Centre for African Studies (LUCAS)

Centre for African Studies
University of Leeds
Leeds LS2 9JT

Tel: 0113 343 5069
Fax: 0113 343 4400

LUCAS Schools Project coordinator

Richard Borowski

Ethnicity Kills? – Einar Braathen, Morten Bøås & Gjermund Sæther

Tagged with the keywords: , , , , , , , , , , , , , ,

[Published in Leeds African Studies Bulletin 63 (2000), pp. 13-25]

Ethnicity Kills? The Political Economy of War in Africa

Einar Braathen, Morten Bøås & Gjermund Sæther

[Paper presented to The LUCAS and The Review of African Political Economy (ROAPE) Millennium Conference, “Africa – Capturing the Future”, April 28-30, 2000, University of Leeds. This paper is a revised version of a contribution to a forthcoming book: Braathen, Bøås and Sæther: ”Ethnicity Kills? Social Struggles for Power, Resources and Identities in the Neopatrimonial State”. In  Braathen,Bøås, & Sæther (eds.): Ethnicity Kills? The Politics of War, Peace and Ethnicity in Sub-Saharan Africa.Macmillan:London (2000)]


The two main ideas behind this paper is to shed critical light on the reduction of African civil wars to ethnic conflicts, and to argue for the emergence of civil wars as the result of political struggles over power, resources and identities. In part one we review some influential pieces of academic work and criticise certain discourses. The construction of Africa as the “other” has entailed that factors commonly used to explain war elsewhere have been neglected in accounts of African civil wars. The writing on war in Africa has to a large extent been trapped within a mind-set developed when the world was constituted by ‘civilised’ and ‘uncivilised’ peoples. In part two we present our own alternative to the ethnic paradigm: A lead thread for doing research on Africa’s civil wars should be the struggle over distribution in society that evolve around the post-colonial state. More precisely, we put forward a thesis whose core is built on the notion of neo-patrimonialism and turning points of the neo-patrimonial state. In part three we discuss the conditions for a post-patrimonial state and transition from war to lasting peace. Our intentions are to present a truly comparative approach that take ethnicity seriously without being one-dimensional or Euro-centric.

Introduction: Ethnicity kills?

In almost all recent dramatic events in Sub-Sahara Africa issues of ethnicity and contested identities seem to be at the heart of the matter. But can really the crisis in countries like Angola, Congo-Brazzaville, Democratic Republic of Congo,  Liberia, Mozambique, Rwanda, Sierra Leone and Somalia be reduced to ethnic cleavages?

Obviously, ethnicity plays a role in most conflicts in Sub-Sahara Africa, in the sense that ethnic affiliations often structure the composition of groups in conflict. Moreover, there is little doubt that one of the main reasons why people kill each other is because of who they are and the identities they represent. We are all to some degree still tied to the identities around which ethnic and national conflicts are fought. The power that binds us to these identities through a process that Foucault (1982) refers to as assujettissement, or subjectification, still operates. This is a process where two meanings of the word “subject” become socially constructed  truths: subject to someone else by control and dependence, and tied to his/her own identity by a conscience or self-knowledge. “Both meanings suggest a form of power which subjugates and makes subject to” (Foucault 1982:212).[1] However, it is possible, and we would even argue necessary, to accept the presumption that most civil wars in Africa are centred around ethnic identities as the subjectification of power, and still question whether this in itself can explain the causes behind violence. Perhaps, it would be more fruitful to search for the basic causes for the conflicts? Should we not focus more upon politics and economics: on how the struggle for power and resources on marginal sites is turned into ethnic conflict as the damnation game of these struggles turn more and more violent?[2] The main assumption in this volume is that even though ethnic affiliations often structure the constitution of armed factions it is too simplistic to characterise war in Africa as tribal conflict. On the contrary, the various conflicting groups and armed factions must be understood in light of the socio-economic context in which they operate, and within this context ethnicity is just one among many variables. Ethnicity is not an “essentialist” attribute of Africans, or Europeans for that matter, but just one of several identities. Ethnicity therefore cannot be discussed outside of its precise historical and geographical context (Chabal & Daloz 1999). Thus, the significance of ethnicity is a function of the circumstances under which it suddenly becomes salient. By focusing on the context of ethnicity we also hope to shed some critical light on the simplifications made inside and outside academia in accounting for civil wars in the region.

Part One: Stereotypes and Trends in Academia

Our point of departure is that the subject of violence in contemporary sub-Saharan Africa must be approached with caution to steer clear of deep-rooted stereotype images of Africa. During colonisation the distance between the Europeans and Africans was confirmed and made scientific through the anthropological gospel of cultural evolutionism. Through the language of “science” a type of scientific primitivism was constructed (Brantlinger 1986:206). Its main representation was the dichotomy between the “modern dynamic nations of Europe” and the “traditional stagnated African tribes.” This gospel is no longer an explicit part of the discourse, but tribal and ethnic identity is still considered as one of Africa’s main problems. All too often it is either seen as a leftover from a by-gone age and a barrier to modernisation or just as a destructive political weapon in the hands of political leaders. However, ethnicity is neither a primordial carryover nor a result of a modern conspiracy. Every ethnicity has a social history, and it is in continous making and remaking (Mamdani 1996:185). Thus, it should not be used as a static concept. Rather, ethnicity is one part of a complex set of dynamic and interactive identities. We all carry with us a flexible set of identities and the ways in which we define ourselves and others are in accordance with sets of beliefs, values and perceptions that are susceptible to change over time. The important issue is therefore not the notion of ethnicity, but why and how it is instrumentalised politically (Chabal & Daloz 1999). The answer to this question is embedded in and conditioned by historical circumstances and processes of political and economic transformation. Subsequently, we will argue that behind every war and political conflict in Africa labelled as tribal or ethnic, there is a complex set of interactive causes.

