Internal governance mechanisms exist outside the formal sphere of state apparatus. Although researchers have questioned the African parastatal mechanisms, especially chiefdoms (Fanthorpe 2005), applying this approach to armed movements in Africa is relatively new and increasingly central. Apart from studies of guerrillas, scientific interest on civil wars’ governance dynamics and on proliferation of national and international non-state actors, is relatively new (Branch and Mampilly 2005) because these issues are usually addressed by state-centred scholars. This evolution is materialized by the notion of ‘governance without government’ which refers to internal mechanisms, and interestingly explains that in weak states, there is more opportunity for the arrival on the political scene of non-state (armed) actors (Reno 2002, Bratton & Van de Walle 1994). In other words, the absence of state does not match the lack of governance. This concept analyses the internal actors who perform governance roles in the absence of or against a central political power. Authors interested in conflict dynamics note that these substitutes take the form of armed groups engaged in socio-political activities (Sawyer 2004) which can complement or compete with the state for locales’ control.
Mampilly defines a ‘governance system’ as structures providing public goods but also ad hoc or bureaucratic governance practices, and therefore ‘when the provision of public goods is bureaucratized, we can refer to an insurgent ‘civil administration’- an apparatus distinct from its military organization’ (2011, 4). Tilly on the other hand considers that the advent and expansion of the state is primarily an economic matter. Accumulating resources to secure a locale, ensure capital and seize power involves governance as a necessity, not as the intended purpose. Waging war requiring funding, ‘bandits’ are forced to not only stabilize their sources of accumulation, but also invest more in their territorial conquest projects (Tilly 1985, 172). These two analyses allow to focus on the realities of governance outside state apparatus in the African civil wars with their diverse and complex actors and strategies.
This paper is a critical analysis of ‘rebelocracy’ (Arjona 2014, 1375) in the territories controlled by the Revolutionary United Front (RUF) in Sierra Leone between 1991 and 2002: how to conciliate the notions of ‘governance’ and ‘rebels’, considering that the former implies social order and continuity, and the latter violence and discontinuity from an existing order? The first section outlines the inconsistencies of the RUF governance, focusing on the use of violence as a strategy to gain or maintain authority, the governance structures’ efficiency and the RUF’s role as a political actor. The second part reflects on Tilly’s claim that ‘war makes states’ (1985, 170) and argues that the RUF did not pursue a traditional guerrilla warfare but rather aimed at controlling territories and populations for three reasons: to increase their capital by exploiting resources, to have the monopoly of violence in exchange of the protection of civilians, and finally to gain legitimacy by normalising interactions. The last argument challenges the rhetoric that the RUF war led to a rupture from the former state and to ‘modern governance’. It gives three reasons why the Sierra Leone case should be compared to a colonial state in its early form: the recourse to traditional leaders, the way power was transmitted and the continuity character of its governance.
Rebels, and especially the RUF, represent everything referring to violence, whereas ‘governance’ suggests functioning institutions guaranteeing stability and economic development. On the one hand, the lack of empirical studies and extensive media coverage of the destructive nature of this war and its fighters led to doubt the RUF’s intention to govern. On the other hand, the fighters militarily occupied areas and found themselves forced to govern those territories’ populations. The RUF adopted socio-political structures normalising life because war was not an end in itself. This entails several inconsistencies both in terms of understanding the structures and in analysing the war. The RUF attempted to administer localities it occupied, not only schools, hospitals and agricultural exploitations, but also mixed structures, including both civilians and military, like the War Council (Ayissi & Poulton 2006, 147) which were used to gather information, make decisions, pass judgments, in other words, to govern.
