Rebel governance: a vibrant field of research

This literature review provides an outline of a comparatively young academic discipline, referred to as Rebel Governance. Individual avenues of research are divided and introduced, and landmark studies are reflected upon and connected with each other. This field of research has grown steadily in recent years, and its relevance is particularly evident as it becomes increasingly visible that rebel groups exercise rule beyond the nation-state with varying degrees of ambition, success, and violence, and that spaces lacking state authority are by no means ungoverned. Following an introduction on the embeddedness of Rebel Governance in International Relations and a general overview of the entire research field, the individual research strands in Rebel Governance are presented and outlined. These include political institutions and organizational structures, the regulation of commercial production, civilian life under rebel rule, social services and legitimacy enhancement, rebel diplomacy, and symbolic governance. This division and the research presented therein results in a broad overview of an increasingly important field of research that can generate explanations for phenomena that are so far insufficiently understood and can provide policy advice for interacting with rebel groups exercising governance. Simultaneously, this review also offers a wide-ranging reflection of the different spheres of Rebel Governance that have been explored so far and can thus be used to consolidate and connect insights in order to increase the accessibilityof this field of study.

To understand how Rebel Governance as a discipline is embedded in IR, it is useful to trace the debates and developments in IR that influenced the emergence of Rebel Governance. Throughout the historical phases which IR has passed as a discipline, many scholars have traditionally placed their primary focus on the behavior of Westphalian nation-states within the international system (Holsti 2004, p. 46;Jackson 2005, p. x; Knutsen 1996, p. 2). Although the discipline has diversified substantially over time and sparked debates on numerous avenues of research, for many scholars, nation-states have remained the main providers of political order and the sole owners of sovereignty (Knutsen 1996, pp. 2-4;Krasner 1999;2001;Mampilly 2011, p. 10). This assumption was aptly highlighted by Munro, stating that "in modern social formations the state is the principal institutional locus of political power" (1996, p. 116) and is "seen as the legitimate provider of specified political goods, over which it has sole and universal jurisdiction on the basis of a national collectivity" (1996, p. 116).
While some saw a need for discussing the existence of de facto states in the international system (Bakke 2011;Coggins 2011;Florea 2014Florea , 2017Florea , 2018, many described contested territories in civil wars, or other areas with limited state authority, as ungoverned spaces or black spots (Stanislawski 2008). The debate on ungoverned spaces is closely linked to debates on failed states, which both had a substantial influence on, for instance, the United States' foreign policy (Kaplan 1994;Rotberg 2002;Lynch 2016). Since then, both ungoverned spaces and failed states, sharing the same narrative "that poor governance elsewhere will negatively affect the American homeland or the United States' allies and global interests" (Keister 2015, p. 2), have become subject to considerable criticism (Clunan and Trinkunas 2010, pp. 17-18;Mitchell 2010, p. 289). Thinking about areas of limited or absent state authority as ungoverned spaces is not only inappropriate but often deceptive, as it "leads analysts and international actors to assume that no political actor worthy of recognition exists and therefore that violence is the only avenue for engagement" (Mampilly 2011, p. 28). Over time, many different insurgent groups have successfully challenged the sovereignty of the states in which they operated, and have in many cases shown that the areas they controlled were by no means ungoverned (Florea 2018;Keister 2015;Hansen 2020, p. 9). By now, the assumption that "ungoverned spaces are actually not ungoverned, but exist under authorities other than formal states" (Keister 2015, p. 2) is also shared by numerous scholars (Mampilly 2011, p. 254;Reno 2015, p. 265;Risse 2011, p. 23;Stanislawski 2008).
To outline the basic ideas and relevance of studies on rebel governance, Arjona et al. (2015) aptly pointed out that "when rebels secure territory, they must decide how they will interact with local residents" (2015, p. 1) and, drawing from their longstanding experience, further stated that "a surprisingly large number of rebel groups engage in some sort of governance, ranging from creating minimal regulation and informal taxation to forming popular assemblies, elaborate bureaucracies, schools, courts, and health clinics" (2015, p. 1). Since the involvement of insurgent groups in governance activities varies widely, Kasfir (2015) advocated for comparative analysis to explore the dynamics of rebel governance and highlighted that "three scope conditions [of rebel governance] are territorial control, a resident population, and violence or threat of violence" (2015, p. 21). An ever-growing body of litera-ture focuses on exactly these issues including numerous studies that delve deep into governance structures and dynamics in various insurgent groups, such as the Taliban in Afghanistan (Terpstra 2020;Terpstra and Frerks 2018), the SPLM/A in Sudan (Branch and Mampilly 2005;Blunt 2003), the RCD in the Democratic Republic of the Congo (Mampilly 2015a), the LTTE in Sri Lanka (Klem and Maunaguru 2017;Mampilly 2011;Stokke 2006), and the FARC and ELN in Colombia (Arjona 2017a;Borch and Stuvøy 2008).
