"Root Causes": The Inversion of Causes and Consequences in Civil War

Abstract

[...] the ethnic enmity that many commentators depict as the reason of armed confrontations may just as well be seen as a product of the conflicts. [...] so-called new, small wars are closely linked up with international political economy (e.g. Mantz 2008, Reno 2001), they are usually part of a regional conflict rather than being circumscribed within national boundaries (e.g. Ero and Ferme 2002), and all of them claim legitimacy in opposition to a centralised, national power and feed on local grievances.\n Yet it is still premature to conclude about the prospects of future coexistence of previously warring factions in this part of the world.

Introduction

It appears as something of a paradox in the social sciences of culture, such as anthropology which is the discipline I represent in these pages, when scholars refuse vehemently to consider the role of culture in causing and sustaining organised, violent conflict2. One major reason for disclaiming culture as a causal mechanism in the outbreak and course of armed conflicts resides in the fact that anthropologists and others take pains to dissociate their approach to culture from the so-called culturalist explanations of the many small, new wars in the post-Cold War era. A number of political scientists and some journalists writing for a larger audience have thus become interested in the impact of culture on political life, including in particular organised, armed conflict (e.g. Chabal and Daloz 1999, Ellis 1999, Huntington 1993, Kaldor 2004, Kaplan 1994). On the other hand, anthropologists are in general wary of providing explanations of cultural and social phenomena and conceive instead of their science as semiotics, preoccupied with questions of meaning and sense-making. Among those few who actually address the issue of causal mechanisms in connection with social violence, the untenable idea that the epiphenomenon of culture can be seen as causal is replaced by notions of rational choice among individual actors (pursuing maximal interests, sometimes through the deliberate "use. of culture) and the role of social institutions, including intergenerational tensions and class conflict.

I sympathize with the view of those scholars who discard culturalist explanations of human behaviour as long as the critical issue of culture's motivational force and implications for human behaviour remains unaddressed in terms of underlying cognitive and psychological processes (Højbjerg 2007). However, the argument of this article is based on the idea that the refusal to consider the impact of "culture. on collective violence and war contains the risk of throwing out the baby with the bathing water. Causes of war are very often blurred. War is profitable to many actors who may have different interests. Take, for instance, the constantly reoccurring rebel insurgency in eastern DR Congo. To the people involved and many commentators alike it is not always clear what is at stake. Is it a competition for resources and gains, ethnic tensions, or a proxy war waged by one or more neighbours of the DR Congo? Not only are causes of war blurred. It is also difficult to disentangle the causes and effects of war. Thus, the ethnic enmity that many commentators depict as the reason of armed confrontations may just as well be seen as a product of the conflicts. As shown by Kalyvas, it is also worth recalling that actions and motivations of combatants in civil war rarely resonate unambiguously with a general ideological program orchestrated by supralocal elites. Top-down run group goals that make use of social violence always intersect with local feuds, personal revenge and sheer opportunism (Kalyvas 2003, cf. Hoffman 2007:647-648).

The apparent confusion of causes and consequences of conflict, especially in connection to the so-called "new wars., has not prevented contemporary analysts of war to craft a number of broad brush explanations of the proliferation and nature of violent conflicts at the end of the Cold War and beyond into the twenty-first century. Such general explanations of the new, small wars privilege either one or a combination of factors that relate to resource scarcity and population pressure, clash of cultures, and global political economy. General insight into the nature of new wars is, however, often gained at the expense of adequate attention for their social and cultural context. I here apply the term context both in the sociological sense as reference to the actors planning and carrying out war (Richards 2005a) and in the sense of cultural ideas and practices associated with violence, identity construction and notions of belonging. It is characteristic of anthropology's contribution to an understanding of collective conflict and civil war in an era of globalisation that it privileges a bottom-up approach; that is, a "grounded. account of the background and scale of conflict which differs from most "broad brush. explanations and large-scale comparative analyses (e.g. Englund 2005). This contextual alternative to mono-causal explanations of war allows for a better grasp of the social dimensions of war, and of the agency of the war-affected in re-inventing peace.

Much attention has also been paid to the cultural, symbolic aspects of violence in order to understand how both victims and perpetrators make sense of, cope with and justify otherwise incomprehensible acts (e.g. Appadurai 1998, Hoffman 2006). As mentioned, this kind of research is interpretive rather than explanatory in orientation. It may rightly be argued that a focus on the micro-level of war entails the risk of blurring causes and effects and of losing sight of the "real issues. and the broader picture. Yet it is at the same time a matter of debate whether sweeping explanations of the nature of small, new wars actually risks misconstruing the complexity of the subject under study. As suggested by a number of anthropologist and political scientists of war and political violence it is rather a question of striking a balance between, on the one hand, macro-level politics and political economy, including the effects of globalisation, and, on the other, micro-level dynamics that influence the course of violent conflict at a local level (e.g. Ferguson 2008, King 2004, Reno 2007). It appears futile also to discuss causes of war without a clear notion of the scaling of war. Thus, so-called new, small wars are closely linked up with international political economy (e.g. Mantz 2008, Reno 2001), they are usually part of a regional conflict rather than being circumscribed within national boundaries (e.g. Ero and Ferme 2002), and all of them claim legitimacy in opposition to a centralised, national power and feed on local grievances. Besides the problem of scale and nature of explanation, the notion of new war itself to designate the rash of armed conflict in the post-Cold War era has become object to a long series of criticism over the last five to ten years (e.g. Cramer 2006, Reyna 2008, Richards 2005a).

