Repertoires of citizen action in hybrid settings

Motivation: In recent years, a growing literature has emerged analysing how organized citizen action can achieve more accountable and inclusive governance. While this literature is mostly focused on relatively open, democratic, or more stable contexts, often in western democracies, only a small proportion of the world's population lives in such settings. Rather, authoritarianism is growing and democratic spaces are closing globally. Purpose: This article explores how hybrid political settings— those which have both democratic and non-democratic features, fragmented authority, and closing civic space— affect strategies of citizen action. Methods and approach: Drawing from a large number of single and comparative case studies of citizen action, mainly from Mozambique, Myanmar, Nigeria, and Pakistan, the article builds on work by Tilly and Tarrow on repertoires of contention. Based on these, the article develops a typology of eight repertoires of citizen action through which citizens may express their grievances and attempt to hold authorities to account. Findings: The literature suggests that there may be little space for contentious social and political action in closed or authoritarian settings. This article finds, however, that a rich and diverse tapestry of citizen action may be seen in the “hybrid” settings (those which have both democratic and non-democratic features). Shaped by political opportunities

forms of popular struggle vary and change from one sort of political regime to another, much less why such variation and change occur (Tilly, 2006, p. 3).
In his book Tilly attempts to fill the gap, arguing that repertoires of contention will vary across regime type.In particular he points to the importance of two factors of the regime: (1) the level of democracy which exists, which in turn affects the possibilities and spaces for action; and (2) the capacity of the state, which affects the ability of authorities to respond.In a later book Tilly and Tarrow (2015, p. 75) develop this framework further, particularly in reference to hybrid regimes, in which there are "systematic segments of democratic and undemocratic rule operating side by side within the same regime." Drawing from a number of case studies and research projects conducted as part of the work of the Action for Empowerment Research Programme (A4EA), this article explores the nature of citizen action in such settings, with a particular focus on Mozambique, Myanmar, Nigeria, and Pakistan, each of which (at the time of the research) reflected qualities of hybrid regimes referenced by Tilly and Tarrow.Tilly and Tarrow-and others-paint a somewhat bleak picture of the nature and possibilities of the action that will occur in such regimes-in contrast to more open democracies with stronger democratic institutions.
However, based on original case studies from the A4EA research programme, this article argues that in fact a rich and diverse tapestry of citizen action may be found in these settings.The article offers a unique typology of eight strategies or "repertoires" of action developed inductively from the A4EA research cases.We argue that these repertoires will vary according to their visibility and their level of "ruliness" or "unruliness," and that they may shift or re-enforce one another over time.While it may appear self-evident, we offer a simple but little-recognized conclusion: in "hybrid regimes" citizen action itself may take multiple and hybrid forms, reflecting both the political opportunities which are available, as well as "trigger events" (Bruun, 2013) which lead to public protest.Moreover, such actions have an important range of impacts which can serve as building blocks for broader change.For those who are interested in supporting pathways to citizen-led accountability, understanding these broader repertories will be important to strengthening and protecting the full "ecosystem" of accountability available (Halloran, 2021), going beyond more commonly understood approaches.
The article is organized as follows: Section 2 examines further the question of how political settings or "regimes" affect strategies of citizen action and offers a framework which focuses on how these settings affect the visibility and "ruliness" of such strategies.Section 3 introduces further the hybrid settings from which our cases are drawn, and how they were chosen.Section 4 then presents a typology of eight repertoires of citizen action that have emerged from studies in the A4EA research programme, describing each more fully.Section 5 provides further discussion about what can be learned from these cases about the nature of citizen action in hybrid settings, what aspects of the regime affect the nature of the action, and the impacts of such actions.The article concludes with a discussion of implications for donors, policy-makers, or other actors who wish to support such action for achieving more accountable and democratic governance.