The field of rebel governance is dominated by neither qualitative nor quantitative research. Instead, both qualitative and quantitative approaches have been commonly employed, and their results have often been integrated into studies with complementary approaches. Mampilly, who focused primarily on qualitative approaches, conducted several in-depth case studies (2011, 2015a, b) and made many valuable contributions by using comparative approaches to develop generalizable hypotheses about factors impacting the effectiveness of rebel governance (2011, pp. 209-230). Mampilly and Stewart (2020) also introduced a conceptualization of political institutions of rebel governance. Many other scholars using qualitative methods focused on delving deep into the governance activities of certain rebel groups, such as the FARC (Arjona 2017b), Taliban (Terpstra 2020), LTTE (Klem and Maunaguru 2017), SPLM/A (Wassara 2010), or Hamas (Berti 2015), or dedicated their research to explaining causal processes (South 2017;Terpstra 2013) or certain dynamics in insurgents' governance behavior (Schoon 2017). Arjona frequently employed quantitative as well as qualitative data collection, resulting in valuable mixed-methods contributions to research on rebel governance (Arjona 2009(Arjona , p. 127, 2014(Arjona , p. 1369. Stewart, regularly applying quantitative approaches to rebel governance, published several studies that revealed previously unknown causal relations between certain aspects of rebel governance, such as the relationship between the extent of an insurgent group's governance activities and its military capacity (Stewart 2018(Stewart , 2020, or the relationship between service provision, killing of civilians, and the international reputation of secessionist insurgencies (Flynn and Stewart 2018). Stewart's latest in-depth work "Governing for Revolution: Social Transformation in Civil War" (2021), also impressively demonstrated how fruitful the combination of quantitative methods and qualitative approaches can be for rebel governance research. Other proponents of quantitative research on rebel governance are Akcinaroglu and Tokdemir (2018), who also introduced the Reputation of Terror Groups Dataset (Akcinaroglu, Tokdemir 2016 1945(Florea 2015 and published several studies on their emergence, developments, survival, and disappearance (2014,2017,2018,2020). Other meticulously elaborated databases on rebel governance or certain aspects of it are the Big Allied and Dangerous Dataset by Asal et al. (2011;Asal and Rethemeyer 2015), the Dangerous Companions Project by San-Akca (2016), and the Foundations of Rebel Group Emergence (FORGE) Dataset by Braithwaite and Cunningham (2019).
One of the main subjects in studies on rebel governance is the relationship between civilians and rebel groups in areas under rebel rule, which derives from acknowledging that "rebels play a central role in defining how civilians live their lives during wartime not only through violence but equally through the develop-K ment of structures and practices of rule" (Stokke 2006). Gowrinathan and Mampilly (2019) also highlighted that "rebels may incorporate civilians into this [governance] project to fulfill essential roles within the civilian administration, [...] or they may choose a mode of governance in which civilian involvement is non-existent" (2019, p. 6). Arjona, who mainly focused on structures and dynamics of rebel governance in the Colombian civil war, introduced the dichotomous distinction between rebelocracy and aliocracy, two forms which insurgent governance can take (Arjona 2014(Arjona , pp. 1374(Arjona -1375(Arjona , 2017a. She argues, that when trying to understand social order, which she also describes as predictability, in civil wars, the key factor to look at is the adherence to a social contract between rebels and civilians, which she describes as a "set of rules that structure human interaction in a given community during wartime, allowing for that predictability to exist" (Arjona 2014(Arjona , p. 1374). If rebels, as well as civilians, adhere to this social contract, Arjona speaks of rebelocracy, in which order exists, and therefore predictability is high (Arjona 2014(Arjona , p. 1374(Arjona , 2017a). The minimal involvement or non-involvement of rebels in governance activities, or the failure of such a social contract, leads conversely to disorder and low predictability of rebel and civilian behavior, a situation that she refers to as aliocracy or the rule of others (Arjona 2014(Arjona , p. 1375(Arjona , 2017a. Rebel groups choosing aliocracy were referred to by Huang as "low institutionalists" (2016, p. 82).

