• civil wars;
  • lineages;
  • forests;
  • land tenure;
  • ethnicity


  1. Top of page
  2. Abstract

This paper examines agrarian issues in civil wars in Côte d’Ivoire and Sierra Leone. Attention is paid to two different ways in which lineage society evolved during the colonial and post-colonial periods. The motivations of fighters are related to these different trajectories of agrarian social change. In Côte d’Ivoire youth militia fought to uphold a lineage-based social order, but in Sierra Leone a comparable group of young fighters sought to overturn it. Large migrant populations on a forest frontier are an important factor in Côte d’Ivoire, while in Sierra Leone significance attaches to an excluded agrarian underclass. Not all African conflicts are ethnic conflicts; autochthony is shown to be a factor in one conflict and class in the other. Approaches to post-war reconstruction based on undifferentiated notions of community should be resisted.


  1. Top of page
  2. Abstract

What part might agrarian factors have played in recent West African civil wars? Here, we pay attention to two such conflicts – Côte d’Ivoire (2002–7) and Sierra Leone (1991–2002). Competition over land clearly contributed to tensions fuelling war in Côte d’Ivoire (Losch 2000; Banégas and Losch 2002; Bouquet 2005; Chauveau 2000). In Sierra Leone the causes of the fighting have been speculatively traced to rootless urban youth (Abdullah 1997; Kandeh 2001; Mkandawire 2002) or the lure of ‘blood diamonds’ (Collier 2000; Smillie et al. 2000). But a randomized large-sample study of demobilized ex-combatants revealed that a majority of fighters (c. 85 per cent) were rural in background and little motivated by minerals (Humphreys and Weinstein 2004), suggesting the appropriateness of an agrarian analytic focus in Sierra Leone also. Comparing two specific regions (central-western Côte d’Ivoire and eastern Sierra Leone), we find belligerents were recruited from among groups detached from or marginalized within local agrarian social institutions. What is different about the two cases, however, is that in central-western Côte d’Ivoire the violence was directed at outsiders and in defence of community institutions, whereas in eastern Sierra Leone it was targeted on local leaders and against community institutions. We seek to explain these differences by reference to an analysis of variations in agrarian organizational context.


  1. Top of page
  2. Abstract

Eastern Sierra Leone and central-western Côte d’Ivoire belong to the western African (Upper Guinean) rainforest block (henceforth UGF). The UGF has been transformed by hunters and farmers over several millennia, and by long-distance trade over several centuries. The (savanna) origins of West African agriculture can be traced back 3,000 years.1 It is assumed the outer portions of the UGF were converted to farming, and influenced by long-distance trade, over a longer period than areas in the core. The core of the UGF – a zone in south-eastern Liberia and south-western Côte d’Ivoire – remained a major forest frontier until the twentieth century. Expansion into the UGF core by planters resulted in Côte d’Ivoire becoming the world's largest producer of cocoa in the second half of the twentieth century.

The distinction between core and outer areas of the UGF is supported by human genetics. Communities on the edge of the forest have high rates of human sickle cell gene mutation (HbS); populations closer to the core have lower rates. Livingstone (1958) suggested that high exposure to malaria (selecting for HbS as a balanced polymorphism) was influenced by the spread of intensive (rice) agriculture, but later argued that HbS gene frequencies might be better explained by association with Islam and trade (Livingstone 1989). Populations characterized by low HbS frequencies, on the UGF forest frontier, would be those resisting the influence of long-distance trade, and escaping the mixing of populations implied by a vigorous internal West African trade in slaves.2

The indigenous communities of the UGF core today retain social characteristics associated with pioneer agriculture in the forest core. Access to land is based on descent from first settlers. Governance structures are decentralized. Elders arbitrate disputes and community organization rests upon a combination of lineage and age-based associations. Reflecting the crucial significance of gang-labour in heavy forest clearing, these associations continue to exercise strong influence over local decision-making processes (Moran 2006, ch 6). There is little to stop the young from forming new communities in the forest. These pioneer communities have a history of seeking to limit the activities of Muslim-Mande traders, e.g. by not allowing them to settle (Ford 1992; Massing 1980).3

Ranked lineages, chieftaincy and a large labouring underclass (originally formed through acquisition and integration of slaves) are characteristic of the agrarian communities of the forest margins (Little 1951; Gibbs 1965; D’Azevedo 1962, 1969–71). Young men combine for agricultural work but their decision-making influence over village politics (reflecting the slave origins of many) tends to be lower than in the UGF core. Political power is monopolized by chiefs and high-ranking officials of male and female ‘secret societies’ (Little 1965). The ranked forest communities encourage long-distance trade; a specific institution, the avunculate (preferential marriage with the daughter of the mother's brother), brings wealthy ‘strangers’ (e.g. traders) into the governance process (Murphy and Bledsoe 1987).