In situations of insurgencies, violence is a necessity for legitimacy but does not rule out the authority and institutional power in the interactions between a guerrilla and a local population (Podder 2014, 221). The society needs order, protection, guidance and guerrillas who remain with the population need its support for recruitment, food etc. Undoubtedly, it becomes less expensive for the guerrillas to regularize these interactions through the establishment of structures managing these issues. An administration imposes itself through interactions and negotiations, therefore governance results from the continued presence among the inhabitants. However, it should be clarified that ‘good governance’ is not implied because violence was at the heart of governance techniques, a central tool in the regulation of social order by the rebels through a wide range of indiscriminate punishment (Wood 2008, 542).
This practice was little capable of organizing behaviour in depth because of the mix between military and civilian rules. Indeed, children went to the RUF schools to learn how to fight as schools were turned into military bases and training centres where child soldiers were training (Leonardi 2007, 398). Social mobility in RUF areas depended on the ability to inflict violence and to inspire fear rather than intelligence or knowledge. The state of youth once seen as a barrier to social mobility became a criterion for promotion (Wood 2008, 550). Finally, the notion of proportionality between crimes and punishments gave way to a more military approach, the RUF trying to recreate a society in the image of a country under a strong authority.
The ‘wartime social order’ (Arjona 2014) assumes that the groups are strategically ‘establishing governance systems that provide collective goods in exchange for civilian consent to rebel rule’ (Mampilly 2011, 8). This is relevant not only to understand the real issues of the Sierra Leone case but also to discuss any civil wars because the similarity is precisely this tendency of many groups to seek control of the central state (Reno 2011, 3) and thus to capture territories where they try to establish governance structures. Mampilly draws attention to the efficiency of armed forces’ governance structures through various case studies in several countries and identifies different levels of effectiveness. For example in Sri Lanka Tamils set up a particularly competent system; in Sudan the SPLM/A have proved moderately effective; whereas in Democratic Republic of Congo (DRC) the Rally for Congolese Democracy-Goma has been an inefficient model. More than the rebels’ ability to govern, their ability to secure a territory and to meet the needs of the population in terms of ‘service provision’ (Mampilly 2011, 112) that is to say health, education, etc., is important. The assessment is no longer about the will or legitimacy to govern of these guerrillas but rather to observe, as in the studies of states, the effectiveness of their governance. In other words, the failure of governance does not mean a lack of will to govern, although the ability to provide public goods does make an armed movement more effective by increasing the population’s loyalty.
The failure of the RUF and its ‘regional warriors’ (Human Rights Watch 2005) was threefold. First it failed to secure controlled territories against the excesses of the combatants in 1991-92, second it did not have the means to respond to the food crises in 1993-95, and third it counter-attacked the ECOMOG intervention in 1998-99 (TRC 2004, 39). The efficiency level of insurgency governance structure has a popular base, therefore the movement’s destructive image, the relative success of the 1996 elections and the new government reinforced the population’s desire for peace although RUF fighters received some popularity at the beginning of the war. This necessity to provide goods, particularly security, is highlighted by Leonardi (2007) who notes that one of the main concerns of rebel recruits is to ensure their and their relatives’ safety. The impossibility of the RUF to change its image contributed to the failure to create a state because adaptation and evolution are key points in the survival of institutions. It was difficult to find trusted people to represent the movement, either at the internal level of the country or at the regional level of the neighbouring states and therefore it was not credible enough. Over time Sankoh lost a lot of credibility and his command style led to doubt that he really controlled his movement (Ayissi & Poulton 2006, 59). Therefore, without massive population support or a remarkably efficient organization, it is by force that the RUF managed to defy a weak government for the control of certain territories in Sierra Leone.
Subnational coexistence of various governance relationships is possible and the level of violence and established order in each of the villages depended on the level of resistance perceived by rebel leaders at their arrival. The fact that ‘cadres had helped villagers find a path to safety before torching their houses’ and that ‘in other instances, victims reported preachy combatants more interested in putting over a political message than fighting’ (Richards 2005a, 389) shows the inconsistent orders because in most cases the RUF simply attacked and burned cities, killed or expelled the population. Richards reports that ‘their concern was to explain to people the nature of their movement and its political message’ (Richards 2005a, 391). Indeed, as the war progressed the RUF understood the need to convince people to follow them and to accept its ‘ideology system’ (Mkandawire 2002, 205) for three purposes: get help from the population, demonstrate that it was a force to be taken seriously, and recruit more members. The strategies modified over time to improve the abilities to win a war (Podder 2014, 219).