Political institutions and organizational structures
One major area of research in studies on rebel governance activities is the development and functioning of political institutions and the dynamics and implications of the different organizational structures of insurgent groups. Studies in this sub-field have often focused on developing general typologies of political institutions and organizational structures of insurgent groups, in order to increase the predictability of the future development of such institutions and structures and to make them more easily accessible for further research (Mampilly and Stewart 2020;Furlan 2020a;Hoffmann and Verweijen 2018;Malejacq 2016;Parkinson and Zaks 2018). Many other studies were centered around how political institutions and organizational structures influenced the decision-making of rebel groups and their military capacities, and how individual rebel-build institutions proved themselves functional or dysfunctional (Stewart 2020;Jackson and Amiri 2019;Parkinson 2013;Provost 2017;Zelin 2020). Other studies also focused on how political order under rebel rule varies within the territory a rebel group controls, and comprehensively examined conditions influencing rebel governance and multiple local orders (Arjona 2009, p. 126) in civil wars (Arjona 2016;Berti 2020;Branch and Mampilly 2005;Blunt 2003).
Introducing a new typology to further conceptualize rebel governance, Mampilly and Stewart identified four dimensions of rebel rule which underlie the political institutions of insurgent groups and lead to six different forms of rebel governance (2020, pp. 7 and 17). The authors define these dimensions as power-sharing with civilians, integration of preexisting civilian institutions, innovation of such preexisting institu-tions, and inclusiveness of governance institutions (2020, pp. 7-15). Applying these dimensions to various insurgent groups led the authors to six types of rebel political institutions, which are martial law (Rwandan Patriotic Front); partial subjugation (Islamic State); status quo, less inclusive (Eritrean Liberation Front); status quo, more inclusive (Gerakan Aceh Merdekafree); transformative, less inclusive (Khmer Rouge); and transformative, more inclusive (African Party for the Independence of Guinea and Cape Verde) (Mampilly and Stewart 2020, pp. 17-23). Besides their contribution to improving the conceptualization of political institutions in rebel governance, what is particularly interesting is that the authors classified the Islamic State as partial subjugation but noted that "partial subjugation was not necessarily the most common political arrangement that IS formed, as other towns experienced forms of rule closer to martial law" (2020, p. 18). Another valuable typology to classify insurgent governance was brought about by Furlan (2020a), who described seven dimensions of governance (2020, p. 2). These dimensions are inclusivity, civilians, generation of compliance, [interaction with] other actors, institutions and personnel, bureaucratization, and executive style (Furlan 2020a, pp. 490-496).
In recent years, the analysis of Islamist rebel governance has become more prominent, although still under-researched, particularly concerning the governance of the Islamic State and the Taliban but also some smaller groups such as Boko Haram or Hay'at Tahrir al-Sham (Hassan 2022;Keser and Fakhoury 2022). In this context, valuable work has emerged on the Islamic State's regional administrative structure (Zelin 2016) and political institutions (Al-Tamimi 2015, 2016), its jurisdiction (Revkin 2016), its security apparatus (Al-Tamimi 2017, 2018), real estate administration (Al-Tamimi 2020), and its zakat and charities system (Al-Tamimi 2021), all of which have extended existing research on rebel governance in a meaningful way. Analyzing administrative documents from the Islamic State, Teiner (2021) also illustrated the Islamic State's political-administrative institutions and how they intervened in civilian life and local economies. A sound comparative work mapping a plethora of different Islamist rebel groups engaged in governance activities has been elaborated by Lia (2015).
With regard to the political institutions of specific rebel groups, Terpstra (2020) extensively researched different phases and dynamics of rebel governance by the Taliban, while Jackson and Amiri (2019) specifically focused on actors in the organizational structure of Taliban governance and investigated how the group plans and implements health and education policies, and how it deals with complaints by civilians. While Terpstra (2013), through extensive fieldwork, also examined Taliban justice provision, Schwab and Massoud (2022) studied insurgent courts in the Syrian civil war and Provost (2017) researched the judicial administration by the FARC in Colombia. All these studies provided deep insight into specific cases of rebel governance and how rebel groups design political institutions, and how these institutions in turn affect their governance. Analyzing if and how rebel political institutions may influence the military capacity of a rebel group. Stewart (2020), drawing from quantitative research, found that "governance appears to have either no relationship with rebel strength and sometimes even a negative and statistically significant relationship with rebel military capacity" (2020, p. 16).
