Insights from the comparative literature

Prominent recent contributions on democratic consolidation have played a key role in reframing the debate, drawing attention to the need for more nuanced conceptualisations of resilience and greater analytical precision in identifying the mechanisms that sustain it (Boese et al. Reference Boese, Edgell, Hellmeier, Maerz and Lindberg2021; Croissant and Lott Reference Croissant and Lott2025; Holloway and Manwaring Reference Holloway and Manwaring2023; Riedl et al. Reference Riedl, Friesen, McCoy and Roberts2025). Although this work has approached the subject from different vantage points, it has collectively advanced several core insights that have shaped the emerging consensus within the field. One major contribution has been the recognition that democratic resilience should not be understood as a fixed attribute or a permanent state (Boese et al. Reference Boese, Edgell, Hellmeier, Maerz and Lindberg2021), but rather as a process of continuous adaptation. In contrast to earlier work that treated resilience as an inherent quality of well-institutionalised democracies, recent research portrays it as a dynamic phenomenon shaped by how political actors, institutions and societies respond to challenges over time. Resilience is, therefore, best understood as a series of ‘patterned adjustments’ (Bourbeau in Holloway and Manwaring Reference Holloway and Manwaring2023: 72), through which democracies seek to manage and recover from political stress. In turn, this emphasis on processes has redirected scholarly attention away from the question of which democracies are resilient towards looking at how resilience is enacted and maintained, emphasising agency and institutional evolution rather than structural preconditions alone.

A second key insight concerns the distinction between internal and external drivers of democratic decline and recovery. Rachel Riedl and colleagues’ recent work has been instrumental here (2025; see also Carothers Reference Carothers2024), emphasising that the forces shaping resilience are often multilayered and cannot be reduced to domestic political dynamics. Internal drivers include elite behaviour, institutional design, party system configuration and the strength of civic and media oversight. External drivers, by contrast, encompass regional and global norms and values, as well as international economic conditions. This distinction matters because it highlights the varying sources of both vulnerability and protection across different contexts. In some states, for example, democratic decline may stem primarily from the internal capture of oversight institutions, while in others, it may reflect dependence on external patrons that enable authoritarian consolidation. These two sets of factors are not mutually exclusive but interact in complex ways: external leverage can reinforce domestic resilience, as when regional organisations defend constitutional norms, but it can also weaken it when international incentives reward authoritarian loyalty. This approach thus challenges overly insular models of democracy that treat regimes as self-contained systems and underscores the need to locate resilience within a web of domestic and transnational interdependencies.

Third, there has been a growing effort to distinguish between the resilience of systems – referring to the institutional and societal features that help democracies endure – and the resistance mounted by individual actors and civil society groups (Gamboa Reference Gamboa2022; Hou Reference Hou, Mehta and Palazzo2020; Tomini et al. Reference Tomini, Gibril and Bochev2023). While resilience is often conceptualised at the systemic level, resistance captures the agency of citizens and organisations. Early analyses of democratic decline often blurred this distinction, focusing primarily on institutional strength or, conversely, on episodes of civic mobilisation, without exploring how the two might reinforce one another. More recent work, particularly that of Marie-Eve Desrosiers et al. (Reference Desrosiers, Cheeseman, Beardsworth, Gyimah-Boadi and Hassan2025), conceptualises these two processes as being interdependent but not interchangeable. In contexts where formal institutions are weak or captured, resistance from below can temporarily substitute for institutional resilience, keeping democratic norms alive through collective mobilisation, protest and advocacy. Yet institutional resilience remains essential for consolidating democratic recovery over the long-term. This interplay between resistance and resilience lies at the heart of explanations of democratic survival.

Fourth, the literature has increasingly made clear that the determinants of democratic resilience are context-dependent. Holloway and Manwaring’s (Reference Holloway and Manwaring2023) meta-review reveals that the most cited factors, such as a state’s degree of economic development, the integrity of electoral institutions and the strength of civil society (see Table 1), do not operate uniformly across cases. Instead, their effects are conditioned by political history, institutional legacies and the nature of the threat(s) to democracy. The effectiveness of different forms of resilience and resistance is also shaped by the relationship between different ‘building blocks’, which may amplify or undermine one another through their presence or absence. The judiciary may play a key role in checking the abuse of power, for example, but this is less likely if judges do not benefit from a strong constitution, vigilant legislature and supportive public to safeguard their security of tenure.

Table 1.

Factors Contributing to Democratic Resilience

Table 1 Long description

The table lists types of factors and explains how each can bolster democratic resilience. Economic development and diversified economies can reduce elite resistance and support social changes such as higher education and a larger middle class. Democratic norms and attitudes among citizens encourage accountability, elections, institutional limits, and peaceful competition. A vibrant civil society supports participation and oversight and can mobilize resistance to democratic erosion, while prior democratic experience provides institutional memory that can be reactivated during stress. Institutional safeguards include rule of law, independent courts, constitutional constraints, empowered legislatures, and competitive electoral and party systems that channel contestation through established procedures and check executive overreach. Transparency, integrity, and oversight mechanisms deter abuses and reinforce public trust. Additional supports include inclusive economic policy that reduces exclusion and grievances, a plural information environment that enables scrutiny and exposes abuses, and regional or international settings that reinforce democratic norms and discourage authoritarian backsliding. The entries describe plausible contributions rather than ranking factors or quantifying their relative impact.

* As reflected in research on democratic resilience. Adapted from Holloway and Manwaring (Reference Holloway and Manwaring2023: 82).

We should, therefore, not assume that institutional reforms that work in one setting, such as constitutional safeguards or electoral commissions, can be transplanted wholesale to another. What emerges from these insights is a rejection of universal or ‘silver bullet’ explanations in favour of a more relational understanding of resilience (Cheeseman et al. Reference Cheeseman, Desrosiers, Cianetti and Gehrke2024): the impact of any single variable depends on how it interacts with others. This insight also calls for greater methodological pluralism, as large-N studies often obscure the contingent and interactive nature of resilience that qualitative and mixed-method approaches are better equipped to capture.