In his impressive and influential work “Ethnic Groups in Conflict”, Horowitz (1985) puts ethnicity at the centre stage of violent conflict. Drawing on an immense source of literature Horowitz was seemingly capable of showing that other factors evolved around ethnicity. The main problem with “Ethnic Groups in Conflict” was not that “ethnicity” was given an important role, but rather that political, historical and economic factors were degraded to a secondary role as they were seen through the lenses of ethnicity. Horowitz (1985:12) asserts that: “In divided societies, ethnic conflict is at the centre of politics (…). Ethnic conflict strains the bonds that sustain civility and is often seen as the root of violence (…).”[3] On the issue of identity and ethnicity Horowitz’ conclusions are basically that while these two can be separated from each other in the West, they cannot in Africa and Asia. Horowitz’ work is thus a true reflection of the colonial discourse in which identity is reduced to ethnicity. With Horowitz’ work a new standard for approaching African civil wars in which ethnicity seemingly was the root of all evils and thus capable of explaining crisis and conflict had been set. Osaghae (1991) is one example of a typical follower of Horowitz as he sees cultural and linguistic factors connected to ethnicity as possible causes for conflict in their own right. [4]

This kind of simplified approach is striking in the Somali case. Based on the writings of the British Anthropologist Ioan Lewis a number of social scientists have emerged in western media explaining the civil war as a clan conflict in which people simply kill each other because their opponent belong to another clan. Moreover, it is asserted that the difference between present day fighting and yesterday’s warfare is basically the AK-47. This view leaves out historical, political and socio-economic factors. Why is the outbreak of wars in the western hemisphere seen as a result of a number of interactive factors while the outbreak of the Somali civil war is seen as a result of Somali culture, something that lies in the “Blood and Bone” to quote a title from Lewis (Lewis 1994)? As in every war there are also in the Somali case distinct aspects, but this cannot explain why the tribal arguments constitute a paradigm in one case while it is left out in others (see Sæther 2000). Wars signify a struggle over the distribution of power, wealth and the representation of identities everywhere. So why this difference in approach?

Deconstructing the stereotypes

Perhaps the differences between approaches to civil wars in Europe and in Africa stem from the fact that we are still trapped within a mind-set developed when the world was constituted by civilised peoples and uncivilised (partially civilised ones). The first group was organised in states – the main characteristic of their civilised status, whereas the other group were scattered in far-off continents such as Africa. The people in the first group, or rather through their main organising principle, the state, ruled the latter group. Their rule was secured by coercive means, but it was also underscored “by narratives of improvement, of the civilising mission and the white man’s burden, which were secured in systems of knowledge which made sense of these narratives, and were, in turn, informed and shaped by them” (Seth, Gandi & Dutton 1998:7).

The West was the subject of history, the colonised appeared as passive in these narratives. This dualism was institutionalised by law. Civil law claimed to speak a universal language, that of rights, but it excluded natives on the grounds they were creatures of habit who needed to be ruled through a regime that would enforce custom, by customary law. While civic power was racialised, the native authority was tribalised. In this way, “bifurcated” states were created  (Mamdani 1996). Subsequently, African societies were neatly divided into different tribes through a Euroamerican system of classification that oversimplified and distorted ethnographic observation (d’Azevedo 1989:100). The ethnic marking of African communities thus reflected the peripheral nature of the powerless (Lema 1993:172). African societies became the subjects of colonial “science” with respect to who they were and how they acted. In the European historical context, Foucault called  processes which transform human beings in this way for objectification (Foucault 1982:208). “The exotic Other always comes out of this operation [the constitution of the Other] as an oddity lacking something – rationality, control, decorum, propriety – and exceeding in something else – violence, sensuality, passion” (Savigliano 1995:81)[5]. This does not entail that there was not any relation between the empirical data and the colonial categorisation of communities. Group identities existed. But the concepts used were certainly not “neutral” and even if the phenomena of group identities was poorly understood the possibility to use them at the service of the colonial power was quickly perceived.

Britain was the first to marshal authoritarian possibilities in native culture, by breeding  “native authority.” Its all-embracing creation had three notable consequences: First, the colonial administration did not use one customary law for all natives, but roughly as many sets of customary laws as there were said to be tribes. Second, although there in each “tribe” in the late-nineteenth-century context was several traditions, the colonial powers privileged the one’s with monarchical, authoritarian, and patriarchal notions of the customary. Third, the colonial rural state was marked by force to an unusual degree; the authority of the “traditional” chief fused in a single person all elements of power: judicial, legislative, executive, and administrative. The colonial power stripped the “native authority” for all the checks and balances that used to surround it (Mamdani 1996:22-23). Another implication of the colonial mapping of societies is the need to underline the modernity of what appear as ethnic groups today. Ethnicity were in many cases a response to colonial administration (e.g. d’Azevado 1989; Vail 1989; Mamdani 1996; Mafeje 1997). However, it is also clear that in other cases one can indeed trace the identity of communities far backwards. In areas of e.g.  Nigeria, Great Lakes Region, Liberia, Sierra Leone and Sudan it is evident that ethnicity has played a complicated role in a long history of cross-community relations. Only a close scrutiny of the history of what today appear as a ethnic group might provide the key to understand the status and workings of ethnicity in any given conflict.