Inconsistently, while trying to promote a non-violent image, Sankoh ordered very destructive actions. He had difficulty maintaining a coherent military strategy and often sacrificed civilians’ protection to feed or motivate his combatants. Nevertheless, the RUF was a key political actor, evidenced by the calls for negotiation both nationally and internationally. The Lomé and Abidjan Peace Agreement contributed to establish their political importance by temporarily sharing the state’s role with the government, and indeed Sankoh previously warlord became Vice-President (Sawyer 2004, 449; ICG 2001, 12). Additionally, the international and regional dimensions are considerable. Demarcating territory and population within borders does not guarantee state sovereignty and implies having relations to gain legitimacy. For example, South Sudan needed such international recognition to become legitimate in a territory that had long been under its control (Branch & Mampilly 2005, 6). Without international law protection, such criminal space risk being attacked by the government and its allies, which in the case of Sierra Leone forced the RUF to take refuge in the mountains. Finally, Nigeria’s arrest of Sankoh in 1997 and the 1998 ECOMOG intervention also indicate the regional context’s importance.
In large-scale civil conflicts, rebels are essential in the daily administration of the local societies. In some cases, they even demonstrated better ability to identify the needs of the population than the official government, a fact widely ignored by international state-builders (Podder 2014, 216). For example, in post-1963-secession Uganda the Rwenzuru ‘government’ remained in place for over twenty years, setting up an elaborate governance system and creating an administration capable of meeting its population’s needs (Mampilly 2011). This governance capacity of non-state groups should not be underestimated. Finally, the RUF case shows that the main weakness of guerrilla remains its inability to regularize or normalize its interaction with the civilians whereas its strength lies in the weaknesses of the state it opposes. Therefore, strengthening a weak state through the creation of governance structures may provide a solution in which responsible local actors participate in a local political system that is no longer in the margin or periphery of globalization. Berger and Borer state that data on states’ fragility depends on ‘if a stricter measure of stability is used, and a serious notion of what the delivery of basic rights and services to the majority of a nation-state’s citizens entails’ (2007, 199), which is interesting because the weakness of a state can provide the opportunity to establish proto-states as explained in the next section.
According to Tilly, ‘wars make states’ (1985, 170) whereas for Olson, ‘the rational, self-interested leader of a band of roving bandits is led, as though by an invisible hand, to settle down, wear a crown, and replace anarchy with government’ (1993, 568). Tilly depicts both war-makers and state-makers as organized criminals branding all of them as ‘bandits’ and descrcibes ‘a social contract, an open market where armies, operators and states offered their services to consumers’ (Tilly 1985, 169). Concerning Sierra Leone, without claiming that the root causes were top-down greed rather than bottom-up grievances (Raleigh & De Bruijne 2015), these guerrillas were considered as rational actors, bandits or thieves rather than as combatants motivated simply by ideology. The local structures allowed it to successfully ‘proclaim control over the means of violence’ (Tilly 1985, 170). However, Tilly wrote in 1985 that the examples he used resulted in strong states whereas the RUF was not in the end able to build an appropriate and well-accepted administration by the population as holders of coercive power in social relations. It seems that the Town Commanders (TC) or the War Council were created for the inhabitants to deal with civil matters, as an administration distinct from the military. However, the governance problem resulted first from the militarization of these territories, and second from the increase in recruitment as protection against the brutality of this same movement (Leonardi 2007; Human Rights Watch 2005, 11).
