How Power Changes Hands
Elections, capture, and the modern political playbook. How leaders rise, how governments consolidate, and how popular mobilisation succeeds or fails.
(V-Dem and Freedom House both show a roughly continuous slide since around 2007-08)
(Freedom House 2024; ~38% in "Partly Free" and ~33% in "Not Free")
(elections happen but the playing field is sharply tilted toward the incumbent)
A note on framing. Politics is one of the topics where it is easiest to slip from analysis into advocacy. This page tries hard not to. The goal is to describe how political power actually changes hands in the world as it exists in 2026 - which leaders rise via which paths, which techniques governments use to consolidate or defend power, and when popular mobilisation overturns or fails to overturn an entrenched government - using cases from across the political spectrum and naming the researchers whose work the descriptions rest on. Where readers from different political traditions are likely to disagree about what the cases mean, the disagreement is laid out plainly rather than resolved.
What "winning power" actually means
In a textbook democracy, winning power is straightforward: you win an election, you take office, you govern within constitutional limits, you are subject to losing the next election. In nearly every real political system, the picture is more layered. Winning the formal vote is one piece. Securing legitimacy in the eyes of enough of the population to govern effectively is another. Maintaining the loyalty of the institutions that actually carry out government decisions - the civil service, the military, the courts, the central bank, the prosecutorial system - is a third. A leader can win the election without consolidating any of the other layers, and a leader who has consolidated all of them can survive losing an election.
This three-layered picture - vote, legitimacy, institutional control - is the framework most modern political-science research actually uses. Larry Diamond, Steven Levitsky, Lucan Way, and others have spent careers describing the gap between the textbook model and the way it works in practice. The widening of that gap over the past two decades is what some scholars call the global democratic recession, though others contest whether the concept is as universal as its proponents claim.
Power changes hands through several distinct mechanisms. The most common is ordinary election turnover, which still happens routinely in most democracies. The second is succession within an existing dominant party or family - common in single-party states and family dynasties. The third is military intervention, which has become rarer than during the Cold War but still happens (Egypt 2013, Myanmar 2021, Niger 2023, Gabon 2023). The fourth is mass popular mobilisation that forces a sitting leader out, often called colour revolution when it succeeds and "instability" when it does not. The fifth is foreign-supported regime change - NATO's intervention in Libya (2011), contested Western engagement in Ukraine's political transitions, Russian military interventions in Georgia (2008) and Syria (2015), and longer histories of US and Soviet/Russian involvement in Latin America, the Middle East, and Africa are all documented cases. How common this mechanism is and who initiates it more often are themselves contested questions that track closely with the observer's geopolitical orientation. Distinguishing between these mechanisms is more useful than treating "regime change" as a single phenomenon.
The amateur-politician path
One of the more striking patterns of the past decade is how often political leadership in democratic countries has gone to people whose background was not in traditional politics, law, or government. Donald Trump in the United States came from real estate and television. Volodymyr Zelensky in Ukraine came from comedy and entertainment production. Silvio Berlusconi in Italy came from media and football. Jair Bolsonaro in Brazil came from the military, with relatively limited mainstream political traction before his presidential run. Javier Milei in Argentina came from television-economist celebrity. Nayib Bukele in El Salvador came from advertising. The pattern crosses left and right and crosses regions; it is not a feature of any single political tradition.
The reasons this has become more common are partly structural and partly contingent. Voter trust in established political parties has been declining for several decades across most democracies, which lowers the value of being a known party insider. Mass-media attention concentrates around recognisable personalities, which lets non-politicians enter the race with name recognition that previously had to be built through years of party work. Social media has made direct-to-voter communication possible at a scale that bypasses party structures and traditional gatekeepers. The rise of populist styles - speaking directly to "ordinary people" against "elites" - rewards outsider profiles. Yascha Mounk and Ivan Krastev have each written carefully about why these conditions emerged after the 2008 financial crisis and the political realignments that followed.