In short, biological, ethnographic and historical data support a distinction between trade-linked slave-based ranked lineage societies on the forest margins, and egalitarian communities characterized by the labour of ‘free youth’ at the core. As we will show, some of these differences are long-lasting, and are now incorporated in the institutional fabric of rural cash-cropping society. They re-surface (we argue) in the motivations of militia fighters, and are thus important in understanding civil war in the region.


  1. Top of page
  2. Abstract

A lineage society is one organized for transfer of rights of access to land and other productive resources by lines of descent.4 In the 1960s and 1970s French Marxist anthropologists engaged in a debate about the Lineage Mode of Production (LMP), i.e. the organization of the economy in lineage society. Different models of the LMP, once seen as competing formulations, will be here treated as regional variants, and used heuristically to organize our data on agrarian grievances.

We focus on two versions of the LMP in terms of the degree of social hierarchy entailed. A core forest version (LMP1) comprises broadly equivalent sets of lineages (i.e. it is an egalitarian social formation). LMP2 (with variants, LMP2a and LMP2b) involves a significant degree of ranking between constituent lineages (i.e. it is an hierarchical formation). Following Douglas (1970, 1993, 2005), we distinguish between ranked and egalitarian institutions in terms of their underlying principles of inclusion and justice. Egalitarian institutions allocate resources according to tests of loyalty. In hierarchical institutions there is trade-off between status and inclusion; the functional hierarchy maintains a place for everybody, however lowly.

The debate by French anthropologists on the LMP drew especially on Western African materials. Its terms are clearly summarized by Seddon:

In France, two points of view emerged . . . First, there were those (e.g. Dupré and Rey) who claimed that class antagonisms existed, since the seniors exploited the juniors and had an interest in maintaining this situation; second, others [including Meillassoux] held that there was no class antagonism properly speaking. The debate remains open. (1978, 159)

Meillassoux (1973) based his conclusions on his analyses of the ethnography of the Gouro, in central-western Côte d’Ivoire, a typical case (when studied in the 1950s) of an UGF frontier society.5 Young men and women were strongly bonded into age-based peer groups through initiation. These age-based groups were the basis for labour mobilization for key tasks, notably clearing high forest. A strong egalitarian ethos prevailed (LMP1). Elders ruled through prestige not force. It was hard to speak of any systematic exploitation of the labour of a younger generation, or of status differentiation between lineages, so long as each lineage maintained a sufficient quota of young men to continue to tackle forest intake.

Dupré and Rey (1973) advocated a different model (LMP2) based on ranked lineages articulated with long-distance trade; here they claim there are systematic relations of exploitation between elders and juniors. With rotational fallow food-crop farming available to all and limited institutionalized coercion (no police, no standing army), the mercantilist6 chiefs of such communities exploited the logic of exchange to prevent ambitious young people from breaking away to found their own groups. In effect, elders created an ‘internal market’ in social incorporation through the manipulation of prestige goods. Only seniors could supply such items. Successful elders accumulated resources for the benefit of their own lineages, and thus sustained a society of ranked lineages.

Maintenance of authority over marriage was a means through which mercantilist lineage heads exercised control over juniors under LMP2. A chief accumulated rights in the reproductive capacities of young women by acquiring female slaves, or by polygynously marrying girls offered by client lineages.7 With marriage partners monopolized by the ‘big men’, a young man could only marry by finding a patron. The patron might help a client by paying bride wealth, or offer one of his own wives as a partner (see below). In either case the client would work off the obligation through farm labour and other forms of service, strengthening the resources and status of the patron's group. Chiefly patronage provided a place for even the poorest young man in the system, provided he worked for the elders. Dupré and Rey (1973) suggest the main sanction against young men seeking to rebel was the threat of being sold as a slave. The slave was admitted to a new group only at the base of the social pyramid (Holsoe 1977) and the threat of such dramatic declassification was a strong inducement to conform.8

