The End of Stability – How Burkina Faso Fell Apart

ABSTRACT Not so long-ago Burkina Faso was considered an “island” of stability in a conflict-prone part of Africa. This is not the case anymore as armed insurgencies have caused widespread insecurity. While spill-over effects from the conflict in Mali clearly play a role, we argue that the sudden demise of the rule and regime of Blaise Compaoré also is an important contributing factor. To decipher to what extent regime transition shaped the current state of affairs, we show that what kept Burkina Faso stable and out of the conflicts in the region was a “big man deep state” of formal and informal networks of security provisions. When this “deep state” vanished with the ousting of Compaoré and his allies, local security providers have sought new solutions, and this strengthened the role of self-defense militias but also led them to compete against each other, at times also violently. This provided fertile terrain for jihadi insurgents. Therefore, this paper is an attempt to provide a conceptual understanding of how weak rulers actually rule, how some succeed in preserving their rule for a lengthy period of time, and what can happen when they eventually fall.


Introduction
Weak states can be affected by the spill-over of violence and conflict from neighboring states. Due to weak infrastructural capacity and a long history of transnational circulation predating the colonial administrative architecture, state borders are porous and can often not do much to prevent the spread of violent conflict. 1 In the Lake Chad Basin, the Boko Haram insurgency that originated in northern Nigeria in 2009 has spread to three neighboring states (Cameroon, Chad and Niger). In the Mano River Basin, Liberia imploded in 1989, and both Sierra Leone and Côte d'Ivoire subsequently had their own civil wars. 2 However, some weak states manage to navigate the bad neighborhood that they find themselves in and avoid such spill-over effects. Guinea, for example, was a seemingly unlikely stable state during the Mano River conflicts. 3 For a long time, Burkina Faso resembled Guinea and seemed remarkably resilient to negative spill-over effects of this kind. Based on Burkina Faso's history of religious tolerance and nonviolence, several analysts and researchers deemed the risk of conflict spreading from Mali to Burkina Faso as very limited in the short and medium-term. 4 Similarly, the UNDP programme launched in 2016 to prevent violent extremism in Africa identified Mali as an epicenter of violent extremism and Niger as the potential spill-over country in this region, while Burkina Faso was not even identified as "at risk." 5 Nonetheless, the first terrorist attack struck the capital of Ouagadougou in January 2016, and in November, the same year, the first jihadist insurgency erupted. Today, Burkina Faso is rivaling Mali as the epicenter of the crisis in the Sahel, with jihadist insurgent groups claiming allegiance, at least rhetorically, to al-Qaeda and the Islamic State, competing for influence and control over the territory. 6 How can we explain this remarkably negative transition from being a state considered to have relatively strong stability to suddenly slipping into devastating violent conflict? Our argument is that when the former President Blaise Compaoré's 27-year long reign came to an abrupt halt in 2014 what Burkina Faso was facing was a complicated and troublesome political transition. When Compaoré was forced to step down, not only did he vanish from the country, but with his departure what also disappeared was a "big man deep state" deeply entangled with formal and informal networks of security provisions. While we certainly are aware of the many administrative flaws and human rights violations of the Compaoré regime, 7 we also argue that Burkina Faso under his rule represents an interesting case. For almost three decades, Burkina Faso showed substantial domestic stability and weathered a number of intrastate conflicts in its region. However, soon after Compaoré's demise in late 2014, Burkina Faso slid into violence, becoming a new battleground in the Sahel crisis that started in Mali in 2012. The question is, therefore, if this was just the case of a neighborhood effect causing spill-over from the battlegrounds in Mali just waiting to happen, 8 or if a more nuanced alternative approach is better suited to explain the sudden demise of security in Burkina Faso?
In this article, we propose a novel conceptual framework that aims to refine and deepen our understanding of the inner workings of weak and resourcepoor states and why they, for a time, can seem remarkably stable, but sometimes also suddenly fall apart. This will be achieved through an analysis of how Compaoré had worked to secure his rule and his country, and what happened when this state of formal and informal security networks was dismantled by the popular uprising that led to his demise.
As this article continues, we start by developing a theoretical framework based on a novel conceptualization of the "big man deep state." In the next step, we use this framework to analyze the Compaoré-state and how it managed to navigate a bad neighborhood. Thereafter, we turn to the question of how the state of Compaoré fell apart and what the outcome has been. In the concluding section, we summarize key findings and what implications this analysis has for the debate about these types of states and what this implies for future research.

Explaining the violence unleashed in Burkina Faso
While the neighborhood effect argument -that is the argument that intrastate conflicts have been shown to often cluster in time and space, increasing the risk of violent conflict in proximate states -carries some empirical clout, 9 we argue that an important factor missing is the rapid disintegration of the country's neopatrimonial security system. This system of rule tied together the political center and the periphery facilitating the navigation of different threats despite the obvious shortcomings of a weak and resource-poor state.