Fifth, and relatedly, the growing attention to context has increasingly led scholars to identify specific pathways of autocratisation. While these pathways may overlap and reinforce one another – and multiple mechanisms can operate simultaneously – their disaggregation has proven both heuristically and analytically valuable (Bermeo Reference Bermeo2016; Waldner and Lust Reference Waldner and Lust2018).

Most frameworks typically identify four or five recurring pathways (Cheeseman and Cianetti Reference Cheeseman and Cianetti2023; Riedl et al. Reference Riedl, McCoy, Roberts and Somer2024; Wunsch and Blanchard Reference Wunsch and Blanchard2022). Table 2 summarises the most frequently referenced pathways, indicating whether they are more likely to take the form of gradual or sudden processes, and whether these processes tend to be depicted as resulting predominantly from internal factors, or are seen as also being impacted by external ones.

Table 2.

Pathways of Autocratisation

Table 2 Long description

The table categorises pathways through which a country may move toward autocratisation. It splits the overall process into two broad types: gradual erosion and political rupture. Under gradual erosion, the listed pathways are state capture, polarisation, and executive aggrandisement. Under political rupture, the listed pathway is violent overthrow. A separate row labels sources as external and internal, indicating that pathways can be associated with pressures or actors from outside the country or from within it. The table provides categories only and does not quantify frequency, severity, or likelihood for any pathway.

One pathway is the violent overthrow of the regime, which encompasses coups d’état, armed rebellions and forceful seizures of power that abruptly terminate democratic rule (Powell and Thyne Reference Powell and Thyne2011). Other pathways are more gradual. These include executive aggrandisement, in which elected leaders slowly concentrate power by weakening legislatures, judiciaries and oversight institutions (Haggard and Kaufman Reference Haggard and Kaufman2021; Huq and Ginsburg Reference Huq and Ginsburg2018). A second example is political polarisation, through which partisan divisions erode both tolerance and democratic norms (Levitsky and Ziblatt Reference Levitsky and Ziblatt2018). Related to polarisation is the rise of information disorder – encompassing disinformation, media manipulation and digital propaganda – which has been identified as a cross-cutting mechanism that can accelerate other forms of decline (Persily and Tucker Reference Persily and Tucker2020). Finally, some researchers have pointed to the importance of state capture, where political elites co-opt state resources and institutions to serve private or partisan interests, hollowing out the substance of democracy while retaining its façade (Hellman Reference Hellman1998; Peiffer and Englebert Reference Peiffer and Englebert2012).

It is important to note that we present the extent to which a pathway is seen to be domestically or internationally constituted as a spectrum to reflect recent research that has argued that contemporary threats to democracy are particularly powerful because they are often ‘interscalar’ (Cianetti et al. Reference Cianetti, Del Panta and Owen2025), i.e. they are shaped simultaneously by local, national and international actors, whose efforts reinforce one another (Cheeseman et al. Reference Cheeseman, Desrosiers, Cianetti and Gehrke2024). Polarisation, for example, may be the product of increasingly violent language between rival candidates, as in the United States, but has often also been driven by cross-border Foreign Information Manipulation and Interference campaigns. Analyses of state capture often unearth the existence of a web of multinational companies through which state assets are corruptly laundered (Chipkin et al. Reference Chipkin, Swilling, Bhorat, Buthelezi, Duma, Friedenstein, Mondi, Peter, Prins and Qobo2018). Meanwhile, sustaining executive aggrandisement in the face of growing domestic and international criticism may be facilitated by financial support from a sympathetic foreign ally, as was the case with India’s support of Sheikh Hasina’s regime in Bangladesh. Even a military coup to depose an unpopular president may be facilitated and sustained from abroad, as with Russia’s backing of the military juntas that have taken power in West Africa in recent years. Given this, the vast majority of episodes of autocratisation are likely to have important domestic and international components, albeit in different admixtures.

Yet, despite this conceptual and empirical progress, comparative work continues to privilege a subset of pathways that are particularly significant in the most high-profile cases, such as the United States, leading to an exaggerated emphasis on the role of polarisation and specific forms of executive aggrandisement – such as the gradual undermining of checks-and-balances institutions (Bermeo Reference Bermeo2016; Levitsky and Ziblatt Reference Levitsky and Ziblatt2018). This focus is understandable but risks oversimplifying a more complex reality. In many other world regions, the literature emphasises different pathways. Violent rupture through military coups and violent seizures of power is particularly important with respect to countries such as Myanmar, Thailand and several countries in West Africa (Albertus and Menaldo Reference Albertus and Menaldo2017; Powell and Thyne Reference Powell and Thyne2011). Even where similar pathways prevail, the precise mechanisms often vary. In parts of Asia, Africa and Latin America, for instance, executive aggrandisement involves term-limit evasion, which transforms time-bound presidents into lifelong rulers and marks a critical juncture in democratic decline (Dulani Reference Dulani2011; Haggard and Kaufman Reference Haggard and Kaufman2021; Posner and Young Reference Posner, Young and Cheeseman2018). By contrast, term-limit evasions are largely absent from Northern-focused accounts.

State capture has also been increasingly discussed in regions such as sub-Saharan Africa (CDD-Ghana 2025) in recognition of the role that non-elected actors can play in hollowing-out democratic institutions from the inside. This term was originally coined to explain the political economies of post-communist states in the 1990s (Hellman et al. Reference Hellman, Jones and Kaufmann2003) and has recently been invoked to describe the case of Hungary under Prime Minister Viktor Orbán, highlighting its cross-regional relevance. Yet despite this, the state capture pathway and its associated literature are rarely mentioned in comparative and North American-focused literature, despite the clear relevance for Western cases of democratic erosion, such as the United States under President Donald Trump. In the case of state capture, then, the divide in the literature can perhaps be said to be as much ‘East/West’ as it is ‘North/South’.

