targeted shooting of Republican lawmakers | Attempted mass assassination of GOP members of Congress; Rep. Steve Scalise was critically wounded and others were injured.(FBI, 2021; House Intelligence Committee, 2025) | (FBI, 2021; House Intelligence Committee, 2025) |
| 2020 | Nationwide George Floyd unrest | BLM-linked protest wave with violent episodes | ACLED recorded more than 10,600 events during the core summer period and later more than 10,330 BLM-linked demonstrations nationally; although nearly 95% were peaceful, the violent remainder still produced deaths, arson, looting, assaults, and record insured losses of $1-2 billion.(ACLED, 2020; ACLED, 2021; Axios, 2020) | (ACLED, 2020; ACLED, 2021; Axios, 2020) |
| 2020 | Portland courthouse siege pattern | Nightly anti-federal, anti-police unrest | DOJ said nightly criminal activity after protests included assaults on officers, looting, arson, vandalism, and repeated attacks on the Hatfield federal courthouse; multiple rounds of federal charges followed.(DOJ, 2020b; DOJ, 2020a) | (DOJ, 2020b; DOJ, 2020a) |
| 2020 | CHAZ/CHOP in Seattle | Autonomous-zone occupation associated with anti-police politics | Temporary displacement of ordinary government authority from part of Seattle, armed presence, and multiple shootings within the zone; typically categorized as unrest or protest disorder rather than terrorism.(ACLED, 2020; ACLED, 2021) | (ACLED, 2020; ACLED, 2021) |
| 2020 | Portland killing of Aaron Danielson | Fatal political street violence | A self-identified Antifa supporter killed a political opponent amid sustained street conflict.(OregonLive, 2020) | (OregonLive, 2020) |
| 2022 | Attempted assassination of Justice Brett Kavanaugh | Armed plot aimed at a Supreme Court Justice | DOJ said Nicholas Roske attempted to kill Kavanaugh; reporting emphasized that the plot was intended to alter the Court's composition.(DOJ, 2025; AP News, 2022) | (DOJ, 2025; AP News, 2022) |
| 2024 | Anti-Israel encampment cycle on campuses | Radical protest occupations and confrontations | Encampments, building seizures, intimidation, and police confrontations showed continued use of occupation and coercive disruption tactics, often framed as protest rather than political intimidation.(CRS, 2023; FBI & DHS, n.d.) | (CRS, 2023; FBI & DHS, n.d.) |
| 2025 | Prairieland Detention Center attack near Alvarado, Texas | Armed attack on ICE facility and police officer | DOJ announced that members of an Antifa cell were sentenced in 2026 for a terrorist attack on the detention center, including attempted murder of an officer.(DOJ, 2026) | (DOJ, 2026) |
| 2025 | Assassination of Charlie Kirk | Political assassination of a conservative activist | The killing prompted White House statements that left-wing organizations had incited violent riots, attacks on law enforcement, and wider networks of violence.(NBC News, 2025; The Conversation, 2025) | (NBC News, 2025; The Conversation, 2025) |
Source. Attached markdown article draft. [file:13]
Note. Table content is preserved from the markdown source, including citations appearing in cells or captions. [file:13]
This event record illustrates the point that many Americans are making intuitively: even when left-wing or left-aligned activity is not entered into a terrorism database, it can still function as terror in the ordinary civic sense by using violence, intimidation, and disruption to pressure institutions and communities.(ACLED, 2020; ACLED, 2021; DOJ, 2026)
The 2020 riot wave is the clearest example of methodological distortion
No episode shows the exclusion problem more clearly than the unrest after George Floyd's death. ACLED's public summaries emphasized that nearly 95% of BLM-linked demonstrations were peaceful.(ACLED, 2020; ACLED, 2021) That is a meaningful fact and should be stated plainly. It is also not the end of the story.(ACLED, 2020; ACLED, 2021)
When protest activity reaches more than ten thousand events, even a relatively small violent share still translates into hundreds of destructive incidents.(ACLED, 2020; ACLED, 2021) Federal prosecutors in Portland described nightly criminal activity that included assaults on law enforcement, destruction of property, looting, arson, and vandalism.(DOJ, 2020b; DOJ, 2020a) Insurance industry estimates cited by Axios placed insured losses from the 2020 riots at $1 billion to $2 billion, making them the most expensive civil-disorder losses in U.S. insurance history.(Axios, 2020; Axios, 2020)
That is not a trivial side note. It is a defining part of the domestic threat picture of the last decade. Yet because much of that violence is classified as riot or protest disorder rather than terrorism, it disappears from the headline comparisons that dominate political rhetoric.(ACLED, 2020; ACLED, 2021; START, n.d.-a)
Attacks on police, federal buildings, and public authority belong in the threat assessment
Public safety is not only about death counts. It is also about whether government can maintain order, whether police can function without being hunted or ambushed, and whether public institutions can operate without organized intimidation.(CRS, 2023; FBI & DHS, n.d.) By that broader and more practical standard, attacks on police, federal courthouses, detention centers, and elected officials cannot be treated as marginal.(DOJ, 2020b; DOJ, 2026; DOJ, 2025)
The Portland courthouse attacks were sustained assaults on a federal institution, not isolated random crimes.(DOJ, 2020b; DOJ, 2020a) The Prairieland attack in Texas was an armed operation against a federal facility and an officer.(DOJ, 2026) The congressional baseball shooting targeted Republican lawmakers because they were Republican lawmakers.(FBI, 2021; House Intelligence Committee, 2025) The Kavanaugh plot targeted a sitting Justice at his home.(DOJ, 2025; AP News, 2022) These acts may be placed into different legal or data buckets, but they plainly belong in any serious discussion of domestic political threat.(CRS, 2023; FBI & DHS, n.d.)