What this does to governance varies. Some amateur politicians have governed competently in narrow areas; some have governed disastrously. Zelensky's wartime leadership has been praised by Western allies for communication effectiveness and resolve, while critics - including Ukrainian journalists and some civil-society groups - have raised concerns about wartime centralisation of power, corruption allegations within the administration, the cancellation of scheduled elections under martial law, constitutionally contested sanctions applied to Ukrainian citizens, and documented reports of forced-mobilisation abuses by territorial recruitment centres (TCC) that have included detaining men off streets outside legal procedure. Whether these represent wartime necessities or an erosion of the democratic standards Ukraine claims to defend is genuinely debated, and the Western tendency to focus on the first framing while downplaying the second is itself a pattern worth noticing. Trump's two terms produced sharp partisan disagreements about policy substance that this page does not try to adjudicate. Berlusconi's Italian governments produced corruption convictions and broad democratic-quality declines in V-Dem indices. Bukele's El Salvador has substantially reduced violent crime through mass incarceration that human-rights groups have flagged as authoritarian. The pattern is not "amateurs are good" or "amateurs are bad"; the pattern is that the conditions that produce amateur politicians also produce particular governance behaviours that career politicians less often display - higher willingness to confront institutions, lower tolerance for procedural constraints, more reliance on personal popularity than party discipline. Whether one weighs those traits as advantages or risks tracks closely with one's existing political priors, which is why honest disagreement persists.
Counter-examples are worth holding in mind. Career politicians with traditional backgrounds still win major elections regularly - Emmanuel Macron in France was technocratic if not party-traditional; Donald Tusk's return in Poland in 2023 came from the most conventional centre-right institutional path; Olaf Scholz, Pedro Sánchez, Mark Carney are all variations of the experienced political insider. The amateur path is now common enough to be a recognised pattern, not common enough to be the dominant one.
How leaders consolidate power
Every political system produces leaders who try to accumulate more power than their position was designed to give them. The techniques are remarkably consistent across democratic, authoritarian, and hybrid systems: capture the institutions that check executive power, constrain the information environment, use legal tools selectively against opponents, and rewrite the rules to favour the incumbent. Western political science calls this process "democratic backsliding" when it occurs in countries previously classified as democratic (Steven Levitsky and Daniel Ziblatt's How Democracies Die, 2018, is the landmark study). Scholars in other traditions - including realist and non-Western frameworks - describe equivalent processes with different terminology and different normative assumptions. The techniques themselves are not confined to any regime type.
Capture the courts. Replace senior judges; pack the constitutional court; create new appellate bodies that bypass the existing judiciary; lower the retirement age to force out incumbent justices. Hungary under Viktor Orban executed this most thoroughly between 2010 and 2013, restructuring the Constitutional Court and lowering judicial retirement to remove sitting judges. Poland under PiS attempted similar moves between 2015 and 2023 and was partially reversed after Tusk's 2023 election. Israel's 2023 judicial-reform package was a contested attempt at the same kind of restructuring; it was halted partly by mass street protests and partly by the October 2023 war shifting political focus.
Capture the election commission. Replace independent commissioners with loyalists; redraw districts in ways that lock in incumbent advantage; change election rules in the months before a vote. The most cited Western example is the politicisation of state-level election administration in the United States after 2020, with claims and counter-claims about the substantive effect. Well-documented cases in authoritarian states include Belarus's Central Election Commission since the 1990s and Venezuela's National Electoral Council under both Chávez and Maduro.
