The Playbook: How Democracies Die Slowly - And What We Can Do About It

April 6, 2026

A Signal & Noise Deep Dive


Part I: The Playbook

Democracies no longer die with tanks in the streets. They die with legal language, constitutional mechanisms, and the slow acquiescence of institutions that were built to resist exactly this. The playbook is public, well-documented, and being executed in real time across the Western world. Understanding it is the first act of resistance.

The Seven Tactics

Drawing on the work of Protect Democracy, Steven Levitsky and Daniel Ziblatt (How Democracies Die), and the Brookings Institution's Democracy Playbook 2025, the modern authoritarian playbook consists of seven interlocking tactics. They don't operate in isolation - they compound. Each one creates cover for the next.

1. Politicize Independent Institutions

Every democracy depends on institutions that operate independently from partisan politics: law enforcement, central banking, election administration, intelligence services, inspectors general. The first move is to capture these. Replace career professionals with loyalists. Reframe nonpartisan civil service as "the deep state." Fire independent watchdogs. Once the institutional immune system is compromised, everything else becomes easier.

The U.S. case: mass firing of inspectors general, purging of career civil servants, pressure on the Department of Justice to protect allies and prosecute opponents, the use of DOGE to restructure federal agencies without congressional authorization.

The international pattern: Hungary's Orban replaced the heads of every major independent institution within his first two terms. Turkey's Erdogan hollowed out the judiciary and intelligence services after the 2016 coup attempt. In both cases, the legal shell of the institution remained - only the independence was removed.

2. Spread Disinformation

All politicians spin. Authoritarians propagate and amplify falsehoods with abandon and strategic intent. The goal is not merely to lie, but to destroy the shared epistemic foundation that democratic deliberation requires. When citizens cannot agree on basic facts, collective action becomes impossible.

This operates on multiple levels: relentless attacks on press credibility ("enemy of the people"), flooding the information space with contradictory narratives so that no single truth can gain traction, and using social media algorithms - which reward engagement over accuracy - as force multipliers. Generative AI is accelerating this further, making high-quality disinformation cheap and scalable.

3. Aggrandize Executive Power

Authoritarian projects cannot succeed without weakening the checks that legislatures, courts, and other institutions provide. This means bypassing Congress through executive orders at unprecedented pace, asserting the "unitary executive" theory (that the president has near-total control over the entire executive branch), and exploiting moments of crisis to claim emergency powers.

The U.S. Supreme Court's 2024 ruling granting broad presidential immunity for "official acts" represents a structural shift - it removed a critical constraint that had kept presidential power within bounds for the country's entire history.

4. Quash Criticism and Dissent

This doesn't require outright censorship. It requires making dissent costly. Pull federal ad spending from critical outlets. Threaten broadcasters through the FCC. Use federal contracts and funding as weapons against law firms, universities, and states that oppose the administration. Sue or investigate critics. Create a chilling effect where self-censorship does the work that formal censorship cannot.

The data is striking: scholars interviewed by NPR asked not to be quoted, fearing their research grants would be slashed. Teachers, lawyers, and donors declined to speak on the record. A scholar at UC Davis found that 14-19% of American adults now agree that the government should arrest ordinary people and reporters who publicly criticize its policies.

5. Marginalize Vulnerable Communities

Authoritarians need scapegoats. By targeting vulnerable groups - immigrants, minorities, LGBTQ+ communities - they create a politics of fear that divides the public and draws attention away from institutional erosion. Enforcement actions are designed not just for their direct impact but for their psychological effect: ICE operations that spread fear disproportionately, policies that signal to supporters which groups are "in" and which are "out."

This tactic serves a dual purpose: it energizes a base motivated by cultural anxiety while simultaneously testing how far institutional norms can be bent before they break.

6. Corrupt Elections

Authoritarians maintain electoral facades while systematically undermining democratic competition. This happens through voter suppression laws (19 U.S. states passed restrictive election laws after 2020), strategic gerrymandering, certification interference, and campaign finance advantages. Hungary's electoral law changes made opposition coalitions nearly impossible while maintaining technical democratic compliance.

The sophistication lies in the plausible deniability: each individual measure can be defended as "election security" or "reform." Together, they create a system where elections still occur but outcomes are structurally predetermined.