In this article focus is on the consequences of armed conflict, rather than on the causes in the sense of first instances. While the reasons for waging war often are many and blurred, the consequences are more clearly identifiable. Local identities and oppositions between neighbouring groups are typically enhanced and new power complexes emerge. On the background of a discussion of current explanations of civil war, I argue for the relevance of adopting an analytical approach that pays special attention to the social and cultural dimensions of war and unstable post-conflict situations. With reference to a war within the civil wars that ravaged the Upper Guinea coastal and forest region from 1990 to 2003 I am going to address the issue of how group identities and oppositions become enhanced during protracted conflict and, then, in the aftermath of war risk generating further armed violence. This is the scenario of consequences of war turning into causes of war. Policy makers and external peace makers are highly aware of this risk and, as I am now going to show, they act accordingly.

Culture War

Nowadays the official ending of civil war entails the intervention of a host of NGOs and international agencies that engage in peace-keeping and state-building activities as well as in community based conflict resolution. Except for the most basic infrastructure, such as roads, medical centers and schools, the economic and material support intended to improve a war-affected population's living conditions only follow at a later stage, since donor agencies consider peace and security as a prerequisite for development. Two years after the belligerent parties in the Liberian civil war (1990-2003) signed the peace accord in 2003 I had the opportunity to attend an official peace-building workshop in the administrative center of Voinjama, a town in the remote area of the Liberian hinterland close to the border of neighbouring Guinea (see map below). Before and after the war two 'distinct' peoples, the Loma and the Mandingo (or Manya) have been coexisting peacefully in this border region. Large parts of the town of Voinjama had been entirely destroyed as a result of fourteen years of war interrupted by shorter or longer periods of relative peace. When I visited the place for the first time in 2005, internally displaced refugees and refugees in exile in neighbouring Guinea were slowly and hesitantly returning in small numbers; the main public school and one medical clinic had been reconstructed by UN agencies; two renovated, religious buildings, a church and a mosque, clearly illustrated the importance of religion in this initial phase of community reconstruction. Amidst ruins, partly repaired houses and the few renovated public buildings, lay the fenced and heavily guarded military camp of the Pakistani UN peace-keeping force.

Both local and international NGOs plan and coordinate peace-building workshops in war-affected communities. The workshops occur quite frequently and despite being initiated and led by outsiders, the organizers profess a sort of bottom-up philosophy which is meant to induce a perception of ownership among the participants to the ideas processed and the decisions made during the course of the event. In the case mentioned here the performance of a NGO-run conflict resolution initiative was a relatively formalized event, involving ecumenical prayer, collective singing, opening and closing speeches and so on. Peace workshops should not be compared to ritual events in a strict sense, though. It is not uncommon to observe participants engage in open dispute during the peace-building sessions which make them rather resemble institutionalised exchanges of political opinion. In the area I refer to here, properly ritualized, peace-making efforts are instead locally organized events that involve one or more warring factions and occur without the participation of any external actors.

The peace-building workshop that took place in the town of Voinjama in early June 2005 was an all-day event, interrupted only by a much awaited free lunch. It may be of interest to note that a local, evangelical church was used to host the meeting, even though the members of one of the two participating groups were Muslim. A national NGO (LCIP), supported by the American development agency USAID, was responsible for the workshop and the participants comprised what the organizers defined as "community facilitators" from among the two major ethnic groups living in town and the surrounding area. In this case the term "community facilitators. was a euphemism for a local elite that consisted of political actors appointed by the government, religious leaders and traditional chiefs, representatives of the elders, the women, the youth, and the ex-combatants. According to the organisers, the workshop was intended to fulfill two related objectives that one may chose to see as constituting a collective catharsis. The first objective was to relieve tensions between the two opposed ethnic groups by making them identify conflict causes and solutions in public and in common. The second objective of the meeting was meant to prepare for cleansing rituals to be carried out at selected sacred sites by each group separately. Here the aim was stated as "cleaning our hearts for animosity towards opposite groups and against people from one's own group."