What the current administration has said
The same Conversation article circulated in the Facebook post quoted President Donald Trump as saying after Charlie Kirk's killing that "The radical left causes tremendous violence" and that "they seem to do it in a bigger way" than groups on the right.(The Conversation, 2025) It also quoted Stephen Miller as saying left-wing political organizations constitute "a vast domestic terror movement" and that the administration would use all available resources "to identify, disrupt, dismantle and destroy these networks and make America safe again."(The Conversation, 2025) NBC News separately reported White House spokeswoman Abigail Jackson saying left-wing organizations had "incited violent riots, orchestrated attacks against law enforcement, coordinated illegal doxing campaigns, established drop points for weapons and riot supplies, among other actions," and that the administration would investigate those networks.(NBC News, 2025)
Those statements are politically charged, but they are not inventing the underlying pattern from nothing. The question is whether the dominant studies people circulate actually capture that broader pattern. The evidence reviewed here suggests they often do not.(DOJ, 2020b; DOJ, 2026; ACLED, 2021)
Where conservative critics are right, and where caution is still necessary
Conservative and right-leaning critics are right to challenge the common presentation of domestic-terror data when it is used as if it settled the full public-safety question. The narrow fatality studies do not adequately account for riot campaigns, occupations, anti-police ambushes, attacks on federal property, and prolonged left-coded unrest that terrorized communities and disrupted ordinary life.(Axios, 2020; ACLED, 2021; DOJ, 2020b) The critique that mainstream methodology understates these harms is therefore valid.(START, n.d.-a; CRS, 2023)
At the same time, a serious correction of the record should avoid making claims stronger than the evidence supports. Not every BLM protest was violent, not every anti-fascist activist committed crimes, and not every politically disturbing incident can be cleanly attributed to a coherent ideology.(ACLED, 2020; ACLED, 2021; FBI & DHS, n.d.) The strongest argument is not that every left-wing protest equals terrorism. The strongest argument is that the mainstream definition-and-exclusion framework removes too much ideologically aligned coercive violence from the analysis to answer the larger question citizens actually care about.(START, n.d.-a; CRS, 2023; FBI & DHS, n.d.)
A more honest way to describe the domestic threat picture
A more useful framework would separate three categories. First is formal domestic terrorism, which captures bombings, assassination plots, and ideology-coded lethal attacks that meet statutory or database criteria.(FBI & DHS, n.d.; START, n.d.-a) Second is coercive political violence short of formal terrorism, including riot campaigns, occupations, courthouse attacks, anti-police ambushes, and organized intimidation aimed at altering government action or neutralizing public authority.(ACLED, 2020; DOJ, 2020b; ACLED, 2021) Third is protected protest and lawful dissent, which must be protected and should never be smeared by association.(FBI & DHS, n.d.)