In many African states, a range of non-state actors operating beyond, in parallel or in tandem with officially sanctioned channels have facilitated the coexistence of bureaucratic rationality and patrimonial norms as a "neopatrimonial" form of rule, which enabled central governments to coopt not only national elites but also regional elites and relevant actors in the peripheries. 10 At the heart of this conceptualization of neopatrimonialism lies the idea that the ruler's authority can be reproduced all the way down to the village level through the president's clients who yet again become patrons for their own set of clients by doling out resources and opportunities by virtue of their formal role in the state system. 11 While accurate in several cases, the weakness of this argument is that at some point in time, almost every African state has been described as neopatrimonial, lumping together a great number of very different states, assuming the phenomenon as an inherently societal issue. On the contrary, we embrace Nugent's suggestion to re-focus the attention on networks and institutions in order to better understand how neopatrimonalism as a system of rule can work. 12 What we suggest is that some states, while displaying several characteristics of neopatrimonial, also develop what we call a "deep state" which dominates the formal state by an organizational body with its own sets of norms and hierarchy.
Indeed, some weak states do not only rule on the basis of loose or informal personal networks and alliances but also through a more institutionalized "deep state." The literature, while acknowledging several types of "deep states," agrees on. some critical elements: that it can be seen as a type of informal actor not part of the state's formal institutions but still operating within it, 13 but also a "political interplay between unacknowledged or unrecognised factions inside and outside the regular government." 14 In essence, it is a network of individuals from the coercive state apparatus and various civilian spheres involved in the regular activities of the state while maintaining a shadow set of activities. 15 Such a network can be called an "autocratic clique" which gathers political support and exerts direct influence on the regular state and society at large through hierarchical ties. This clique can take the form of a "security community" composed by "those elements of the regime most directly involved in the planning and execution of repression, intelligence gathering, interrogation, torture, and internal clandestine armed operations." 16 Autocratic cliques are semi-formal institutions because they lack formal recognition despite their official operation at large, and often acts with general impunity. This can often be seen in the deep state's symbiotic relationship with racketeering and low-intensity warfare in territories that provide safe havens for extracting resources through trafficking and other shadowy economic activities. 17 Such armed violence is not a necessary condition for a deep state to emerge but can be a decisive moment for its assertiveness. 18 "Deep states" or "security states" have been identified in Western liberal democracies, 19 especially during the Cold War-era. Today, this type of rule is most commonly associated with defective democracies 20 in non-Western states, 21 and especially where state security takes absolute precedence. 22 Defective democracies describe states where either the military is devoid of any control or is under non-democratic control, in which the latter consists of networks of patronage. 23 Holding key positions in the state, the members of the "deep state" bypass the state's formal institutions to promote their interest, secure material assets, economic privileges, and their own political power. 24 The deep-state conceptualization can help us to refine and increase our understanding of weak states using Burkina Faso as a case. Therefore, our aim and ambition are to move beyond "neopatrimonialism" and the muchcriticized "fragile-states approach" that assumes fragility without explaining how such states actually work by offering an alternative conceptualization of security governance in weak states. 25 In the case of Burkina Faso, for lack of a better term, we chose to heuristically employ the expression of a "big man deep state"; a form that can condense the main characters of Compaoré's regime and the kind of internal workings that allowed such a regime to survive for decades. The embeddedness of security and military forces into economic and policy-making circles in power under Compaoré bears similarities to a "deep state" at the head of which rested the skillful military-turned-president ruler. Thus, a Big Man who embodies power and carries the power of the regime in himself as the embodiment of both state and regime. The words and orders of the Big Man can matter more than the country's official constitution and laws, but this does not mean that such rule by necessity has to be despotic, completely autocratic and unpredictable. The kind of Big Man we have in mind is therefore not the same phenomenon as the personal ruler of Jackson & Rosberg whose rule is based on despotism and unpredictability. 26 Instead, while the Big Man of Marshall-Sahlins is powerful, 27 his power is also based on those that recognize his Big Man status, meaning that the Big Man is also indebted to them. This suggest that what this conceptualization capture is a relationship based on reciprocity and while what Driscoll calls "Big Man Governance" has an informal character it must also have a degree of predictability and thus being rule-bound. 28 As Utas reminds us, 'the Big Man do not generally control followers. 29 Quite the opposite: it is in the interests of followers to maintain ties with the Big Man.
Compaoré's unquestioned personalized rule rested on his fearing fame as a successful army man, although he maintained a sober and discrete public persona, far from the image of the lavish Big Man present in some of the literature. 30 However, it is the ability to attract and maintain a following that makes the Big Man and not charisma per se, and the relatively unbiased nature of the redistribution of dividends with regards to kinship and ethnicity was unquestionably an important part of Compaoré's Big Man status. 31 Most importantly, however, acting at the intersection between political negotiations and economic arrangements, Compaoré sustained -and was supported -by a small autocratic clique of key security officers, mainly belonging to the Régiment de sécurité présidentielle (RSP), of whom he was the Big Man. Among this select group, strategic individuals such as Gilbert Dienderé 32 and Djibril Bassolé, 33 allowed the entrenchment of business, security and politics in a network of alliances carefully built by Compaoré, who muscled his way at the top of the state and acted as the Big Man and by being so also became the very hub of the "deep state." Through this position, he was an irreplaceable cog in the governance of Burkina. This "big man deep state" was therefore the carefully built and apparently resilient structure that kept Compaoré's rule together, but it also rested on the rather fragile foundations of his political persona. Once Compaoré was removed by street protests in 2014, and key members of the military and the RSP were imprisoned, the "big man deep state" started to crumble. In this very moment the neopatrimonial arrangement working for the "deep state" revealed its vulnerability that had been there all along, but only now came into stark visibility.