At the stage of state formation, ‘stationary bandits’ (Höijer 2004, 25) generally agree to normalize social relations in order to improve production for their own benefit through the control of material resources and goods. According to Olson, a bandit who ‘takes his theft in the form of regular taxation and at the same time maintains a monopoly on theft in his domain’ will create ‘an incentive to produce’ (1993, 568). Here clearly lie the rational reasons for the RUF to establish local administrations to create what might be called a ‘proto-state’. Nevertheless, the initial and long-term costs of RUF banditry or taxation weighed on the population, which persisted despite Sankoh’s promise that the self-reliant movement would pay back with the population after the first days of occupation (RUF-SL 1995). For example, the RUF set up large farms to cover combatants’ food needs, operated by the local population who, coercively subject to this ‘free’ or forced labour, hoped to continue to live quietly and keep some food for themselves (Peters 2011, 106-107). Moreover, agricultural productions were also taxed which, in a food crisis, was experienced as a heavy and unprofitable burden. In areas far from the administration of the central state, Freetown is only a name, thus to replace an absent central government by a too present movement represented a crucial change in the local life.
Poverty was high before the movement’s arrival but increased because the insurgents’ attempt to exercise permanent control and establish a system of taxation distorted rather than built ties in populations that were not used to having a strong state operating in daily management. Thus, the RUF tried to stabilize its local projects of extractions (agricultural and mining) by involving civilians and particularly important local traditional chiefs (Fanthorpe 2005, 31). By partly demilitarizing its extraction projects, the RUF created autonomous systems, transferring the responsibility of extracting resources to the civilians themselves.
The weakness of this rationalist approach lies firstly in the general shortage from taxation and the disturbance of trading systems between the capital and the countryside, which were attributable to the presence of the RUF and therefore developed popular discontent. Secondly, cutting people’s access to major border markets of neighbouring countries impoverishes agrarian populations (SCSL 2006, 7). The RUF wanted to maintain both military capacity increase and recruitment through taxation and territorial control, men for agricultural production and recruitment needs and women for marriages. This extirpation of the population’s resources led to a monopolization of the means of production, allowed the accumulation of rents by local RUF-chosen businessmen, which greatly improved their capacity to establish local governance structures. Nevertheless this caused popular disillusionment because the RUF had to either ally itself with local traders, or designate some of them as trusted people, or ‘business men’. The local economy’s control and the centralization of the means of extraction and commercialization fuelled this image of an independent and autonomous proto-state as companies were indebted to the state and under its control. The social costs of warlords were coming from their perpetual ‘appetites for military power, international prestige, and larger domains’ at the international level (Olson 1993, 569). Indeed the more territories the RUF controlled, the more it wanted to go on conquering. As a result they spent little time in a new territory before beginning the taking over of another. These high levels of exploitation of economic and human resources show that bandits (in the sense of Tilly and Olson) also seek to protect their territories and people.
For Tilly the term ‘protection’ is used in two opposite senses, namely comfort and disaster (Tilly 1985, 170). From the former, protection refers to images of shelter from the dangers offered by a more powerful actor. On the other hand, for the latter, protection refers to a racket system practiced by a locally strong man, against traders to whom he guarantees protection for financial purposes against the threats of their own ‘war of terror’ (TRC 2004, 40). The difference of degrees is important because the protection of a population by the RUF seems paradoxical. Various instruments were developed by the RUF to protect the populations such as the G5, the War Council and the Internal Defence Unit (Peters 2011, 131-132). However in the name of protection the RUF put an end to the freedom of movement of these populations by imposing a system of passes and safe-conduits, and played a dominant role in the commercial activities of the population as no one had the right to trade without authorisation.
Regarding this dual aspect of protection, Olson adds that control and profits are maximized ‘by selling protection against crimes that it could itself commit and against those that could be committed by others’ (Höijer 2004, 26-27). Consequently, the question of the exploitation of resources is necessarily linked to the notions of protection and monopoly of violence. The RUF needed a permanent and regular source of rent which is why in 1992 the RUF began to fight against its NPFL allies who indiscriminately plundered and destroyed. RUF officials seemingly understood that without protection, they could not secure a population and it became impossible to control territories or resources available.