By extrapolating from a small number of unrepresentative cases, much of the comparative literature risks mistaking one regional experience for a universal model of autocratisation. The implications are both analytical and practical: theory-building risks overgeneralising from atypical experiences, and policy advice derived from these cases may travel poorly where state capture, violent rupture or term-limit evasion are more salient. A broader comparative agenda that systematically incorporates these neglected trajectories is therefore essential both for explanation and for effective intervention. It also creates greater space for alternative conceptions of democratic resilience emerging from the Global South, particularly those that place greater emphasis on the capacity and willingness of citizens to defend democracy. One example is the work of the Innovation Foundation for Democracy, an initiative led by Achille Mbembe that seeks to develop new democratic models rooted in African realities. In the Foundation’s Bringing Society Back In White Paper, Jean-Éric Biem (Reference Biem and Mbembe2026: 31) advocates a more substantive conception of democracy based on ‘endogenous democratic models’ rooted in African experiences, local knowledge and everyday forms of participation. Echoing key themes developed here, Biem argues that democratic resilience comes from society itself, and that institutions should be built around and sustained by citizen participation rather than imposed from above.

Finally, the comparative literature has begun to recognise that resilience is specific to the mode of autocratisation it confronts. The pathways through which democracies erode shape the forms of resilience most likely to succeed. This argument builds on Desrosiers et al.’s call to link typologies of autocratisation to corresponding modes of resistance and institutional adaptation (Reference Desrosiers, Cheeseman, Beardsworth, Gyimah-Boadi and Hassan2025). A democracy facing gradual executive aggrandisement, for instance, may require a long-term strategy centred on judicial independence and civic vigilance, while one confronted with the threat of a coup may depend more heavily on military professionalism and regional deterrence. The literature has therefore moved towards a more differentiated understanding of resilience, one that treats it not as a uniform response but as a set of contextually specific repertoires (Cheeseman et al. Reference Cheeseman, Desrosiers, Cianetti and Gehrke2024; Holloway and Manwaring Reference Holloway and Manwaring2023). This refinement represents an important conceptual advance, as it begins to bridge the gap between comparative and regional research traditions, and underscores that the study of resilience must be grounded in the empirical realities of how autocratisation unfolds in different political environments.

Moreover, the interaction between different forms of resilience and resistance is itself shaped by the pathway of autocratisation underway and by the existing capacity of political institutions to respond. In contexts where executive aggrandisement has successfully eroded institutional autonomy, civic resistance often becomes the final line of defence. In such cases, resilience depends less on institutional integrity than on the capacity of social actors to generate pressure from below, raising political costs for repression and generating incentives for restoration. This pattern is especially evident in settings that are less democratic to begin with, where decline takes the form of growing authoritarian entrenchment. The experience of Bangladesh, where student movements have repeatedly mobilised against authoritarian excess and electoral manipulation, demonstrates how organised resistance can keep open limited spaces for contestation in restrictive environments. Similarly, in Malawi, the mass protests that followed the disputed 2019 elections helped embolden the judiciary to annul a controversial presidential vote. Such cases highlight both the potential for resistance to substitute for resilience under certain circumstances, and the potentially reciprocal relationship between them. Where formal institutions falter, societal mobilisation can help sustain them, and where citizens succeed in defending democratic norms, they can provide the foundation for institutional renewal once political space begins to reopen.

Taken together, the recognition that democratic resilience is dynamic and context-dependent – shaped by the interplay of internal and external forces; cannot be reduced to universal factors; emerges from the interaction between institutions and civic agency; and varies according to the nature of the threat it confronts – represents a maturing of the field. These insights also provide a valuable foundation for a closer examination of the specific pathways of autocratisation and forms of democratic resilience that remain underexplored in the comparative literature, which is the focus of the next section.

Executive aggrandisement: nuancing ‘northern’ narratives

Of all the pathways of autocratisation, ‘executive aggrandisement’ is perhaps the most widely recognised. Popularised by Nancy Bermeo in her seminal article (Reference Bermeo2016), this term refers to the capacity of incumbents to systematically erode formal constraints and concentrate power in the executive, undermining accountability (Khaitan Reference Khaitan2019; Laebens Reference Laebens2023). Since the publication of Bermeo’s piece, which focused on democratic backsliding, others have used it to describe the same process in semi-authoritarian and authoritarian settings. Some of the main forms that aggrandisement takes are legislative and judicial subversion, the expansion of the powers and prerogatives of the executive through plebiscite, or simply bypassing rules, norms and institutions (Riedl et al. Reference Riedl, McCoy, Roberts and Somer2024). Freeman specifically speaks of three processes: the colonisation of existing institutions, the duplication of these by creating alternate favourable ones and their outright evasion (Freeman Reference Freeman2020: 35). In turn, this implies a gradual process combining incremental changes, but as we explain below, some developments occur much more rapidly.

Executive aggrandisement has rapidly become the focus of scholarship because it has been one of the most visible vectors of autocratisation in cases that have become paradigmatic for the study of regime change, such as Hungary and Poland (Cianetti et al. Reference Cianetti, Dawson and Hanley2018; Davies Reference Davies2018; Enyedi and Mikola Reference Enyedi and Mikola2024; Ilonszki and Dudzińska Reference Ilonszki and Dudzińska2021). The case of Orbán in Hungary, for example, has been used to show how legal institutions have been subverted to increase the power of the prime minister. Similarly, autocratisation in Poland under the Law and Justice Party has regularly been used to illustrate how judicial independence can be eroded through legislation and institutional reform, court packing and the broader delegitimisation of the judiciary in public discourse.

Related processes have also been observed in a range of Southern-focused scholarship, but this literature has tended to look at a much wider set of mechanisms – some of which are considerably less gradual. This includes, for example, the removal of legal constraints on term-limits for incumbent presidents and electoral manipulation. While the removal of term-limits does not directly undermine other democratic institutions, a number of mechanisms have been identified through which it can generate a negative ‘step-change’ in the prospects for democracy (Cheeseman and Desrosiers Reference Cheeseman and Desrosiers2026 forthcoming), significantly undermining the prospects for democratic ‘bounce back’ (Nord and Lindberg Reference Nord and Lindberg2025). These include higher levels of manipulation in elections that feature the incumbent and the curtailing of lame-duck effects, which weaken presidents in their final term. By openly defying the rule of law, term-limit removal also reveals its fragility and creates permissive conditions for further transgressions, accelerating forms of hyper-presidentialism (Baturo Reference Baturo2014; Dulani Reference Dulani2011). Term-limit removal can therefore delegitimise the political system (Diallo et al. Reference Diallo, Bassah, Kolokosso, Coliko and Agblegoe2024) and undermine legal measures specifically instituted to prevent hyper-presidentialism that has been prominent in regions such as Asia, Africa and Latin America (Alegre and Maisley Reference Alegre, Maisley, Hübner Mendes, Gargarella and Guidi2022; Diallo et al. Reference Diallo, Bassah, Kolokosso, Coliko and Agblegoe2024).