Ethnographic and historical sources suggest the model fits especially well with the pre-colonial ranked lineage societies of eastern Sierra Leone and north-western Liberia.9 Chiefs (as predicted) accumulated young women both as labourers and as marriage partners for clients and male slaves, and ceded reproductive rights in women only upon payment of marriage prestations. A client repaid his patron's help in meeting these requirements for a wife through bride service (often a lengthy period of farm work). Wealth (for a man) was measured not only in terms of money and goods, but also in terms of his control over women and children, a system Bledsoe (1980) terms ‘wealth in people’. Gibbs (1965, 215), writing about Kpelle society, in north-western Liberia, refers to men as ‘wife givers’, ‘wife takers’ and ‘wife borrowers’, noting that this defined three classes (the freeborn, slaves and pawns).

Sources also confirm that (again as the model predicts) to seek marriage outside the patronage system was to invoke severe sanctions. Taking a partner without approval of the male elder deemed to have rights in the woman (‘woman damage’) today results in steep fines in local courts (see below), but in pre-colonial times might lead to imprisonment, being sold into slavery or death. A mid-nineteenth-century account by a visitor to the court of King Gendema, on the Sierra Leone–Liberia border (Jones 1983, 189), reports large numbers of young men held in chains for ‘woman damage’. In 1568, a man fleeing for his life from a liaison with a wife of the King of Zambulo, on the Sierra Leone peninsula, offered himself as a slave to the English merchant venturer John Hawkins, in return for protection (Hair 2000, 59).

Maintenance of order in ranked lineages also depended, significantly, on the work of ‘secret societies’. The western flank of the UGF is sometimes referred to as the Poro belt (D’Azevedo 1962), due to the widespread occurrence of this male power association. A women's society (Sande) is generally found alongside. Poro and Sande initiate children and prepare them for an orderly adult life. In effect, Poro and Sande monopolize knowledge of adulthood and make it available in initiation. Poro and Sande ‘juniors’ then form cohorts for a range of community tasks, but the upper echelons of the associations closely guard the secrets of political power and reproductive health (Bledsoe 1980; Little 1965). Only men and women of substance progress towards the higher ranks. Harris (a mid-nineteenth-century British trader in the Galinhas, and a Poro member) claimed the organization was dominated by the chiefs, and that no slave would be admitted even to junior ranks (Jones 1983). The ritual ‘secrets’ are ‘prestige’ items accessible only to members of the appropriate grade.

As will be explained further below, colonial authority in Sierra Leone recognized the right of elders to fine young men for marriage infringements and respected the privileges of Poro and Sande. Thus in effect, the British left in place the means to reproduce ranked lineage society. Shifts in political economy in the post-colonial period triggered a process of involution in the system. Denied state resources, rural chiefs intensified control over marriage and reproduction to a point where many young people felt they no longer had any stake in society. Respect for hierarchical authority failed, and rural class conflict erupted.

Within Côte d’Ivoire we need to take account of a more diversified economic system on the eastern flank of the UGF offering greater scope for underclass social advance. The history of the Akan forest societies of Ghana10 and eastern Côte d’Ivoire (Terray 1995) shows long-distance exchange accompanied by accumulation of slaves and pawns (for domestic production and reproduction), but also by (i) strong integration of slave descendants into descent groups, facilitated by a matrilineal kinship system, and (ii) the weakening of the monopoly of aristocratic chiefs over long-distance trade (Wilks 1993; Mikell 1992).11 This suggests that in a context of greater economic success (see below), the relationship between juniors and lineage elders was able to evolve along less confrontational lines than in the western UGF. This variant (LMP2b) was discussed by Marxist Africanists in France, though ‘African mode of production’ (Coquery-Vidrovitch 1969) is too general a term.

Something should now be added about the colonial and post-colonial evolution of the lineage institutional forms captured within the LMP model. LMP1 communities of the core were less dependent on slaves than in the peripheral parts of the UGF. Strong traditions of egalitarian ‘gang’ organization encouraged young people to adapt to a variety of labouring tasks under the colonial economy, and much volunteer labour migration ensued. These out-migrants retained expectations of re-settling in their home communities in respectable old-age. In western and central-western Côte d’Ivoire the loss of young people to urban migration after World War 2 was made good by immigration of settlers under the tutorat (see below). But in adverse economic conditions at the end of the twentieth century, many young men were forced to return from city labouring jobs, and now encounter strangers on the land. They blame elders for failing to protect the customary system, and target resentment on migrants, who serve as scapegoats.