Nevertheless, the four succeeding governments’ forces were the first enemy of the RUF to prove locals that it alone could protect them. After deconstructing the local state apparatus, local RUF representatives rather than the police or traditional chiefs were responsible for protection matters. These representatives’ legitimacy grew and the bandits progressively became the ‘monopolists of violence’ in a ‘predator-protector model’ (Höijer 2004, 30), the true sources of authority guaranteeing protection. Significantly, after the civil war and under a Disarmament, Demobilisation and Reintegration program, some were able to join political parties and even won local elections as legitimate representatives in the new post-conflict democratic system, although their candidate for the 2002 elections only won 1,7% (Reno 2011, 185).
Tilly claims that the bandit seeks to protect himself from within rather than from external threats therefore ‘the construction of the State produces lasting instruments of surveillance and control within the territory’ (Tilly 1985, 181). Local governance is either based solely on local self-governing methods (such as TCs) or a more hierarchical mode with institutions to maintain a relationship of cordial domination vis-a-vis the inhabitants. Although both modes of governance relied on occasional and selective violence, the warlord’s personal behaviour remained important to build confident relationships. Nevertheless the most common empirical model is a mix of the two when guerrilla forces sometimes entrust the local governance to local elites, as the studies on RENAMO in Mozambique show (Branch & Mampilly 2005). In Sierra Leone, the RUF installed TCs generally from the city in control but serve the RUF’s interests. This undemocratic order is paradoxical for a movement claiming to democratize the country (RUF-SL 1995). In fact, a local leader such as TC proved necessary to manage populations that often belonged to different tribes and spoke different languages.
The formation of states results from continuous interactions and negotiations and requires ‘agreements on protection’ (Tilly 1985, 186) which are bound to evolve continuously. Consequently, a civil war can be an extension of social disputes, a claim to social rights or a criticism of the existing social contract and indeed, African civil wars are generally demands from the bottom for better governance or democratization (Podder 2014). In Sierra Leone, the population was not dissatisfied at first with the presence of the RUF. Indeed it originated in territories under one-party system built along ethnic lines long before the war, as the lower level of abuse shows (Humphrey & Weinstein 2006, 436). Therefore the RUF, in calling for the restoration of democracy and the multiparty system, also used local ethnic issues.
The RUF as a proto-state no longer centred its speeches on what it was going to do, but logically on what it was doing. However, its tribunals, War Councils and other structures only strengthened and served the RUF’s power. Moreover, this movement was not able to break away from this patrimonial logic that characterized the opposing regime, because time and events did not allow to engage in protracted processes of negotiation or renegotiation with the local population, required for the creation of institutions. Instead, the RUF simply applied its laws using selective violence and targeting dissented groups. Even the 1996 mass mutilation campaign had a broad but defined target which was to prevent them from voting (TRC 2004, 44).
Finally, Tilly’s analysis shows that when states start hiring violent fighters especially in a country where the concept of nationality is not necessarily understood, it only increases the level of violence in civil wars (1985, 173). To conclude on this brief incursion into the attempt to create proto-states, it is necessary to address the colonial character of the RUF war as reflected not only in its style of governance but also in the means it has used to go to war.
The notion of modern governance is useful not only to illustrate the will of the RUF to govern but also to question the claim of rupture with the former state and the modern character of its governance, compared to the socio-political structures existing in these societies before its intervention. For Kooiman, it is mainly the increasingly complex and disparate interactions between governors and governed that must be observed. There are several forms of governance, namely ‘self-governance capacities of local systems, co-governing arrangements and hierarchical governing’ (Kooiman 2003). This modern governance, composed of several socio-political actors, private as well as public, implies that this new reflection on governance results from changes and new, diverse and dynamic complexities that have upset modern societies and therefore require a new way of understanding governance. For Kooiman, the ‘modern’ qualifier is linked to the changes experienced by societies during their socio-political developments, including the opening of the political sphere. The state is no longer thought of as a hierarchical, autonomous and independent structure of society. The discourse of modern governance brings to light the complexity of the relations that can exist between rulers and governed.