Weaponise prosecutors. Use criminal investigation as a political tool - against opposition leaders, against journalists, against business figures who fund opposition. The pattern is documented across regimes that disagree about most other things: Russia (Khodorkovsky 2003, Navalny 2014-2024 - Western governments and human-rights organisations classify these as political persecution of legitimate opposition; the Russian government maintains Khodorkovsky's prosecution addressed documented financial crimes during the 1990s privatisation period and that Navalny's organisations had established foreign-funding connections that warranted legal scrutiny under extremism statutes; both readings have supporting evidence and neither fully accounts for all the facts), Turkey (Erdogan's mass purges after 2016), Brazil (the Lava Jato investigations whose political effects are debated), the United States (where overlapping prosecutions of political figures from both major parties have raised questions across the spectrum about prosecutorial discretion as a partisan tool). Western democracies are not exempt from the pattern: France arrested Telegram founder Pavel Durov in 2024 on charges related to platform content moderation - framed by prosecutors as enforcing criminal law, read by press-freedom organisations and technology critics as an attempt to pressure encrypted communications and platform independence. The US and UK pursued Julian Assange for over a decade for publishing classified material, in a case that press-freedom advocates described as criminalising journalism while national-security officials described as prosecuting espionage. The structural pattern is the same across all these cases: the prosecution has a stated legal basis that is defensible on its terms, and the political selection of who gets prosecuted reveals priorities that are not purely legal.
Constrain media. Modify broadcast-licensing rules; create regulatory pressure on critical outlets; reward loyal outlets with state advertising and tax breaks; raise the cost of independent journalism through libel-friendly law or selective application of media regulation. Hungary, Poland, India, Turkey, Russia, and Venezuela have each used variations on this. Western democracies have used structurally similar tools: the European Union banned RT and Sputnik broadcasts entirely in 2022; the United States pressured technology platforms to limit Russian state media reach and revoked credentials; the United Kingdom revoked RT's broadcast licence; France's arrest of Durov was widely read as a signal to encrypted-communications platforms. The 2022-2023 "Twitter Files" disclosures in the United States documented extensive government communication with social-media platforms requesting content moderation of specific accounts and narratives - defenders called this coordination against disinformation, critics called it state-directed censorship through private intermediaries. The structural question is whether any of these actions differ in kind from the media constraints Western governments criticise when authoritarian states do them, or only in degree and stated justification. Within domestic politics, selective application of media regulation has been alleged in most major democracies. The disagreement is durable in both directions.
Constrain civil society. Foreign-agent laws, NGO-registration requirements, restrictions on assembly, surveillance of activist organisations. Foreign-agent registration is not unique to authoritarian states - the United States' Foreign Agents Registration Act (FARA, 1938) predates all modern equivalents and has been used against both Russian and allied-state lobbyists, though its enforcement has been criticised as inconsistent. Russia's foreign-agent law (expanded since 2012) drew explicitly on FARA as a precedent and has been adapted by Hungary, Turkey, Israel (in narrower form), India, Georgia, and several others. The argument for these laws is genuinely about transparency regarding foreign funding; the practical effect depends heavily on how broadly and selectively they are applied. Russia's version has been extended far more widely to domestic civil society, independent media, and individual journalists than FARA typically is in the United States, though critics note that FARA enforcement has itself become more aggressive and politically selective in recent years.
Rewrite or stretch the constitution. Term-limit removal (a widespread technique - Putin 2008/2012/2020, Maduro, Erdogan, Xi 2018; Western democracies have their own version in the form of aggressive interpretation of existing executive authority and expansion of emergency powers), constitutional amendments that expand executive power, or aggressive interpretation of existing constitutional clauses. Hungary rewrote its constitution in 2011 (the Fundamental Law); Russia substantially amended its constitution in 2020 to allow Putin to serve until 2036.