In the ranked lineage communities of the western forest flank (LMP2a) the children of chiefs comprised a class of freeborn farmers, while former slaves became clients. The labour of clients supported plantation development for a ‘peasant’ export-oriented agriculture. Underclass youth were freed from formal ownership, but continued to work for elders in fulfilment of bride service, or via ‘customary’ requirements to make plantations or build a farm access road for the chief. Strong controls over marriage continued to be enforced, but now through means recognized by the state (e.g. fines for ‘woman damage’ imposed in ‘customary’ courts and paid in labour). After emancipation some ex-slaves left and sought work as migrants. D’Azevedo (1969–71) reports mass desertion of Gola chiefs in north-western Liberia after emancipation in 1930. But this amounted to complete rejection of the agrarian social system. Return requires explicit acceptance of chiefly ‘custom’.12

Among the matrilineal Akan on the eastern flank of the UGF (LMP2b) an indigenous diffusion of tree-crop family agriculture in a context of land abundance took place as early as the 1910s, after gathering of wild rubber had come to an end. It resulted in the emergence of a class of autochthonous ‘big planters’, many of them (but not all) originating in aristocratic lineages (Rougerie 1957; Stavenhagen 1969; Gastellu and Affou Yapi 1982; Boone 2003). In this respect, the Akan Southeast stands in contrast to the central-western and western regions of the forest zone, where ‘big planters’ were, and still are at present, non-autochthons (migrant farmers, particularly Akan, and urban elites). Colonial employment opportunities and the opening of the western pioneer front to settlement in the 1950s, when the regions of the east of Côte d’Ivoire began to experience land pressure, offered many opportunities to rural youth originating in Akan communities. Access to these new resources contributed to a release of the intra-family tensions in eastern Côte d’Ivoire, in particular those related to matrilineal inheritance. Contrary to the situation in the western UGF, young men were encouraged to create their own resources, provided this served to enhance accumulation of wealth in matrilineal family groups. This meant that rural youth from Akan communities found allies and not competitors in their elders (Gastellu 1989), particularly in accessing land on the western agrarian frontier.13

A recent development in Côte d’Ivoire, addressed further below, is that confrontation has today broken out between ‘core’ communities (LMP1) and those configured along LMP2b lines. A notable percentage of immigrants to the west, whose presence fuels the resentment of autochthonous youth with respect to their elders, are in fact young men originating in the eastern forest margins whose migration facilitates the reproduction of successful lineage society in their home communities. ‘Local anchorage’ of the state via the mobilization of rural youth and the instrumentalization of their frustrations increases scope for manipulation. A politics of ethnicity flourishes. A second factor of importance in Côte d’Ivoire but not found in Sierra Leone is the presence on the cash-cropping frontier of large numbers of foreign migrant labourers, who become a scapegoat group for the frustration of indigenes, as will be explained further.

In summary, therefore, we have three potential fault lines along which agrarian conflicts now develop in the UGF. In the core, youths fight to protect tradition against ‘lax’ elders and scapegoat the migrant foreigners the elders have admitted. In the western margins ‘class’ conflict between ranked lineage and underclass interests results in attacks on the symbols and personnel of the ranked lineage system itself. In a third case, migrant youths from the expansive communities of the eastern forest margins clash with local youth on the forest frontier in episodes of inter-ethnic violence. We will now substantiate this three-fold differentiation with detailed evidence from two case studies.


  1. Top of page
  2. Abstract

Neither Peace nor War: A Summary of Armed Confrontation in Côte d’Ivoire

Côte d’Ivoire experienced an authoritative single-party regime from independence, in 1960, until 1990. Thereafter, the succession struggle after the death of Houphouët-Boigny in 1993 and the use of ‘ivoirité’ as an ideology against Allassane Dramane Ouattara instituted ethno-nationalism as a component of the multi-party system (Losch 2000). In December 1999, a mutiny of young soldiers led to a coup d’état to the benefit of General Robert Gueï, former chief of staff dismissed by President Konan Bédié. The takeover of the country by a Comité de salut public degenerated into a confrontation between the major political leaders. Insecurity was fuelled by violence on the part of the armed forces and by civil war in neighbouring Liberia. After stormy elections and Robert Gueï's unilateral attempt to declare himself the victor, Laurent Gbagbo and his party the Front Populaire Ivoirien (FPI – Ivorian Popular Front) were recognized as winners. The organization of a Forum of National Reconciliation at the end of 2001 did not prove effective. Nevertheless, a coalition government took office and international financial aid resumed.