However, it is worth noting that voices from the Francophone academic community diverged from the mainly Anglophone trend described above. Mbembe (1990:8-9) points out how the accounts of war and violence based on the role of opposing community affiliations have been unable to explain earlier phases of equilibrium produced by the very same cultures which now are seen as the root cause of all-out and destructive wars. Consequently, some types of changes must have taken place compared to earlier warfare. Thus, what would have been evident factors in explaining civil wars elsewhere has too often been neglected when it comes to Sub-Sahara Africa. In a pathbreaking book on the causes of  the armed conflict in Mozambique, Geffray (1990) pursues an “anthropologie d’une guerre civil” based on an analysis of how the actions of the post-colonial state increased latent cleavages within rural social structures. A social basis for armed conflict came out of this process, which could not be conceptualised in ethnic terms, although the emerging civil war created sub-ethnic dividing lines as a result of the rivalry for support from the chiefs and other traditional leaders by the warring parts.

We agree with Chabal (1992) and Bayart (1993) that the process of understanding politics in Sub-Saharan Africa, is conceptually and in practice similar to the process of understanding politics elsewhere. Ethnicity is an important factor in many civil wars, but it must be put into its proper political, historical and economic context. Wars has a social setting and the contemporary setting is clearly different from the pre-colonial setting. “Far from being a secretion of primitive societies, it is instead around the power game on the arena of modernity that the political competition intensifies and where it [the competition] triggers off the clashes between groups with different origin or obedience” (Lemarchand 1991:202).[6] The key to understanding of this “arena of modernity” is the state and the fight to gain control of state resources, power and possibilities. It is our view that in the end civil wars are the results of political conflicts in which the struggle over distribution is fundamental. As such, the main focus for analyses should be the interplay between structural factors (mainly the political economy) and identities which are mutually constituted around the marginal site of the African state. In short, our approach is to focus on the object of fighting – the post-colonial state.

Part Two: Neo-patrimonialism as an Explanatory Framework

The number one formula when attempting to write on civil wars in contemporary Sub-Saharan Africa must be “to see violence within its social setting, to appreciate its roots in social conflict, and to understand how and why people turn to it” (Crummey 1986:3). In developing a framework to account for civil wars we will argue for the importance of a state-centred approach, but not a top-down approach. We think in line with Bayart (1993) that in order to understand the nature of the power-game played out over control of the post-colonial state we have to take on a deliberate “politics from below” perspective, that is the political behaviour of the dominated actors; reintroducing in this way the historical dynamics of African societies. However, in order to grasp the dynamics at play in the post-colonial state, such an approach have to be combined with an approach that is deliberately oriented toward the top, but in close interaction with the bottom. In Médard’s writing on neo-patrimonialism, in Migdal’s work on state power and social forces and in Mamdani’s inquiry of citizens and subjects in Africa we find building blocks for such an approach (see Médard 1982; 1996; Migdal 1988, Mamdani 1996). The very basic point we would like to draw attention to is that there are permanent tensions and a recursive relationship between state and society. There are a multitude of arenas for potential conflict, and the state (whether cleptocratic, developmental, neo-patrimonial or predatory), penetrates into people’s daily life. Most often, there is resistance to this penetration expressed by counter-elites in the periphery or on the marginal sites. In Mamdani’s perspective, the penetration into the rural or “tribal” areas started with the colonial institution of indirect rule, and in the post-colonial era it escalated under radical regimes who replaced “reactionary” native authorities with direct rule (Mamdani, 1996:26-27).

Neo-patrimonialism is a mixed type combining in various degrees differentiation and lack of separation between public and private spheres. From a structural point of view, the African state is differentiated, but weakly so from a functional perspective. In neo-patrimonial societies like the post-colonial African state, bureaucratic and patrimonial norms coexist. Although this kind of state is far from what it pretends to be, it is able to extract and redistribute resources, but this extraction and redistribution is privatised. In redressing the colonial legacy of racially inherited privilege, the independent states created a specific patrimonial path of redistribution  which divided the indigenous majority along regional, religious, ethnic and at times familial lines.   This privatisation of the public has three consequences. First, political-administrative power, instead of having the impersonal and abstract character of legal-rational domination specific to the modern state, is a personal power. Second, politics becomes a kind of business, because it is political resources that give access to economic resources. Third, mass politics are structured around vertical clientelistic relationships.[7] In sum, the advantage we see in the use of the notion of neo-patrimonialism is that it subsumes many different social and political practices that we observe in Africa, and which have in common the confusion between the public and the private.

It is precisely the personalisation of power combined with the “businessification” of politics that offers rulers and warlords the opportunity to create power images of the other and to ask for the securing of the self. We would therefore like to point out that in sub-Saharan Africa as elsewhere there is much to be found in Clausewitz thesis that war is the continuation of politics with different means. With a less rationalistic approach, we could say that politics in conditions of democratic-civic decay  tend to be reduced to armed politics.

However, this angle can not alone explain civil wars. No monocausal analysis can grasp the complexity of the factors behind such tragic events as civil wars. Nevertheless, approaches can be carved out that is much more grounded in real-life complexities than the “ethnic” approaches of Horowitz and his followers. Our guiding question should be: Under which circumstances do neo-patrimonial politics become both ethnicised and militarised ?