It would be incorrect to believe that if an insurgent group gets involved in governance activities, or even has been a de facto government in a region for many years, the political order it creates would not vary within the controlled territory. Arjona (2009Arjona ( , 2016 has raised awareness to this issue by exploring the phenomenon of "multiple local orders" (2009, p. 126) where an armed group in some cases can be "nothing but a violent invader that victimizes and harasses the population, [while] in others civilians interact with it as their ruler" (2009, p. 126). Analyzing governance activities of the FARC, Arjona illustrated the group's varying rule with the village of "Librea, [where] the rebels ruled over the political, economic, and social lives of the population, while in Zama, civilian leaders remained the ultimate authority" (2016, p. 99). Variation of rebel rule in different regions can often depend on the aforementioned factor of civilians' resistance to or cooperation with rebel rule (Arjona 2016, pp. 115-116). Branch and Mampilly (2005), referring to the example of SPLM/A rule in Sudan, also raised a concern that ethnic conflicts may shape variation in different regions by the same rebel group (2005, p. 1). The authors argue that the conflict between Dinka and Equatorians partially shaped the structure of local governance of the Dinka-dominated SPLM/A (2005, p. 5). The fact that the SPLM/A addressed this conflict led to ongoing negotiations and talks that culminated in "constitutional and legal reforms that would guarantee the self-rule of Equatoria free from Dinka dominance" (2005, p. 5). Branch and Mampilly's study gives a useful indication that a wide array of factors may lead to varying extents of rebel rule in different contexts.

The regulation of commercial production
Many rebel groups consider it particularly advantageous to regulate economic activities in the regions they control. The main reason for this is the opportunity to establish new ways of financing in order to safeguard the survival of the rebel group, support its political ambitions, and expand its military capacities (Kasfir 2015, p. 37). Looting, robbery, protection racketeering, kidnapping for ransom, or trade with illegal goods are well-known ways in which numerous rebel groups in history have financed their operations (Kasfir 2015, p. 37;Chojnacki and Branovic 2011, p. 92). Involvement in rebel governance considerably broadens the forms of financing and the level of potential income. Forms of financing of rebel groups that require some level of involvement in rebel governance include comprehensive tax collection or the forced takeover of companies, such as agricultural, oil, gas, or mining companies (Kasfir 2015, p. 37;Chojnacki and Branovic 2011, p. 92;Conrad et al. 2021;Mampilly 2021;Le Billon 2021;Johnston et al. 2019, p. xi;Revkin 2020a;Thurston 2021). Whether or not some of the above-mentioned forms of financing need to be adapted or abolished also depends primarily on whether a rebel group wants to achieve support from civilians living in the areas it controls (Uribe 2017, p. 15). Chojnacki and Branovic argued that rebel groups have to "choose between the institutionalization of a political order [...] or a violence-mediated state of conflict, in which the civilian population is used as spoils, or as an extractable resource" (2011, p. 99).
Several studies have looked more specifically at the changing forms of financing of rebel groups, and how their involvement in rebel governance has influenced their methods of financing. The institutionalization of the regulation of commercial production in the organizational structure of a rebel group was presented by Suykens, showing that the Government Committee of the Naxalites in India consisted of eight departments, of which the three for finance, agriculture, and forest production were directly involved in the regulation of the economy (Suykens 2015, p. 140). Borch and Stuvøy have shown how the FARC in Colombia has benefited significantly from cocaine trafficking, while at the same time it engaged in economic activities such as "mining (gold, emeralds, etc.), agriculture, [and] stockbreeding" (2008, p. 108) in areas of little government influence, and levied taxes on the local population (2008, p. 108).