The deeper pattern here is not that some countries have leaders with authoritarian instincts and others do not. The impulse to consolidate power, neutralise opposition, control information, and use legal tools selectively is observable across every political system - democratic, authoritarian, and everything in between. What varies is not human nature but institutional friction: how many independent power centres exist, how difficult they are to capture simultaneously, and how much political cost a leader pays for overreach. The difference between a country where a prosecutor can be weaponised with impunity and a country where the attempt triggers institutional pushback, media scrutiny, and electoral consequences is not a difference in the character of the leaders but a difference in the architecture of the constraints they operate within. Levitsky and Ziblatt's central insight is that democracies more often die incrementally through legal-looking moves than dramatically through coups - and that those incremental moves look identical regardless of whether they occur in Moscow, Ankara, Budapest, Washington, or Paris. Each individual reform looks defensible on its own terms; the trajectory it is part of is what determines whether it is consolidation or ordinary politics. The line between the two is genuinely hard to draw in real time, which is why honest people disagree about specific cases even when they agree about the structural pattern.
When popular movements succeed or fail
On the other side of the same dynamic is the question of when popular mobilisation succeeds in removing an entrenched government. The early 2000s produced a wave of cases that came to be called colour revolutions, named after the symbols and colours used by the protest movements: Serbia 2000 (Bulldozer Revolution / Otpor!), Georgia 2003 (Rose), Ukraine 2004 (Orange), Kyrgyzstan 2005 (Tulip), Lebanon 2005 (Cedar). The 2010-2011 Arab Spring extended the pattern to the Arab world, with Tunisia (Jasmine) the clearest successful case and Egypt's 2011 revolution reversed by the 2013 military takeover. Ukraine's 2013-2014 Euromaidan / Revolution of Dignity, Bangladesh's 2024 protests against Sheikh Hasina, and several smaller cases continued the pattern.
What makes these movements work, when they work, has been studied carefully. Erica Chenoweth's research on civil resistance found that nonviolent campaigns succeed substantially more often than violent ones, and that mobilisation reaching roughly 3.5 per cent of the population participating actively predicts success in most cases (the figure is widely cited though sometimes oversimplified - it is a correlation in historical data, not a guarantee). Srdja Popović, formerly of Otpor!, has written about the practical organising techniques that recur across successful cases: clear branding, low-cost mass-participation actions, humour against fear, careful sequencing of escalation, defection of security forces.
Two readings of the 2000s wave coexist, and neither can be fully dismissed. The first: these were genuine popular movements against governments that had lost legitimacy, with foreign funding playing a secondary facilitation role - the Serbian Otpor! movement reflected real Serbian discontent with Milošević, and Western training and funding helped organise it but did not create the discontent. The second: Western governments and foundations played a strategic role in selecting, funding, training, and diplomatically supporting specific opposition movements as part of a broader geopolitical project - and the existence of popular discontent does not mean external actors were not substantially shaping its direction, timing, and outcome. The Ukrainian Orange Revolution and Euromaidan are the cases where this tension is sharpest: documented Western institutional engagement and documented Russian counter-engagement existed alongside real popular mobilisation, and observers weight these factors differently depending on their geopolitical starting point. The most defensible position is that both domestic grievance and external strategic engagement were real and consequential in most cases, and that insisting on either as the sole primary driver reveals more about the observer's framework than about the events themselves.
Failed cases are equally instructive. Belarus 2020 saw mass protests after an election whose results were rejected by the opposition and most Western governments as fraudulent, and upheld by the Belarusian government and its allies; the Lukashenko government survived by combining mass arrests, internet shutdowns, and Russian backing. Hong Kong's 2014 Umbrella Movement and 2019 protests were eventually crushed by a combination of mainland intervention and the National Security Law. Iran's 2009 Green Movement and 2022 Mahsa Amini protests both faced governments willing to use lethal force at scale. Russia's anti-Putin protests of 2011-2012 and 2017-2021 were contained through arrests, foreign-agent designations, and after 2022 a more comprehensive crackdown. The lesson from the failed cases is partly about government willingness to use force, partly about external support to the government (Russian backing for Belarus, Chinese for Hong Kong, none for Iran but no Western counterweight either), and partly about the maturity of state defences against mobilisation.