Utas interestingly argues that the civil wars in Liberia and Sierra Leone were modern wars, and specifies that the leaders used modern means -whether armaments or means of communication- and that the purpose was to take possession of the states described as modern, both in terms of their socio-political formation and their mode of societal reproduction. Based on a political and institutional dimension rather than on tradition, for the author violence expresses aspirations to power and recognition (Utas 2007, 38-39). On the other hand, Hobsbawm specifies the modern context of states in their present form, thus all the present institutions are necessarily modern, the so-called traditional aspects coming from ‘invented traditions’ (Hosbawm & Ranger 1992). Although both Utas and Hosbawm clearly use the term ‘modern governance’ and without denying the modern context, the governance system was rather rudimentary.
Because discourse and realities are different, it seems more significant to assimilate the Sierra Leonean situation with a colonial state, to its early rather than late form, oriented towards developmentalism, because of the following reasons. First, the actors involved in local management were traditional leaders as well as some civilians gathered in War Council. It was even led by a civilian who worked with several representatives in the district in order to improve the interactions and thus facilitate the governance. The RUF wanted to demonstrate its break with traditional methods of local governance which, in the absence of the State, was often seen as the responsibility of traditional leaders. However, these representatives were part of the RUF ‘network of control’ (Raleigh & De Bruijne 2015, 9) rather than elected, and thus served more as the right hand of the RUF to tax the population. This post-colonial state with a single party also tended to reduce the costs of governance, hence the interest of choosing civilians who could speak the local language within its War Council to consolidate its local presence.
The second reason is that governance by the War Council or the TC was badly perceived and received by the former local authority representatives, i.e. the traditional chiefs, because the RUF only transferred power from one chief to another. In some cases, the RUF’s appointed agents were even more powerful because they possessed powers that previous leaders did not have (Fanthorpe 2005). Moreover, this tense relationship changed over time as shows the increasing use of diamonds to finance the war (TRC 2004, 107). With the first military defeats, it was forced to rethink its strategies, including encouraging traditional leaders to stay in cities to inform the people that the RUF was a people’s movement. Nevertheless, most of the violence committed against the leaders was taking advantage of the chaos opportunity to take personal revenge on a leader. Finally, the colonial state as well as the post-colonial state often appealed to the chiefs, those who refused to collaborate were replaced. Certainly, the movement had difficulty convincing the chiefs to remain in these territories because of the perception of the enduring violence. However, the RUF still needed local district chiefs to convince their populations to contribute voluntarily to national reconstruction, for instance by working in large-scale farming projects or mines (Bangura 2000, 572).
The third reason is that apart from the RUF rhetoric to fight the practices of the past (RUF-SL 1995), the democratisation of the country seemed badly engaged. In reality, the movement did not break away from traditional practices, defined as ‘normally governed by overtly or tacitly accepted rules and of a ritual or symbolic nature, which seek to inculcate certain values and norms of behaviour by repetition, which automatically implies continuity with the past’ (Hobsbawm 1992, 1). Hobsbawm notes that revolutions and progressive movements are by definition intended to break with ties of the past. Richards explains that these wars have the ability to reconstruct because the ‘focus upon ritual aspects is as applicable to the analysis of war fought with sophisticated modern weapons as with knives and cutlasses’ (2005a, 378). He adds that one of the key factors was the RUF’s desire to create a new society, to emancipate itself from the past and to offer a ‘generational inversion of authority’ (Wood 2008, 550). However, its observable way of governing was not modern as shows the patrimonial nature of the governance system in RUF territories, the privatization of public goods, the corruption by RUF executives, and Business Contractors. Finally, the previous regime type has influenced rather strongly the following ones, generated by demands ‘from below’ (Bratton & Van de Walle 1994). In cases of real regime change, African countries have instead evolved into a neo-patrimonial logic. Therefore, given the situation of RUF territories this movement could have evolved towards a political system like that described by Bratton and Van de Walle.