| REPERTOIRES OF CITIZEN ACTION IN HYBRID SETTINGS
In their seminal work on social movements, Tilly and Tarrow (2015, p. 231) argue that: routine interactions between governments and political actors produce political opportunity structures that channel what forms of contention different potential claim makers can actually initiate… At the same time, governmental action and popular contention interact to form repertoires of contention -limited arrays of known, feasible ways to make collective claims -that also limit possible forms of contention in any regime.
Using this framework, Tilly (2006) and later Tilly and Tarrow (2015) argue that in low-capacity democracies and non-democracies we will be more likely to see violent forms of contention-including civil wars and revolutions-whereas in high-capacity democratic settings we will see the emergence of other forms of social movements, with less violence, partly because there is a belief that the state has the capacity to respond, as well as the range of political opportunities and democratic space in which people can mobilize.Similarly, in their work on violent and non-violent forms of protest, White et al. (2015) argue that in settings with autocratic rule and limited political space: the opportunity structures in countries with large rural populations, low individual resources, backward economies, and poorly developed state penetration are more likely to encourage violent challenges.Dissent is more likely to take on nonviolent forms in more urban and developed states, which have better prospects for mass mobilization and poor prospects for covert action.(White et al., 2015, p. 486) Other scholars have questioned this view, arguing that there has been very little research on opportunity structures and citizen mobilization in non-democratic settings, and that the literature reflects a bias towards studies done in more democratic settings (Alimi, 2009;Corduneanu-Huci & Osa, 2003;Osa & Schock, 2007).In response to this conceptual gap, and to growing examples of protests and other forms of citizen action in authoritarian regimes across the world (Carothers & Press, 2020;Ortiz et al., 2022;Youngs, 2019), a number of studies now also examine forms of citizen resistance and action in such settings (Bruun, 2013;Chen & Moss, 2019;Schedler, 2018).Questioning Tarrow and Tilly's argument that "contention greatly increases with democratization" (Tilly & Tarrow, 2009, p. 448), Schedler (2018, p. 44) argues that "rather than rare, disruptive events that provoke existential crisis, or reactive moves that aggravate such crises, recurring protests seem to form an integral part of political normality in many authoritarian regimes."However, between these two bodies of literature-those that focus on citizen action in more open democracies and those focused on more authoritarian regimes-very few studies still look at forms of such action in hybrid regimes, which contain both democratic and non-democratic elements.As Tilly and Tarrow state: "what has remained unspecified is whether these regimes produce forms of contention that are distinct from both their authoritarian cousins and from true democracies" (2015, p. 75).
While some authors have argued that we can understand hybrid regimes as a subset of authoritarian regimes on the one hand, or of democratic regimes on the other, this paper accepts the argument of Mufti (2018, p. 112) that "hybrid regimes are not transitional phases but in fact political regimes that manifest a combination of both authoritarian and democratic tendencies that ought to be examined in comparison to each other and not against the standards of democracy."As such, it follows that if we argue that the context affects the nature of citizen action, then we need also to further understand how such action might manifest itself differently in hybrid regimes than in either democratic or authoritarian settings.What is needed, Mufti suggests, is further research, including "both quantitative and qualitative research…driven by single-n case studies, or paired comparisons based on in-depth research with the intention to advance our contextual knowledge of these regimes and facilitate mid-range theory building" (Mufti, 2018, pp. 112-113).
Drawing on a number of in-depth case studies produced by the A4EA research in Mozambique, Myanmar, Pakistan, and Nigeria, this article further explores this challenge.Through grouping multiple empirical case studies, we arrive inductively at a range of repertoires of citizen action in the countries we studied, including both contentious and less contentious forms.Such repertoires are shaped both by political histories, as well as by contemporary political opportunities and events, which we argue will affect the nature of citizen action in at least two ways.
The first has to do with what we might call the degree of visibility-in particular to powerful authorities-which repertoires of engagement will take.In settings characterized by authoritarianism, fear of authorities, closed spaces, the risks for direct challenge to those in power may be great (Márquez, 2016, p. 220;Schedler, 2018).But awareness of risks does not necessarily mean that action does not occur-it simply may be taking place in ways that are less visible to authorities.Indeed, there is a rich literature which explores the nature of hidden forms of resistance (Fu, 2017;Justino, 2019;Scott, 1985Scott, , 2008)).
The second feature which shapes the dynamics of contention in low democracy settings has to do with the extent to which engagement is more ruly or more unruly.If we are dealing with settings in which there are few channels for effective participation, where state structures are either absent or fragile, or where the channels are perceived as closed or ineffective, then action may occur outside these formal structures.Indeed, Tilly and Tarrow (2015, p. 62) talk about the difference between contained and transgressive contention.Other scholars have similarly distinguished between the ruly and unruly forms of citizen politics (Khanna, 2012;Khanna et al., 2013).As Shankland et al. (2011, p. 1) elaborate: unruly politics, as we define it, is political action by people who have been denied voice by the rules of the political game, and by the social rules that underpin this game.It draws its power from transgressing these rules -while at the same time upholding others, which may not be legally sanctioned but which have legitimacy, deeply rooted in people's own understandings of what is right and just.
Building upon these two features, we suggest that within totally closed, repressive regimes, repertoires of engagement might tend towards the more hidden and more unruly forms, as much of the literature suggests.However, in more hybrid regimes, in which authoritarianism, and some spaces or channels for democratic engagement sit side by side, then we might expect the repertoires of action to be far more varied, both hidden and visible, ruly and unruly, depending on the opportunities available at any given moment.In such settings, rather than the absence of action, we indeed find a diverse and complex set of repertoires which citizens may put to use in hybrid fashion, with the ability to shift repertoires over time as opportunities arise or weaken.Moreover, in line with Bruun (2013) and della Porta (2020), we argue that, in settings where repression and political histories have shaped more hidden forms of action or inaction, key "trigger events" can also lead to the rapid emergence of more visible and unruly actions, even without the clear emergence of new political opportunities.
While this range of repertoires of citizen action may not produce regime change, we suggest that their outcomes are important as building blocks for more accountable and democratic governance, and that the potential of meaningful, non-violent citizen action in such contexts should not be overlooked.