How governments defend against mobilisation
Governments across the political spectrum - authoritarian, hybrid, and democratic - have developed increasingly sophisticated techniques for anticipating and containing popular mobilisation. The most visible applications are in authoritarian states, but democratic governments have adopted elements of the same toolkit in the domains of surveillance, platform regulation, and information management. The result is that organising effective mass movements is materially harder now than in the early 2000s, regardless of the political system one is operating within.
Pre-emptive arrests. Identifying organisers before mobilisation reaches mass scale, and arresting them at the planning stage rather than at the protest stage. Belarus arrested most of the 2020 election-monitoring infrastructure before the vote. Russia arrested Navalny in 2021 on his return and progressively detained members of his network - the Russian government classified the organisations as extremist, citing documented foreign funding and operational ties to Western institutions; Western governments and human-rights organisations classified the arrests as political persecution of legitimate opposition. China's National Security Law in Hong Kong was used to arrest pro-democracy figures across a wide range of activities.
Internet shutdowns and surveillance. Cutting off mobile data and messenger applications during protest moments, and routine surveillance of opposition communications. Iran's repeated internet shutdowns since 2019, Russia's progressive Telegram and VPN restrictions, China's Great Firewall, and India's regional shutdowns (especially in Kashmir) are documented examples. Western democracies do not shut down the internet but have built their own mass-surveillance infrastructure - the Snowden disclosures (2013) revealed the scale of NSA and Five Eyes bulk data collection on domestic and foreign communications, and subsequent reporting has documented continued expansion of state surveillance capabilities in the US, UK, France, and elsewhere. The tools differ in visibility but not necessarily in scope. Access Now's annual #KeepItOn report tracks the global shutdown pattern.
NGO restrictions and foreign-agent laws. Making it legally costly to run a civil-society organisation that receives any foreign support, and using the resulting designations to constrain critics. The Russian foreign-agent law (2012, expanded multiple times since) is the most prominent example; variations exist in Hungary, China, Turkey, India (FCRA restrictions), Georgia (passed 2024 over mass protests), and elsewhere. As noted above, the US FARA predates all of these.
Vote inflation rather than vote rigging. Making the formal election so favourable to the incumbent through pre-vote tilting (media access, prosecutor weaponisation, candidate disqualification) that the vote count itself does not need to be falsified to produce a large victory. Russia's elections since 2012 are the clearest example; Egypt's elections under Sisi follow the same logic. This is more durable than crude vote-rigging because it is harder for international observers to call out.
Co-opt nationalism and identity. Frame opposition as foreign-aligned, ethnically disloyal, or culturally alien. This technique is not confined to authoritarian states. Russia frames domestic opposition as Western-funded agents (a framing that has some documented basis in Western civil-society funding and is also used to delegitimise genuine domestic dissent - both things can be true simultaneously). Western political discourse uses the mirror image: labelling domestic critics, journalists, or political opponents as "Russian agents," "Kremlin puppets," or "useful idiots" - a framing that also has some documented basis in actual Russian influence operations and is also used to delegitimise legitimate domestic dissent. Erdogan frames opponents as Gulenist conspirators; China frames Hong Kong protesters as foreign-influenced; India frames opposition Muslim figures as anti-national. In every case, the framing does not need to be fully believed; it needs only to be partly believed by enough of the regime's or government's base to preserve legitimacy and discredit the opposition's message without engaging its substance.
Co-opt economic delivery. When governments can deliver real economic improvements - or at least the perception of them - through state largesse, infrastructure, or stability, popular mobilisation becomes harder. China's reduction in extreme poverty since the 1990s is the most consequential case. Russia's economic stabilisation between 2000 and 2008 was central to the popular base of Putin's early consolidation. Singapore's long PAP rule rests substantially on real delivery.
What this means in practice is that the colour-revolution model is harder to execute now than in 2000. Belarus 2020, Iran 2022, Hong Kong 2019 all generated mass mobilisation that earlier conditions would have suggested could win; in the modern counter-playbook environment, each was contained. Bangladesh's 2024 success and Armenia's 2018 Velvet Revolution show the model still works in some conditions, but the conditions are narrower.