As I will return shortly to some of the consequences of the armed conflict for intercommunal relationships, I jump straight to the main causes of armed conflict as they were perceived by the workshop participants. As it appears from the "conflict tree" (see photo below) that the facilitators used as a working tool in the peace meeting to illustrate causal relationships, the "root causes" of the civil war implied the role of the "book people", "tribalism", "lack of respect for cultural heritage", "lack of respect for elders", "kinship", and finally the issue of "citizenship". While both Mandingo and Loma agreed on the negative effects for intercommunal relations of so-called educated people and ethnic identity formation, they differed in their emphasis on other aspects that clearly mirrored the present circumstances of each group taken separately. For instance, the Loma would express their frustrations concerning the Muslim Mandingos. disrespect of their traditional religious practices and the Mandingo were equally heard complaining of their stigmatized identity as being "Guineans" rather than Liberians. The local perceptions of the major causes of civil war can be rendered as referring to the role of educated community leaders and state authorities; ethnic differences; the recognition of norms and values associated with traditional, secret societies and Islam; generational conflict; inter-marriage between the opposed groups; and the claim of some Mandingos to be recognized as citizens of the Liberian nation. Ironically, one of the workshop coordinators promptly refused to recognize the issue of citizenship as to be of any relevance in the present case. As mentioned, this attitude went against the foundational didactical principle of acknowledging the participants. right to define problems and solutions.

Before considering in further detail the issues raised by members of the conflicting parties, I would like to point out one conspicuous aspect of the causes of conflict mentioned. With one or two exceptions that suggest an explicit dimension of power; that is, the elders versus youth and the role of the "book people., the depicted "root causes" of conflict refer first of all to matters of culture and in particular to cultural difference. There is no mentioning here of the major conflict generating factors evoked by other external observers, such as competition over resources, including diamonds, or access to land in rural areas or in commercial centers. Nor is there any allusion to long-term exploitation and domination by a state-supported, local elite, or by some traditional, gerontocratic political culture. To evoke the role of the "book people" may suggest popular resentment towards political leaders, though it is unclear whether the category refers to the "black colonialism. of the descendants of the Americo-Liberian settler elite who ruled the country of Liberia for more than a century, to the subsequent, oppressive "indigenous regime. of Samuel Doe that preceded the outbreak of war, or to any educated peoples associated with one faction or another during the fourteen years of civil war.

In the present case there seems to be clear evidence that local victims and perpetrators of armed conflict perceive of the lack of respect for cultural differences as a major factor causing and perpetuating war between neighbours. This local viewpoint on the cause of a contemporary conflict sounds almost too familiar. It reflects the way media in the North usually represent civil wars in the global South. Moreover, it sustains the famous theses by Kaplan (1994) and Huntington (1993) that explain the nature and origins of wars in the twenty-first century in terms of culture clashes. The analyses of Kaplan and Huntington have been subject to much debate. Critics have among other things questioned the notion of culture (and civilization) employed in these culturalist approaches to conflict and which conceive of culture as a bounded, essentialised and homogenous entity. In contrast, anthropologists and others have long since employed a more pragmatically defined notion of culture as differentiated, negotiable and subject to politicized manipulation. Critics also raise the essential question as to how culture can be causal and which may also be addressed to a majority of students of culture, including anthropologists, who tend to take agency and culture's impact on human behaviour for granted. In addition, one should not forget to apply a methodological skepticism and not take the word of one's subjects of study, the victims and perpetrators of war, at face value. On the other hand, it would be analytically misleading (and morally unjustifiable) to disregard local representations of conflict. Even if the perception of cultural difference may not be the primary cause of conflict outbreak, significant differences in language, religion, marriage practices, etc. that are important for identity construction may be enhanced during times of conflict and thus become critical for the perpetuation of conflict. As already suggested, the cultural competition thesis is not the only prevalent model of explanation of protracted West African armed conflicts. Most scholars are critical of the culturalist approach and have instead associated the 'root causes' of civil war with 'crisis of youth' (e.g. Hoffman 2006 and Richards 2005b), so-called 'failed states' and general political economy (e.g. Reno 2001, Sawyer 2004), and 'greed and grievances' (e.g. Keen 2005).

New Wars and Root Causes

For some time a metanarrative about the new world order has served to explain the proliferation of wars in the global South. As the story goes, the end of the Cold War put an end to the super powers. monopoly of violence which frequently manifested as the waging of proxy wars in the worlds periphery. Undemocratic regimes ceased to receive financial and logistic, military support. During the same period international donor agencies imposed structural adjustment programs that reduced state control of the economy and affected state sovereignty. By the same token the living conditions of impoverished populations decreased even further while their numbers were increasing. Western donor countries moreover required political reforms in response to development aid. Along with economic and political reform, the expanding, economic globalization facilitated privatization processes which put an increasing pressure on state authority in many African states and elsewhere; a process that also created new opportunities for 'privatised states'. More generally, the neoliberal, post-Cold War world order is held to have caused the erosion of the state and state sovereignty. Incapable of controlling their territories, weak states then withdraw or collapsed. A rash of small wars irrupted in the interzones between failed nation states in Africa, the Balkans and other parts of the former communist world. In this new era inter-state wars were replaced by intra-state wars in "ungoverned zones. (e.g. Engel and Mehler 2005; Richards 2005a; Reno 2003).