The main weakness of a neopatrimonial system, even with a more institutionalized big man rule at the center of a "deep state," is that it can become an engine for perpetual crisis, regardless of past stability, as the lack of more formally institutionalized structures can create fragmentation that sustains itself into even deeper levels of fragmentation. 34 As Bøås and Dunn point out, if the patronage systems fail to deliver on the promises embedded in them, neopatrimonialism becomes a source of instability. 35 When neopatrimonial practices become unstable, as in Burkina Faso in the 2000s, the established modality of governance is thrown into question and begins to fray.

Methodology and research methods
Methodologically, the article is based on the authors' long engagement with Burkina Faso and on the literature and other types of open access research material. Due to Covid-19 fieldwork planned for 2020 was delayed with over a year. Intensive fieldwork was conducted in October 2021 in Ouagadougou, allowing ten interviews with key informants, in addition to surveys, research notes and field reports with interviews conducted by the authors' scientific partners in Burkina Faso. Our scientific partners carried out fieldwork in the spring of 2021 interviewing 21 persons from the northern region, Soum, and 30 persons in the southern region, Center-Est. The interviewees included state officials, farmers, herders, traders, civil society leaders, traditional and religious authorities, and NGO-workers. 36 A more sustained and long-termed work on the terrain, however, was not possible due to the ongoing Covid-19 pandemic and the deteriorating security conditions in the country at large.
In the analysis of the qualitative data and secondary literature, we searched for recurring patterns. Specifically, we identified who the various members of the autocratic clique are, and their interconnections between civilian spheres, the "security community," and organized crime. We then sought to identify the ways in which the "big man deep state" asserted its political authority and became intrinsically intermeshed with state security, for example, through the co-optation of key elements of the security apparatus, engaging in clandestine operations in areas of low-intensity warfare, maintaining relations with organized criminals, and establishing a mediation business. In short, we investigated whether "regime security" and "state security" had become "two sides of the same coin." 37 The exact nature of the "big man deep state" and the full extent of its impact on Burkina Faso's security provisions and ability to withstand a jihadist insurgency in the likes of today is difficult to determine. However, if we can better elucidate these informal actors, we can also be more precise in grasping their role in the current security predicament and even draw potential parallels with other cases.

The Compaoré-state
Emerging from the French colonial empire, Burkina Faso (then Upper Volta) was like many other African countries characterized by a combination of administrative weakness, political and economic fragility that led to the typical unsettlements of post-colonial African states. The reach of the state was limited, and authority perpetuated a form of indirect rule through local chieftaincies from the colonial era. The limited loyalty toward state authorities in Ouagadougou was based on patronage networks diverting scarce resources, which created a highly personalized political leadership. 38 The societal landscape during the country's first period of independence could be described as that of a "web-like" society, presenting fragmented and heterogeneous forms of socio-political control. 39 Burkina Faso's first decades were therefore turbulent, characterized by a weak single party-dominated state and a series of military regimes. The most noteworthy of these was the short-lived one led by Thomas Sankara who, in companion with Compaoré, attempted to strengthen state functions and public authority. 40 This did not last for long as Blaise Compaoré staged a coup against Sankara in 1987 and gradually built up a "big man deep state" as the regime went through a controlled democratic transition in the early 1990s. During Compaoré's rule, in a pyramid-like structure, networks of patrons and clients stretched from the top-echelons of the government down to the village level. Compaoré's system was very flexible in the sense that a local patron could himself be a client to some district patron who, in turn, could support a regional or national figure. 41 The state did not have to provide services directly to everyone, nor did it need their direct support since everything flowed through intermediaries. Those holding central state office could utilize the patronage system to acquire local patrons, which extended their influence into areas they would otherwise not be able to penetrate. 42 In this way, the central government was indirectly but firmly connected to the peripheries.
The "big man deep state" of Compaoré was therefore built around a set of networks and alliances that were situated in between formal institutions and informal areas, including both the military, Compaoré's political party, support groups and clientelist structures connecting the formal and the informal, creating a powerful informally institutionalized structure in-between these.