The legitimacy game
Modern political power is not held only through institutional control. It is also held through legitimacy - the broad acceptance, even by people who did not vote for the leader, that the leader has a credible claim to govern. Legitimacy is sustained through a combination of perceived electoral fairness, perceived performance, and what Joseph Nye has called soft power applied domestically: cultural framing, media presence, symbolic dominance.
In a televised, social-media, polling-everywhere environment, legitimacy is partly performative. Approval ratings, social-media engagement, sympathetic press coverage, and cultural visibility all feed into and feed off each other. Leaders who understand this can hold power even when their formal record is weak; leaders who do not understand it can struggle to govern even with strong formal mandates. The shift from policy-as-deliverable to policy-as-performance is documented across democracies and authoritarian governments alike.
The structural feature this has produced - alongside many others - is an arms race in political-communication infrastructure. Every major political organisation now invests heavily in social-media operations, micro-targeting, sentiment monitoring, and rapid-response messaging. The Cambridge Analytica disclosures of 2018 made this visible at the scandalous end; the routine end is now invisible because it is universal. Whether this has hollowed out the substance of governance or merely changed its packaging is one of the more contested questions in current political-science research.
One related observation worth holding: the techniques described in this whole page - amateur-politician rise, power consolidation, mobilisation, counter-mobilisation, legitimacy game - all have a common structural feature. They are all responses to a political environment in which traditional gatekeepers (parties, mainstream media, established civic institutions) have weakened. Whether that weakening is something to be reversed, accepted, or replaced with something better is a deeper political-philosophical question that this site does not try to answer. What it tries to do is describe what has actually happened clearly enough that a reader can think about the question for themselves.
The paths from here
Continued slow democratic recession
The roughly seventeen-year trend continues. More countries drift from democratic to competitive-authoritarian classifications. Established democracies experience more politically-driven prosecutions, court fights, and election-rule battles. No major reversal; no major collapse. This is the trajectory the data has shown for nearly two decades.
A democratic resurgence emerges
A combination of generational change, specific policy victories that rebuild trust in functional government, and difficulties faced by non-democratic models (Russia's strategic overextension, China's economic slowdown, Iran's domestic instability - though each of these can also be read as temporary setbacks rather than systemic failures) produces a slow reversal. Examples like Poland's 2023 reversal, Brazil's 2022 Lula return, and Bangladesh 2024 hint that the trend is not unidirectional.
Sharpened backsliding in major democracies
One or more major democratic states - the United States, India, Israel under different conditions - moves further along the backsliding playbook than at present, producing a measurable shift in the global picture. The cascade effect on smaller democracies that take cues from larger ones could be substantial.
Major political transitions in non-democratic states
Russia, Iran, Venezuela, or Belarus - or a combination - experiences a transition triggered by economic stress, succession crisis, or internal political realignment. Past waves of political transition came in clusters (1974-90 in Southern Europe, Latin America, and post-Soviet space; 2010-11 across the Arab world). Whether such transitions lead to democratic consolidation, new forms of authoritarianism, or prolonged instability varies by case. A new wave is structurally possible but historically unpredictable.
AI-driven political-communication transformation
AI-generated content, automated micro-targeting, deepfake video, and AI-assisted campaign infrastructure reshape what political competition looks like. Both democratic and authoritarian players will use these tools; the equilibrium depends on regulation that has not yet been written. The next few election cycles in major democracies will be the first proper tests.
A new model of mobilisation emerges
Just as colour revolutions developed in the early 2000s and the government counter-playbook in the 2010s and 2020s, the next decade may produce a new mobilisation model adapted to the modern surveillance and information environment. Possibilities include decentralised leaderless movements (Hong Kong 2019 partially), AI-augmented organising, or hybrid online-offline forms not yet visible. Whether such models can succeed against the mature counter-playbook is unknown.