| THE SETTINGS AND THE APPROACH
As mentioned earlier, the starting point of this article is the question of citizen action in hybrid regimes, defined by Tilly and Tarrow (2015, p. 75) as those with "systematic segments of democratic and undemocratic rule operating side by side within the same regime."While there is now a vast literature on the meaning of hybrid regimes, "it is difficult to find consensus among scholars over what hybrid regimes actually are" (Mufti, 2018, p. 113).Some argue that such regimes may be understood better as "competitive authoritarianism" (Levitsky & Way, 2010), or simply weaker forms of democracy (Merkel, 2004;Puhle, 2005), or as "transitional" regimes, between democracy or authoritarianism (Armony & Schamis, 2005).Following Mufti (2018), this article accepts the argument that hybrid regimes can be understood as an independent, not a residual, category, and that they need to be understood as "multi-dimensional" focusing not only on formal democratic indicators such as electoral competitiveness (Levitsky & Way, 2010;Schedler, 2016), but also on other dimensions of power by unelected bodies, such as the military, monarchs or religious authorities ("tutelary interference"), and the existence or lack of civil liberties for citizens to engage (Gilbert & Mohseni, 2011).
Mozambique, Myanmar, Nigeria, and Pakistan, the key countries of focus in the A4EA research programme, all offer examples (at least over the period covered by this study) of hybrid regimes, with both democratic and non-democratic elements along the dimensions of electoral democracy, fragmented forms of authority, and limited civic space and civic liberties.In a world of diminishing democratic space, the four countries are in the lower 50% on the V-Dem scale of liberal democracy-while they do not represent the most closed or autocratic settings, neither are they the most open (Alizada et al., 2021).At the time of the research, each country had at least nominal constitutional provisions for electoral democracy (though this changed in Myanmar with the coup in February 2021). 1 Each country has historical experience of military or authoritarian rule, with powerful non-state actors continuing to play a role in governance, whether ethnic militias, religious groups or traditional authorities (for further discussion of such fragmented authority, see Joshi, 2023;Anderson et al., 2023).On the OECD Fragility Index all are ranked as highly fragile in terms of security issues, meaning threats of violence still loom large (OECD, 2020).And each country has experienced limited civic space, which worsened further under COVID-19 (McGee, 2023).In 2021, the CIVICUS Monitor, which measures civic space, characterized Mozambique, Nigeria, and Pakistan all as "repressed," while Myanmar was downgraded from "repressed" the previous year to "closed" (CIVICUS, 2021).
Over five years, the A4EA action research programme investigated dozens of examples of citizen action within these countries.These studies were linked to particular sub-themes within the A4EA work, and used a variety of innovative methods, including qualitative case studies, governance diaries, and experimental design.(For more on the methods see the introduction to this special issue and Mohmand & Anderson, 2023.)Each study was derived from the core question of the programme, which focused on understanding the forms of social and political action related to empowerment and accountability in fragile, conflict-and violence-affected settings.From this larger set of studies, this study chose cases which (1) documented the range of strategies and repertoires used; (2) had enough information to understand the dynamics and impacts of these; and (3) provided a sampling across the four countries.
Using this approach, we are able to draw on in-depth and original single and comparative case studies to examine the range of types and repertoires of action in these settings, using established approaches to theory building from case-study research (Miles et al., 2020;Yin, 2013). 2 In offering this range of cases, our intent is not to say they are the only possible strategies, nor that they will always occur; rather it is to use these multiple cases inductively to develop a more complete understanding of the multiple strategies used in such hybrid settings.

| CITIZEN ACTION IN "HYBRID REGIMES"
Drawing from A4EA research, this paper identifies a unique typology of eight differing repertoires of civic action that have emerged in these contexts.The focus is both on contentious and non-contentious forms of action, which emerge in more visible and more hidden ways, reflecting the opportunities available.A search of the literature has found no comparable empirically based study which illustrates the range of actions in such settings.

| Cultural resistance
Much has been written about the use of cultural forms of expression, such as political song, as vehicles for resistance and contention (Márquez, 2017).This was certainly found to be the case in Mozambique, where researchers explored protest songs to provide "alternative understanding of demands for public accountability" (Taela et al., 2021, p. 1).
The authors of this study observe that protest song has a long tradition in colonial societies and has long been used as a way to express grievances against the state in more coded forms.Musicians use different strategies to "disguise the content of their own songs in order to avoid censorship in state controlled public media" (Manhiça 1 The V-Dem regime type index in 2021 categorized Mozambique, Myanmar, and Pakistan as electoral autocracy, while Nigeria just scraped into the "electoral democracy" category (Alizada et al., 2021, p. 31).Electoral autocracies are those where authoritarian governance and significant limits to democratic freedoms and participation sit alongside democratically elected institutions.
2 For a similar study outlining the outcomes of citizen engagement across 100 cases, see Gaventa and Barrett (2012), which also offers further discussion of the validity of a similar multiple-case method.et al., 2020, p. 33).These included sharing songs with close friends to test their political sensitivity, editing out or muting certain words which could trigger response, and using rhetorical devices such as metaphor "to avoid naming and direct confrontation and a powerful way to convey culturally coded messages that resonate with large audiences" (Manhiça et al., 2020, p. 34).

GAVENTA
At key moments, triggered by anger over concrete issues such as the economy, corruption, or police brutality, music can also be used to give visibility to suppressed grievances: when examined as a collective body of work, the messages and performances of protest songs can provide understandings of demands for public accountability and political action which expand the repertoire of the limited occasions when citizens' views and actions are expressed in public protests… The lyrics emerge as a vehicle to amplify subdued voices in a context where critique of those in governments can result in reprisals.(Taela et al., 2021, p. 2) While sometimes disguised or deflected away from authorities, and at other times used to amplify grievances, such cultural resistance is also unruly, often used in protests or to incite collective action outside the normal political channels for engagement to "denounce, demand, and mobilise for social and political change" (Taela et al.,p. 5).