In part because term-limits were often introduced following what has been called the ‘third wave of democratisation’, these regions have seen an acceleration of evasion attempts as incumbents approached stipulated limits in recent years. Between 2000 and 2020, for example, there were 60 such attempts, 23 of which took place in sub-Saharan Africa, 17 in South and Central America and 12 in the Caucasus and Asia.Footnote ⁵ By contrast, there were no such attempts in Western and Central Europe or North America, which explains why this issue is rarely given prominence in scholarship emanating from these regions, while it is so prominent elsewhere.

Electoral manipulation is another dimension of executive aggrandisement that tends to receive relatively less attention in comparative work, but is central to struggles to defend democracy globally. Concerns about election rigging and the manipulation of elections have grown in recent years as strategies behind electoral fraud have expanded (Cheeseman and Klaas Reference Cheeseman and Klaas2019; Freedom House 2024). Elections may be manipulated on the day through ballot fraud but are often controlled well in advance via a range of strategies, including censorship, intimidation, the misuse of state resources and the exclusion of popular opposition candidates (Levitsky and Way Reference Levitsky and Way2010). These strategies are not exclusive to the Global South. Redrawing the boundaries of electoral districts for partisan gain, i.e., gerrymandering, has a long tradition in the United States, for example, and has been noted in executive aggrandisement scholarship. However, the form electoral manipulation takes, and the extent to which it undermines political competition, varies markedly. In more developed and institutionalised democracies, direct vote buying and the use of physical violence are less frequent, for example. By contrast, these strategies, and others such as election-day fraud, are remarkably common in Africa, Asia, the Middle East and post-Soviet states (Cheeseman and Klaas Reference Cheeseman and Klaas2019; Nord et al. Reference Nord, Medzihorsky and Lindberg2024: 3).

The role of electoral manipulation is particularly notable because, in addition to undermining vertical accountability, a number of researchers have argued that it erodes support for democracy and in doing so may help to legitimate unconstitutional transfers of power. In sub-Saharan Africa, for example, democracy remains the most favoured political system for most citizens, but developments such as corruption and vote rigging help to explain why ‘support for elections has dropped by 8 percentage points across 30 countries’ in the last decade (Afrobarometer 2024: 5).

The insights provided by these Southern perspectives demonstrate the importance of understanding the full range of strategies that leaders use to consolidate power. In doing so, they also suggest the need to theorise more carefully which forms of aggrandisement are likely to represent a step-change for autocratisation, and call into question the usefulness of ‘executive aggrandisement’ as a category given its great breadth and, hence, lack of analytical precision (Cheeseman and Desrosiers Reference Cheeseman and Desrosiers2026 forthcoming).

Because executive aggrandisement is seen as such a chronic threat, Southern-focused scholarship also has much to say regarding how institutional strengthening, and in particular constitutional reform, can enhance democratic resilience. Efforts to entrench judicial independence, insulate oversight bodies and professionalise electoral commissions have often emerged from long-running campaigns for constitutional reform rather than from routine institutional strengthening (Riedl et al. Reference Riedl, McCoy, Roberts and Somer2024). Latin America, for example, has seen extensive efforts to reform constitutions to both curb executive power – especially across the multiple roles a president can play, such as head of the state and head of the ruling party – and strengthen the judicial system and supervisory bodies (Negretto and Couso Reference Negretto and Couso2018; see also Uprimnyi Reference Uprimnyi and Rodríguez-Garavito2014), as ongoing efforts in Chile illustrate. Similarly, the 2010 Constitution of Kenya was the culmination of decades of struggle to restrain executive dominance and to embed structural checks on power through an expanded Bill of Rights, a more independent judiciary and devolution designed to dilute presidential centralisation (Gathii Reference Gathii, Graber, Levinson and Tushnet2018; Murunga et al. Reference Murunga, Okello and Sjögren2014). Similar movements in Ghana, Malawi and South Africa have sought to codify judicial autonomy and bolster the credibility of electoral management bodies, viewing constitutionalism itself as a form of preventive resilience (Fombad Reference Fombad2013; Gebeye Reference Gebeye and Gebeye2021).

Importantly, Southern struggles over constitutional reform and institutional strengthening offer lessons that are highly relevant for Northern countries. Despite the growing threat of power being captured by non-democratic and far-right leaders, there has been remarkably little discussion within policy or academic circles about the need to better insulate democratic norms and rules from subversion. Yet recent efforts in countries such as Norway and Germany to strengthen judicial safeguards suggest a growing recognition that long-standing democracies also require renewed institutional defences (Blauberger and Kelemen Reference Blauberger and Kelemen2017; European Commission 2024). The Southern emphasis on constitutionalising independence – through explicit protections for judges, term-limits and oversight bodies – thus provides a useful model for thinking about preventative resilience in Northern contexts. The South – and other overlooked regions such as Eastern Europe – are therefore not merely a site of particular democratic fragility but rather a crucible of innovation from which broader insights into sustaining liberal democracy globally can be drawn.

At the same time, because institutions are often seen as inherently vulnerable to executive manipulation, Southern-focused literatures offer distinctive insights into the role of citizens and civil society to defend them in contexts where systematic resilience is low. In several countries, popular mobilisation has compelled courts, legislatures and electoral bodies to assert their autonomy in moments of political strain. As noted above, in Malawi citizen protests and sustained pressure from civil society were instrumental in encouraging the judiciary to annul President Mutharika’s victory in the 2019 elections, reinforcing the idea that public mobilisation can embolden institutions to resist executive encroachment (Dionne Reference Dionne2024). This focus, therefore, enhances our understanding of the relationship between resistance and resilience, and the importance of social defence – i.e., citizens’ willingness to make institutional compliance politically salient and costly to violate (Desrosiers et al. Reference Desrosiers, Cheeseman, Beardsworth, Gyimah-Boadi and Hassan2025). In doing so, it reveals that civic agency can serve both as an external check and as a catalyst that reactivates latent institutional capacities during periods of democratic stress.