The turning points of the neo-patrimonial state

When accumulation is closely tied to politics, a dynamic of violence may occur due to repression in combination with exclusion of some groups from the state, and thereby accumulation. When the elites in power feel that they are being subverted from within by their own politics of exclusion which narrow their support base, violence becomes a tool of repression on behalf of the state. The structuring of state violence is above all a reflection of the wish to keep power and thereby possibilities for accumulation. These contests on the arena of the state can thus provoke violent confrontation between groups with different locality and/or community affiliations. In particular, these situations tend to emerge when the neo-patrimonial state is forced to retreat due to lack of financial resources. In Sierra Leone, for instance, long term patterns of accumulation of forest and mineral resources have fed a political economy dominated by patrimonial distribution (Reno 1995). The political elite builds support through distributing resources on a personal basis to followers, whereas relatively few resources are distributed in accordance with principles of bureaucratic rationality and accountability (Richards 1996). Then in the 1980s, at least partially due to Structural Adjustment Programmes (SAPs) and tightening of overseas aid budgets, patrimonial redistribution in Sierra Leone went into sharp decline. Thus, due to resource shortages the ability of the state to control its peripheral regions was weakened, and the loyalty, of in particular the younger generation, was tested. Subsequently, the retreat of the Sierra Leone neo-patrimonial state from the Liberian border region allowed dissidents to enter the country from Liberia and deploy methods of violent social destabilisation that they had learnt during the Liberian civil war (Richards 1996). The war in Sierra Leone is one obvious case, but also experience from other war-zones in Africa suggest that the cause of conflict is not ethnicity, but the expansion and then retreat of a neo-patrimonial state, followed by widespread social exclusion. Our suggestion is therefore, not only to focus on the obvious source of fighting – the post-colonial state, but more precisely to emphasise two central turning points of the trajectory of the post-colonial state.

The first turning point is the halt of the expansion of the neo-patrimonial state. In accordance with Médard (1982; 1996), we argue that patrimonial logic is an integrated part of the state, it is present and co-exists with other kind of logics such as the legal-rational bureaucratic logic or  the capitalist logic. A combined social and political crisis emerges when balance is lost, when  the patrimonial logic becomes (i) too dominant and/or (ii) looses its integrating/institutionalising/legitimating aspects and mechanisms. According to Bayart (1993), this happens when the reciprocal assimilation of elites has been exhausted. Due to various forms of economic hardship the neo-patrimonial state is forced to retreat. In such instances, reform that emphasises economic and political liberalisation will only further decrease the incentives of those who control the state to pursue conventional strategies for maximising power through generating economic growth and state revenues. Rather, as the state retreats, new opportunities become available both to individual officials and other strongmen whose interests often run counter to that of the ruler (Reno 1998). The state becomes criminalised, or increasingly permeated by sets of practices which contravene national and international legal codes and morals (Bayart, Ellis & Hibou 1999)[8] At this moment, the capacity to solve the political crisis is crucial. If civic order  is on the brink of collapse, then the  time factor is critical, and the demand/supply of non-violent conflict-solving mechanisms is decisive. If  the major leaders are unwilling to agree consensually on emergency solutions, or unable to stick to ’national’ agreements and give peace a chance, the stage is set for the damnation game with various strongmen pitted against each other.

The second turning point occurs when these socio-economic and political struggles get militarised.  This turning point is made up of several interlinked events, like when (i) the struggles are translated into new frontlines (on marginal sites), (ii) new or old frontlines are guarded by armed people on both (or all) sides, and (iii) the shooting and killing begins. It is at the second turning point that conflict is socialised by competing elites and counter elites (potential warlords) that mobilise and recruit the henchmen that are willing to take up arms. It is only then, when the stage is set for the damnation game that ethnic identities are mobilised and play is crucial role in civil wars. However, this stage will carry over some of the structural characteristics and effects of the more or less collapsed state. Here, one might observe inscribed effects of regime differences. First, neo-patrimonial states vary according to their degrees of social inclusiveness. Bratton and van der Walle (1994, 1997) show that the more inclusive a regime, the more capable to solve conflicts peacefully. Second, neo-patrimonial states vary according to their degrees of centralisation and institutionalisation of radical measures to introduce direct state rule in the rural areas. Politically, conservative regimes like Mobutu’s Zaire have been more decentralised than their radical adversaries. “What holds Congo together is not as much the civic power in Kinshasa and Kisangani, and so on, but the hundreds of native authorities that control the bulk of the population in the name of enforcing custom” (Mamdani 1998). This opens the national stage for local despots with ethnic recruits. Economically, there are different regimes of land tenure. In times of economic hardship and/or collapse in formal economy, access to self-subsistence land becomes a matter of survival. However, under conservative regimes this access has depended on membership in an ethnic community, by virtue of customary tenure. Customary tenure causes a division between the peasant in the customary home and the migrant (stranger) peasant (Mamdani 1996:183-4). “This was seen in Uganda and later in Congo, where migrants from Rwanda caused local social conflicts escalating into political “ethnic” conflicts.” This is “a dilemma that arises wherever there are substantial numbers of immigrants and where the state inherited from colonialism makes a structural distinction between two kinds of citizens: Those indigenous, and those not” (Mamdani 1998).