It was in this relatively favourable economic context that, on 19 September 2002, a group of military insurgents, many exiled in Burkina Faso, triggered a coup d’état. The attempt failed, at the cost of the assassination of several political leaders, but the rebellious Mouvement Patriotique de Côte d’Ivoire (PMCI – Patriotic Movement of Côte d’Ivoire) managed to assume control over the northern half of the country. It was quickly joined by two groups from the west (including from the region of origin of General Gueï, killed during the confrontations in Abidjan): the Mouvement pour la Justice et la Paix (MJP – Movement for Justice and Peace) and the Mouvement Populaire Ivoirien du Grand Ouest (MPIGO – Ivoirian Popular Movement of the Greater West). These groups formed a politico-military alliance called the Forces Nouvelles (FN – New Forces) led by Guillaume Soro (a former member of a pro-Gbagbo student movement). The FN were supported by the then Liberian president, Charles Taylor, an old ally of General Gueï, and benefited also from the support of President Blaise Compaoré of Burkina Faso.

It was mainly during these disorders that loyalist ‘young patriots’ mobilized. Activity in Abidjan received the greatest media coverage and attention in academic studies (Banégas 2006, 2007). Nonetheless, this mobilization by rural young people is a significant element in the wider conflict. As we will see, their activity highlights the agrarian roots of conflict in Côte d’Ivoire around questions of access to land, the value placed on the work of young men of rural origin, and their extreme mobility between town and country. Mobilization initially took the form of communal and village self-defence, including roadblocks and checkpoints which rapidly degenerated into simple racketeering until their prohibition a few months after the opening of hostilities (Chauveau and Bobo 2003). Thereafter, youth mobilization was maintained by the national or regional organizations of ‘young patriots’, of which the COJEP14 has been the most active in rural areas. The most active, durable and violent youth activism has been experienced in the forest societies of the west and the mid-west of Côte d’Ivoire (in communities we consider to be configured on lines of the LMP1 model). This is especially the case in the forests of the extreme west, on the Liberian border, where rebel and governmental forces used Liberian and Sierra Leonean mercenaries, and where pro-government local political leaders raised ethnic militia (International Crisis Group 2003, 2004, 2005). But even in these zones, this militia remained under the control of political leaders and local neo-traditional authorities (again this contrasts with the Sierra Leone case), in spite of the distrust occasionally expressed by the youth regarding administrative and political authorities.

After the Linas-Marcoussis Agreements (near Paris) in January 2003, several attempts to resolve the conflict resulted in breached peace agreements. A UN peace-keeping force (including French soldiers) was put in charge of a ‘Zone of Confidence’ separating belligerent forces from north and south. Direct confrontations between governmental and rebel forces was limited thereafter.15 In this situation of ‘neither peace nor war’, rural mobilization of youth was relatively mild when compared to the violence seen in Sierra Leone.

A peace agreement was recently negotiated and signed without the aid of international authorities, but with brokerage by President Blaise Compaoré of Burkina Faso, between President Gbagbo and the chief of the FN, Guillaume Soro (Ouagadougou Agreement, 4 March 2007). Some of its provisions have already been implemented; these include a new transition government led by Prime Minister Guillaume Soro and including representatives of opposition parties, a central integrated military command comprising rebel and government armies, the dismantling of the buffer zone and the redeployment of administrative staff in the northern zone. But at least two of the provisions of the new agreement have encountered important obstacles, at the national political level and in terms of the local balance of power. One concerns the resumption of the process of citizen and resident identification involving ‘open audiences’, to determine electoral registration and issue of national ID cards. The other is the process of disarmament, demobilization and rehabilitation (DDR) of the combatants of the FN in the north and pro-government militia in the south (especially in the south-west). In both cases these provisions risk reigniting, in different ways, the grievances and demands of rural youth.