Southern-facing scholarship has also generated fresh insights into the importance of the interaction between popular attitudes, ruling-party disunity and legislative checks in constraining specific forms of executive aggrandisement. Research in Africa and Latin America, for example, shows that parliamentary resistance – often emanating from within the ruling coalition – has sometimes been a key source of resilience (Gamboa Reference Gamboa2022; Gloppen and Rakner Reference Gloppen and Rakner2025). With regard to term-limit evasions, in Zambia and Nigeria, presidents failed to secure legislative support for constitutional amendments extending their tenure partly because members of their own parties harboured presidential ambitions and resisted efforts to remove term-limits. In Zambia, these internal constraints were reinforced by popular opposition and coordinated civic action, most notably the OASIS Forum, which mobilised cross-sectoral resistance to President Chiluba’s third-term bid (Sishuwa Reference Sishuwa2025). Such cases underscore how resilience can emerge from the interplay of elite fragmentation and civic engagement, rather than from institutional robustness alone. By contrast, where legislatures are wholly subordinated or where ruling parties operate as instruments of personal power, the prospects for such internal resistance diminish sharply.

Finally, looking at cases in which democratic norms and practices are often still emerging or are fragile brings home the significance of what has been termed ‘democratic stock’: the accumulated habits and expectations that arise from long-term democratic experience (Boese et al. Reference Boese, Edgell, Hellmeier, Maerz and Lindberg2021). The longer a democracy endures, and the more its citizens, elites and institutions internalise democratic reflexes that can be reactivated when challenged, the harder autocratisation becomes. Where voters expect leaders to step down, and where courts and oversight bodies retain credibility, executives face stronger constraints on attempts to aggrandise power. As Daniel Posner and Daniel Young (Reference Posner, Young and Cheeseman2018) note, in countries where leaders initially respected term-limits, these norms have proven self-reinforcing, while in contexts where elections have yielded genuine alternations of power, citizens are more likely to defend electoral integrity (Beardsworth et al. Reference Beardsworth, Siachiwena and Sishuwa2022). These findings suggest that resilience is path-dependent and cumulative, building over time through both institutional compliance and collective memory.

Taken together, these insights also make an important broader point: democracy is always at risk, and so resilience must be cultivated in non-crisis periods and not only addressed in moments of political stress. Indeed, preventive investment in institutional strength, civic capacity and political trust is more effective than reactive defence once erosion is underway (Desrosiers et al. Reference Desrosiers, Cheeseman, Beardsworth, Gyimah-Boadi and Hassan2025). This is a particularly important message for Northern states in which democracy has not yet come under sustained attack: strengthening checks and balances, ensuring judicial independence and fostering an informed, organised citizenry remain the most durable safeguards against the forms of democratic decay that have defined the current era of autocratisation.

State capture: the fundamental importance of a ‘hidden’ driver

State capture remains one of the more neglected pathways in global comparative scholarship, despite the fact that it has been a prominent theme in work on autocratisation in Eastern Europe, Asia and sub-Saharan Africa. Because analyses of democratic backsliding – particularly those focused on major Northern cases such as the United States – tend to privilege formal institutions and state actors, what has been referred to as ‘democracy capture’ or the rise of a ‘shadow state’ (Albertus and Menaldo Reference Albertus and Menaldo2017; CDD-Ghana 2025; Hellman et al. Reference Hellman, Jones and Kaufmann2003) has tended to be overlooked. These processes are important, however, as they highlight how private or non-state actors can turn the state into an instrument for their own benefit, undermining its democratic foundations.

Echoing scholarship on the role of ‘elite pacts’ in authoritarian endurance (Brownlee Reference Brownlee2007), the notion of state capture highlights how the rent and power-seeking interests of non-elected actors – cartels, oligarchs, entrepreneurs, political middle-men, traditional leaders and foreign backers (CDD-Ghana 2025: 7; Gyimah-Boadi Reference Gyimah-Boadi2021; Ismail Reference Ismail2025) – can erode accountability, transparency and representation. State capture is therefore distinct from executive aggrandisement because, while it can ‘open the door and abet the rise of an aggrandized executive’ (Riedl et al. Reference Riedl, McCoy, Roberts and Somer2024: 22), it is not necessarily the result of the executive’s actions and may take place across an extremely wide range of institutions including regulatory bodies. Through this process, private interests and ruling elites come to build a stable ‘distributional’ system that is designed to benefit their interests rather than the public good (Albertus and Menaldo Reference Albertus and Menaldo2017: 26).

As a vector of autocratisation, state capture is generally a gradual process that comes to entrench the dominant position of private or non-state interests over public institutions, through systemic corruption and patronage, based on informal norms and agreements (CDD-Ghana 2025). Critically, this process can lead to the concentration of wealth in the hands of a small coterie of regime allies, uncontrolled corruption and the growing influence of anti-democratic forces. This group of elites – which may enjoy leverage over legislatures, judiciaries and the executive itself – then use their power to block reform efforts and shield one another from investigation and prosecution (Ismail Reference Ismail2025).

The term ‘state capture’ was first coined by scholars of post-Soviet states attempting to make sense of how the twin processes of political and economic liberalisation were being manipulated by powerful new politico-business networks in ways that undermined both free voting and the free market. The result, according to influential authors Joel Hellman and Danny Kaufmann, was ‘oligarchs manipulating policy formation and even shaping the emerging rules of the game to their own, very substantial advantage’ (Reference Hellman and Kaufmann2001: 31). Countries seen to have high state capture – and hence to suffer negative political and economic consequences – included Azerbaijan, Bulgaria, Croatia, Kyrgyzstan, Latvia, Moldova, Russia and Ukraine. One potential consequence of these processes was democratic stagnation or erosion, in part because the ‘organization and the nature of political parties will change and run contrary to the principles of democratic organization’ (Pavlović Reference Pavlović2023: 965).