The realisation that the new wars represented a security threat to the world as a whole soon entailed new forms of intervention in the war-affected regions, especially by UN-led peace-keeping missions, which are sometimes preceded sometimes succeeded by a host of humanitarian agencies and INGOs. Some commentators have questioned the effectiveness of these intervention forces and suggest that rather than serving to put an end to wars, they may produce a reverse effect (e.g. Duffield 2001). On the other hand, international military intervention efforts to stop armed conflict and sustain peace within recognized boundaries signals a consensus that the state has not withered away. International political culture demands that states continue to be territorially based (e.g. R.B. Ferguson 2003; Kapferer 2004, see also Ferguson 2006:39). Mentioned in passing, it is likely that the above metanarrative about neoliberal global (dis)order will be going through revision in light both of the recent military intervention of one nation-state, Russian, into another, Georgia, and the outbreak of the current global, financial crisis.

I shall leave aside the question whether Twenty-first century wars are really new in the sense of being unprecedented (Kaldor 2004), or whether they rather resemble the low-intensity, non-state wars fought in pre-Westphalian Europe (Reyna 2008). For the purpose of this paper, it suffices to indicate how a majority of commentators explain the nature of contemporary, non-state armed conflict. According to the anthropologist Paul Richards, for instance, three kinds of explanation of "new war. emerged during the 1990s. He labels them as "Malthus with guns., "the new barbarism., and the "greed not grievance. debates. The first, exemplified by the political scientist Homer-Dixon, favours the explanation of violence in developing countries as a consequence of resource scarcity and competition over resources; the second, exemplified by Kaplan and Huntington's famous thesis concerning culture competition, explains new war in terms of endemic hostilities between peoples endowed with immutable, bounded cultural identities; the third mode of explanation, exemplified by the work of the geographer Le Billon (e.g. 2001) and the economist Collier (e.g. 2004), see internal wars as a result mainly of economic considerations (Richards 2005a: 6-11). In their effort to simplify even further the dominant readings that have emerged about the "root causes. of violent conflict in Africa, the political scientists Engel and Mehler identify two major lines of argument. One concentrates on ethnicity and identity (basically, but not exclusively, the "grievance. argument) and the other focuses on the political economy of conflict (basically, but not exclusively, the "greed. argument) (2005:90). Thus, while there may not be a consensus on the reasons behind the emergence of the "new wars., most explanations converge on the point of construing such violent conflicts as apolitical (Cramer 2007:76, Hoffman 2006:12, Kaldor 2004, Richards 2005b:573). To reject the political subjectivity of armed actors is a questionable assertion that rests on a narrow definition of the political from a state-centred perspective.

I agree with Richards's objection to the notion of "new war. as a "mindless response" to resource scarcity and population pressure, cultural competition, or trans-boundary political economy. None of these arguments, says Richards, "offers a convincing explanation of why war happens when and where they do, offering only an explanation how war is intensified or prolonged (...) [W]ar does not break out because conditions happen to be "right., but because it is organised (...) [W]ar is inescapably sociological." (Richards 2005a:4; see also R.B. Ferguson 2008:42-43). I part company, however, with Richards. alternative sociological explanation of war in the Mano River region in West Africa as a product of rural youth's grievances against "customary. chiefs and local elders (Richards 2005b:586; Chauveau and Richards 2008). Although it is presented as "Emile Durkheim's theory of civil war" because of its association of an alleged forced division of labour and revolt of farm slaves, this explanation reads more like a classical Marxist description of an alleged class conflict in West African lineage-based societies. There are two reasons for objecting to Richards. conceptualisation of the recent civil wars in Sierra Leone and Liberian as well as beyond in Guinea and Cote d.Ivoire. First, as mentioned by another critique of Richards. analysis, the empirical evidence sustaining his thesis appears inadequate and the interview-based methodological approach to be of limited value (Fanthorpe 2005). Second, ethnographic data that I have gathered in rural and semi-urban communities in northwestern Liberia and southeastern Guinea also does not lend credit to the notion of civil war in the Mano River region essentially as a revolt of an exploited and disillusioned rural youth. Neither in the past nor at present does competition over agrarian resources appear to have constituted a major cause of war. This second point does not preclude, however, that Richards. Marxist-Durkheimian analytical framework applies to specific parts of war-torn West Africa, such as the Sierra Leone Hinterland, but I raise doubts about the general value of this approach. My questioning of Richards. view on the West African subregional conflict does not mean that I do not consider the impact of local grievances and rather have a preference for the 'greed, not grievance' explanatory model, which has gained currency in connection with this West African war zone. In accordance with this model 'blood diamonds' constitutes the simplified, popular image of the driving force behind the conflict.