The political-military regime
The regime of Compaoré was characterized by a tight relationship between military and political elites. The army, led by Compaoré and his companions, was the main instrument of coercion and repression against real and perceived opponents, but also a state agency that was deliberately weakened by divisions and patronage. These divisions were created by giving priority to some branches over others. While this protected the ruler and his regime from the emergence of possible contenders to power in the army, it also resulted in grievances and riots in the armed forces. 43 Harsch argues that given the regime's reliance on coercion to shore up power and Compaoré's origin in the army, it was not surprising that his patronage network would penetrate the security forces which blurred the lines between security personnel and those in the ruling party. 44 Lieutenant Colonel Djibril Bassolé, the Security Minister (later Minister of Foreign Affairs) controlled the gendarmes while building a base of his own within the ruling party. 45 The most prominent client of Compaoré was General Gilbert Diendéré, in charge of the elite regiment drawn from the most able in the military forces, the RSP. 46 Despite its official role as a security service for the President of Burkina Faso, the RSP was the "big man deep state" extended arm into the "security community" enjoying impunity for engaging in corrupt activities and violent exertion of influence. In other words, the RSP was involved in regime preservation activities but also in formal and informal networks of security provisions. While being a military branch directly under the president, the RSP also operated as Burkina Faso's intelligence service. While reputed as one of the best intelligence services in the region, it rested on the informal networks of key ally General Diendéré. It was a network cultivated over years based on formal and informal connections with regional Big Men and clandestine operations within and outside Burkina Faso in areas of low-intensity warfare in the neighborhood. 47 The RSP was both the most capable fighting force in the military as well as the state's main intelligence service. This means that the organization was to some degree indispensable to the security of the state, whilst being part and parcel of a hidden and unaccountable "big man deep state."

The political party
The second pillar that Compaoré's regime rested upon was the political party, Congrés pour la Démocratie et le Progrés (CDP). It was an instrumental component of a neopatrimonial state that came to represent a monopolized system of governance and patronage, though not without internal factional struggles. The CDP, with a nation-spanning clientilist system, not only created a hegemony in Burkinabe political life but also reached into the bureaucracy, armed forces, business circles, traditional chiefdoms, religious groups and civil society. 48 Harsch describes the Compaoré-state as an encompassing "party state" that functioned on two levels: one was official, based on the rule of law and accountability through pluralist elections and constitutionalism, representing a "shadow play," partly, to please donor agencies and Western partners. 49 The ruling party could always subvert, manipulate and simply ignore formal institutions whenever it needed to. The second part of this state was an unofficial realm of politics in which power and at times more arbitrary forms of rule prevailed. While achieving a monopoly in the political realm, with time, the power politics within the CDP resulted in internal struggles between a pro-Compaoré faction and an oppositional faction: the FEDAP-BC (Fédération associative pour la paix avec Blaise Compaoré) centered around Compaoré's brother, François, and, in opposition, some of the so-called "party-barons," among them the former president Roch Marc Kaboré.

The chiefs and business circles -the clientelist system connecting center and periphery
The traditional chiefdoms were the third pillar of the Compaoré-regime. The chiefs provided local support to the regime by forming a second, informal level of administration and extension of its power outside the capital. Most importantly, the chiefs represented the bulk of the clientelist system. The system never existed formally, but before every important election, chiefs would secretly issue voting instructions to the people. Traditional chiefs supported Compaoré not only as an expression of gratitude for being allowed to retain power over local constituencies but also because chiefs could claim the spoils this system of rule created. 50 However, some nuance must be added as this only served as a pact of stability between the center and the periphery as long as these local chiefs were respected in their local rural areas and could mediate in and resolve local disputes. 51 Thus, through local chiefs, a web of alliances could work both to neutralize threats to Compaoré's authority as well as defuse underlying community-based tensions. 52 Business circles were an integral part of the regime's allies, providing it with financial resources. Oumarou Kanazaoé, who died in 2011, is an example of an important patron close to Compaoré. He allied himself with every government in Burkina Faso and made a fortune on public contracts. He played a significant role in the Compaoré-regime as he guaranteed support for the CDP in his native region of the Nord, funded the party and many of its infrastructure projects, and he also helped to defuse conflict as the head of the Muslim community where he united different currents. 53 Business circles moreover also established associations to support (or negotiate with) the regime. It is the case of the FEDAP-BC that could count on the support of rich businesspeople like Alizéta Ouédraogo and Lassine Diawara. Elected officials, the CDP, chiefs and economic actors all had a stake in preserving the political status quo and made use of both repression and co-option of dissidents in direct contact with central authorities. 54 What this leads us to see through the "big man deep state" lens is a state where formal institutional structures as the military apparatus and the regime's political party are fused with other informal structures of the state apparatus. Together this creates a web-like but still a pyramidical social structure that enables regime stability, and which presents features of a "deep state." This clearly points to Erdmann and Engel's argument that studies of neopatrimonial practices has placed too much emphasis on patrimonialism and neglected the state part of the puzzle, on which the theory initially aimed to draw attention to. 55 However, what we aim to show as we continue our review of the Compaoré state is that if we are to understand how rulers of weak states actually rule, we need to move beyond the tendency to ignore regime preservation strategies. This is important as rulers of weak states do not just muddle through; they strategize as everybody else. It is just that some are more successful than others. Compaoré certainly were one of these for a long time as he skillfully combined his "deep state" resources both domestically and in his regional neighborhood.