Concerning tax collection by rebel groups, Wickham-Crowley has shown in detail how Sendero Luminoso in Peru has benefited greatly from taxing local Coca farmers and offered protection from state repression in exchange (2015, p. 58). The Taliban have also protected both Afghan poppy farmers and heroin smugglers from state interference in areas under their control and have demanded fees for such protection (Uribe 2017, p. 21). This practice has made the Taliban quite popular among the local population, as it has allowed the farmers to increase their income significantly (Uribe 2017, p. 21). It is therefore debatable whether such agreements should be seen as protection rackets or as mutually agreed arrangements. With an in-depth examination of the Islamic State, Revkin (2020a) has also shown how extensive taxation by wellfinanced rebel groups can be explained. Another noteworthy aspect of rebel tax collection was highlighted by Barter, who investigated how the Indonesian GAM cooperated with Islamic leaders who were able to extensively collect taxes for the group since they enjoyed a high reputation and were considered honest (Barter 2015, p. 233).

Civilian life under rebel rule
Considering the basic question of how rebel groups interact with civilians in the territories they control, various studies that researched the relationship between rebel groups and civilians have conceptualized and shown empirically how rebels decide how they interact with civilians (Kasfir 2005 Revkin and Ahram 2020;Vale 2020) or focused on civilian resistance to insurgent governance (Bamber and Svensson 2022;Barter 2014Barter , 2015Arjona 2015Arjona , 2017bMasullo 2020Masullo , 2021. Keister and Slantchev (2014) were able to develop a framework in which they conceptualized various possibilities of how rebel groups interact with civilians. They argued that "coercion and service provision help rebels solve their need for civilian compliance" (2014, p. 21), and that rebel groups at some point have to choose between these. Similar to Arjona's finding of multiple local orders in FARC-controlled regions, lo-K cally varying rebel responsiveness towards civilians in Côte d'Ivoire was extensively studied by Van Baalen (2021).
Keister and Slantchev presented a model of rebel decision-making that makes it possible to measure the factor of ideology, which has before been described as important but only abstract in its influence (2014, p. 21). In this model, ideology has three effects, which are a "direct effect on the level of civilian support, [an] indirect effect through its impact on the effectiveness of service provision relative to coercion, and another effect on the rebels' value of ruling" (Keister and Slantchev 2014, p. 21). Successfully bringing ideology and rebel governance into a measurable relationship revealed a "trade-off between power and ideology" (2014, p. 21) that rebel rulers have to face.
Focusing on the repression and resistance of civilians under rebel rule, Gowrinathan and Mampilly (2019) researched women's activism under LTTE rule in Sri Lanka which intended to oppose repression and the role of religious leaders, thereby finding out that the LTTE killed numerous critics of their rule, but accepted criticism by others (Gowrinathan and Mampilly 2019, p. 2). Gowrinathan and Mampilly identified a key variable, that whether or not a critic is perceived as a traitor strongly impacted the group's response, because "traitors undermine the overarching construction of the political collective that armed groups seek to develop" (2019, p. 12). Criticism that was perceived as betrayal was frequently met with the torture and killing of those critics in response (2019, pp. 2 and 12). The authors conclude that if criticism of LTTE rule was perceived as coming from loyal constituents, the critics did not have to fear harm, but those whom the LTTE perceived as traitors experienced brutal responses (2019, p. 10). They also found that while "enemies living within rebelgoverned areas are unable to resist and face repression on a collective basis" (2019, p. 10), victims of brutal treatment by rebels had opportunities to hold their punishers accountable (2019, p. 10). Through their conceptualization of violent or non-violent responses to criticism, Gowrinathan and Mampilly provided useful analytical tools to research such behavior in other cases. Another important area of research that has become more prominent in recent years, partly due to the atrocious treatment of women under Islamic State rule, is the study of sexual violence in rebel governance, which has been extensively studied by Revkin and Wood (2020) and with regard to sexual slavery by Al-Dayel et al. (2020).