As a matter of fact, many students of conflict in the area I refer to, especially among political scientists, seek to strike a balance between the greed and grievance perspective. William Reno and David Keen are among the best known analysts who take account of both predatory behaviour of armed groups, international political economy, and the political revolt and popular grievances that constitute the complexity of the interconnected civil wars in Sierra Leone and Liberia (Keen 2004, Reno 2001, 2007, Richards 1996). R.B. Ferguson suggests the neologism 'identerism' to depict the so-called identity-linked new wars for which we possess no general term. In opposition to "some current explanatory divides that stress either self-interest or identity issues (sometimes framed as 'greed vs. grievance'), 'identerism' highlights the point that practical interests and self-identities are very commonly fused into one (...) Calling a conflict 'identerest' creates a question of identities and interests that must be answered..." (Ferguson 2008:43). The anthropologist Gunther Schlee, a specialist on conflict in east African pastoral societies, proposes a theory of conflict that also intends to move beyond the analytical dichotomy between greed and grievance (Schlee 2004). Schlee acknowledges the importance of resource-orientated, economically or ecologically inspired theories that account for conflicts caused by the competition for resources and gains. Notwithstanding, the issue of identity work and group identifications that occur in situations of violent conflict is claimed to remain poorly understood. Who is fighting who and why is less evident than it appears and it cannot be explained alone by taking the legitimizing accounts of participants at face value. According to Schlee, "there are still deficiencies in our understanding of the ways in which people in specific conflict situations may make and break alliances and which patterns of identifications they follow." (2004:135)3.

There are clear merits to an approach to conflict analysis that builds on rational choice theory and theories of group identification. I have my doubts however, whether this integrated, explanatory model of conflict actually meets its own demands. Despite being action-centred the theory lacks a sustained notion of agency that helps understand the full rationale of people's actions in situations of conflict. What is needed, I will argue, is a notion of agency that also takes in the motivational force of memories and emotions, rather than a notion which rests ultimately on people's calculation of costs and benefits (including the unintended consequences of being passively included in or excluded from a given social unit) and the semantic constraints of social categories. New research on the means and ends of war, and on organised political violence in general, ought instead to proceed from a theoretical perspective that works toward articulating a micro level, which addresses questions of intentionality, memory, emotions and intersubjectivity, and a macro level, which focuses on structural conditions, social and cultural context.

Culture War Recontextualized

It is high time to return to the major themes of the peace-building meeting I reported at the beginning of this paper. My aim in the following is to suggest how intercommunal tensions, which I consider as a consequence of protracted, violent conflict, may in turn during so-called post-conflict time become causes for further, or sustain ongoing collective, violent interaction. In contrast to culturalist assumptions about the causal role of essentialised culture, the following section is meant to illustrate how under certain circumstances the epiphenomena of culture and second-order causes turn into first-order, or "root causes" of violent conflict.

The rural urban settlement that constitutes the empirical background of this article is situated in the northwestern hinterland of Liberia close to the border of Guinea. Located at the heart of a trouble-ridden region, between two nation-states, the town of Voinjama constitutes a typical border zone that affords a great variety of (un)regulated economical, political and military opportunities. This was the case to an even greater extent during the war characterised by a high frequency movement of people, both refugees and rebels, and illegal border transaction of goods, including arms and looted property, by entrepreneurial individuals, marauding rebels and government soldiers from both Liberia and Guinea. There are good reasons to conceptualise the protracted violent conflict that ravaged the area as part of a regional war without frontiers (Højbjerg 2008). On the other hand, the conflict dynamics in nortwestern Liberia and across the border into southeastern Guinea are too complex to be seen as simply deriving from the general, subregional political deficiencies that sparked civil wars during the final decade of the twentieth century.

In the Voinjama area and beyond a minority of people labelled ethnically as Mandingo currently find themselves in an antagonistic relationship to an amalgam of ethnic groups (e.g. Loma, Mano, Gio, Kpelle, etc.) who refer to themselves as autochthonous. These "people of the land. classify the Mandingo as "strangers from Guinea" and sometimes pejoratively as "Dingos". Within a greater zone of Mandingo-autochthonous cohabitation conflict dynamics vary due to different historical trajectories of Mandingo migration and settlement patterns. For instance, the Mandingo in northeastern Lofa, who name themselves Manya, have long since been recognised as the guardians of their own territory. They are usually considered less "strangers" than the Mandingo in the adjacent Nimba County, who often refer to themselves as Konyanka, and who are looked upon as descendants of stranger traders. Ethnogenesis and the possible common origin of the Manya and Loma in the Voinjama/Macenta border region is a prominent issue among scholars as well as among local people, though it remains a moot point4. The question of origin is also prevalent among the inhabitants of Nimba County, but here identities have always been declared and ascribed in less ambiguous terms. Tensions mainly evolve around land rights and property relations, especially in urban centres and in diamond extraction areas. According to my own findings in both ethnically mixed and homogenous rural settings around Voinjama in norteastern Lofa county, access to cultivable land does not seem to constitute a major conflict cause, except in cases where rice swamp cultivation has been introduced. In comparison, conflict over access to farm land represents a more urgent issue among the same groups of people in the adjacent and more densely populated areas in southeastern Guinea. From a more general perspective, then, intercommunal violent conflict in the Liberia-Guinea border region is comparable to other conflicts in the Upper Guinea forest area, which, according to Chauveau and Richards (2008) are triggered, or sustained by competition over agrarian resources. Having said that, it is important to recall that the present case does not to imply the class-based violence reported in relation to adjacent regions by the mentioned authors. In addition, war did not occur spontaneously in the Liberian and Guinean hinterlands, but was obviously brought there from the outside world by the warlord and later president Charles Taylor who was heading a small rebel army composed of mercenaries recruited from several West African countries. Once the fighting between rebels and governments forces had begun, local peoples started adhering to Taylor's 'cause' and the war took on its own course during which causes and consequences, private motives and collective interests became blurred.