Navigating a bad neighborhood
As the head of a resource-poor nation, Compaoré started out as a regional troublemaker befriending multiple rebel leaders but later chose to turn his country into a regional diplomatic powerhouse, making mediation Burkina Faso's trademark. The regime made itself seem indispensable by promoting the image of a poor but enterprising and well-administered country capable of resolving regional crises as well as utilizing its networks to negotiate the release of international hostages taken by insurgents active in the regional neighborhood. 56 The regime also maintained, for the most part, good relations with Western countries and donor agencies on which it was heavily financially dependent. However, the role of Compaoré and his senior officials in the region was much more ambiguous than what their regular peace envoys and international summits in Ouagadougou would suggest. The interests of the regime in these informal arrangements were all bound up with the preservation and prosperity of the regime and not the strengthening of the institutional design of the state. 57

Unofficial diplomacy, covert operations and profiteering
The Compaoré-regime was deeply involved in informal regional diplomacy creating arrangements with several different non-state actors. Idrissa argues that the raison d'être of the regime was to serve the interests of the ruling elites and maintain the loyalty of sections of the population which were key to its regime preservation strategies. These interests expanded beyond Burkina into troubled spots of West Africa, in Liberia and Sierra Leone in the 1990s and Côte d'Ivoire and Mali in the 2000s. While Compaoré engaged officially in several of these countries' various crises, he and his senior commanders and top foreign affairs personnel also engaged in unofficial activities that served several purposes, among them enriching top elites in the "big man deep state" and feeding the patronage machine. 58 Internally, Burkina Faso provided little opportunity for military graft and profiteering, but the political economy of war created by various conflicts in Africa provided opportunities for the regime. In the Mano River wars, while denying any involvement in the Liberian civil war, Compaoré and his trusted few supplied Charles Taylor with arms, ammunition and troops, and opposed the military intervention by ECOWAS. In 1991, Compaoré admitted having sent 700 troops to Liberia but claimed that Burkina Faso's involvement in the conflict was now over. However, in 1994, ECOWAS renewed allegations of supply of arms and training of mercenaries. 59 As the Mano River wars expanded, so did Compaoré's covert operations. A UN panel found conclusive evidence of Burkinabe authorities' involvement in illegal arms dealing in Sierra Leone. Arms were legally purchased in Ukraine by the Ministry of Defense with a document signed by Diendéré stating that Burkina Faso would be the sole user of the weapons. However, the arms were smuggled through Liberia and into the hands of rebels in Sierra Leone in exchange for diamonds. 60 Another UN panel described Burkina Faso as a safe haven for the Angolan rebel group UNITA led by Jonas Savimbi. The panel found it highly likely that arms and ammunition destined for Burkina Faso were diverted to UNITA in Angola during the 1990s. Ouagadougou was also the preferred place of UNITA to conduct its diamond transactions. In exchange for the hospitality and protection served by Compaoré, Savimbi made direct personal payments to the president, contributed financially to his political campaigns and on two occasions provided funds for Burkina Faso's state coffers. 61 Burkina Faso's involvement in the Ivorian civil war provides a good example of the deep state's duality as it officially operated as a peace envoy securing the Ouagadougou Political Agreement in 2007 but also unofficially profiting from the conflict through supporting the rebels, the Forces Nouvelles. Compared to Compaoré's involvement with Taylor, UNITA and the RUF, the situation in Côte d'Ivoire also seemed to involve significant political interests. Burkina Faso rarely exported its internal problems to the outside world, but it did export, mainly to Côte d'Ivoire, millions of migrant workers. The Ivorian civil wars and anti-Burkinabe sentiments led the Compaoré regime to side with the Forces Nouvelles. With the southern port in Abidjan unavailable to northerners in Côte d'Ivoire, Burkina Faso became the main recipient of most of the north's exports. A UN group of experts found evidence of arms and ammunition being transported from Burkina Faso territories to rebel-held territories in Côte d'Ivoire but no evidence connecting it to the authorities. 62 However, with the regime's previous engagement in arms dealing, it is not unlikely that at least parts of the autocratic clique in the "big man deep state" had a stake in such transactions. This is an excellent example of how good the Compaoré-regime was at strategically managing various regional crises in a way that benefitted the regime and its supporters. The regime succeeded in both profiteering from the conflict and avoided a possibly destabilizing refugee crisis of people of Burkinabe origin returning form Côte d'Ivoire. It is also an example of the mediation "industry" that made Burkina Faso an unlikely diplomatic powerhouse in West Africa and, to outside powers, an island of stability in an otherwise unstable and troubled region.