| Crafting solutions beneath the radar
Another way that citizens may keep their actions beneath the radar is through solving problems for themselves, through building their own alternative solutions rather than confronting authorities (Kashwan et al., 2019;Spade, 2020).For instance, research at the very local level in Myanmar found that in general villagers preferred not to take issues they faced to the authorities, preferring to "either not do anything about the issue faced or address the issue on their own" (Myanmar research team, 2021, p. 40).This reluctance was reinforced, researchers report, by strong cultural norms, especially "those of wanting to make problems non-existent or small, and not wanting to 'stand out' or to escalate issues."This is linked to the Burmese concept of "Anade", a strongly embedded social norm which places emphasis on not taking actions which "would cause embarrassment, offense or inconvenience to others" (Myanmar research team et al., 2021, p. 32).In addition to these cultural norms, for both historical and contemporary reasons, fear, especially of the military, also militated against public action other armed units or public officials.This went hand in hand with a reported sense of hopelessness and fatalism (a belief that nothing could be done about the situation), very low expectations that making claims would be successful, and a view that public actions would be too risky.
And yet these cultural attitudes of not causing trouble, fear of reprisals, or low expectations from the authorities did not mean that collective action did not take place.A4EA work found a number of examples of villagers taking action for themselves rather than making claims to authorities.Such actions were ruly, in that they did not necessarily involve contestation, and they were not done in secret, but they were often low key, done informally so as not to appear to be disruptive.Examples include communities building bridges or roads, or paying for teachers, or raising their own funds to provide much-needed services, such as education or electricity.Yet, in some cases these forms of self-reliance take a hybrid approach, in which the government provides some resources, and the community contributes others as a form of "co-production" (Anderson et al., 2023;Joshi & Moore, 2004).

| Indirect claim-making via intermediaries
Where political space is limited, and where we see legacies of low expectations or legacies of fear, marginalized groups may also choose to express claims via intermediaries, well-connected family, or community members who may be in a better position to speak on their behalf to others.Sometimes such indirect claim-making or negotiation is done on behalf of a collective.

GAVENTA
In the first phase of its work, the A4EA programme used governance diaries with marginalized communities to understand how these groups interacted with authorities, and on what kinds of issues.Across Pakistan, Mozambique, and Myanmar, the research found that in general chronically poor and marginalized citizens in these settings rarely engage directly with the state, even when pressed with poor services or perceived injustices (Chaimite et al., 2021;Loureiro et al., 2021).Once again, however, this did not mean that they were not taking action-rather that such action was often taken through intermediaries, local fixers, brokers, and informal leaders at the community, village, settlement, or neighbourhood level.In turn, in a context of multiple authorities, these intermediaries would often approach differing groups, e.g.ethno-regional authorities or military groups in Myanmar, or a mining company in a remote part of Mozambique, often "forum shopping" to resolve grievances in the most efficient way (see Anderson et al., 2023).When quieter forms of networking and appeals to authorities failed to get a response, these intermediaries might take to the media and social media networks to put public pressure on authorities, or even mobilize the occasional protest (Anderson et al., 2022).

| Protests
While we have thus found a number of ways in which marginalized groups may take action in hidden or "beneath the radar" ways, we have also seen in our case countries a number of more visible and direct forms of engagement.These can be thought of as more contentious episodes, which are discrete, spontaneous protests, or they can emerge into more sustained social movements or campaigns, which diffuse across time and scales (Tilly & Tarrow, 2015), as we shall see in the next section.For instance, while research using the governance diaries approach in Myanmar demonstrated the hesitancy to confront authority directly over day-to-day governance issues and services (as discussed earlier), at the same time the military coup of February 2021 triggered an outpouring of public protests and civil disobedience by medical workers, students, and civil servants across the country, especially in more urban areas.In late 2020, Nigeria saw the eruption of the #ENDSARS protests, among the largest in its recent history, which spread quickly but then faced heavy police crackdown, leading to the multiple deaths on Lekki Bridge Lagos.Even before this large event, protests in Nigeria had grown from 99 documented instances in 2010 to over 600 in 2019-a six-fold increase (Akomolafe, 2020).
Even though the A4EA countries which we studied have low levels of women's participation in formal politics, and often face strong patriarchal and repressive norms, our research found that women have often been at the forefront of protest actions and movements, especially on issues related to sexual harassment (Tadros & Edwards, 2020), women's rights (Khan et al., 2019;Khan & Naqvi, 2018), and on issues of security (Aina et al., 2019;Nazneen, 2023).Khan and Taela (2023), studied two key highly localized women's protests-the Hazara women's protests against ethno-sectarian killings in Pakistan and the Chiango women's protests for road safety and infrastructural development in Mozambique.In both cases, in response to particular events which threatened the safety of women and their families, women used local public protests to "disrupt gendered and political habitus to make claims for the state to protect and provide" (Khan & Taela, 2023).
While the cases show the importance of very localized protests, in other cases, we found the escalation of protest activity, often with the possibility of disrupting or at least getting the attention of the regime.In Pakistan, Nigeria, and Mozambique, for instance, protests around access to affordable fuel and electricity for meeting basic needs of cooking, travel to work, communications, etc. were found to be particularly significant.National-level fuel protests in Nigeria (Atela et al., 2021) and Mozambique have been among the most significant demonstrations in either country, while in Pakistan our research documented 456 protests between 2007 and 2015 on access to electricity, contributing to making this a national issue (Javed et al., 2021).In the absence of more open channels through which citizens could raise their deeply embedded grievances regarding the right to energy, citizens in these settings resorted to more unruly-but highly visible ways-of making their grievances known (Hossain et al., 2021).In both cases-the protests of women and the protests about energy-the exclusion of particular voices or concerns from more regularized channels of participation or political opportunity structures were met not by inaction, but in some cases, significant, disruptive, protest actions.

