Major reviews and classic theory treat the repression–dissent relationship as a persistent puzzle because coercion can be associated with deterrence, escalation, substitution, backlash, or mixed effects. Endogeneity matters: authorities respond to perceived threats while challengers adapt to expected state behavior.
CF-01 Comparative dossier
Coercion does not have one programmable effect.
Verified scholarship supports a conditional feedback model: repression can deter participation, provoke backlash, alter legitimacy, reshape media and public interpretation, or produce several of these effects at once.
StatusPublishedsource-verification pass · 2026-09-20
Research question
When does coercion suppress dissent, and when does it change the conflict in ways authorities did not intend?
This dossier tests the project’s recursive-feedback dimension. The central finding is not “repression radicalizes” or “repression works.” The more defensible claim is that the effect is conditional on who is targeted, how force is applied, how legitimate it appears, how participants and bystanders interpret it, what tactics the movement uses, and what institutional and information environment surrounds the event.
Movement boundary: this is comparative evidence. It does not establish that the Unprogrammable movement or its participants are repressed, radicalized, violent, anti-system, or revolutionary.
What the evidence supports
In the 2013 Gezi campaign, selective arrests away from demonstrations were associated with lower protest rates while indiscriminate violence during demonstrations was associated with higher protest rates. That is evidence for differential mobilization effects in one case—not proof that selective repression is broadly cost-free or universally effective.
A natural-experiment study in Uganda found selective repression reduced support for police, especially among opposition supporters. This means a tactic may suppress or incapacitate activity yet still impose institutional legitimacy costs.
Panel evidence around Zimbabwe’s 2018 election found exposure to local repression increased willingness to protest among opposition supporters and nonpartisans despite increased perceived risk. Relational and emotional responses were more consistent with the short-run backlash than simple cost calculations alone.
Historical U.S. evidence from 1960–1972 links nonviolent Black-led protest—especially when met by repression—to civil-rights framing and shifts in public and elite attention, while protester-initiated violence shifted attention toward social control. This is historically specific evidence about agenda and framing, not a universal moral ranking of tactics.
A 2024 comparative analysis of deadly repression in Bolivia found movement success after lethal repression in many cases and concluded that backfire did not depend on strict nonviolence or nonlethality. That directly limits stronger versions of the “political jiu-jitsu requires nonviolence” claim.
A 2026 survey experiment in Istanbul found that declaring a protest banned increased public approval of subsequent police repression and stronger intervention. Legal framing can therefore affect whether coercion is interpreted as abuse, order maintenance, or something in between.
Recent review work emphasizes that repression includes more than visible police force: surveillance, law, administrative controls, arrests, organizational disruption, and other mechanisms can interact. A useful model therefore tracks the whole sequence rather than coding a single “repression” switch.
Mechanism map
A recursive sequence, not a one-way treatment.
Authority interprets threat
Institutions choose legal, administrative, informational, or physical responses based on their own expectations and constraints.
Participants update risk and meaning
People may fear costs, perceive injustice, strengthen group identification, withdraw, substitute tactics, or remain unchanged.
Bystanders interpret legitimacy
Target identity, protest tactics, legal framing, prior trust, and available evidence influence whether observers view coercion as legitimate, excessive, or ambiguous.
Networks transmit the event
Organizations, peers, media, and digital networks can amplify, mute, reinterpret, or contest what happened.
Institutions respond again
Authorities may escalate, concede, investigate, reform, prosecute, change tactics, or hold course—creating another round of feedback.
Competing explanations & limits
What could make the same evidence mean something else?
Selection and endogeneity
Governments often repress movements they already perceive as threatening. A post-repression increase in protest may partly reflect pre-existing mobilization capacity rather than a treatment effect.
Different outcomes
Protest frequency, willingness to demonstrate, police legitimacy, media framing, institutional reform preferences, system-replacement preferences, and support for violence are different dependent variables.