The mediation business
Compaoré and his associates built up a mediation "industry" that provided prestige, networks for intelligence gathering and openings for Burkinabe businesspeople. As with the regime-controlled democratic transition, the mediation business bolstered the regime's international legitimacy helping to secure good donor-relations and to preserve domestic stability by successfully handling destabilizing external factors such as refugee inflows from the various conflicts that erupted along the country's borders. The regime hosted several international summits and conferences -the Franco-Africa summit in 1996, the Organization of African Unity (OAU) meeting in 1998, and the 2005 meeting of Francophone countries. As peace envoys, regime officials mediated political crises and violent conflicts in Togo Niger, Côte d'Ivoire, Guinea and Mali during the 1990s and 2000s. 63 The mediation business rested upon the personal connections and networks of a few important elite individuals in the "big man deep state" rather than the formal organization of the Burkinabe Foreign Ministry. Compaoré was often present as a peace envoy; for example, serving as a representative for ECOWAS-missions and hailed as the craftsman of successes such as the Ouagadougou Political Agreement during the Ivorian civil war. 64 Apart from the president, particularly Djibril Bassolé played an important role. Emerging from the security forces, Bassolé made a career as a minister serving in different posts and became Compaoré's most trusted diplomat leading mediation missions in Togo, Niger, Côte d'Ivoire and Mali. When the current civil war in Mali erupted, Bassolé traveled to Northern Mali to meet with today's leader of the jihadist group, JNIM, 65 Iyad ag Ghaly. 66 Another important figure was the Mauritanian Moustapha Ould Limam Chafi, who served as a presidential advisor to Compaoré. Chafi had intimate knowledge of the Saharan borderlands and its peoples and was one of the most important unofficial diplomats in Burkina Faso. Through his connections, he helped Compaoré manage the Tuareg crisis in Niger (2007-09); and served as a liaison between Ouagadougou and the Forces Nouvelles rebellion during the Ivorian crisis (2002-7). He was also a part of the ECOWAS negotiation team in northern Mali and worked to free Western hostages kidnapped by AQIM. 67 Under Compaoré, Bassolé and Chafi were two "big men" situated in between formal and informal institutions serving the formal state of Burkina Faso but also the unaccountable "big man deep state" that worked to their own personal benefit, but also for the survival of the regime and domestic stability of Burkina Faso.
Acquiring diplomatic prestige served as an external regime preservation strategy. This strategy allowed Compaoré to place many of his allies and clients in significant regional and international organizations. From a regime -preservation perspective, this meant that Compaoré could both reward or remove important regime allies and increase his external influence. For example, in 2000, Ablassé Ouédraogo, a former Foreign Minister, was appointed Deputy Director-General of the World Trade Organization, 68 and former Prime Minister Kadré Désiré Ouédraogo, was appointed president of ECOWAS in 2012. 69 Second, Western donors critical of Compaoré's meddling in African conflicts during the 1990s came around due to Burkina Faso's apparent switch to more peaceful diplomacy.
There are two things that can be seen in Burkina Faso's unofficial and official involvement in the region under Compaoré. First, the covert involvement in Angola, Liberia, Sierra Leone and Côte d'Ivoire shows a "big man deep state" that is hidden and unaccountable, seizing the opportunities found in territories with low-intensity warfare providing safe havens for extracting resources through smuggling and other shadowy economic activities. Such profiteering can be seen as a preservation strategy for the regime as it lavishly rewards the top echelons of the neopatrimonial system and contributes to fulfilling the promises embedded within it by feeding the patronage machine.
Second, the formal state makes itself visible in its engagement in peace talks and international mediation. This adds prestige, while it also puts a veil over the hidden and unaccountable deep state's involvement. It was not the formal state with its formal institutions per se that was involved but rather the selected few from the autocratic clique in Compaoré's "big man deep state." The consequence of which was not only that these informal institutions guided Burkina Faso's diplomatic decisions but also that Compaoré and his selected few made themselves indispensable to the Burkinabe formal state's navigation of its bad neighborhood. We argue that this shows how deeply entangled the "big man deep state" was with formal and informal networks of security provisions and how skillfully Compaoré combined his "deep state" resources both domestically and in his regional neighborhood.

The "big man deep state" that fell apart
The managing of clients and allies of the "big man deep state," however, was based on the maneuvering of the various factions composing it. The regime survived several mutinies in the army (1999,2003,2006,2007,2011), often in combination with popular protests (1999, 2011), through the brokering of deals between the senior military leadership close to Compaoré and the RSP and younger generations of officers. Such pacts, however, were only successful in the medium term, as their fragility could not prevent the progressive falling apart between factions. Internal struggles within the ruling circle and the need to quell street protests pushed the regime to lose oversight over many rural areas where banditry went rampant. To the latter the government responded through the loosening of its tight overseeing and the promotion of local security initiatives, fostering the emergence of parallel security governance systems. Meanwhile, the 2011 mutiny among junior ranks of the army overlapped with popular protests over the death of a student while in police custody. 70 This led to the first major crack in the "deep state" as the RSP for the first time joined the mutiny, officially because of unpaid allowances and housing. 71 Compaoré promptly rushed to give into the demands of the RSP and regained their loyalty. This however provoked further protests in parts of the army that felt treated as a second-class unit, turning their anger against the RSP which was deployed to disarm the mutineers.