| Social movement campaigns
While contentious episodes may be short-lived, when they are more sustained they may become a social movement or develop into more focused campaigns using a variety of tactics (Tilly & Tarrow, 2015).While Tilly argues that these occur "mainly in democratic regimes" (Tilly, 2006, p. 187), our research documented a number of examples of social movements for accountability in less democratic settings (Joshi, 2019), including, for instance, an anti-corruption movement in Guatemala (Flores, 2019), the Right to Food movement in India (Joshi & Chowdhury, 2019), and the lawyers' movement in Pakistan (M.Khan, 2019).
One example which moved from a single protest event to a long-term movement was the Bring Back our Girls Campaign in Nigeria (Aina et al., 2019).This movement emerged in a context of growing concerns about state security and citizen safety in Nigeria, particularly related to the Boko Haram insurgency in the North-East, and its abduction of women and girls.The kidnapping of 276 girls from the community of Chibok in April 2014 was a trigger for collective action, beginning with protests and the hashtag #BBOG, which many thought would be short-lived.But unlike many other protests, this one grew over time, to become a sustained, well-organized movement, with high visibility both nationally and internationally.
Unlike some of the earlier strategies, which sought to disguise identity and voice from those in power, in this case the #BBOG movement used visibility itself to gain legitimacy and power.It built widespread international alliances with a wide range of groups, appealing to international human rights standards, and gaining support from highly visible international leaders, including Michelle Obama.The movement was one of the first to extensively use online media, with its hashtag #BBOG quickly spreading to 69 countries in the first month alone (Aina et al., 2019).While having a strong social media presence, it combined this with offline protests, vigils and demonstrations, especially in Abuja and Lagos, and also kept the story in the eyes of the mainstream media.Throughout, the movement maintained a fluidity of goals, expanding from the initial concern with the Chibok girls to include other issues of gender-based violence, corruption, security, and more, which also enabled it to build broad networks of support with others.

| NGO-led advocacy
While social movements and campaigns can be quite unruly, other forms of advocacy are also used to express and channel grievances, through more ruly and institutionalized means, often organized by or with non-governmental organizations (NGOs) (Fox & Halloran, 2016).In each country in which we worked, NGOs have played important roles as advocates for citizens, as watchdogs and monitors, and as protectors of key rights and policies.
For instance, in Pakistan, NGOs such as the Aurat Foundation, Shikat Gah, ASR Resource Centre, Strengthening Participatory Organization, South Asia Partnership, and the Human Rights Commission played a critical role in efforts to establish and protect reserved seats for women in the legislature (Khan & Naqvi, 2018).In Nigeria, the NGO Spaces for Change has focused on monitoring human rights abuses, as well as the regulatory framework affecting civic space itself (Spaces for Change, 2020).In Mozambique, groups like CIP have focused on demanding transparency of revenues linked to extractives, a strategy also used by many groups such as Extractive Industries Transparency Initiative or Publish What You Pay in other countries (Awortwi & Nuvunga, 2019).
Research on closing civic space under COVID-19 found that, even as civic space grew even more constricted, such groups still played vital roles, often shifting to new areas of focus made more salient by the pandemic, such as health, livelihoods, or education (McGee, 2023).Across the countries, professional associations of doctors and nurses demanded greater protection and delivery of health services.In Mozambique, a coalition of civil society groups known as the Budget Monitoring Forum (Fórum de Monitoria do Orçamento) mobilized to monitor relief funds from donor governments and the World Bank, as well as workers' associations and trade unions, and made new demands for the protection of livelihoods, while other groups emerged with a focus on education and more.The point is that even in these settings, which became even more hostile under COVID-19, NGOs and other formal, registered organizations are an important part of the landscape for demanding accountability and advocating for new policies, using such tactics as public pressure, lobbying, or monitoring and exposing corruption behaviours.

| Dialogue and deliberative approaches
While social movement campaigns or advocacy often use more contentious approaches, we also find many examples of attempts to create spaces for more ruly dialogue and deliberation between citizens and authorities. 3In these settings, where there were few deliberative or participatory channels in existence, these were often introduced by external actors, such as international donors.
Perhaps a good example was found in Myanmar, where, with funding from the European Union (EU), a large international NGO4 launched a multi-year programme to deepen social accountability, working across four sub-regions.In a context of little trust between citizens and authorities the programme used public hearings on key issues to create spaces for dialogue, while also helping to equip citizens to enter these spaces with greater knowledge of their rights, as well as of the responsibilities of those in authority.The programme found that "the combination of approaches taken has developed new spaces in which ordinary people and authorities can have dialogue and agree actions for the public good" (Anderson et al., 2019).
Similarly, in Nigeria the large externally funded Partnership to Engage, Reform and Learn (PERL) programme created forums for citizens and government officials to engage on budgets and other issues (Aremu, 2022), while in Pakistan, the AAWAZ Voice and Accountability Programme also created new elected forums from village to district to provincial levels, and supported women and excluded groups to raise their demands with public officials (Khan & Qidwai, 2021) In a broader study of six large accountability programmes funded by the UK government, Anderson et al. (2020) find that these programmes used one or more "multi-scalar strategies" to link across levels of governance, including conveying citizens' demands upwards to those responsible for public services, linking civic organizations at the grassroots with larger-scale civil society lobbying and activist initiatives, and developing horizontal networks (Anderson et al., 2020).They find that these had positive effects, including activating accountability at different levels and building multi-level coalitions for reform.