Rwanda and Mozambique represent two different types of civil war. The different outcomes of conflicts might be explained by the prior regime characteristics. In Rwanda, social structural conflicts escalated into ethnic war and genocide. The prior conservative regime had cultivated ethnic rights and regulated customary tenure; the conflict became largely rurally based and inter-ethnic (See Lema 2000). In Mozambique, ethnic citizenship and other “tribal” institutions were suppressed. Customary or other exclusive land rights had been effectively downplayed by the radical Frelimo regime. On the other hand, the regime exacerbated the rural-urban divide.  For the rebel movement, Renamo, the civil war was a struggle for civic citizenship and political changes, not for land or ethnic privileges. With the Frelimo government embedded in corporativistic or inclusive ideology of governance, a negotiated political settlement of the war was within reach. Although both parties employed certain ethnic cards in their struggle, the conflict could not be defined as an inter-ethnic conflict (see Cahen 2000).

The interplay with external factors

We therefore have to ground our analyses in the localised aspects of Sub-Saharan Africa’s civil wars, but simultaneously we have to take into consideration that both states and rebel groups are located within a wider setting of regional and global politics (see Clapham 1998). Many of the civil wars in Sub-Saharan Africa have been influenced by outside forces, be it other African states, the US, France or other European countries. The end of the cold war and the concomitant reduced geopolitical significance of Africa shows that also the lack of interference have an effect on the ground. Barre could still have been the leader of the Somali state if the cold war had continued.

In addition, political, historical, cultural and economic ties between foreign actors and groups in conflict are factors that must be taken into account. This includes the overall reduction of the status and influence of the state as a privileged actor within its territory (Badie 1997). According to Clapham (1996:222), the retreat of the state is, “in a sense withdrawing parts of Africa from the formal scope of international politics, but in the process creating a new international relations of statelessness.” More precisely, we will argue that the relationship between rebel groups/war-lords/state rulers and foreign and regional firms are of vital importance if are to understand the dynamics of war in Africa.[9] It is impossible to understand the dynamics of the Liberian war if we do not take into account the relationship between the various warlords and foreign firms, and it is impossible to understand the war in the Democratic Republic of Congo (former Zaire) if we do not understand the relationship between local, regional and international mining firms and the various factions (see Reno 1993; 1998; African Business May 1998). Likewise, armed intervention by governments in neighbouring countries and other regional power wielders contribute to both militarization and prolongation of conflicts, as depicted by Morten Bøås in this volume. Moreover, we need to come to terms with how the foreign policy agenda of powerful external players like France and the US affect the on-going conflicts, and how humanitarian intervention and humanitarian aid affect an ongoing war.

The support of outside forces to groups in conflict might take various forms such as “neutral” humanitarian aid, soldiers, weapons, the import of commodities like timber or diamonds etc. The support is not only important militarily, but it is also a source of accumulation for parties to the conflict. In this manner outside backing can often contribute to entrenchment of those interests who benefit economically from a civil war. The patron-client relationship between actors in civil war and external actors should not be underestimated. Kabila’s march to power, and subsequent battle to retain it is an excellent illustration of this point. In fact, we will argue that the lack of trust and of shared projects between the state holding group and colonial/neo-colonial actors, in particular in the field of industrial and economic transformation, add crucial factors to our understanding of what has been coined by Zartman (1995) as state collapse in Sub-Saharan Africa. The strong external linkages of the governments of Angola and Mozambique through investment/donor-projects to Portugal and the European Union who could mobilise the UN on behalf of their client governments, can at least partially explain why these countries avoided the fate of Liberia and Somalia. Nevertheless, despite the importance of outside support, it is still the state and the struggle for economic resources on the state arena which is at the centre of conflict, and subsequently should constitute the centrepiece of our analyses.

Part Three: Towards a Post-patrimonial State: transition from war to peace

Many, both with a general focus and concerned with Africa only argue for the weakened role of the state in the age of globalisation.[10] There is little doubt that states everywhere are facing new challenges and their regulatory role in society has decreased. In the introduction we emphasised the role of non-state actors if we were to understand the background of civil wars. Nevertheless, we would like to stress that the state remains the main arena and framework for political struggle. As long as the international community at large base their relation with each other on formal recognition of statehood, the state will keep its role as the focal point for conflict and conflict resolution within its territory. Thus, for actors whose aim is political power, control over the state apparatus is still the best way to dominate the political order. Even for a highly successful war-lord like Charles Taylor, who did very well for himself in Greater Liberia, the yielding of state power was the ultimate aim. We do not believe that state power has such an attractiveness only because of the formal status. It also gives testimony to the importance of the state as a tool for political control and administration even if the state in question is a “quasi-state” (Jackson 1990) or a “shadow state” (Reno 1995), and despite a process of state displacement (as observed by Clapham 1998). Thus, as we turn to the question of transition to peace from war, we yet once more return to the source of fighting – the post-colonial state and the two central turning points in the trajectory of the post-colonial state.