Arjona argued that civilian resistance depends partially on how rebels choose to rule in the territories they control, and partially depends on the political institutions in place before rebel rule (2015, p. 198). Arjona also stated that "governance limited to the spheres of public order and tax collection tends to trigger only partial resistance" (2015, p. 198), while forms of governance beyond these spheres may result in full resistance from civilians (2015, p. 198). She argues that "communities with high-quality institutions are more likely to engage in full resistance [...], while communities with low-quality institutions are likely to engage in partial resistance only" (2015, p. 198). Further conceptualizing the cooperation or non-cooperation of civilians with rebel rule, Arjona (2017b) introduced an analytical framework in which cooperation can take the form of obedience, spontaneous support, or enlistment, while non-cooperation can take the form of disobedience, resistance, or defection (2017b, p. 761). Masullo (2020), focusing on ideational factors that influence civilian resistance to rebel rule, also raised the concern that "normative commitments can restrict civilian contention to nonviolent forms of action, while exposure to oppositional ideologies can push civilians toward more confrontational forms of noncooperation with armed groups" (2020, p. 1). Barter, focusing on conflicts in Southeast Asia (2014) and the Indonesian GAM in particular (2015), argued that in some cases of war, there is the potential for civilian agency that allows civilians to pursue their own strategies (2014, p. 6). Barter presented evidence that GAM allied with religious leaders, who served as judges for GAM rebel courts, and urban student activists, in order to enhance support for the group and its rule, leading to a constellation in which they mutually influenced each other (2015, pp. 226 and 234). The peacebuilding potential of civilians in conflicts, including under rebel rule, has also been intensively studied by Autesserre (2021) and Mac Ginty (2021), who both published authoritative works on the subject.

Social services and legitimacy enhancement
Two further central issues that were subject to several studies on rebel governance are the provision of social services by rebel rulers, and their search for options to enhance the legitimacy of their rule (Duyvesteyn 2017;Furlan 2020b;Podder 2017;Gawthorpe 2017;. These issues, again, naturally overlap with other areas of research on rebel governance, such as rebel groups' relationship with civilians, the development of political institutions, and the symbolic processes of rebel governance and rebel diplomacy, which are discussed later. Grynkewich (2008) argued that the provision of social services can be an effective tool to enhance support for an insurgent group and therefore helps to successfully challenge the authority of the state in which the group operates. Similarly, Szekely (2015) found that Hamas in Palestine was able to alter its public perception "[so] that they are not merely soldiers or ideologues, but capable bureaucrats and managers as well" (2015, p. 275). Terpstra (2013) even found that civilians under Taliban rule favor the group's informal justice system over the formal justice system of the Afghan state because "informal bodies are still able to offer quick and needs-responsive solutions that are appreciated by community members" (2013, p. 65). In another study, Terpstra and Frerks (2017) analyzed how the LTTE used strategies to enhance the group's perceived legitimacy, which were based on "Tamil nationalism, tradition, charismatic leadership, sacrifices made by LTTE cadres and the people's need for protection" (2017, p. 279), but ultimately failed to achieve that goal. Schoon (2017), focusing on how rebel groups try to generate legitimacy in the territories they control, analyzed the popular evaluation of the PKK in Turkey, while Kitzen (2017) examined rebel leaders from a Weberian perspective by categorizing their legitimacy-building efforts as "through either rational-legal ways or by co-opting local power-holders who hold a position as traditional or charismatic leaders" (2017, p. 853), thereby providing helpful analytical tools to access the strategies which rebels use to enhance their legitimacy. Khalaf, who examined legitimacy as one of the three cornerstones of Keister and Slantchev's governance model with regard to the Islamic State in Syria, has shown that the organization has sought to increase its own perceived legitimacy primarily through the continuous propagandistic justification of its ideology and diplomacy with various communities (Khalaf 2015a, b). Flynn and Stewart (2018) looked at the provision of social services by rebels and how it influences the international public perception of the legitimacy of a rebel group's rule. Astonishingly, they found that "social service provision can allow rebels to decrease-and, in some cases, eliminate-the public costs of killing civilians" (2018, p. 7). The authors also stated that service provision to supporters as well as to non-supporters, which they call "inclusive service provision" (2018, p. 7), was especially influential, and "allowed both groups in our studies to completely eliminate the negative effect of killing civilians on legitimacy" (2018, p. 7).