Regime and case scope
Evidence from Turkey, Uganda, Zimbabwe, Bolivia, and historical U.S. civil-rights conflict cannot be merged into one universal coefficient. Regime type, institutions, political alignment, policing structure, and history differ.
Time horizon
Short-run backlash can coexist with longer-run exhaustion or organizational decline; immediate deterrence can coexist with later legitimacy costs. Dossier conclusions must specify when the outcome is measured.
Relevance to unprogrammability
Coercion changes the system it is trying to control.
This is a strong example of recursive adaptation: state action does not simply impose a fixed effect on a passive population. It changes risk, legitimacy, identity, information, coalition boundaries, and strategic expectations; those changes then alter the next institutional response.
For the Living Movement Study, the implication is methodological rather than diagnostic. If participants ever report coercive or exclusionary institutional encounters, study the event context, perceived legitimacy, prior trust, tactical change, network interpretation, and subsequent participation separately. Do not infer “radicalization” from controversy, protest, anger, or self-identification.
Verified source basis
9 checked sources, kept inspectable.
The owner-supplied coercion report was used as a research lead. The claims above were rewritten after checking the external scholarship below. Inclusion means the source informs this dossier; it does not imply that every proposition in the source has been adopted.
01Peer-reviewed literature review External scholarship — claim-level source checkedState Repression and Political OrderChristian Davenport · 2007
Foundational review documenting a mature but unresolved repression literature and persistent puzzles in how coercion interacts with dissent. Used to support the dossier’s core claim that there is no context-free repression effect.
Citation: Davenport, Christian. “State Repression and Political Order.” Annual Review of Political Science 10 (2007): 1–23. https://doi.org/10.1146/annurev.polisci.10.101405.143216
repressioncoerciondissentpolitical order
02Peer-reviewed research article External scholarship — claim-level source checkedDeterrence or Escalation? The Puzzle of Aggregate Studies of Repression and DissentMark Irving Lichbach · 1987
Early formal account showing why repression can produce substitution, deterrence, or escalation rather than one uniform response. Included as intellectual background, not as a final empirical verdict.
Citation: Lichbach, Mark Irving. “Deterrence or Escalation? The Puzzle of Aggregate Studies of Repression and Dissent.” Journal of Conflict Resolution 31, no. 2 (1987): 266–297. https://doi.org/10.1177/0022002787031002003
repressiondissentdeterrenceescalation
03Peer-reviewed literature review External scholarship — claim-level source checkedLayers of Political Repression: Integrating Research on Social Movement RepressionJennifer Earl and Jessica Maves Braithwaite · 2022
Supports treating repression as layered and multidisciplinary rather than reducing it to visible police violence alone.
Citation: Earl, Jennifer, and Jessica Maves Braithwaite. “Layers of Political Repression: Integrating Research on Social Movement Repression.” Annual Review of Law and Social Science 18 (2022): 227–248. https://doi.org/10.1146/annurev-lawsocsci-050520-092713
repressionsocial movementslawpolicing
04Peer-reviewed research article External scholarship — claim-level source checkedThe Effects of Selective and Indiscriminate Repression on the 2013 Gezi Park Nonviolent Resistance CampaignTijen Demirel-Pegg and Karen Rasler · 2021
Event-data study finding selective arrests away from demonstrations were associated with reduced protest rates while indiscriminate violence during demonstrations was associated with increased protest rates in the Gezi campaign. The result is case-specific, not a universal rule.
Citation: Demirel-Pegg, Tijen, and Karen Rasler. “The Effects of Selective and Indiscriminate Repression on the 2013 Gezi Park Nonviolent Resistance Campaign.” Sociological Perspectives 64, no. 1 (2021): 58–81. https://doi.org/10.1177/0731121420914291
repressionselective repressionindiscriminate repressionGezi Parkmobilization
05Peer-reviewed research article External scholarship — claim-level source checkedHow Repression Affects Public Perceptions of Police: Evidence from a Natural Experiment in UgandaTravis Curtice · 2021
Natural-experiment evidence that selective repression reduced support for police, especially among opposition supporters. Important counterevidence to any claim that targeted repression necessarily avoids legitimacy costs.