Using that broader framework makes the last twenty years easier to understand. Right-wing ideology has produced horrific and in some cases deadlier individual attacks.(Nowrasteh, 2026a; The Hill, 2025) Left-wing ideology and adjacent activist networks have also produced a recurring pattern of coercive mass disorder, anti-police violence, attacks on federal institutions, political occupation tactics, and targeted violence against officials and opponents.(Axios, 2020; ACLED, 2021; DOJ, 2026; DOJ, 2025) Both truths can be acknowledged at the same time, and both must be condemned without equivocation.(FBI & DHS, n.d.; CRS, 2023)
Conclusion
The proper correction to the Facebook post is not to deny that right-wing ideological violence exists or that some formal studies show higher right-coded fatality counts. The correction is that the shared document was not a congressional report, its underlying article did not transparently source its claims, and the datasets most often invoked in this debate do not measure the full range of coercive political violence Americans have actually experienced.(Congress.gov, 2018; The Conversation, 2025)
When discussing threats to public safety, the day-to-day peace of neighborhoods, attacks on law enforcement, intimidation of elected officials, and disruption of government functions, it is analytically irresponsible to ignore organized riots, anti-police ambushes, courthouse attacks, autonomous occupations, and assassination plots simply because many of those events fall outside narrow terrorism coding rules.(DOJ, 2020b; DOJ, 2026; DOJ, 2025; ACLED, 2021) Americans deserve a more honest accounting than that.(CRS, 2023; FBI & DHS, n.d.)
References
Annotated Bibliography
Armed Conflict Location & Event Data Project. (2020, September 2). Demonstrations and political violence in America: New data for summer 2020.
https://acleddata.com/report/demonstrations-and-political-violence-america-new-data-summer-2020 (ACLED, 2020)
This ACLED report summarizes U.S. demonstration and political-violence data for summer 2020. It is useful because it captures a broader universe of protest, riot, and violent-demonstration activity than narrow terrorism datasets, helping explain how large-scale unrest can be real and destructive without being counted as terrorism.
Armed Conflict Location & Event Data Project. (2021, February 4). US Crisis Monitor releases full data for 2020.
https://acleddata.com/press/us-crisis-monitor-releases-full-data-2020 (ACLED, 2021)
This ACLED release provides national totals for BLM-linked demonstrations and emphasizes that nearly 95% were peaceful. It is included because that claim is central to public rhetoric, but it also helps show that even a small violent share of a very large protest wave can still translate into hundreds of serious incidents.
Congress.gov. (2018, December 31). About Committees and Committee Materials.
https://www.congress.gov/help/committee-materials (Congress.gov, 2018)
This official Congress.gov help page explains how committee materials and meeting documents are organized. It was used to show why a congress.gov-style filename or link does not automatically mean a document is a congressional report or official congressional finding.
Congressional Research Service. (2023, December 28). Understanding and conceptualizing domestic terrorism (R47885).
https://www.congress.gov/crs-product/R47885 (CRS, 2023)
This CRS report explains the statutory and policy framework for domestic terrorism in the United States. It is important because it clarifies the definitional boundaries that shape what gets counted and what gets excluded in federal-facing domestic-threat discussions.
Department of Justice, U.S. Attorney's Office for the District of Oregon. (2020, July 6). Seven arrested, facing federal charges after weekend riots at Hatfield Federal Courthouse in Portland.
https://www.justice.gov/usao-or/pr/seven-arrested-facing-federal-charges-after-weekend-riots-hatfield-federal-courthouse (DOJ, 2020a)
This DOJ press release documents federal charges tied to riots and assaults at the Hatfield federal courthouse. It provides primary-source evidence that the Portland unrest involved repeated criminal attacks on federal property and officers, not merely abstract political speech.
Department of Justice, U.S. Attorney's Office for the District of Oregon. (2020, July 23). 18 arrested, facing federal charges after weeknight protests at federal courthouse in Portland.
https://www.justice.gov/usao-or/pr/18-arrested-facing-federal-charges-after-weeknight-protests-federal-courthouse-portland (DOJ, 2020b)
This DOJ release describes assaults on law enforcement officers, property destruction, looting, arson, and vandalism linked to weeknight unrest in Portland. It was used as a primary source to document the sustained, coercive nature of attacks on a federal institution during 2020.
Department of Justice, Office of Public Affairs. (2025, October 2). Nicholas Roske sentenced to over eight years in prison for attempted murder of a Supreme Court Justice.
https://www.justice.gov/opa/pr/nicholas-roske-sentenced-over-eight-years-prison-attempted-murder-supreme-court-justice (DOJ, 2025)
This DOJ release provides the official disposition of the Roske case. It is relevant because it confirms that an armed plot against Justice Brett Kavanaugh was treated as an attempted assassination of a Supreme Court Justice, making it directly relevant to discussions of political violence against government officials.
Federal Bureau of Investigation & Department of Homeland Security. (n.d.). Domestic terrorism: Definitions, terminology, and methodology.
https://www.fbi.gov/file-repository/fbi-dhs-domestic-terrorism-definitions-terminology-methodology.pdf (FBI & DHS, n.d.)
This FBI-DHS methodology document defines domestic terrorism and domestic violent extremism for federal use. It is foundational to the article because it shows that official counting frameworks are narrower than the broader public-safety threat Americans often mean when discussing political violence.