To understand the decision-making of insurgent groups which sometimes aims to improve, or sometimes results in the neglection of, their public perception, Akcinaroglu and Tokdemir (2018) studied which terrorist groups sought to enhance their reputation in their constituency and abroad, and which terrorist groups did not try to do so. Investigating a total of 443 terrorist groups between 1980 and 2011, Akcinaroglu and Tokdemir developed five hypotheses that help to explain how terrorist groups, decide how they treat civilians in the territory they control (2018, pp. 362-366). They argue that if ethnicity or religion are the driving forces of a terrorist group, it is more likely to seek to build a positive reputation than purely ideological groups (2018, p. 362). The authors also state that radical groups are more likely to build a negative reputation than non-radical groups and terrorist groups that want to negotiate about political change (2018, pp. 364-365). Furthermore, Akcinaroglu and Tokdemir observed that terrorist groups with territorial control are more likely to want to build a positive reputation in that territory, and groups with crossborder support are more likely to want to build a negative reputation than terrorist groups with no support (2018, pp. 363-364). In summary, choices between coercion or service provision to ensure civilian compliance to rebel rule can be influenced by several key factors such as ideology and pragmatism, the extent of radicalism, aspiration to engage in political negotiations, territorial control, and the extent of cross-border support for an insurgent group.

Rebel diplomacy
Rebel diplomacy, which focuses on if, how, and why insurgent groups try to get involved in negotiations with the government(s) of the state(s) in which they operate, non-state actors, or the international community, is another area of particular interest in studies on rebel governance. This sub-field is also heavily related to the topics of political institutions, the generation of legitimacy, and the treatment of civilians in rebel-controlled areas. Most studies that have investigated the relationship between rebel governance and diplomacy have focused on how rebel governance can enhance a group's reputation and bargaining power in negotiations with international actors or the state government(s) in which it operates (Worrall 2017;Malejacq 2017;South 2017;Staniland 2012Staniland , 2015Staniland , 2017Staniland , 2021. The sub-field of Rebel Diplomacy has also been innovatively expanded, for instance, by Schwab (2021), who studied negotiations between armed non-state actors such as Jabhat al-Nusra and other rebel groups in the syrian civil war and made a significant contribution in this area. The issue of service provision is also connected to the diplomatic engagement of insurgent groups. In a quantitative approach to rebel diplomacy, Heger and Jung (2015) showed that rebel groups, which provide social services, engage more often in conflict negotiations; that states are more willing to engage in negotiations with such rebel groups; and that negotiations of service-providing groups are more stable than those of non-providing Heger andJung (2015, 1203).
In a study on Taliban policymaking, Jackson and Amiri (2019) found that the group fruitfully negotiated with several non-governmental organizations, which were mainly aid and health agencies (2019, pp. 25-26). These groups negotiated with the Taliban in secret for a variety of reasons, which included a fear of cuts to funding if the Afghan government or donors found out (2019, pp. 25-26). The secret nature of those negotiations prohibited agencies from collectively negotiating, with the result that the Taliban "were able to play various NGOs off against one another" (2019, p. 27) and made higher demands for access from weaker NGOs (2019, p. 27). While this example demonstrated the potential for negotiations with insurgents, Mampilly (2009) analyzed how foreign aid after the Asian tsunami in 2004 affected peace negotiations with the LTTE in Sri Lanka and highlighted "that post-tsunami relief efforts closed the door to a negotiated settlement by contributing to the insurgency's failed play for autonomy from the Sri Lankan state" (2009, p. 302).

Symbolic governance
Alongside the variety of research avenues in studies on rebel governance, several scholars have acknowledged the importance of symbolic aspects of rebel governance, which can take the form of ritualized processes or symbols, signs, songs, and many more things that reflect the identity of an armed group. Symbolisms are mostly derived from the group's ideology and may also include hints to specific moments of the group's evolution, the movement from which a group may originate, or other historical references (Mampilly 2011, pp. 4 and 56). Furthermore, Mampilly highlights that "symbolic processes reduce the need for a rebellion to use force to ensure compliance" (2015b, p. 74). Mampilly divides symbolic processes into two strands, of which the first includes processes that reflect and support the "coercive and bureaucratic power of the political authority" (2015b, p. 79), and the second is intended to "strengthen [the] identification between the political authority and the civilian population" (2015b, p. 79), while both serve the purpose of enhancing the compliance to and legitimacy of a rebel group's rule. Mampilly provided several tools for accessing symbolic processes in rebel governance. The "set of symbolic processes available to rebel groups" (Mampilly 2015b, p. 82) is referred to as symbolic repertoire, while other relevant factors for using that repertoire are the sources and audience of such processes, which in combination are referred to as a symbolic register, meaning the "timbre at which a particular symbolic repertoire is calibrated" (Mampilly 2015b, p. 84).