7. Stoke Violence

The final tactic. Authoritarians either deliberately look the other way or intentionally inflame politically useful violence. Such outbreaks provide political cover for restrictions on civil liberties or expansion of coercive security measures. They suppress opposition voter turnout while energizing supporters. And they undermine the norms of trust and negotiation that democracies depend on.

The Meta-Pattern: Salami Tactics

Hungarian communist leader Matyas Rakosi coined the term "salami tactics" - slicing away at democracy a sliver at a time, each cut too small to trigger massive resistance, until only the shell remains. This is the operative principle. No single action is the breaking point. The accumulation is.

As Protect Democracy puts it: "There is no longer a singular bright line that countries cross between democracy and authoritarianism. But the outcome is still the same."


Part II: The Signals - And the Noise

Separating authoritarian moves from normal political hardball requires a framework. Here are the questions that cut through the noise.

Three Calibration Questions

Drawn from Protect Democracy's framework for journalists, these questions apply equally to citizens:

1. How significantly does this action deviate from modern precedent?

Normal politics involves hardball. Threatening to defund universities for political noncompliance, firing independent inspectors general across the government simultaneously, or a Supreme Court ruling that the president is immune from prosecution for official acts - these have no modern precedent.

2. To what degree is the authoritarian tactic being implemented?

A single instance of political pressure on a media outlet is concerning but could be an isolated event. A systematic campaign of economic coercion against every law firm, university, and corporation that opposes the administration is a pattern - and patterns are the signal.

3. Does this action present a systemic risk to democracy?

Individual actions that are supported by institutions or political parties represent far greater threats than those taken by individual actors alone. When Congress abdicates its oversight role in the face of openly illegal behavior - as Levitsky argues happened with the Department of Justice - the system itself is failing.

The Scorecard: Where Things Stand

The numbers tell a stark story:

  • Bright Line Watch (500+ political scientists): U.S. democracy rated at 67/100 after November 2024. Weeks into the second term, it dropped to 55 - the biggest single decline since tracking began in 2017.
  • The Century Foundation Democracy Meter: From 79/100 in 2024 to 57/100 by end of 2025. A 28% collapse in a single year. State institutions dropped from 22/30 to 10/30.
  • V-Dem Institute: Global democratic advances made over 35 years have regressed to 1986 levels. 72% of the world's population now lives in autocracies.

These scores place the United States alongside Serbia, Turkey, Hungary, and India - countries that maintain elections but where the playing field is so tilted that the term "competitive authoritarianism" applies.

The Signals to Watch

Based on the research, these are the leading indicators - the canary-in-the-coal-mine metrics:

Institutional Compliance Without Resistance When powerful institutions preemptively comply with authoritarian pressure - Jeff Bezos pulling Washington Post editorial positions, Paramount settling, Columbia University capitulating on funding threats, major law firms agreeing to provide pro bono work for the administration under pressure - this is the most dangerous signal. It means the informal norms that hold democratic societies together are dissolving faster than formal institutions can protect them.

Congressional Abdication When the legislature stops performing its oversight function in the face of executive illegality, the most important structural check has failed. Levitsky identifies this as one of the four major developments of democratic backsliding.

Chilling Effects on Speech Not formal censorship but the measurable withdrawal of public criticism: scholars avoiding quotation, donors going quiet, journalists self-censoring. When the cost of dissent rises to the point where rational actors choose silence, the information ecosystem that democracy depends on is breaking down.

Politicization of the Military and Law Enforcement The military swears an oath to the Constitution, not the president. When military leaders are replaced for insufficient loyalty, or when enforcement agencies (ICE, DOJ) are deployed as political instruments, a critical firewall is breached.

Internal Party Fragmentation - Or Its Absence Levitsky notes that authoritarian regimes die when they split internally. The Indiana State Senate's refusal to redistrict on Trump's request, and increasing instances of Republican resistance, are positive signals. Their absence would be among the most alarming indicators of consolidation.

The Noise to Ignore

Not everything is a five-alarm fire:

  • Partisan outrage that doesn't connect to systemic institutional change
  • Political hyperbole from both sides that conflates normal policy disagreement with authoritarianism
  • Individual scandals that are embarrassing but don't erode structural checks
  • Predictions of imminent collapse - competitive authoritarianism is a slow process, and fatalism is itself one of its most effective tools

Part III: What We Can Do - The Democratic Renewal Toolkit

The scholars are clear: neither denial nor despair is appropriate. The threshold has been crossed, but the door back is still open. Here is a concrete toolkit - from the hyperlocal to the systemic - for strengthening democratic resilience.