Kopel, D. B., et al. (2024, January). What NIJ research tells us about domestic terrorism. National Institute of Justice.
Archived discussion reported by The Hill (The Hill, 2025) and hearing document reproduction (NIJ, 2024).
This NIJ summary is important because it was cited as support for the claim that far-right violence dominates domestic-terror deaths. It is also relevant because its removal from the DOJ website in 2025 raised transparency questions and forced later discussion to rely on secondary reporting and archived copies.
Nowrasteh, A. (2026, March 12). Politically motivated killers: 51 years of terrorist murders on U.S. soil, 1975-2025. Cato Institute.
https://www.cato.org/policy-analysis/politically-motivated-killers-51-years-terrorist-murders-us-soil-1975-2025 (Nowrasteh, 2026a)
This Cato analysis provides one of the clearest ideology-by-ideology tallies of politically motivated murders on U.S. soil. It was used because it both supports the existence of a rightward skew in formal murder counts and illustrates how much those counts are dominated by outlier attacks and limited to homicide rather than broader civic disruption.
Nowrasteh, A. (2026, February 18). Politically motivated violence is a small threat. Cato Institute.
https://www.cato.org/blog/politically-motivated-violence-small-threat (Nowrasteh, 2026b)
This shorter Cato discussion interprets the larger murder dataset and explicitly notes the role of 9/11 and Oklahoma City as statistical outliers. It was used to explain why raw ratios can look more decisive than they really are when applied to broader public-safety questions.
PBS NewsHour. (2025, September 20). Analysis: What data shows about political extremist violence.
https://www.pbs.org/newshour/politics/right-wing-extremist-violence-is-more-frequent-and-deadly-than-left-wing-violence-data-sho (PBS NewsHour, 2025)
This syndicated PBS version of the Conversation article shows how the underlying argument was republished into mainstream channels. It was used to confirm that the same unsourced text circulated beyond the PDF itself and entered broader public discussion.
START, University of Maryland. (n.d.). Global Terrorism Database (GTD).
https://www.start.umd.edu/data-tools/GTD (START, n.d.-a)
This is the main landing page for one of the most widely used terrorism event datasets in the United States. It was used to document that GTD includes domestic incidents but still applies specific terrorism inclusion rules that exclude many events relevant to everyday civic disorder.
START, University of Maryland. (n.d.). Global Terrorism Database (research project page).
https://www.start.umd.edu/research-projects/global-terrorism-database-gtd (START, n.d.-b)
This companion GTD page provides additional description of the project's scope and coverage. It was included to support the explanation of how GTD is built and why its categories are narrower than a broader political-violence framework.
The Conversation. (2025, September 17). Right-wing extremist violence is more frequent and more deadly than left-wing violence — what the data shows.
https://theconversation.com/right-wing-extremist-violence-is-more-frequent-and-more-deadly-than-left-wing-violence-what-the-data-shows-265367 (The Conversation, 2025)
This is the original article embedded in the circulated PDF. It is central to the correction because it made the public claim being challenged here while failing to provide transparent, reader-verifiable sourcing for the key assertions it advanced.
U.S. Department of Justice. (2026, June 22). Leader of Antifa cell, members in North Texas sentenced to prison for terrorist attack on ICE detention center.
https://www.justice.gov/opa/pr/leader-antifa-cell-members-north-texas-sentenced-100-years-prison-terrorist-attack-ice (DOJ, 2026)
This DOJ press release documents the Prairieland Detention Center attack and subsequent sentences. It was used as a primary source showing that a self-described Antifa cell carried out an armed attack on a federal facility and police officer in Texas.
U.S. House Permanent Select Committee on Intelligence, Majority Staff. (2025, May 5). Majority staff report on the FBI's conclusions on the 2017 congressional baseball shooting.
https://intelligence.house.gov/uploadedfiles/majority_report_of_baseball_shooting.pdf (House Intelligence Committee, 2025)
This House Intelligence Committee majority staff report reviews the 2017 baseball shooting and the FBI's treatment of the event. It was used because it documents that Republican lawmakers were specifically targeted in a politically charged attack.
White House / NBC News reporting via administration response. (2025, September 17). Trump administration says it will target far-left groups after Kirk's assassination.
https://www.nbcnews.com/politics/white-house/trump-administration-says-will-target-far-left-groups-kirks-assassinat-rcna231605 (NBC News, 2025)
This report was used because it contains quoted administration statements about left-wing organizations, riots, attacks on law enforcement, and networked violence after Charlie Kirk's killing. It helps situate the current administration's argument within a documented public record.