In conclusion, this study analysed the interactions between rulers and ruled and demonstrated that these governance relationships are not only intended to ensure the continuity of the war but also to govern the local population. Thus the existence of interactions, negotiations and other forms of participation of civilians in structures established by the RUF (such as the War Council, the G5 or the Internal Defence Unit) have governance for purpose. However intentions do not always correspond to the practices or results and the effectiveness of these structures was questioned without denying neither their existence nor their importance locally.
The RUF did not in the end elaborate a state, meant to last and be stable because its control over territories did not last long enough. The theories of Tilly and Olson on state-formation are useful to analyse rebel-held territories and to explain how and why rebels created proto-states complemented by governance structures and mechanisms. There is no consensus about the exact intentions of the RUF and some even argue that it had ‘no meaningful political constituency’ (ICG 2001, ii). Nevertheless it still conquered territories, controlled populations, and monopolized violence which gradually became legitimate through the establishment of governance structures and institutions of socio-political regulations.
The previous focus on non-state actors inevitably leads to address the issue of the state as an actor in civil wars. Indeed in many African countries, absents or inoperative states lead to the multiplication or strengthening of various governance options on which local populations rely. Studies on the relationship between the presence of guerrillas and states’ weakness show that non-state guerrilla governance is the corollary of ‘collapsed states’ (Reno 2002, Berger & Borer 2007). However, this state-centric approach ignores several relevant aspects of the rise of guerrillas, parameters described in detail by Richards (2005b). Other analyses specific to the Sierra Leone state (Fanthorpe 2005; Bangura 2000) demonstrate the causes and effects of its collapse. The RUF had a plan to govern because it undeniably did try to build schools, health centres or to set up specific structures to manage relations with civilians, fight corruption etc.
Finally the semi-urban character of this armed movement shows that despite military setbacks, this guerrilla always had an ‘urban agenda’ (Mkandawire 2002, 207). Therefore, with a strong Sierra Leonean state, the cities would have remained under the control of Freetown central power. Generally in guerrilla warfare the rebels control or park in the bush, forest or countryside and the regular army, the state instrument, controls the cities, but the RUF did control some urban areas. The fact that a force that had enough fighters to steal and loot chose to control vast rural towns, areas, territory and population (Richards 2005b) shows that the interactions between the RUF and the inhabitants ought to be better understood.
Arjona, Ana. ‘Wartime Institutions’. Journal of Conflict Resolution 58 (2014): 13601389.
Ayissi, Anatole and Poulton, Robin. Bound to Cooperate: Conflict, Peace and People in Sierra Leone. United Nations Publications UNIDIR, 2006.
Bangura, Yusuf. ‘Strategic Policy Failure and Governance in Sierra Leone’. The Journal of Modern African Studies 38 (2000): 551-577.
Berger, Mark T. and Borer, Douglas A. ‘The Long War: Insurgency, Counterinsurgency and Collapsing States’. Third World Quarterly 28 (2007): 197-215.
Branch, Adam and Mampilly, Zachariah. ‘Winning the War, but Losing the Peace? The Dilemma of SPLM/A Civil Administration and the Tasks Ahead’. The Journal of Modern African Studies 43 (2005): 1-20.
Bratton, Michael and Van de Walle, Nicolas. ‘Neopatrimonial Regimes and Political Transitions in Africa’. World Politics 46 (1994): 453-489.
Fanthorpe, Richard. ‘On the Limits of Liberal Peace: Chiefs and Democratic Decentralization in Post-War Sierra Leone’. African Affairs 105 (2005): 27-49.
Hobsbawn, Eric, and Ranger, Terence. The Invention of Tradition, Cambridge: Cambridge University Press, 1992.