In any transition from war to peace in Sub-Saharan Africa, the major step is the construction of a post-patrimonial state where the levels of businessified politics, social exclusion and impoverishment are drastically reduced and a sustainable economy is reconstructed. However, such a transition has to start where the neo-patrimonial state ended: at the battlefield.  The transition to peace has to pass the same turning points as the transition to war, but in reverse direction: First the military-political, then the socio-economic turning point. First the political and socio-economic struggles have to be demilitarised. The shootings and killings have to stop, and bridgeheads between the frontlines have to be established. It is only then that a genuine reciprocal assimilation not only of elites, but of the whole population can start. National reciprocal assimilation will have to constitute the backbone of the post-patrimonial state and it is upon this kind of reciprocal assimilation that new all-embracing mechanisms of integration, institutionalisation and legitimisation will have to be built. Mobilisation for development that does not take into account relevant social formations or neglect the existing social capital might lead to new civil war. In this context, the relationship between ethnicity and nation-building has to be rethought, and multiculturalism and democracy should be regarded as prerequisites for a truly developmental state. The people must participate in a process where citizenship based on equal political, cultural, and socio-economic rights is acknowledged and constructs links between the rural and urban population. Only popular-“national” and constitutional agreements of this type can create new collective identities that can give peace a new chance for real and help bring about an economy of opportunity that is socially inclusionary instead of exclusionary. However, such a process has to be built on trust among the parties concerned. Thus, we would like to emphasis not demilitarisation as the most crucial factor, but the ability to construct bridgeheads between the frontlines. Military solutions to civil wars are rarely possible. Of course it is possible to win a civil war, but such solutions are most often not sustainable.

Lessons from Mozambique and Angola

Thus, our plea is that peace is not a naive dream, but the only realistic alternative in the long run. Bloodshed often precedes unity, and out of Holocaust emerged the United Nations and the European Union. Civil wars are not inevitable in Sub-Saharan Africa, and the lesson from Mozambique is that even well-developed damnation games can be stopped if the major parties to a conflict agrees to. When Frelimo and Renamo settled on the peace deal in October 1992, the fighting stopped immediately. At the general elections two year later, Renamo won 38% of the vote to Frelimo’s 44%. Instead of returning to the bush to fight on, as Angola’s UNITA rebels did in similar circumstances, most of Renamo’s leaders took their seats in parliament. The following mandatory period of five years saw nearly no political violence. Although the results of Mozambique’s second general elections in December 1999 were strongly contested by Renamo, who threatened with a campaign of civil disobedience untill a recount of the votes or new elections were carried out, there has been no suggestions to return to war (See Braathen 2000).

Important lessons can be drawn from comparing Mozambique and Angola. The complete failure of the UN intervention in Angola and the success of the peace process in Mozambique shows that it is still the state and the internal struggle for economic resources on the state arena which is at the centre of conflict. The two Lusophone countries differed in two types of capacities influencing the transition to peace. First, the varying access to endogenous economic resources. In Angola, MPLA financed 95 per cent of its state and war machinery by oil revenues, and UNITA has been one of Africa’s largest traders in diamonds and ivory. This makes them economically capable of continuing the war endlessly. In Mozambique, however, the Frelimo state was nearly entirely financed by foreign donors, and Renamo was a “beggar’s barefoot army” compared to UNITA, and thus more vulnerable when drought ocurred. Second, the inclinations of the rebel movements to become reintegrated differed in the two countries. UNITA regarded itself as a national liberation movement, born in the anti-colonial struggle as authentically as the ruling party. UNITA’s national project was a competing project, which rallied ethnic forces behind a civil war. Renamo, by contrast, was born after independence and its objectives were to be a recognised part of the ruling party’s nation. Hence, Renamo was easier to move to a negotiated settlement (See Cahen 2000).

While the UN found very little real co-operation in its peace operation in Angola, there was a genuine support for demilitarisation and demobilisation in both parties in Mozambique.[11] UN became a forceful, legitimate and autonomous political actor in Mozambique but not in Angola.[12] However, the success of the UN operations depended on a number of internal and external contingencies that may not be repeated.

Concluding remarks

Instead of the high profile operations of the international donor and NGO community who often just prolong conflict (Duffield 1994; de Waal 1995; 1999) what is needed is support for concrete, low-level positive developments to be encountered in what de Waal (1995) terms the “Aid Free Zone.” Our best bet is therefore perhaps to opt for what Richards (1996) coins as “smart relief.” Richards’ (1996:77) point is that we have to realise that “the process of bridging, reaching out and re-making society are as dependent on internally generated capacities for communication, brokerage and social enlargement as they are on syncretic capacities resulting from external interventions.” In conflict intervention the international community should therefore rather provide support for local forces that aim for reconciliation and political renewal (the first step towards the creation of a post-patrimonial state) instead of supporting operations such as the ECOMOG experience that promoted the status of war entrepreneurs (in government and among the rebels) whose political and economic benefits depend upon the prolongation of violent conflict. NATO’s military intervention in the conflict between Kosovo-Albanians and Serbs in ex-Yugoslavia might confirm the universal validity of this experience, with one difference being that civil wars in Sub-Saharan Africa tend not to attract the same level of attention and resources as for example the conflict in Kosovo.

Nevertheless, this cannot excuse concerned world citizens from pushing forwards the alternative agenda of political, legal and economic restructuring that can prevent civil wars and solve conflicts peacefully everywhere on the earth. Efforts to secure lasting peace in Sub-Saharan Africa must therefore be concerned with social justice and combined with measures to fight poverty, locally and globally. However, if the international community want to make an impact, it is perhaps best achieved if one shifts emphasis away from material support in the form of bulk items and towards knowledge-intensive assistance. In doing so the important element is to strive to complement and support, but never substitute local efforts to restructure the state and solve the basic problems of reconciliation and social justice.