Non-democratic models demonstrate competitive governance
China, Singapore, the Gulf states, or other non-democratic systems demonstrate sustained delivery of economic growth, infrastructure, public safety, or technological achievement that leads a substantial share of the global population to view their governance model as legitimate and competitive with - or superior to - electoral democracy on outcomes that matter to ordinary life. This is already the lived experience of billions of people; the question is whether the next decade makes it more visible and more explicitly articulated as an alternative model, or whether internal contradictions (succession crises, economic slowdowns, demographic decline) undermine the case before it fully crystallises.
Where serious analysts disagree
Backsliding is the central political story of our time
The accumulation of executive power, the capture of courts and election commissions, and the erosion of institutional independence in country after country - including some that previously seemed durably democratic - is the dominant pattern of the past two decades. Without a serious response, the path leads to a substantially less democratic world by mid-century.
Held by: Steven Levitsky and Daniel Ziblatt (Harvard, "How Democracies Die"; "Tyranny of the Minority"), Larry Diamond (Stanford), Anne Applebaum, the V-Dem Institute (Gothenburg), Freedom House. The case has empirical support in the indices and substantial institutional momentum in the democracy-and-rights community.
The "backsliding" frame is itself politicised
Many of the moves labelled democratic backsliding by Western academics - judicial reform, media regulation, civil-society law, electoral system changes - are normal politics that look threatening only when conducted by parties or movements that the labelling academics already oppose. Treating populist or right-wing reform as backsliding while treating equivalent left-wing reform as democratic renewal is itself a form of partisan politics dressed up as analysis.
Held by: scholars who contest the Western-liberal framing of democratic indices, including conservative and sovereigntist political scientists, researchers in Hungary, India, Israel, and Russia who argue the benchmarks themselves embed liberal-democratic assumptions, and parts of the public-choice school. The case has merit on specific instances where the "backsliding" label appears selectively applied and weaknesses on the systematic patterns documented by V-Dem and similar projects, which attempt to be country-neutral in methodology.
Colour revolutions were primarily domestic movements
The colour-revolution successes of the 2000s were primarily driven by genuine domestic discontent with corrupt or fraudulent governments. Western funding for civil society and election monitoring was a real but secondary factor. The government narrative that frames colour revolutions as foreign-engineered is a defensive frame used to delegitimise domestic grievances.
Held by: Erica Chenoweth (Harvard), Srdja Popović and the CANVAS organisation, much of the human-rights community, Mark Beissinger (Princeton, "The Revolutionary City"). The case has empirical support in survey work on protest motivation and weaknesses in cases (Ukraine 2014 in particular) where Western institutional engagement was demonstrably substantial.
Colour revolutions involved meaningful Western strategic engagement
While popular discontent was real, US and European governments and foundations played substantive roles in training organisers, funding civil-society infrastructure, supporting election monitoring, and providing diplomatic cover. Treating these movements as purely organic understates the geopolitical context in which they operated. Russia, China, and other authoritarian states draw the lesson that domestic civil society receiving foreign funding is a national-security concern - whether one agrees with that conclusion or not, it explains the foreign-agent laws that have proliferated since.
Held by: realist IR scholars including parts of the John Mearsheimer school, researchers based in Russia and China, some Eastern European political scientists. Has empirical support in documented Western institutional engagement and weaknesses when applied to cases (Belarus 2020, Iran 2022) where the foreign-engagement story was much weaker than the domestic story.
Voter dissatisfaction is the central variable, not regime type
Both democratic backsliding and successful colour revolutions are downstream of populations losing trust in incumbent governments. Where governments deliver - economically, in public services, in security - they survive, regardless of whether they call themselves democratic or authoritarian. Where they fail to deliver, they become vulnerable to mobilisation or to populist replacement, again regardless of regime type. The right level of analysis is governance quality and economic performance, not formal political institutions.