Agrarian Context

In forested Côte d’Ivoire the smallholder coffee and cocoa economy expanded from the 1920s. Its development relied on the arrival in areas of low population of many immigrants from drier regions unsuited for coffee and cocoa cultivation (Upper Volta, Mali, and the northern and central savannah districts of Côte d’Ivoire – especially Baoulé[Bawle] people). They came as wage labourers but also looked for forested land to create plantations (Chauveau and Dozon 1985; Chauveau and Léonard 1996). Currently, the western part of the Ivorian forest belt supplies most of the country's cocoa exports.

The Oumé district (sous-préfecture), located in the centre-western part, is a former pioneer front where coffee and cocoa cultivation expanded from the 1950s. Indigenous communities (Gban and Gouro) represent less than one quarter of the rural population. Baoulés account for more than one third, and non-nationals about one third, mainly originating in Sahelian countries (people of Burkinabé origin alone make up about a quarter). The Gban (the population under study) were a politically loosely articulated segmentary society, with village chiefs and district headmen (chefs de canton) introduced during the colonial period. Access to land depended on belonging to or dealing with localized exogamous patrilineages (gligba, sing.) and lineage segments. At present, the lineage cults related to land have largely fallen into disuse, although ‘negligence’ of them is regularly cited as one of the causes of ‘misfortunes’ striking lineages or villages (Chauveau 2006).

The relationship between autochthons and ‘strangers’ (whether Ivorian or of foreign origin) regarding access to land therefore structures the land issue in Southern Côte d’Ivoire (SCI).16 This relationship was most often characterized by the social embeddedness of land right transfers in the tutorat institution (Colin 2005; Chauveau 2006). The tutorat,17 commonly found in rural societies in West Africa, is an agrarian institutional device for regulating first comer–late comer relations.18 It belongs to a moral economy in which one cannot refuse an outsider who needs land as a means of subsistence. The ‘stranger’ must respect the bundle of duties associated with his social incorporation in the local community and consequently contributes to the reinforcement of the community under the locally prevailing social order. The delegation of rights on a plot of land by the tuteur is accompanied by a patron–client relationship. The stranger owes his tuteur perennial gratitude (an obligation transferred to his heirs), expressed through the gift of agricultural products, the contribution to his tuteur's expenses at times of obsequies, schooling expenditures, etc. A key point to underline is that land transfers and the socio-political dimensions of the tutorat are intimately entangled; as a social institution, the tutorat regulates both the transfer of land rights and the incorporation of the ‘strangers’ in the local community.

Tensions and conflicts between autochthons and settlers are well documented for southern Côte d’Ivoire (Raulin 1957; Dupire 1960; Köbben 1963; Hecht 1985; Chauveau 2000). They are often rooted in disputes over the content of land rights and duties transferred to the settlers, whether as ‘gift’, ‘purchase’ or ‘lease’. Under the tutorat, payment of money does not conclude the transaction or relationship; in fact it institutes or perpetuates it. As will be explained below, the conflicting interpretations of rights and duties embedded in the tutorat relationship were kept under control during the colonial period as well as during the first decades of Independence to the detriment of the autochthons, as public policies tended to favour the settlement of strangers.

The issue has resurfaced openly and widely in the past ten years, with autochthons contesting earlier land transfers in order to recover a ‘land due’ or reclaim the land. In this process, young autochthons lacking economic prospects play a leading role. The political debate has taken up the question of land property rights in a major way, in a context in which public authorities no longer support indiscriminately the settlers’ interest. The era after the death of the pro-settler President Houphouët-Boigny marks ‘the return of the autochthony’ in the guise of an ideology of ‘ivoirité’ (Losch 2000). The 1998 land law, voted unanimously, clearly goes that way by excluding foreigners from land ownership. But the law also implicitly privileges autochthony as the main source of legitimate entitlement to ownership rights, opening up the possibility of an exclusion of Ivorian Baoulé and Dioula19 from legal registration (Chauveau 2002). The land issue has become a burning question in the current socio-political context, especially since the onset of civil war.