Braathen is a researcher at the Norwegian Institute for Urban and Regional Research. Bøås is a research fellow at the Centre for Environment and Development, University of Oslo. Sæther works for the Royal Norwegian Ministry of Foreign Affairs




d’ Azevedo, Warren L (1989) “Tribe and Chiefdom on the Windward Coast.” Liberian Studies Journal, vol, XIV, no, 2. 90-116.

Badie, Bertrand (1997) “Les Facteurs Internationaux de l’effondrement des Etats.” Paper presented at the XVII World Congress of the International Political Science Association.

Bayart, Jean-François (1993) The State in Africa. The Politics of the Belly. London: Longman.

Bayart, Jean Francois, Stephen Ellis & Béatrice Hibou (1999) The Criminalisation of the State in Africa. Oxford: James Currey, 1999.

Brantlinger, Patrick (1986) “The Genealogy of the Myth of the Dark Continent.” In Henry Louis Gates Jr. [ed.] Race, Writing and Difference. Chicago: University of Chicago Press. 185-222.

Braathen, Einar (2000) “Mozambique: The 1999 Parliamentary and Presidential Elections . Report from the Norwegian Election Observation Mission”. Working Paper x/00, Norwegian Institute of Human Rights, Oslo

Bratton, Michael & Nicolas van de Walle (1994): “Neopatrimonial Regimes and Political Transitions in Africa.” World Politics, vol. 46 (July). 453-89.

Bratton, Michael and Nicolas van de Walle (1997):  Democratic Experiments in Africa: Regime Transitions in Comparative Perspective. Cambridge: Cambridge University Press, 1997.

Bøas, Morten (1997) “Liberia-the Hellbound Heart? Regime Breakdown and the Deconstruction of Society.” Alternatives, no. 22. 353-380.

Bøås, Morten (2000): “Nigeria and West Africa: from a Regional Security Complex to a Regional Security Community?” In Braathen, E., M. Bøås and G. Sæther (eds.) (2000) Ethnicity Kills? The Politics of War, Peace and Ethnicity in Sub-Saharan Africa, Macmillan, London.

Cahen, Michel (2000) ‘Transitions from War to Peace. Nationalism and Ethnicities: Lessons from Mozambique’, in Braathen, E., M. Bøås and G. Sæther (eds.) (2000) Ethnicity Kills? The Politics of War, Peace and Ethnicity in Sub-Saharan Africa, Macmillan, London.

Chabal, Patrick (1992) Power in Africa. London: Macmillan.

Chabal, Patric & Jean Pascal Daloz (1999) Africa Works. Disorder as Political Instrument. Oxford: James Curry and Indiana University Press.

Clapham, Christopher (1996) Africa and the International System. The Politics of State Survival. Cambridge: Cambridge University Press.

Clapham, Christopher (1998) “Degrees of Statehood.” Review of International Studies, vol. 24, no. 2. 143-157.

Crummey, Donald (1986) “Introduction: The Great Beast.” in Donald Crummey [ed.] Banditry, Rebellion and Social Protest in Africa, London: James Currey. 1-32.

Duffield, Mark (1990) “War and Famine in Africa.” Oxfam Research Paper. No. 5.

Foucault, Michel (1982) “The Subject and Power.” In H.L. Dreyfus & P.R. Rabinow: Michel Foucault. Beyond Structuralism and Hermeneutics. New York: Harvester Wheatsheaf. 212.

Geffray, Christian (1990) La Cause des Armes au Mozambique. Anthropologie d’une Guerre Civil. Paris-Nairobi: Karthala-Credu.

Horowitz, Donald (1985) Ethnic groups in conflict. Berkeley: University of California Press.

Human Rights Watch (1995a) Africa, April 1995. New York: Human Rights Watch.

Jackson, Robert (1990) Quasi-states: Sovereignty, International Relations and the Third World. Cambridge: Cambridge University Press.

Keller, Edmond J. (1996) “Introduction: Toward a New African Political Order.” In Edmond J. Keller & Donald Rothchild [eds.] Africa in the New International Order: Rethinking State Sovereignty and Regional Security. Boulder: Lynne Rienner Publishers. 1-14.

Lema, Antoine (2000). ”Causes of Civil War in Rwanda:The Weight of History and Socio-Cultural Structures”, in Braathen, E., M. Bøås and G. Sæther (eds.) (2000) Ethnicity Kills? The Politics of War, Peace and Ethnicity in Sub-Saharan Africa, Macmillan, London

Lemarchand, René (1991) “La Violence Politique.” In Christian Coulon & Denis-Constant Martin [eds.] Les Afriques Politiques. Paris: Éditions La Découverte. 200-212.

Lewis, Ioan (1994) Blood and Bone: The Call of Kinship in Somali Society. Lawrenceville NJ.: The Red Sea Press.

Mafeje, Archie (1997) Ethnicity and Intra-Class Conflict in Africa. Research Paper submitted to Codesria.

Mamdani, Mahmood (1996) Citizen and Subject. Contemporary Africa and the Legacy of Late

Colonialism. Princeton N.J: Princeton University Press.

Mamdani, Mahmood (1998) “Why Foreign Invaders can’t help Congo.” Mail & Guardian, 2 November

Mbembe, Achille (1990) “Pouvoir, Violence et Accumulation.” Politique Africaine, no. 39. 7-24.

