Keep Factually independent
Whether you agree or disagree with our analysis, these conversations matter for democracy. We don't take money from political groups - even a $5 donation helps us keep it that way.
Comparison of far-right and far-left domestic terror groups in the US
Executive summary
Recent research and government reviews present competing pictures of U.S. domestic terrorism: long-run empirical reviews and government summaries show far‑right extremists have produced more fatalities and a larger share of lethal attacks since 1990 (NIJ summary: far‑right 227 events, >520 lives; far‑left 42 events, 78 lives) [1]. More recent 2025 analyses from CSIS and reporting note a short‑term rise in far‑left incidents in the first half of 2025 — in raw incident counts, far‑left plots and attacks outpaced far‑right ones for the first time in 30+ years — but these incidents are described as far less lethal overall [2] [3].
1. Historical lethality: far‑right dominance in fatalities
Longitudinal government summaries and scholarly reviews consistently report that far‑right violence has accounted for the majority of ideologically motivated homicides and the bulk of fatalities in recent decades; the National Institute of Justice summary states far‑right extremists committed 227 events that killed more than 520 people since 1990, versus 42 events by far‑left actors that killed 78 people [1]. Academic analyses reaching back to 1990–2020 similarly find far‑right attacks are more likely to be lethal than far‑left ones [4] [5].
2. Recent shift in incident counts, not lethality
CSIS’s 2025 dataset and coverage report that, midway through 2025, the number of far‑left attacks and plots exceeded far‑right incidents for the first time in decades — a shift driven partly by a decline in right‑wing incidents rather than a dramatic surge in lethal left‑wing actors — and CSIS emphasizes left‑wing attacks remain mostly non‑lethal [2] [3]. Reporting from NPR and Axios highlights caveats: the relative uptick in left‑wing incidents is small in absolute terms and may reflect data‑collection and definitional issues as much as a true structural change [6] [2].
3. Methodology matters: definitions, inclusion, and time windows
Scholars and critics warn that different datasets use different definitions of “terrorism,” ideological labels, and inclusion rules, producing divergent headline results. CSIS’s findings prompted scrutiny over which incidents were counted and how disrupted plots versus completed attacks were classified; critics say inconsistent inclusion can change comparative tallies dramatically [7] [6]. The PNAS/START study that compared left, right and Islamist violence underscores that global and domestic inferences vary with sampling and coding choices [5].
4. Violence type and intent differ by ideology
Several analyses find systematic differences in tactics, targets, and lethality: far‑right actors in recent U.S. history have tended toward fatal attacks and mass casualty events, whereas far‑left incidents (including anarchist or environmental militants) have been more frequent in certain periods but far less lethal overall — historically accounting for roughly 10–15% of incidents and under 5% of fatalities in some summaries [8] [1] [3]. University of Maryland–led work finds left‑wing attacks are significantly less likely to be violent or to cause fatalities compared with right‑wing attacks (left 0.33 vs right 0.61 probability of violent act in one dataset) [9] [10].
5. Geography, organization, and recruitment differences
Older law‑enforcement reviews note spatial and organizational contrasts: left‑wing groups have tended to cluster in urban, eastern areas and often lack the same religious or fundamentalist framing seen in some right‑wing violent movements, while right‑wing networks have roots in rural and suburban areas with different cultural and recruitment dynamics [11]. START and other researchers link right‑wing and Islamist extremism to traits such as closed‑mindedness and dogmatism that correlate with higher lethality in some analyses [5].
6. Political context and interpretive agendas
Interpretations of recent shifts show clear political stakes. CSIS analysts attribute the 2025 change partly to policy and political dynamics (e.g., how administration actions affect grievances), while critics and commentators on both sides accuse each other of selective emphasis; independent analysts have noted removal or archiving of certain government summaries has fed disputes about institutional bias [2] [6] [12]. Media and opinion outlets differ sharply: some stress a left‑wing “surge,” while others rebut that the long‑term data still point to far‑right lethality [13] [8] [7].
7. What the evidence supports—and what it does not
Available sources support two firm points: [14] over the long run (since 1990), far‑right extremists have been responsible for more ideologically motivated homicides and deadlier attacks [1]; [15] in the first half of 2025, incident counts for far‑left actors exceeded far‑right counts in one CSIS dataset, but those left‑wing incidents remained much less lethal [2] [3]. Available sources do not mention a definitive causal mechanism that makes left‑wing violence intrinsically more or less likely to become lethal beyond the empirical patterns cited [5] [10].
8. Practical takeaway for readers and policymakers
Policymakers should weigh both long‑run lethality trends and short‑term fluctuations: prioritize resources against the actors who cause the most deaths historically while monitoring emerging upticks and the definitional challenges that can distort comparisons [1] [2] [6]. Independent, transparent datasets and clear inclusion rules are essential to avoid politically charged misreading of the threat landscape [7] [6].