Investigating how the LTTE in Sri Lanka tried to improve their authority and legitimacy, Terpstra and Frerks (2018) examined the symbolic governance of the group in terms of "narratives, performances, and inscriptions" Frerks 2018, p. 1001). Evaluating these forms of symbolism in rebel governance, Terpstra and Frerks found that LTTE created and successfully deployed "a politico-historical narrative on the oppression of the Tamil people by the Sinhalese (state) and the necessity of a liberated Tamil Eelam" (2018, p. 1042). Another narrative that was wellimplemented in the population which the LTTE controlled consisted of the "struggle and the heroic status of LTTE cadres fighting against the government-and dying in the fight" (2018, p. 1042). Terpstra and Frerks also observed that "cemeteries, commemorations of the 'heroes', flags, and other national symbols were the visible inscriptions of the struggle and the sacrifices made, and were widely respected within the Tamil community" (2018, p. 1042).
Another form of symbolic governance that was observed in studies on various rebel groups is the development of a new, unique currency to prove that a rebel group is capable of successfully governing its territory, and therefore to further legitimize its rule (Lokmanoglu 2020(Lokmanoglu , 2021Mampilly 2011Mampilly , p. 215, 2015b. Other forms of symbolism include the distribution of resources by insurgent groups like the FARC (Borch and Stuvøy 2008, pp. 106 and 108) and the distribution of zakah, an "obligation for any Muslim with the financial means to do so as an act of pious giving" (Blannin 2017, p. 18) which was regularly collected and distributed among the residents of the Islamic State's territory (Alkhouri and Kassirer 2015, p. 17). The above-mentioned analytical tools are useful to investigate symbolisms in rebel governance in systematic and comprehensive waysand can serve as solid guidelines to show where and in which forms symbolism in rebel governance may occur and how it is strategically deployed by armed groups.

Conclusions
All the avenues of research on rebel governance presented in this literature review are directly interconnected and, in most cases, mutually dependent. Nevertheless, the division of the sub-fields of Rebel Governance made here helps to systematically grasp and map this field of research. The early debates on rebel governance mainly took their impetus from common perceptions in IR that territorial authority and the assumption of political administration were invariably owned and exercised by nation-states.
The fact that most armed conflicts today are intra-state rather than inter-state made it all the more necessary to point out that spaces in which a nation-state has lost its authority are not ungoverned, but authority is usurped by other actors. The observations of many scholars that rebel groups informally appropriate state authority and exercise governance for different reasons, with varying degrees of ambition and success, supports the argument that governance does not necessarily imply being exercised by nation-states and has resulted in a vibrant field of research. An interesting circumstance that strongly benefits the field is the lack of dominance of one or a small number of methodological perspectives. Research on rebel governance shows a great diversity of methods, in which qualitative and quantitative approaches regularly complement each other in useful ways.
As Rebel Governance has evolved as a discipline, the strands of research reviewed here have become increasingly nuanced within the field. In the sphere of political institutions and organizational structure, valuable typologies have been elaborated for the classification of different groups' governance in general, while also intensive case studies of the administrative activities of rebel groups through more or less professional institutions have been carried out. Interference in the economy and forms of financing of rebel groups in the context of rebel governance have also been well studied for certain rebel groups. Repression of civilians, their compliance, or resistance have now also been studied frequently. This field of research, however, also needs further elaboration due to its sheer complexity. Arjona's fundamental studies and the excellent case studies presented herein will serve as a solid basis for further research. Legitimacy and trust-building are central elements of rebel governance. Providing social services to suggest effective governance and to increase the perceived legitimacy of a rebel group has a prominent role in this regard, even partially compensating the reputation-damaging violent rule of a rebel group. In the context of rebel diplomacy, research has also been presented which has shown that social services also have an impact in this respect, as they are related to more durable peace negotiations. Symbolic governance is another phenomenon that should always be considered in studies of rebel governance. It reflects the identity and ideology of a rebel group, suggests effective governance, and can even help to increase civil compliance.
Overall, the field of Rebel Governance is relatively young compared to other disciplines in social sciences but has grown steadily in recent years. The absence of dominance of particular methodological directions and the mutual recognition between researchers with complementary approaches is also a welcome condition. The discipline is, in its current state, ideal for conducting innovative research, as a rich canon of literature is now available as a foundation, but many phenomena regarding rebel governance remain highly under-researched.
Funding Open Access funding enabled and organized by Projekt DEAL.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4. 0/.