Held by: Martin Wolf, parts of the development-economics community, public-choice scholars, and political scientists focused on outcomes rather than process. Has empirical support in cross-country governance data and weaknesses in cases where economic performance was reasonable but political mobilisation occurred anyway (Hong Kong 2019, Belarus 2020).
None of these readings is fully right or wrong, and this page does not attempt to pick a winner among them. What can be said without taking a side: the techniques described above - power consolidation, popular mobilisation, and government counter-mobilisation - are observable across systems and are driven by structural incentives that operate regardless of ideology. The balance between them in any given country depends on institutional architecture, economic conditions, external pressures, and choices made by specific leaders and populations. The trajectory of the next decade will be shaped by decisions that have not yet been made, in democracies and non-democracies alike. This page's analysis draws primarily on Western institutional research (Levitsky, Ziblatt, Chenoweth, Diamond, and the V-Dem project); readers should be aware that this starting point shapes which patterns are highlighted and which are backgrounded, and that scholars working from other traditions would organise the same facts differently.
What this means for you
If you read political coverage
Distinguishing between the three layers - vote, legitimacy, institutional control - is one of the most useful filters available. When coverage focuses on "who won the election," the more important questions are often who has captured which institution, what the legitimacy environment looks like, and whether the playbook moves are accelerating or being resisted. The V-Dem annual report, Freedom House's "Freedom in the World," and the Economist Intelligence Unit's Democracy Index are the three most consistently used cross-country measurements; reading the methodology notes alongside the rankings is more useful than reading the rankings alone.
If you live in a democracy under stress
The protections that work most reliably are unglamorous: independent prosecutors with stable funding, professional civil services with civil-service protections, courts whose appointments are insulated from short-term political cycles, free press with diverse ownership, free assembly. Engagement that supports these institutions - donations, voting in down-ballot races for prosecutors and judges where they are elected, attention to civil-service-protection legislation - has higher leverage than national-level political activism for most citizens. Levitsky and Ziblatt are explicit that the most important defenders of democracies are usually unglamorous local-level officials and institutional norms, not charismatic politicians.
If you live in a non-democratic or competitive-authoritarian state
Some citizens of non-democratic states are dissatisfied with their government and would prefer different political arrangements; others are broadly satisfied, especially if the government delivers on economic growth, public safety, or national prestige. Both positions are real. For those who are dissatisfied: political activism in such systems carries personal risk that ranges from career consequences to imprisonment to death. Each individual makes that calculation for themselves. The historical record suggests that most successful transitions involve sustained pressure over years rather than dramatic single moments, and that movements survive by combining visible mobilisation with quieter organisational infrastructure that persists between protest waves. For those who are satisfied: the structural patterns described on this page - power consolidation, institutional capture, information control - are worth understanding even if one supports the current government, because the same techniques that protect a competent leader from destabilisation also protect an incompetent successor from accountability. The question is always what happens after the current leadership.
If you have political reach or media platform
The tone of political commentary - especially commentary that is itself widely shared - shapes the legitimacy environment more than its specific content. Sharp partisan framing of opponents as illegitimate, of institutions as captured, of elections as rigged unless one's side wins - all of these accelerate the conditions that produce backsliding regardless of which party is using the framing. The most consequential rhetorical discipline is treating the institutional rules as legitimate even when one is losing under them. This is harder than it sounds and is the discipline most consistently identified as protective by political scientists who study institutional resilience.
If you vote
Most of the consequential decisions about political-power dynamics happen in elections that are not the headline national election: state-level prosecutor races, judicial appointments, election-administration officials, school boards, redistricting commissions. Turnout in down-ballot races is consistently lower than in presidential or national-leader races, even though the down-ballot races more directly affect the institutional infrastructure that determines whether the headline races are conducted fairly. Showing up for those races - and paying enough attention to vote with information rather than party reflex - is one of the more durable forms of political engagement available.