Metaphorical Kinship, Religion and Settler-Stranger Relationships

Over time the long-term conflict between co-existing Mandingo and Loma and other ethnic groups has also been nurtured by the symbolic dimensions of local culture that inform the relationship between co-existing ethnic groups. A key symbolic aspect of local culture that is often alluded to as a contemporary conflict cause concerns the general refusal among Mandingo to recognise the metaphorical use of matrilateral kinship to conceive of the relationship between Mandingo and various self-declared autochthonous peoples. The methaphor of matrilateral kinship embodies an entire political culture of alliance among groups of descent (lineages) that stand in hierarchical relationship to each other as landowning 'mother's brothers and latecoming sister's sons, or as wife-givers and wife-takers. Matrilateral alliance translates as perpetual kinship in the sense that the political organisation at any time ideally reflects a real or mythical exchange between a male landlord and male stranger. Affinal kinship between mother's brothers and sister's sons implies rights and obligations, both at the individual and at the community level, including deference, reciprocal support and ritual collaboration. Historically, matrilateral alliance has at the same time been projected onto entire ethnic groups in the culture area comprising the forest region in Guinea, parts of Sierra Leone and Liberia. In that process the Mandingo (Manya and Konyanka) have been ascribed the status as latecoming 'nephews' or sister's sons, and, as a consequence, as "allochthons" in relation to the landowning 'uncles' or mother's brothers of Loma, Mano, Gio and Kpelle origin who claim "autochthony".

Unsurprisingly, religious identity is another aspect of local culture that has nurtured long-term conflict between co-existing ethnic groups. This is notably the case in connection with the increased difference between a professed Mandingo Muslim identity and the power associations or secret, ritual institutions of those who consider themselves as autochthonous. Until the mid-twentieth century and in some cases later a majority of the Mandingo population on both sides of the Liberia-Guinea border still adhered to the widespread Manding power association known as Koma, or took part in the Poro ritual association of the co-existing Loma and Kpelle population. In the Nimba region further to the west Islam appears to have been anchored much longer and more firmly among the local Mandingo of Konyanka origin. In contrast to people's relative tolerance and respect of religious differences in the past, the conflicting parties currently emphasise differences in religious traditions and sometimes stage them in an aggressive fashion with an ill-concealed political purpose (Højbjerg 2005, 2007). During the NGO-sponsored peace-building workshop reported in the first part of this article protagonists typically phrased the conflict cause as a lack of respect for each other's cultures, implying that culture is synonymous with religious practice.

Objectification of Culture and Changing Politics of Neigbourliness

The intercommunal relationship I have just described reads as a progressive schismogenesis between so-called Mandingo "strangers" and "autochthonous" forest peoples. A formal, ritualized relationship of reciprocity and mutuality has been replaced by an antagonistic relationship expressed in terms of exclusive, ethnic and religious identities. As social scientists we have learned the importance of being wary of people's self representations. If we apply this methodological skepticism to the present case, it follows that the Mandingo and the 'forest people's' description of themselves and their neighbours may not correspond entirely with the way these groups actually interact at different levels ranging from whole communities to the level of families and individuals. There is considerable evidence, though, suggesting that collective narratives about the "Other. have informed in one way or another intercommunal relations, including violent events, in the conflict-ridden border area between Guinea and Liberia during the past two decades (Højbjerg 2009). Granted that I have provided a correct description of the change that is taking place in local conceptualizations of intercommunal relationships, it is relevant to ask whether we are witnessing an irreversible development, or if local strategies of co-existence suffice to cope with the problem of difference and polarization between neighbouring peoples?