Despite the clear evidence of what state capture could do to democratic governance, this term has generally been confined to scholarship on countries in Eastern Europe, Africa and Asia, including Hungary, India, Malaysia and Sri Lanka (David-Barrett Reference David-Barrett2021; Ismail Reference Ismail2025; Leitner and Meissner Reference Leitner and Meissner2017). Perhaps most notably, it was revived from 2015 onwards by South African researchers seeking to explain the impact of big business interests, such as the Gupta brothers, on corruption and governance in their own country (Buthelezi and Vale Reference Buthelezi and Vale2023; Ismail and Richards Reference Ismail and Richards2023). Bringing together prior work on post-communist states with Africanist accounts of similar processes – such as Will Reno’s analysis of ‘shadow states’ in West Africa (Reference Reno2003) – and traditional class analysis, a new generation of scholars sought to demonstrate how economic policy and key parts of the state had been captured by an emergent ‘power elite’. According to Ivor Chipkin et al. (Reference Chipkin, Swilling, Bhorat, Buthelezi, Duma, Friedenstein, Mondi, Peter, Prins and Qobo2018: ix), echoing Reno, this enables the formation of a shadow state, i.e. a ‘relatively well-structured network of people located in government, state institutions, state-owned enterprises (SOEs), private businesses, security agencies, traditional leaders, family networks and the governing party’ who are distinguished by their privileged ‘access (either consistently or intermittently) to the inner sanctum of power to influence decisions’ (Reference Chipkin, Swilling, Bhorat, Buthelezi, Duma, Friedenstein, Mondi, Peter, Prins and Qobo2018: ix).

Put another way, state capture in South Africa represented a contemporary incarnation of the kinds of neo-patrimonial politics – where public authorities govern through clientelistic networks in parallel to formal institutions – that have long been one of the factors the Africanist literature has cited as undermining the prospects for democratisation (Gyimah-Boadi Reference Gyimah-Boadi, Basedau, Erdmann and Mehler2007). This explains why the Ghana Center for Democratic Development (CDD) refers explicitly to a process of ‘democracy capture’ (CDD-Ghana 2025) with significant implications for accountability. According to CDD research that reviewed 370 government agencies and institutions across 10 countries, only 40% were not subject to at least some level of capture (CDD-Ghana 2025).

This pervasiveness explains why the literature on resilience to state capture typically argues that this pathway is particularly difficult to reverse because it undermines the very mechanisms through which autocratisation might otherwise be resisted – such as a free media, an independent judiciary and a responsive legislature (Gyimah-Boadi Reference Gyimah-Boadi2021; Papada et al. Reference Papada, Pavlova and Lindberg2022; Riedl et al. Reference Riedl, McCoy, Roberts and Somer2024). Partly as a result, improvements in institutional design alone are unlikely to be effective, and require other factors to also be in place, such as mass mobilisation in favour of the independence of anti-corruption institutions and effective prosecutions (Mungiu-Pippidi and Johnston Reference Mungiu-Pippidi and Johnston2017). The literature also highlights the importance of investigative journalism and civil society coalitions in exposing the networks underpinning capture and mobilising public pressure for reform (CDD-Ghana 2025; Chipkin et al. Reference Chipkin, Swilling, Bhorat, Buthelezi, Duma, Friedenstein, Mondi, Peter, Prins and Qobo2018). These actors often serve as the last line of defence where formal institutions have been co-opted but are themselves vulnerable to repression.

Research on state capture also offers important insights into contemporary processes of autocratisation in other regions. On the one hand, by centring the role of the business community and economic elites, work on state capture provides insights into the risks that the second Donald Trump administration – which has conferred vast political influence on a wide range of unelected figures such as Elon Musk – poses to American democracy. On the other hand, by emphasising the international connections of shadow states, which typically feature cross-border financial flows, shell companies and international financial institutions (Buthelezi and Vale Reference Buthelezi and Vale2023), this literature serves as an important reminder that the threat to democracy often comes from the interaction of domestic and international forces.

One important implication of this analysis is that anti-capture strategies must extend beyond national reforms. It has therefore highlighted the role of global frameworks, such as the Financial Action Task Force (FATF), and advocacy organisations such as the Tax Justice Network, which have pushed to close loopholes and enhance transparency in tax havens and financial secrecy regimes. Although imperfect, FATF processes and beneficial-ownership disclosure initiatives can play a role in exposing the international infrastructure that sustains capture, for example. Research on British government investigations and asset-recovery efforts that targeted the corruption networks of political leaders in Nigeria and Zambia, which were made possible because British corporations and financial institutions were implicated, has come to similar conclusions. In particular, it has emphasised both the extent to which international actors contribute to domestic state capture (Beqiraj and Mackenzie-Gray Scott Reference Beqiraj and Mackenzie-Gray Scott2022; Rose Reference Rose2024), and the importance of ensuring that international accountability mechanisms reinforce domestic ones by way of response.

Violent rupture: a growth threat?

Violent ruptures, such as coups, armed insurrections and foreign interventions, are less common than gradual forms of autocratisation but are relevant to many countries where state capacity is low (Powell and Thyne Reference Powell and Thyne2011). The tendency to overlook this pathway in comparative research reflects some of the biases noted above, such as the propensity of northern narratives to over-emphasise institutional erosion and populist capture while treating violent seizures of power as exceptional or peripheral concerns. Yet, a growing number of states face significant risks from coups and/or insurgencies, even if they have not occurred yet. This includes countries with a history of conflict and coups such as Nigeria and Thailand, as well as states such as Brazil, where former President Jair Bolsonaro was accused of orchestrating an attempted coup to overturn the 2022 election results and is now facing prosecution for his role in encouraging the January 2023 insurrection.