Citation: Curtice, Travis. “How Repression Affects Public Perceptions of Police: Evidence from a Natural Experiment in Uganda.” Journal of Conflict Resolution 65, no. 10 (2021): 1680–1708. https://doi.org/10.1177/00220027211013097
repressionpolice legitimacyUgandapublic opinion
06Peer-reviewed research article External scholarship — claim-level source checkedRepression and Dissent in Moments of Uncertainty: Panel Data Evidence from ZimbabweAdrienne LeBas and Lauren E. Young · 2024
Panel-data evidence that local exposure to repression increased protest willingness among opposition supporters and nonpartisans despite higher perceived risk; relational and emotional mechanisms were more consistent with the backlash effect than simple risk updating.
Citation: LeBas, Adrienne, and Lauren E. Young. “Repression and Dissent in Moments of Uncertainty: Panel Data Evidence from Zimbabwe.” American Political Science Review 118, no. 2 (2024): 584–601. https://doi.org/10.1017/S0003055423000230
repressiondissentZimbabwemobilizationpolitical identity
07Peer-reviewed research article External scholarship — claim-level source checkedAgenda Seeding: How 1960s Black Protests Moved Elites, Public Opinion and VotingOmar Wasow · 2020
Historical U.S. evidence that protest tactics and repression interacted with media framing, elite discourse, public opinion, and voting. Useful for showing that public interpretation is part of the feedback loop; not generalizable as a universal protest law.
Citation: Wasow, Omar. “Agenda Seeding: How 1960s Black Protests Moved Elites, Public Opinion and Voting.” American Political Science Review 114, no. 3 (2020): 638–659. https://doi.org/10.1017/S000305542000009X
protest tacticsrepressionmedia framingpublic opinionUnited States
08Peer-reviewed comparative case analysis External scholarship — claim-level source checkedWhen Does Lethal Repression Fail? Unarmed Militancy and Backfire in Bolivia, 1982–2021Carwil Bjork-James · 2024
Qualitative comparative analysis of deadly repression events in Bolivia finds backfire did not require strict protester nonviolence. This directly limits stronger versions of the “political jiu-jitsu requires nonviolence” claim.
Citation: Bjork-James, Carwil. “When Does Lethal Repression Fail? Unarmed Militancy and Backfire in Bolivia, 1982–2021.” Journal of Latin American Studies (2024).
repressionbackfireBoliviaunarmed militancyprotest
09Peer-reviewed research article External scholarship — claim-level source checkedDo protest bans make police repression more acceptable? Experimental evidence on the effects of legal repression from TurkeyMert Arslanalp, Ulaş Erdoğdu, and T. Deniz Erkmen · 2026-09-02
Survey experiment in Istanbul showing that declaring a protest banned increased approval of subsequent police repression and stronger intervention. Demonstrates that legal framing can alter the public-legitimacy environment around coercion.
Citation: Arslanalp, Mert, Ulaş Erdoğdu, and T. Deniz Erkmen. “Do protest bans make police repression more acceptable? Experimental evidence on the effects of legal repression from Turkey.” Journal of Peace Research, published September 2, 2026. https://doi.org/10.1093/jopres/xjag060
legal repressionprotest banspolice repressionpublic opinionTurkey
Review state
What this dossier is—and is not.
Publication: 2026-09-20. Author: Unprogrammable.org Editorial Project. Independent reviewer: none recorded. Status: editorial source-verification pass; the cited studies are externally peer reviewed, but this project synthesis is not itself peer reviewed.
Revision trigger: revise when stronger cross-context causal evidence materially changes the conditional model, when a cited source is corrected or retracted, or when the dossier begins making claims beyond the evidence states listed here.