A. Citizen Assemblies and Sortition

The single most promising institutional innovation in democratic governance is the citizens' assembly: a group of people selected by lottery (sortition), broadly representative of their community, who spend significant time learning and deliberating to form collective recommendations for policymakers.

Why this matters now:

  • Citizens' assemblies bypass the pathologies of partisan politics entirely. Members are randomly selected, not elected, so there's no campaign finance, no party discipline, no incentive to perform for a base.
  • OECD research shows their use has been increasing since 2010, with 189 organizations across 55 countries now working in this space.
  • They have a proven ability to reduce polarization. Research shows that when people from diverse backgrounds are given time, information, and facilitated space to deliberate, they find far more common ground than traditional political forums or media narratives suggest.
  • They have real policy impact: Ireland's citizens' assembly led to the legalization of same-sex marriage and the reform of abortion law. France convened 185 randomly selected citizens to deliberate on end-of-life legislation. Switzerland's 2024-2025 national assembly addressed healthcare costs.

How to start locally: You don't need government authorization to run a citizens' assembly. The Sortition Foundation has run roughly 300 democratic lotteries worldwide. DemocracyNext's Assembling an Assembly Guide breaks the process into nine modules. MASS LBP in Canada has led 55 assemblies, delivering 500,000 invitations.

The key insight from AEI's Daniel Stid: "The only way to ward off a top-down democratic decline is through a bottom-up civic renewal." He advocates using sortition at the community level - in bridge clubs, civic associations, neighborhood councils - not just in formal governance structures.

B. Digital Deliberation Platforms

A new generation of open-source civic technology tools makes democratic participation scalable in ways that weren't possible even five years ago.

Polis (compdemocracy.org) The most important tool in this space. Polis is an open-source platform where participants submit short statements and vote agree/disagree/pass on others' statements. Machine learning algorithms then cluster participants into opinion groups and - critically - surface areas of consensus rather than division. It has been used with groups ranging from 40 to 40,000 people.

Polis was instrumental in Taiwan's vTaiwan initiative, where 26 national technology policy issues were deliberated and 80% led to government action. Audrey Tang, Taiwan's digital minister, described it as "social media that instead of polarizing people to drive engagement, automatically drives bridge-making narratives." Twitter's Community Notes was directly influenced by Polis. Anthropic, Meta, and OpenAI have all used Polis-style approaches for AI governance deliberation.

Decidim (decidim.org) Catalan for "we decide." A free, open-source platform for participatory democracy developed originally by Barcelona's City Council. It supports multi-phase participatory processes: budgeting, planning, public consultations, citizen initiatives with commenting, deliberation, secure voting, accountability tracking. Over 3 million users across 500+ organizations worldwide. Barcelona's 2019 participatory budget saw 40,000 voters and 55,000 votes cast.

CONSUL (consuldemocracy.org) Originally developed by Madrid's City Council. Enables direct citizen participation through proposals, voting, debates, and participatory budgeting. Open-source, used in dozens of countries.

adhocracy+ (liqd.net) Developed by Liquid Democracy e.V. in Berlin. A SaaS platform for modular participation processes: collaborative decision-making, consultation, deliberation, and participatory budgeting. Open-source.

Loomio (loomio.com) Collaborative decision-making tool designed for groups. Supports proposals, polls, and discussion threads. Used by organizations from cooperatives to city councils.

C. Liquid Democracy

Liquid democracy is a hybrid between direct and representative democracy. Voters can either vote directly on issues or delegate their vote to someone they trust - and can revoke that delegation at any time. Delegates can further delegate (metadelegation), creating dynamic chains of representation based on expertise and trust rather than geography and party affiliation.

The key principles:

  • Flexibility: vote directly on issues you care about, delegate on issues where someone else is more informed
  • Accountability: delegation is always revocable
  • Transparency: delegate decisions are public, individual votes are private
  • Specialization: enables expertise-based representation without fixed hierarchies

Real-world experiments:

  • Germany's Pirate Party used LiquidFeedback for internal governance
  • Google Votes experimented with liquid democracy for internal decisions
  • Blockchain governance systems (including those at Anthropic's scale) are actively testing delegated voting mechanisms
  • Major investment firms like BlackRock and Vanguard now allow shareholders to delegate votes through "voting choice" programs

The challenges (be honest about these): Stanford's Andrew B. Hall identifies key structural risks: low participation rates, vote concentration in "superdelegates," and weak civic incentives. These mirror the problems of representative democracy at a different scale. The technology works; the social infrastructure around it needs development.