Coups d’état involve the removal of the government through force by the military, while autogolpes or ‘self-coups’ occur when incumbents dissolve or neutralise civilian institutions through coercion. As Rachel Riedl et al. (Reference Riedl, McCoy, Roberts and Somer2024: 19) note, these often take place under emergency rule, when executives cancel elections or dissolve legislatures. Both coups and autogolpes suspend the constitutional order and fuse coercive and political power, eroding civilian checks and centralising authority in military hands. Sub-Saharan Africa has been at the centre of the recent upsurge in military intervention (Akinola and Makombe Reference Akinola and Makombe2025; Opalo Reference Opalo2024), with 30 successful coups since 1990, 10 of which have occurred since 2020, including in Madagascar in 2025. Rising fragility and insecurity – especially in West Africa and the Sahel – have created fertile ground for military interventions (Mbulle-Nziege and Cheeseman Reference Mbulle-Nziege and Cheeseman2022). Civilian failures to provide protection and effective governance have fuelled popular frustration, which juntas have exploited to legitimise their seizures of power. Afrobarometer reports that opposition to military rule has declined by 11 percentage points across 30 countries, with support for juntas rising sharply in Mali and Burkina Faso (Afrobarometer 2024: 5). Yet, as Ken Opalo (Reference Opalo2024: 93) argues, coups are rooted in specific political contexts shaped by weak institutions, the decay of state capacity and failures of electoral politics. Where militaries have historically been politicised, or perceive threats to their corporate interests, interventions become more likely (Wilén Reference Wilén2024).

Insurrections constitute a related but distinct form of violent rupture in states that lack a monopoly on the use of force. Rather than military takeovers from within the state, they involve non-state armed groups seizing or contesting power through sustained conflict. The Taliban’s reconquest of Afghanistan or the capture of power by rebel groups, as has happened in countries such as Rwanda or Uganda (Jalali Reference Jalali2025; Makara and Wang Reference Makara, Wang, Arriola, Rakner and van de Walle2022), represents the most striking examples, replacing the existing regime through force while bypassing formal coup dynamics. Armed insurgencies have also destabilised political systems in countries such as Ecuador and Haiti, eroding democratic institutions even when they fall short of changing the government.

Southern-focused scholarship suggests that resilience to violent ruptures depends not only on the strength and professionalism of the security forces but also on the political and institutional environment that shapes civil–military relations. The literature on coup-proofing emphasises that the creation of parallel command structures, elite units and loyalty networks can reduce the immediate risk of military takeover, but often at the cost of professionalism and cohesion (Wilén Reference Wilén2024). As Erica De Bruin (Reference De Bruin2020) warns, efforts to counterbalance the military by creating parallel security units, such as presidential guards, can backfire – centralising power and fuelling mistrust. Where coup-proofing is based on ethnic favouritism or patronage, it can also undermine military competence and politicise the security sector, creating new vulnerabilities. Militaries tied to partisan networks, foreign patrons or independent economic interests are more prone to intervene. By contrast, the most effective strategies link military autonomy to clear civilian oversight and constitutional accountability – approaches that have proven most durable in states such as Ghana, Botswana and South Africa, where defence institutions are insulated from partisan politics and operate under democratic norms (Adebajo Reference Adebajo2010).

A growing body of work has also shown that falling government legitimacy can exacerbate coup risks. Empirical studies from across Africa show that administrations associated with corruption, economic mismanagement and electoral manipulation are more likely to face military intervention (Afrobarometer 2024). In contrast, regimes that perform credibly – holding competitive elections, reducing graft and maintaining public trust – are better able to sustain civilian control over the armed forces. This perspective underscores the point that coups are as much a symptom of political decay as they are of military ambition. By strengthening democratic performance and ensuring that elections are perceived to be fair, governments can make interventions less justifiable among citizens and soldiers alike (Opalo Reference Opalo2024).

International dynamics also matter. In the 1990s, there was a strong focus on ‘anti-coup’ norms, including among bodies such as the United Nations and regional groupings such as the Economic Community of West African States. These norms have weakened in recent years, however, and Nina Wilén (Reference Wilén2024) describes resilience as being undermined by a ‘permissive environment’ in the Sahel, where juntas reinforce one another rhetorically and diplomatically. Naunihal Singh (Reference Singh2022) adds that the weakening of post-Cold War norms against unconstitutional rule has reduced the reputational costs of intervention. To build resilience against such ruptures, scholars emphasise entrenched civilian oversight, professionalised security forces and strong normative commitments to constitutional rule.

Finally, research on insurgencies highlights the critical role of state capacity. Weak service delivery, limited territorial control and poor governance in peripheral regions often create fertile ground for armed mobilisation (Mbulle-Nziege and Cheeseman Reference Mbulle-Nziege and Cheeseman2022). States with stronger bureaucratic and coercive capacity – particularly those able to deliver security and justice impartially – are far better placed to contain insurgent threats. Yet, as Southern-facing analyses remind us, coercive power alone is insufficient: sustainable resilience requires a combination of credible governance, inclusive development, and local legitimacy (Menocal Reference Menocal2020). The broader implication is that the prevention of violent ruptures depends less on militarised responses than on the quality of democratic institutions, the integrity of political competition and the capacity of the state to command both loyalty and trust. Put differently, the state acts as a critical ‘context’ that can both generate or constrain opportunities for anti-democratic projects, as well as shape the possibilities for responding to them (Cheeseman et al. Reference Cheeseman, Cianetti and Desrosiers2025). Enhancing different components of state capacity – coercive, extractive, administrative – is therefore critical to efforts to reduce the risk of violent ruptures.

Future directions for the study of democratic resilience

The comparative study of democracy is undergoing a period of renewal. A growing body of work offers nuanced analyses of the challenges democracies face and how democracies endure, adapt and even regenerate under pressure. Research on pathways of autocratisation has highlighted the diverse mechanisms that have driven democratic decline and how they vary across the world. Meanwhile, scholars of resilience have shed light on the institutional and societal conditions that sustain democratic rule and studies of resistance have deepened our understanding of how individual actors, civic organisations and social movements defend democracy from below. These two strands of inquiry have, in different ways, expanded and enriched the field beyond the older focus on transition and consolidation.