| Political participation via electoral and parliamentary means
A final strategy for action which emerged in the A4EA programme is through elections and participation in established parliaments and representative institutions.Here, work from the A4EA programme in Pakistan is instructive, especially in relationship to the role of women in electoral politics.On the one hand, what is striking is the absence of women as voters-Pakistan has one of the world's largest gender gaps in urban voter turnout (Cheema et al., 2021).
Exploring the reasons for this gap, their research points to household norms, especially the attitudes of husbands towards participation of women, as well as the failure of political parties to recruit women into the competitive political process.The result, Cheema et al. (2019) conclude, is that women are "invisible citizens," due to "subtle 3 Similarly, Guerzovich et al. (2022, p. 20) make an interesting comparison between what they call "resistance" pathways, and "resonance" pathways to accountability.
processes that socialise women into non-political roles, and result in a 'gendered psyche' that makes women feel invisible and irrelevant to the electoral process" (Cheema et al., 2019, p. 3).
On the other hand, despite women's low level of participation in elections, other research suggests that when women do gain office, their voice can have substantial influence, and shift policies in a way that support women's rights and equity issues, in some circumstances (Khan & Naqvi, 2020).In Pakistan, quotas for women "at all levels of all levels of elected bodies (33 per cent in local government and 17 per cent in elected assemblies) have increased their political representation dramatically," though they still fall below quota levels (Khan & Naqvi, 2020, p. 1).From this position, despite multiple challenges, women representatives have successfully shifted policy between 2002 and 2018 on a number of gender-related issues, including the quota law itself, honour killings, domestic violence and sexual harassment, marriage laws, and more.In part these gains were achieved due to support from activist organizations outside government, as well as the collaboration of women across party lines, led by groups such as the newly established National Commission on the Status of Women.In general, a "shared history of women's rights activism" was a key factor in being able to work collaboratively across parties "to build political consensus for gender equality legislation" (Khan & Naqvi, 2020, p. 5).

| DISCUSSION
In the above pages, we have explored multiple strategies or repertoires of citizen action in hybrid regimes through case examples drawn from four countries (Mozambique, Myanmar, Nigeria, and Pakistan).Counter to expectations that in such settings there will be little citizen action or that when it occurs it will be more violent due to the absence of other channels of dissent (Tilly & Tarrow, 2015;White et al., 2015), in fact, we have found a rich tapestry of non-violent social and political action across a wide range of repertoires.Working inductively from A4EA research cases, we have identified and presented eight such repertoires, including cultural resistance, crafting solutions beneath the radar, claim-making via intermediaries, protest events, campaigns and social movements, NGO-led advocacy, dialogue and deliberation, and political participation via electoral and parliamentary means.
We have argued that such repertoires will range from the hidden to the visible, and the ruly to the unruly, depending on the channels and opportunities available.On the visibility dimension, we have identified several examples of more hidden forms of engagement.These range from the use of cultural expressions such as political song to express grievances in slightly coded or indirect ways, to the aversion of conflict through action to create self-reliant or alternative institutions, to approaching authorities indirectly, through intermediaries, based on social capital, kinship, or other networks.But at other times, gaining visibility for actors or issues which have long been hidden is in itself an important strategy.For instance, political song in Mozambique was also used quite visibly to build public awareness of issues of corruption and to "publicise and amplify a collective sentiment" (Manhiça et al., 2020, p. 26).The women's protests in Mozambique and Pakistan brought concrete local issues to the public arena, disrupting patriarchal norms and exacting at least some recognition by government authorities (Khan & Taela, 2023).The Bring Back our Girls movement helped to keep the abduction of the Chibok girls in the public eye over a period of years, and so on.
We have also seen variation in the "ruliness" or "unruliness" dimension of these repertoires.Where there is an absence of official channels for engagement, citizens may engage in more unruly ways, for instance through spontaneous or sustained protest movements, employing a variety of contentious tactics and performances, as we saw in the women's protests in Mozambique and Nigeria, or the energy protests across several countries.On the other hand, actors may seek to approach intermediaries or engage in official channels in slightly more ruly ways, ranging from institutionalized advocacy, dialogue and deliberation, and employing acceptable electoral strategies.The hybrid nature of the regime means that both strategies are possible, deployed at different moments and circumstances.
A simple though important point emerges: in hybrid settings, which combine features of democratic and non-democratic regimes, citizen action will also be hybrid, sometimes adopting strategies for hidden and less visible action as often found in more repressive settings, and other times becoming more visible, as often found in more democratic settings; sometimes becoming more unruly and, at other times engaging in less contentious ways.
Despite being presented as a typology, however, these forms of action are not static.They may evolve and accumulate over time and must be understood as being in dynamic relationship with one another.Indeed, the most effective forms of action are often those that build links across these strategies over time.For instance, the gains of the women elected officials in Pakistan were linked directly to their histories in the women's social movements, as well as the advocacy work of NGOs, which further strengthened their ongoing alliances between actors in different spaces (Khan & Naqvi, 2018).
In the absence of other literature on the nature of citizen action in hybrid regimes, we suggest that this typologybased on a series of original case studies-is itself an important contribution to the literature.However, it immediately raises at least two other significant questions, which we shall consider briefly.Why do certain forms of action appear at certain times and not others?And second, when such action occurs, what impact does it have, especially on the empowerment and accountability outcomes (which are the focus of this special issue)?