D. Local Democratic Infrastructure

Not everything requires technology. Some of the most effective democratic renewal happens in physical space, at the smallest possible scale.

Burgerrate (Citizen Councils) Germany has been running Burgerrate at the municipal and federal level. Randomly selected citizens deliberate on specific policy questions - from climate to nutrition - and deliver recommendations to elected officials. The format is spreading across German-speaking countries and the EU.

Participatory Budgeting Give citizens direct control over a portion of the municipal budget. Porto Alegre, Brazil pioneered this in 1989. It has since been adopted by thousands of municipalities worldwide. The evidence is clear: participatory budgeting increases civic engagement, improves public trust, and produces spending decisions that better reflect community needs.

Community Land Trusts and Cooperatives Democratic governance doesn't have to mean government. Community land trusts, housing cooperatives, energy cooperatives, and platform cooperatives all practice democratic decision-making at a scale where participation is meaningful and consequences are tangible. These institutions build the civic muscle that national democracy depends on.

Local Journalism The collapse of local news is one of the most underappreciated drivers of democratic erosion. Areas that lose local newspapers see lower voter turnout, less informed citizens, and more corruption. Supporting local journalism - through subscriptions, community-funded newsrooms, or civic information platforms - is a direct investment in democratic infrastructure.

E. The Institutional Reform Agenda

For those with access to policy levers, the Brookings Democracy Playbook 2025 and the Protect Democracy coalition recommend:

  • Protect judicial independence: resist court-packing schemes, defend lifetime tenure for federal judges, strengthen state-level judicial appointment processes
  • Reform electoral systems: ranked-choice voting, independent redistricting commissions, automatic voter registration, and campaign finance reform all reduce the structural advantages that authoritarians exploit
  • Strengthen legislative oversight: Congress must reassert its constitutional role - the power of the purse, confirmation oversight, and subpoena authority
  • Protect press freedom: legal protections for journalists, antitrust enforcement against media consolidation, public funding for independent journalism
  • Build coalitions across ideological lines: as AEI's Stid argues, democratic reform should not be a partisan project. Conservative and progressive cases for institutional resilience both exist - and the authoritarian threat affects both sides

Part IV: The Core Insight

Levitsky captures it precisely: the two greatest dangers are complacency ("it can't happen here") and fatalism ("it's already over"). Both are wrong, and both serve the authoritarian project equally - the first by preventing action, the second by making action seem pointless.

Competitive authoritarianism succeeds when fear, exhaustion, and resignation outpace commitment to democratic norms. That is precisely what it is designed to produce.

The playbook is public. The signals are measurable. The tools exist. The question is whether enough people will use them before the window closes - or whether the slow drip of normalization will do its work.

The answer to "salami tactics" is not a single heroic act. It is a thousand small ones: a citizens' assembly in your municipality, a Polis conversation in your community, a subscription to your local newspaper, a refusal to self-censor, a willingness to show up.

Democracy is not a state. It is a practice. And practices only survive if people practice them.


Sources and further reading:

  • Protect Democracy - The Authoritarian Playbook
  • Protect Democracy - Authoritarian Action Watch Threat Tracker
  • Brookings Institution - Democracy Playbook 2025
  • Levitsky & Way - "America Is Now a Competitive Authoritarian Regime" (Foreign Affairs)
  • Senator Jeff Merkley - Ten Rules of Trump's Authoritarian Playbook (January 2026)
  • The Century Foundation - Democracy Meter
  • Bright Line Watch - Expert Surveys on Democratic Performance
  • DemocracyNext - Assembling an Assembly Guide
  • Polis / The Computational Democracy Project
  • Decidim - Free Open Source Participatory Democracy
  • Sortition Foundation
  • People Powered - Best Digital Participation Tools 2025
  • Harvard Kennedy School - Democracy in 2025 (January 2026)
  • Stanford / Hoover Institution - Liquid Democracy (PolicyEd, May 2025)