My provisional answer to this question is of a double nature. On the one hand, there is little doubt that a majority of the population throughout the region wish to see an end to a long-term conflict, which obviously has no clear winners. Quite the contrary in fact, as losses and suffering has been immense among both sides of the conflicting parties. Therefore, many local initiatives have been taking place with the aim of reconciling neighbour enemies, often in the form of ritual and religious ceremony comprising adherents to different belief systems. Parallel to such spontaneous, local strategies exist the many top-down initiatives introduced by national, interfaith associations, political elites with local roots, the entire security and development aid industry involving the UN, humanitarian agencies, and NGO's, and finally the national governments, including the Truth and Reconciliation Commission in Liberia. There is no reason not to believe, on the other hand, that these peace-building initiatives do not have a positive impact on people's decision to return from exile and reconstruct their lives and accept the presence of neighbours who may have killed members of one's family and looted one's property.

Yet it is still premature to conclude about the prospects of future coexistence of previously warring factions in this part of the world. Political economy aside, one major reason that leads me to conclude in this cautious way relates to the widespread atmosphere of mistrust of the "neighbour. that reigns underneath an official and political correct discourse about reconciliation and peaceful co-existence. In addition, the omnipresent rights discourse only enhances a collective sentiment of exclusive group identities. Whether it is part of the project of peace-building and community reconstruction, or resulting from some other globalizing trend, community leaders indiscriminately profess, "We have the right to practice our culture", or "They don't respect our culture". It is well known that once culture becomes an object of reflection and evaluation in essentialising terms, it is already an object of purposeful manipulation; sometimes with harmful results.

Conclusion

I began this article by noticing the paradoxical fact that students of culture are sometimes reluctant to consider the causal role of culture in collective armed conflict while a number of political scientists and others seem to have opted for the opposite perspective in their analysis of so-called 'new wars'. This article has provided empirical evidence in the sense of local perceptions of conflict causes of the apparent importance of culture, and notably cultural difference, for the onset and continuation of civil war. However, cultural and social scientific analysis of social phenomena, such as civil war, should avoid perpetuating 'essentialised' local representations. Instead of taking them at face value, the role of science is to contextualize local knowledge and put it into perspective by studying how it is produced, distributed and cognized.

Against the dominant broad brush explanations of civil war and collective armed conflict I have argued that it is often most difficult to dissociate causes and consequences in ongoing 'small wars'. If one looks for causes in a too narrow sense there is always a risk of misrepresenting the object of study and it may, furthermore, entail negative effects for peace-making interventions, reconciliation initiatives and community construction. People do not fight each other simply because they are culturally different, but cultural identities and notions of belonging is sometimes endowed with importance in the course of events and may serve consciously or unconsciously to inform collective action. Contrary to standard knowledge within and outside the social scientific study of culture we still do not possess sufficient insight into how 'culture' in the sense of ideas, memories and values affect violent collective action. This article has tried to respond to this demand by focusing simultaneously on the micro and the macro level of a particular case of ethnic enmity and civil war and on consciously held notions of conflict causes.

Footnote

2 Funding for recent fieldwork mentioned in this article has been granted by The Danish Council for Development Research (FFU). I am grateful to the anonymous reviewer and the participants at "The Roots of Civil War and Conflicts..." conference at Alsion for their constructive comments on an earlier version of this article. Thanks also to the Upper Guinea Coast research group at the Max Planck Institute for Social Anthropology, Halle/Saale, for providing me with ideal working conditions to write this article.

3 See also Kalyvas 2003 and King's description of the 'micropolitical turn' in the study of social violence within the disciplines of comparative politics and international relations (King 2004).

4 See Højbjerg 2007: 80-88 for a discussion of Loma and Manya ethnogenesis.

References

Bibliography

Chabal, Patrick & Jean-Pascal Daloz. 1999. Africa works: disorder as political instrument. Oxford: James Currey, Bloomington: Indiana University Press, in association with International African Institute.

Chauveau, Jean-Pierre and Paul Richards. 2008. West African Insurgencies in Agrarian Perspective: Cote d.Ivoire and Sierra Leone Compared. Journal of Agrarian Change 8 (4):515-552.

Cramer, Christopher. 2006. Civil War Is Not a Stupid Thing. Accounting for Violence in Developing Countries. London: Hurst & Co.

Duffield, Mark. 2001. Global Governance and the New Wars. The Merging of Development and Security. London & New York: Zed Books.

Ellis, Stephen. 1999. The Mask of Anarchy. The Destruction of Liberia and the Religious Dimension of an African Civil War. London: Hurst & Co.

Engel, Ulf and Andreas Mehler. 2005. 'Under construction': Governance in Africa's new violent social spaces. U. Engel, Ulf & G. R.Olsen (eds.): The African exception, pp.87-102. Aldershot: Ashgate.

Englund, Harrry. 2005. Conflicts in context. Political violence and anthropological puzzles. Vigdis Broch-Due (ed.) Violence and Belonging. The quest for identity in post-colonial Africa, pp. 60-74. London and New York: Routledge.