| Opportunities and triggers
The question of what triggers or opportunities contribute to action arising at certain times has been widely discussed in social movements and contentious politics literature-with no clear consensus having been reached (Corduneanu-Huci & Osa, 2003;Gamson & Meyer, 1996;Meyer, 2004).For Tilly and Tarrow, as observed earlier, social action is shaped by the existence of political opportunities, leading them to argue (simply put) that because democracies offer more structured opportunities for engagement then we are likely to see more engagement within them than in non-democratic settings.This view has been challenged by an emerging group of scholars who apply the political opportunity approach to non-democratic settings.As Osa and Schock argue, for instance, in democracies political opportunities may arise "from the liberal state's own institutions…In non-democracies, on the other hand, political opportunities arise at those junctures where (or when) the regime's coercive mechanisms malfunction" (Osa & Schock, 2007, pp. 127-128).
Our work suggests several further insights into understanding opportunities and triggers of action in hybrid settings.First, if we are to understand hybrid settings in a "multi-dimensional way" (Gilbert & Mohseni, 2011), then we need also to understand political opportunities not only by looking at those that arise through the formal structures of the state but also at those that emerge in relation to non-state actors.For instance, in Myanmar, citizens formed alliances to solve local problems not only with local government officials, but also with military or religious leaders, often forum-hopping to see where their grievances could best be resolved (Anderson et al., 2023).In other instances, the intervention of non-state actors such as international donors, also created new opportunities for engagement, echoing the finding by others that in more closed settings, external allies may be very important (Osa & Schock, 2007).Second, while political opportunity analysis often examines high-level and somewhat visible opportunities which allow for public and contentious action to emerge, we have also seen the importance of less contentious forms of action for resolving grievances, often carried out in more hidden and indirect ways, and at the subnational and local levels.Third, while a key tenet of the political opportunity structure approach is that repression will trump or preclude social action (Bruun, 2013;Moore, 2000;Olzak et al., 2003), our work, along with that of a number of other scholars (Davenport & Armstrong II, 2004;Osa & Schock, 2007), finds that sometimes repression actually provokes or escalates such action, fuelling peoples' anger and indignation, as we saw in the early days of the #EndSars campaign in Nigeria.
Finally, and perhaps most importantly, we have found numerous examples where citizen action erupted with no apparent change in the political opportunity structure, but rather as a response to key events which triggered subsequent action, fuelled by moral indignation and outrage.For instance, in Nigeria, the kidnapping of girls by Boko Haram stimulated a new citizen campaign led by women allies initially situated in the capital city, not in the region directly affected; in Pakistan and Mozambique, reaction to local issues-failures of security and infrastructure-provoked previously silent women to take public protest action; hikes in fuel and electricity prices in Nigeria, Mozambique, and Pakistan similarly provoked popular response, which sometimes escalated quickly.This gives support to those who have argued that in more repressive societies such "trigger events", rather than more rational calculations of political opportunity, may also be at play (Bruun, 2013, p. 259).For della Porta (2020), such events can also trigger critical junctures, which in turn create further cracks in the system, leading to further action.
While none of these variables gives us a predictive theory on when citizen action will emerge, we should not, on the other hand, assume that in less democratic settings it will not emerge.When it does, it is likely to be the result of a variety of factors-including shifts in formal and informal opportunity structures at different levels, trigger events which unleash anger and indignation, and the emergence of local and external allies, all of which may come together to create new fissures and opportunities in complex and unpredictable ways.

| The impacts of citizen action
But what does this review of strategies of citizen action tell us about the impacts of these strategies, particularly, on the ways in which organized citizens gain accountability from authorities?The study of the outcomes of social movements and social action is generally acknowledged to be difficult (Giugni & Bosi, 2012)and mostly focused on democratic settings, rather than on hybrid or authoritarian regimes (Amenta et al., 2019).The literature on non-democratic settings is mostly concentrated on studies of transition and the downfall of authoritarian regimes instead of more intermediate types of changes (Schedler, 2018).By contrast, work by Gaventa andBarrett (2012, p. 2407), which included a review of 100 cases of citizen action across types of "polities," found that "engagement can make positive differences, even in the least democratic settings." Similarly, the cases in this study point to important outcomes of citizen action, which may seem small, yet are not insignificant.First, in settings with long histories of democratic closure and exclusion, gaining visibility itself becomes an important win, as we saw in the case of political song in Mozambique, or the women's protests in Mozambique and Pakistan.Second, in settings where citizen action has been constrained historically, such actions can help to create a sense of agency, an awareness of rights, and skills and capacities for public engagement that may previously have been lacking, such as we found from the #BBOG movement, or from the NGO-led social accountability programme in Myanmar.
Third, with greater visibility, as well as greater local empowerment, came numerous examples of increased responsiveness from authorities on concrete issues as a result of citizen action, whether through resolving local problems of service delivery in Myanmar, gaining the release of some of the kidnapped Chibok girls in Nigeria, improving energy access in Pakistan, or advancing women's political presence and power in national and provincial parliaments in Pakistan.Finally, the studies also point to a number of important examples of norm change.These include, for instance, increased expectations of transparency, new forms of interaction between citizens and authorities in new democratic spaces, and modelling new processes for public engagement, or changes in norms of inclusion, especially in relation to gender.
Generating responses on concrete issues, creating visibility, building political capabilities, and strengthening norms of inclusion and accountability all represent important changes, especially in contexts where these have been lacking or suppressed.Such "successes" may be limited to specific events, or may be fleeting, yet as Bruun (2013, p. 259) argued in the case of China, even single events can be "eye openers, encourage participation, raise political consciousness, and facilitate learning" which at the structural level "contribute to the emergence of more complex forms of contention, such as to turn local, specific or thematic concerns into general or abstract ones." In our work, such cases as the Bring Back our Girls campaign, which emerged spontaneously, but then was sustained over time through a well-organized multi-scalar campaign, or in the struggle in Pakistan over decades to convert the energy of localized women's protests into parliamentary gains, offer examples of how short-term protests evolved into more complex and sustained movements.