The poor relationship between the regular army and the RSP was especially evident in the chaotic months following the demise of Compaoré when a power struggle at the center eventually fragmented and dissolved the deep state with its security apparatus, regional diplomacy and rural networks. At first, the army Chief of Staff claimed himself the head of state before being side-lined by the second-in-command in the RSP, Colonel Yacouba Zida. 72 As one of the main cornerstones of the old regime, Colonel Zida's nomination was highly contested, and after sustained pressure from civil society, trade unions and many other societal actors, he stepped aside for a former diplomat, Michel Kafando, to be the head of a transitional government. However, in a matter of days, Zida was proclaimed prime minister, showing the continued influence of the RSP.
The power struggle at the political center reached its peak in September 2015, when the transitional government decided to dissolve the RSP, with Zida's approval, and strike a decisive blow to the RSP's ambitions to keep the grip over the state apparatus and its resources. This provoked a coup by the RSP and General Diendéré, seizing the capital and removing the transitional government. However, the regular army intervened and ended the coup in a week. Bassolé did not publicly denounce the coup, which gave it the look of being both an attempt at RSP self-preservation and counterrevolution. 73 With the RSP and former regime elites out of the way, a rush to reform followed, and democratic elections were held. But, for those who had hoped for a clear break with the past, the new president Roch March Kaboré, a former big man in the CDP, showed the continued influence of the old political cadres.
The political transition not only led to an internal political power struggle but also had wider consequences as the "big man deep state" fragmented. Tensions within the army, the fall of Compaoré, the dissolution of the RSP and the removal of former regime elites went hand in hand with a breakdown in how state security had been handled for over two decades. As much as the RSP represented a threat to the democratic transition, its dissolution also meant that the state was bereft of its most capable fighting and intelligence force. The fragmentation of the former "big man deep state" meant that the security forces lost their key managers in Diendéré and Bassolé, greatly weakening and eventually fragmenting the security forces and the state. The gradual promotion during the transition of officers and agents from the "second row" that during Compaoré never had the chance to benefit from EU or US-led military training programs or, more simply, to be engaged in actual security management, could not meet the new urgent needs provoked by a decaying security situation. 74 Not only were "second rows" not prepared to handle the security of the country, but they also lacked the knowledge, skills and resources to handle the many local security initiatives established by the Compaoré "deep state" to deal with banditry in rural communities. Over the years, ethnic-based armed militias started to clash violently, for example Mossi militias clashing frequently with Fulani militias, leading to an increased ethnic polarization. Following a recurring pattern, jihadi entrepreneurs mingled in these tensions, hijacking local grievances and extending their grip over these communities (and their resources), thus far spared from jihadi violence. The Kaborégovernment even accused Compaoré of having a non-aggression pact with jihadist groups -which apparently ended after his ousting. 75 In fact, the Compaoré-regime had maintained tacit agreements and dialogue with jihadist groups and smugglers. Bassolé met Iyad Ag Ghali in 2012 while Compaoré hosted meetings with Ansar Dine senior representatives in Ouagadougou. The senior advisor Chafi was central in developing these ties during the 2000s. 76 It is by no means certain that the security apparatus under Compaoré was significantly scaled to handle such an expansive conflict zone that is seen today. 77 However, we argue that the disintegration of the "big man deep state" and its formal and informal networks of security provisions weakened the state's ability to prevent and handle a full-fledged jihadist insurgency.

The emergence of violent entrepreneurs
After the regime change, the state's monopoly over violence was severely undermined and even challenged by various non-state armed actors. Competing modalities of governance connected to community and ethnic militias emerged out of the fragmentation of the security realm after regime collapse. The informal but institutionalized part of Compaoré's deep state did not necessarily completely disappear but was left in an economic and political vacuum. As "big men" with powerful networks and followings like Dienderé and Bassolé "have been put in a garage" and "in position not to harm [. . .] at least not too much" by the Kaboré-regime, 78 new competing security governance providers over key resources (minerals, lands, waters) have emerged.
As the state's security apparatus became ineffective, it was increasingly replaced by various local security initiatives. While the RSP and the regular army were known for their brutal handling of security, the involvement of ethnic-based and community militias was a real turning point, as these were difficult to hold accountable for their operations against "bandits" and soon pushed the latter to join ranks with jihadi entrepreneurs, now interested in posing as security providers.
The Compaoré regime, already in 2003, issued the Law 32/2003, 79 which allowed local community policing, but only under the supervision of administrative agencies. Vigilante-styled self-defense groups then multiplied, following the model of traditional hunter's associations, such as the Dozos, 80 which often represent the majority of the members.
Under Compaoré, local security initiatives had increased, attempting to fill the void created by intra-army tensions. But, with the waning of the deep state, they have gained more autonomous power. In the Western part of the country, the most prevalent are various local Dozos initiatives, while the Central and Eastern parts are mainly policed by the Koglwéogo (Compaoré & Bojsen 2020). 81 Since 2015 the Koglwéogo play a crucial role in many Eastern territories in response to rising insecurity and banditry, 82 often employing abusive methods. Likewise, the Rougha, Fulani self-defense militias, 83 also emerged as security providers in this fragmented political landscape.