Höijer, R. ‘Theft by Bandits and Taxation by Kings: A critique of Mancur Olson on State Formation’. Political Studies Review 2 (2004): 24-38.
Human Rights Watch. 2005. ‘Youth Poverty and Blood: The Legacy of West Africa’s Regional Warriors’. Available at: http://www.hrw.org/en/node/11795/section/1
Humphreys, Macartan and Weinstein, Jeremy. ‘Handling and Manhandling Civilians in Civil War’. American Political Science Review 100 (2006): 429-447.
International Crisis Group (ICG). ‘Sierra Leone: Time for a New Military and Political Strategy’. Freetown/Brussels/London: International Crisis Group, 2001.
Kooiman, J. Governing as Governance. London: Sage Publications, 2003.
Leonardi, Cherry. ‘Liberation or Capture: Youth in between ‘Hakuma’, and ‘Home’ during Civil War and its Aftermath in Southern Sudan’. African Affairs 106 (2007): 391-412.
Mampilly, Zachariah. Rebel Rulers, Insurgent Governance and Civilian Life during War. Cornell University Press, 2011.
Mkandawire, Thandika. ‘The terrible toll of post-colonial ‘rebel movements’ in Africa: towards an explanation of the violence against the peasantry’. Journal of Modern African Studies 40 (2002): 181-215.
Olson Mancur. ‘Dictatorship, Democracy and Development’. The American Political Science Review 87 (1993): 567-576.
Peters, Krijn. War and the Crisis of Youth in Sierra Leone. Cambridge University Press, 2011.
Podder, Sukanya. ‘Mainstreaming the non-state in bottom-up state-building: linkages between rebel governance and post-conflict legitimacy’. Conflict, Security & Development 14 (2014): 213-243.
Raleigh, Clionadh & De Bruijne, Kars. ‘Where Rebels Dare to Tread: A Study of Conflict Geography and Co-option of Local Power in Sierra Leone’. Journal of Conflict Resolution (2015): 1-31.
Reno, William. ‘The Politics of Insurgency in Collapsing States’. Development & Change 33 (2002): 837-858.
Reno, William. Warfare in Independent Africa, Cambridge University Press, 2011.
Richards, Paul. ‘West-African Warscapes: War as Smoke and Mirrors: Sierra Leone 1991-2, 1994-5, 1995-6’. Anthropological Quarterly 78 (2005a): 377-402.
Richards, Paul. ‘To Fight or to Farm? Agrarian Dimensions of the Mano River Conflicts (Liberia and Sierra Leone)’. African Affairs 104 (2005b): 571-590.
RUF-SL, 1995. ‘Footpaths to Democracy: Towards a New Sierra Leone’. Available at: http://www.fas.org/irp/world/para/docs/footpaths.htm
Sawyer, Amos. ‘Violent conflicts and governance challenges in West Africa: the case of the Mano River basin area’. Journal of Modern African Studies 42 (2004): 437-463.
SCSL (Special Court for Sierra Leone) RUF Trial Transcripts 3 April 2006. Available at: http://www.rscsl.org/Documents/Transcripts/RUF/RUF-040306.pdf
Tilly, Charles. ‘War Making and State Making as Organized Crime’. In Bringing the State Back In, edited by Peter Evans, Dietrich Rueschemeyer, and Theda Skocpol, 169-186. Cambridge University Press, Cambridge, 1985.
Truth and Reconciliation Commission of Sierra Leone (TRC), 2004. ‘Final Report’. Available at: http://www.sierraleonetrc.org/index.php/view-the-final-report/table-of-contents
Utas, M. ‘Fighting (in) the Badlands of Modernity: Youth Combatants and the Liberian Civil War’. Journal of Social Sciences & Management 1 (2007) 37-44.
Wood, Elizabeth. ‘The Social Processes of Civil War: The Wartime Transformation of Social Networks’. Annual Review of Political Science 11 (2008): 539-561.