The aim of this review article has been to encourage a richer conversation between insights developed through analyses of Northern or middle-income democracies and those focusing on Southern (and Eastern) cases. There is much to learn from both sets of experiences. The insights generated by work on political polarisation, disinformation and executive aggrandisement in high-profile cases such as the United States complement those derived from research on state capture, violent rupture and term-limit evasion in Africa and Asia, for example. The challenge moving forward is not to choose between these perspectives, but rather to integrate them into a broader comparative framework that can better account for the diversity of democratic experiences and the multiple levels – local, national and international – at which democracy is contested. There is some evidence that this is beginning to happen, as greater comparative attention has been paid in recent years to high-profile events in countries such as Brazil, Hungary, Poland and South Korea (Bustikova and Guasti Reference Bustikova and Guasti2023; Croissant and Waldner Reference Croissant and Waldner2025; Kelemen Reference Kelemen2023; Levitsky Reference Levitsky2024). Nevertheless, significant gaps remain, suggesting seven broad priorities for future research.

First, there is much we still do not know about authoritarian pathways, including which combinations are the most damaging to democracy, and whether some pathways are becoming more or less common over time. There is also a need to refine and disaggregate existing pathways, most notably executive aggrandisement. Instead of being an analytically precise and parsimonious category, this concept has expanded to encompass a wide array of different phenomena, from the purging of judiciaries to the erosion of term-limits. Future research should seek either to distinguish sub-types of aggrandisement or to adopt a more granular vocabulary that can specify the mechanisms at play. Doing so would facilitate clearer comparisons across regions and provide greater insight into why some forms of aggrandisement constitute temporary distortions while others generate enduring authoritarian entrenchment.

Second, future research should more systematically recognise the relationship between pathways of autocratisation and forms of resilience and resistance. The literature suggests that the sources of resilience are rarely generic. The strategies that prevent executive aggrandisement may differ sharply from those needed to counter violent rupture. A more structured understanding of these linkages would allow scholars to move beyond static typologies and develop pathway-specific theories of resilience. In practical terms, this means identifying the forms of institutional reform, civic mobilisation or international engagement most likely to succeed given the dominant threat to democracy in a particular setting. Such work would also enhance the analytical dialogue between the study of autocratisation and of recovery – fields that remain too separate despite their shared concerns.

Third, it is important to recognise that democratic resilience is not a linear process but one punctuated by critical junctures – moments when institutional trajectories are redefined by decisions that fundamentally alter the balance of power (Cheeseman and Desrosiers Reference Cheeseman and Desrosiers2026 forthcoming). In other words, certain strategies or episodes of autocratisation, such as the removal of presidential term-limits, can represent step-changes that close off or at least significantly hamper the possibility of recovery. Understanding resilience in these terms helps explain why similar pressures can have radically different outcomes across countries. The durability of democracy often depends on whether political actors resist or reinforce these pivotal moments of transformation. While the existing literature has provided considerable detail on how erosion takes place, there has been insufficient focus on which kinds of decline most undermine the prospects for re-democratisation.

Fourth, scholars should be cautious of the temptation to search for universal or single-factor explanations of resilience. The impact of any given factor is shaped by the surrounding political and institutional environment. For example, public support for democracy may be both more important and less immediately influential as a source of resilience in contexts lacking competitive elections and a free press. The challenge, therefore, is to understand how combinations of factors interact to produce resilience or fragility in specific settings. This calls for greater attention to process-tracing, comparative historical analysis and mixed methods designs that can capture complexity rather than reduce it.

Fifth, and relatedly, the literature must deepen its analysis of how resilience and resistance interact. There are two components to this. On the one hand, the growing use of the terms ‘resilience’ and ‘resistance’ has introduced some conceptual ambiguity. Scholars often apply these terms in overlapping but inconsistent ways – to describe institutional strength, civic mobilisation or societal adaptability. Greater clarity and shared usage are therefore needed. A more consistent vocabulary would enable clearer comparisons and prevent the conflation of distinct yet related dimensions of democratic defence.

On the other hand, resilience can operate at both structural and behavioural levels, and its endurance often depends on the capacity of social actors to defend and revitalise democratic norms. Resistance, in turn, may become more effective when embedded within resilient systems that provide channels for mobilisation and oversight. Yet we still know relatively little about how these forces reinforce or weaken one another across different stages of democratic crisis. Comparative work could usefully explore the conditions under which citizen action fortifies institutional autonomy and, conversely, when strong institutions create space for citizens to mobilise safely and effectively. Understanding these reciprocal dynamics is essential for explaining why some democracies not only survive threats but emerge stronger.

Sixth, scholarship must better account for the multilevel nature of both democratic decay and democratic defence. Too much comparative work still treats the state as the primary unit of analysis. Yet, as the challenges to democracy in countries such as Moldova and Burkina Faso increasingly demonstrate, attacks on democracy often originate from, or are mediated through, transnational networks of finance, technology and information (Cheeseman and Cianetti Reference Cheeseman and Cianetti2023). Similarly, local actors and subnational institutions frequently play decisive roles in resisting authoritarianism – whether through municipal coalitions, local courts or grassroots movements. Understanding resilience requires a multi-scalar approach that connects the local, national and international arenas.

Finally, the field would benefit from more integration. The recent literature provides rich conceptual innovation but often remains siloed by region, method or disciplinary tradition. Comparative democratisation could profitably draw more explicitly on the insights of related research programmes: studies of state capture and corruption, analyses of information ecosystems and scholarship on social movements, for example. The field would also benefit from more longitudinal research tracing how resilience evolves over time. The dialogue between theory and practice also needs to be strengthened. Building democratic resilience requires learning across levels, sectors and divides (CEDAR 2023), integrating the perspectives of researchers, policymakers and practitioners so that findings inform real-world efforts to protect democracy.

Taken together, these directions point towards a more coherent, inclusive and dynamic research agenda. The study of democratic resilience and resistance, and the challenges they face, has made remarkable progress, but its next phase demands greater synthesis. The goal should not be to displace existing accounts but to enrich them through engagement with the distinctive experiences of the Global South and other regions where alternative pathways of autocratisation prevail. Bridging these literatures will make it possible to identify both cross-regional mechanisms of democratic endurance and the context-specific adaptations that sustain democracy under pressure. In so doing, scholars can contribute not only to explaining why democracy survives but also to understanding how it can be renewed.