The proliferation of local security forces started in the peripheries of Burkina. However, it was only a question of time before self-defense groups started to emerge also in larger towns. "It had been thought that we did not a need a self-defence group in the city of Tenkedogo. But the thieves hunted in the villages and other cities" [inhabitants] took refuge here without the police and gendarmerie reacting promptly. [. . .] So, we felt the need to set up our selfdefense group'. 84 As such community-based initiatives on security governance have multiplied without real supervision, the new government in 2016 tried to fence them in by what is known as decree 1052. 85 This decree seeks to establish institutional structures and legal boundaries to supervise local self-defense groups, also delimiting their jurisdiction, circumscribing their scope and fixing them to specific territories. Though commendable, such initiatives have come too late and are structurally inadequate to sanction potential breaches, as in villages or semi-nomadic communities, Koglwéogos, Dozos and Roughas are the only armed authorities. Numerically speaking, self-defense groups sometimes outweigh state police forces: in the Boulgou province, for instance, members enlisted in the groups amount to about 12,000. 86 In the entire Center-Est region, they are estimated at 20,000 circa, with an almost capillary presence in virtually every town and village. This shows how the waning of the deep state opened up a new "security market" that different new 'security providers fight violently over. For example, violent conflict between Koglwéogos and Dozos over specific territories or, more urgently, between Koglwéogos and Fulani militias.
While the Decree 1052 should provide, on paper, a formal legitimationand therefore accountability -of self-defense groups, many Koglwéogo local militias refuse to abide by such rule and refer their own internal organization. The Koglwéogo organizational structure mirrors that of the state, with a set of officials, advisers and spokespersons. The latter are supervised by a "general" and his "assistant," while officials are civilians with liaison functions between the provincial and state level. 87 Non-abidance to state laws is not only a channel to bypass state control but is increasingly viewed by self-defense groups and local communities as a necessary mean to ensure community protection and security. "The law is stupid. It should not require thieves to be protected. [. . .] They do not hesitate to kill or wound [and] strip [people] of all their property. They are murdering. How can we give rights to such people?" 88 The tensions during Compaoré's last period in power, the waning of the deep state and the fragmentation of governance in the security realm has been fatal for Burkina. What has happened is a turn from security and stability provision by a deep state orchestrated from the very top of the state to a situation where a new government is attempting, but not succeeding, to reform the security sector and to fence in local security arrangements that undoubtedly engine an escalation of violence.
The sudden fall of Compaoré, and the progressive demise of the "big man deep state" he dominated, strongly shook the security sector in Burkina. Without Compaoré and his key allies in the RSP, the military and the police, security on the ground was progressively left in the hands of community-based militias, whose loyalty (and dependence) to state institutions varies in intensity and degree. This has left Burkina, and its population, with a broken security sector of which jihadi insurgents have proven very apt at taking advantage of.

Conclusion
Our aim with this article has been to move beyond the often diluted and decontextualized versions of the concepts "fragile states" and "neopatrimonialism." This is important if we are to provide better explanations of the strategies employed by weak states to navigate bad neighborhoods. We have therefore shown that the Compaoré-state, while both fragile and clearly neopatrimonial, navigated its bad neighborhood, secured itself and preserved the regime through more than mere informal alliances. It utilized networks and carefully picked clients to the patronage system, but it did this through an institutionalized "big man deep state." It was the combination of the roles of formal organizations as the CDP and the RSP, the mediation "industry" established by the state, and the many informal involvements and arrangements domestically and abroad that spared Burkina Faso from attacks and insurgencies.
The political transition ending with the election of Kaboré following the failed coup d'état saw the remnants of the "big man deep state" led by Compaoré effectively removed. Again, recognizing the shortcomings of the Compaoré-state, we argue that the disintegration of a security system deeply entangled with the former regime weakened the state and paved the way for Burkina Faso's remarkably swift transition into violent conflict a year later. The lack of a more formalized institutional structure ensured that the former modality of governance that monopolized political life fragmented into competing networks, facilitating the proliferation of non-state armed actors as alternative providers of violence and security. Since the transition, the security situation has become increasingly complex and unstable as jihadist insurgent groups have not only taken the fight to the state security forces but also increasingly to its "proxies" and civilians perceived to support the creation of local security initiatives.
The fall of the "big man deep state" does not comprehensively explain the current crisis, however, it does offer an important nuance that complements explanations emphasizing the role of grievances and conflict spill-over. In this case helping us to understand why this led to an outcome where competing informal regimes of power, locally and nationally, set in motion a complex, competitive struggle among current and aspiring Big Men to become the very nodal points in emerging shadow-like semi-hidden informal networks of governance and control. 89 What this suggests is that conducting combined political democratization, economic liberalization and administrative decentralization in a weak state runs the risk of turning the state into an easy prey for a combination of national elites and regional Big Men. 90 While our paper has put some light on how weak states work, we need more and better-situated studies of how weak rulers actually rule, and why some succeed in preserving the rule and their regime, why others fail, and what happens when long-time rulers, as Compaoré, eventually fall.