| IMPLICATIONS
In hybrid settings-where democratic and non-democratic features sit by side-paying attention to this breadth of repertoires for action and how they interact with one another will be important to shift power relations and to build more democratic and accountable systems over time.This review suggests a far more diverse ecosystem of empowerment and accountability actions than is often recognized, either in the academic contentious politics literature or in the world of donor aid and democratic assistance.Moreover, it suggests that we cannot assume that either the strategies used in more open settings, or those used in more closed settings, will be appropriate for hybrid regimes.
These findings have important implications for those seeking to support and strengthen citizen-led actions for accountability (Anderson et al., 2022).Studies by CIVICUS (2021) on "civic space," find that only 3% of the world's populations live in so-called "open societies," and 25% live in "closed" societies.The remaining 72% of the population live in settings which fall between these stools, in settings in which civic space is defined as "narrowed," "obstructed," or "closed".Yet, on the whole, support for accountability programmes tends to focus on those strategies that are most visible and the most ruly, occurring through institutionalized channels, as would be found in more open societies.
There are good reasons for this-these are the most easily found and the "safest" bets for donors, especially those from the outside.However, there are clearly limitations and missed opportunities in doing so, as these more institutional spaces easily open and close, and the gains made here can easily be lost.Moreover, we have found the more hidden or more unruly forms of action to be important in and of themselves, as well as in how they in they interact and create synergies with activities in the more formal spaces.Among other things, the more hidden and more unruly forms of engagement provide social learning and building blocks which can contribute over time to broader accountability and governance changes.
This finding adds weight to the argument in the accountability literature in recent years for the need for more holistic strategies, which pay attention to the broad "ecosystem of accountability," rather than to focus on any single strategy.The accountability ecosystem approach argues for the importance of analysis and mapping of accountability systems, strategies that emphasize an integrated approach, strategic use of varied and complementary tactics, and politically "informed practice that focuses on addressing and shifting power relations that underpin accountability" (Halloran, 2021).The typology provided here provides one approach to mapping the broad citizen-led repertoires and strategies which are part of this accountability ecosystem and could be critical in the avoidance of assumptions drawn from either more fully democratic or non-democratic settings.
For those in the development community who argue for the importance of "thinking and working politically" (Booth, 2015;Carothers & de Gramont, 2013;Dasandi et al., 2019), these findings suggest that political analysis needs to search not only political opportunities and alliances found in formal structures, but also pay attention to those "beneath the radar," as well as the role of informal governance actors.They also point to the important role of trigger events, which can unleash forms anger and indignation, leading to further action, serving as potential critical junctures (della Porta, 2020), or as leverage points for broader change (Meadows, 1999).For donors and allies of democratic action, identifying such key events, which may arise from a single issue or grievance, and figuring out how in turn they can be sustained, grow, and lead to more complex forms of contention, is critical.At the same time, for those who seek to deepen and extend democracies in such hybrid settings, protecting and preserving what spaces are available for action, fostering alliances, and sustaining support over time are important strategies that go beyond the normal recipes for democratic assistance which focus on fostering electoral competitiveness and institution building.
Rather than measuring the outcomes of democratic interventions at the high level of regime or institutional change, in settings with long histories of military rule, authoritarianism, and violence, and the accompanying legacies of fear and weak channels of democratic engagement, perhaps we need to be more modest in what we consider "successful" governance outcomes, especially over the short term.More attention needs to be paid to understanding the building blocks of larger and more sustainable change, and how the positive outcomes which are seen are understood as intermediate processes in a larger strategy.The paths to change are not linear, and setbacks will be experienced as spaces open and close, but, over time, the importance of small steps which change norms and expectations, create capacities and networks, and create new governance models, are not to be dismissed.As Tilly (2006, p. 211) reminds us: "in the short run, the repertoire and opportunity structure interact to constrain the frequency, location, and character of collective claims.But in the medium run, claim-making alters the regime."We have seen multiple examples of how repertoires and opportunity structures interact to shape strategies of collective claim-making, at least in the short term.What is now needed is a better understanding of how these, over time, cumulatively and collectively change the larger regimes of power.