Médard, Jean-Francois (1982) “The Underdeveloped State in Tropical Africa: Political Clientelism or Neopatrimonialism?” In Christopher Clapham [ed.] Private Patronage and Public Power. Political Clientelism in the Modern State. London: Frances Pinter. 162-192.

Médard, Jean-Francois (1996) “Patrimonialism, Neo-patrimonialism and the Study of the Post-colonial State in Subsaharan Africa.” In Henrik Secher Marcussen [ed.] Improved Natural Resources Management – the Role of Formal Organisations and Informal Networks and Institutions. IDS: Roskilde University Press. 76-97.

Migdal, Joel S. (1988) Strong Societies and Weak States: State-Society Relations and State Capabilities in the Third World. Princeton: Princeton University Press.

Reno, William (1993) “Foreign Firms and the Financing of Charles Taylor’s NPFL.” Liberian Studies Journal, vol. XVIII, no. 2. 175-188.

Reno, William (1995) Corruption and State Politics in Sierra Leone. Cambridge: Cambridge University Press.

Reno, Wiliam (1998) Warlord Politics and African States. Boulder: Lynne Rienner Publishers.

Reyntjens, Filip (1994) L’Afrique des Grand Lacs en Crise. Paris: Karthala.

Richards, Paul (1996) Fighting for the Rain Forest: War, Youth and Resources in Sierra Leone. Oxford: James Currey/Heinemann.

Richburg, Keith (1997) Out of America. New York: New Republic Books.

Savigliano, Marta E. (1995) Tango and the Political Economy of Passion, Boulder: Westview Press.

Schafer, Jessica (1998) “A Baby Who does not Cry will be Suckled: AMODEG and the Reintegration of Demobilised Soldiers.” Journal of Southern African Studies, vol. 24, no. 1. 207-222.

Seth, Sajay, Leela Gandhi & Michael Dutton (1998) “Postcolonial Studies: A Beginning…” Postcolonial Studies, vol. 1, no. 1. 7-11.

Synge, Richard (1997) Mozambique: UN Peacekeeping in Action, 1992-94. Washington D.C.: US Institute of Peace Research.

Sæther, Gjermund (2000) “The Political Economy of State Collapse in Somalia and Liberia”, in Braathen, E., M. Bøås and G. Sæther (eds.) (2000) Ethnicity Kills? The Politics of War, Peace and Ethnicity in Sub-Saharan Africa, Macmillan, London

Vail, L [ed.] (1989) The Creation of Tribalism in Southern Africa. Berkeley: University of California Press.

Vines, Alex (1998) “Disarmament in Mozambique.” Journal of Southern African Studies, vol. 24, no. 1. 191-205.

Waal, Alex de (1997) Famine Crimes: Politics & The Disaster Relief Industry in Africa. Oxford: James Curry/Indiana University Press.

Zartman, I. William [ed.] (1995) Collapsed States. Boulder: Lynne Rienner Publishers.


[1] Foucault’s concept of subjectification is part of his understanding of the modern European state, the state of “governmentality.” However, as Mamdani argues, the European colonial state in Africa managed to incorporate even the rural peasants and make them subjects, through indirect rule, as opposed to the “free” citizens in the urban colons (Mamdani 1996). We may interpret this African type of incorporation into “modernity” as a process of subjectification. Due to the colonial legacy of this rural-urban divide, the post-colonial states operate with a peculiar bifurcated citizenship: One civic, the other ethnic. Civic citizenship is a consequence of membership of the central state, specified in a constitution and is the basis of rights. In contrast, ethnic citizenship is a consequence if membership in the native authority; it is the source of a different category of rights, mainly the socio-economic right to use land as a source of livelihood (Mamdani 1998).

[2] For the background of the term the damnation game see Bøås’ (1997) article on the civil war in Liberia.

[3] It is interesting to note that there is a clear parallel between “Ethnic Groups in Conflict” and “The Ethnic Origin of Nations” by A.D. Smith (1987). The focus is in both cases upon frozen ethnic identities, being it in the past of the West or in the present of elsewhere.

[4] An excellent representative for the plethora of more superfluous literature focusing on ethnicity and tribalism is Richburg (1997).

[5] We should, however, keep in mind that just as the colonised were subjected to transformation, the colonisers too were transformed by the colonial encounter. Profits, and the experience of administration and exploitation of the colonies transformed the West’s sense of itself, and created new forms and regimes of knowledge: whole new disciplines were born, such as anthropology.

[6] Authors’ emphasis and translation.

[7] According to Mamdani (1996:20) patrimonialism is “in fact a form of politics that restored an urban-rural link in the context of a bifurcated state, albeit in a top-down fashion that facilitated the quest of bourgeois fractions to strengthen and reproduce their leadership.”

[8] This observation of the state and definition of criminalisation is taken from CODESRIA/The Goree Institute: “The Criminalisation of the State in Africa. Call for research proposals” June 1998. See also Bayart, Ellis & Hibou (1999).

[9] See Reno (1998) for additional arguments and empirical evidence of these processes.

[10] For a general argument about the weakened role for the state in the age of globalisation see, for instance, Badie (1997). For two very different approaches, but whom both are concerned with Africa see Keller & Rothchild (1996) and Clapham (1998).

[11] See Vines (1998) on disarmament in Mozambique and Schafer (1998) on reintegration of demobilised soldiers in Mozambique.

[12] See Cahen (2000), and Synge (1997) on the UN peacekeeping operation in Mozambique.

© Copyright Leeds 2017