Ero, C. et M. Ferme. 2002. Liberia, Sierra Leone et Guinee: une guerre sans frontieres. Politique africaine 88:5-12.

Fanthorpe, Richard. 2005. On the limits of liberal peace: chiefs and democratic decentralization in post-war Sierra Leone. African Affairs 105:27-49

Ferguson, R. Brian. 2008. Ten Points on War. Social Analysis 52:32-49.

Ferguson, R. Brian. 2003. Introduction: violent conflict and control of the state. R.B.

Ferguson (ed.) The State, Identity and Violence. Political disintegration in the post-Cold War world, pp. 1-58. London and New York: Routledge.

Ferguson, James 2006. Global shadows: Africa in the neoliberal world order. Durham: Duke University Press.

Hoffman, Danny. 2006. Disagreement: Dissent Politics and the War in Sierra Leone. Africa Today 52 (3):3-22.

Hoffman, Danny. 2007. The Meaning of a Militia: Understanding the Civil Defence Forces of Sierra Leone. African Affairs 106:639-662.

Huntington, Samuel P. 1993. The Clash of Civilizations? Foreign Affairs 72 (3): 22-49.

Højbjerg, Christian K. 2005. Masked Violence: Ritual Action and the Perception of Violence in an Upper Guinea Ethnic Conflict. Niels Kastfelt (ed.) Religion and African Civil Wars, pp. 147.171. London: Hurst & Co.

Højbjerg, Christian K. 2007. Resisting State Iconoclasm Among the Loma of Guinea. Durham, NC: Carolina Academic Press.

Højbjerg, Christian K. 2008. Le "Master General" et "Madame le Maire": structuration d'un ordre emergent de gouvernance dans l.hinterland liberien. Politique africaine 111:68-89.

Højbjerg, Christian K. 2009. Victims and heroes: Manding historical imagination in a conflict-ridden border region (Liberia-Guinea). To appear in Jacqueline Knorr and Wilson Trajano-Filho (eds.) The Powerful Presence of the Past. Historical Dimensions of Integration and Conflict in the Upper Guinea Coast. Leiden: Brill.

Kaldor, Mary. 2004. New and Old Wars: Organized Violence in a Global Era. Cambridge: Polity Press.

Kalyvas Stathis N. 2003. The Ontology of "Political Violence": Action and Identity in Civil Wars. Perspective on Politics 1:475-494.

Kapferer, Bruce. 2004. Introduction: Old Permutations, New Formations? War, State, and Global Transgression. B. Kapferer (ed.) State, Sovereignty, War. Civil Violence in Emerging Global Realities, pp. 1-15. New York, Oxford: Berghahn Books.

Kaplan, R.D. 1994. The Coming Anarchy. How Scarcity, Crime, Overpopulation, Tribalism, and Disease are Rapidly destroying the Social Fabric of our Planet. The Atlantic Monthly, February pp. 44-76.

Keen, David. 2005. Conflict & Collusion in Sierra Leone. New York: International Peace Academy, Oxford: James Currey.

King, Charles. 2004. The Micropolitics of Social Violence. World Politics 56:431-455.

Le Billon, Philippe. 2001. The political ecology of war: natural resources and armed conflicts. Political Geography 20:561-584.

Mantz, Jeffrey W. 2008. Improvisational economies: Coltan production in the eastern Congo. Social Anthropology / Anthropologie Sociale 16 (1):34-50.

Reno, William. 2001. How sovereignty matters: international markets and the political economy of local politics in weak states. Thomas Callaghy, Ronald Kassimir, and Robert Latham (eds.) Intervention & Transnationalism in Africa, pp. 197-215. Cambridge: Cambridge University Press.

Reno, William. 2003. African Conflicts, Colonialism, and Contemporary Intervention. ASAAP Conference. Key Note Address.

Reno, William. 2007. Patronage Politics and the Behavior of Armed Groups. Civil Wars 9:324-342.

Reyna, Stephen. 2008. Global Warring Today. "Maybe Somebody Needs to Explain". Social Analysis 52:50-71.

Richards, Paul. 1996. Fighting for the Rain Forest. War, Youth & Resources in Sierra Leone. IAI, Heinemann, James Currey. Oxford, Portsmouth

Richards, Paul. 2005a. New War. An Ethnographic Approach. Paul Richards (ed.) No Peace No War. An Anthropology of Contemporary Conflict, pp. 1-21. Oxford: James Currey and Athens: Ohio University Press.

Richards, Paul. 2005b. To fight or to farm? Agrarian dimensions of the Mano River conflicts (Liberia and Sierra Leone). African Affairs 104:571-590.

Sawyer, Amos. 2004. Violent conflicts and governance challenges in West Africa: the case of the Mano River basin area. J. of Modern African Studies 42:437-463.

Schlee, Günther. 2004. Taking Sides and Constructing Identities: Reflections on Conflict Theory. J. Roy. Anthrop. Inst. (N.S.) 10:135-156.