Explaining ‘Young Patriot’ Mobilization: An Argument

After the peace agreement signed in March 2007, Oumé district now experiences ‘neither peace nor war’. It is to a large extent less insecure than areas closer to the Liberian border. For instance, migrant farmers have remained in the villages since the beginning of the ‘war’. But in contrast with the Akan variant of LMP2b in the eastern part of the forest belt, social tensions have developed during the conflict between indigenous populations and settlers, and also among indigenes, between young people and village and family authorities. These latter tensions are related to power struggles on both local and national levels (resignation or dismissal of Gban village chiefs to the advantage of ‘young chiefs’; constitution of self-defence associations of young patriots during the first months of the war, almost exclusively recruited among autochthons; sporadic manifestations of hostility against the high-rank civil servants and governmental soldiers at this moment).

Efforts made by governmental political leaders to mobilize the population at the beginning of the conflict gave a recognized role to rural Gban youth as patriots defending the homeland (comparable to the CDF in Sierra Leone). As a result, their position in the village political arena was reinforced. Self-defence roadblocks gave them, for a while, economic dividends from patriotic mobilization (in particular by extracting money from strangers, even if resident in the same village). For the most part, these young people were in a difficult situation. Having failed as urban migrants, they had only marginal access to family land, much of which was occupied by settlers. However, the young ‘road blockers’ remained dependent on the authority of village chiefs and notables. Public authorities and chiefs later prohibited road blocking and, despite an atmosphere of mutual mistrust, Burkinabé and Dioula ‘strangers’ were able to resume their activities (Chauveau and Bobo 2003). But the self-defence associations of rural young patriots remain in the limelight, even if the peace agreement signed in March 2007 seems to contribute to appeasement.20

The problem of ‘strangers’, non-Ivorians and Ivorians alike, constitutes a kind of common denominator behind the civic mobilization of these young people. Strangers are blamed for monopolizing ancestral land, and suspected of supporting rebels from the North or, if Baoulé, of being partisans of the PDCI-RDA,21 the former ruling party fostering settler immigration from the 1960s. The ‘old’ chiefs, longstanding supporters of the PDCI, are accused of collusion with strangers to the detriment of the youth who have little or no land. The reference to the ideology of ‘autochthony’ thus resonates with the nationalistic and patriotic agenda, but also with desires for restoration of the customary moral economy that elders are thought to have betrayed.

The mass pro-governmental mobilization of Gban youth is related to the economic vulnerability of rural youth. Fierce political competition makes young people easy prey for political factions offering violence as an outlet for frustration and as a source of income. These elements certainly come into play. But they suggest that the behaviour of these youth is merely a reaction to crises and external events. In our view, the explanations behind the mobilization of youth and the social strain that it expresses, particularly in terms of inclusive/exclusive definition of ‘belonging’, are to be found in the internal dynamics of the society. We need to ‘re-socialize’ and ‘re-historicize’ the ‘problem of the youth’ (Chauveau 2005a, 2005b; see Richards 2005b for a generalization of such an approach).

Positioning youth in relation to elders, and reconnecting generational-based differentiation with other registers of social inclusion and exclusion were precisely what the notion of lineage mode of production (LMP), used as a pragmatic concept, can help us to do. Regarding the ‘Gban homeland’ and more generally central-western Côte d’Ivoire, our argument is twofold. First, an ‘egalitarian version’ of the LMP prevailed during the agrarian history of forest conversion and encapsulated a specific trajectory of ‘non-class-based’ agrarian development in this region. Second, the agrarian trajectory underpinning conflict can explain, among other factors, why, when very adverse economic conditions occurred at the end of the twentieth century, the customary moral economy still retained the respect of youth despite the growing tensions inside families, leading to a range of claims about a purified tradition, and to the victimization of strangers (in sharp contrast with the case of ‘class-based’ agrarian trajectory in eastern Sierra Leone as well as in eastern Akan region). We describe now the main periods of agrarian change that led to the present situation and fed the conflict.22

Gban Communities Face Migrant Farmers: The Programme of an Emerging Political Elite

The first cocoa farms around Oumé, the district capital, were established from the 1920s by European and Syrian-Lebanese farmers, Baoulé clerks and Dioula traders. The years of economic recovery, following the recession of 1929–1933, saw a new influx of immigrants, both Baoulé and migrants from Upper Volta (Burkina Faso). Among the Gban, weakly marked by pre-colonial mercantilism and forced labour, the first cocoa and coffee farms generally belonged to village headmen and ‘big men’ whose close links to colonial authorities gave them access to forced labour (cf. Sierra Leone case study below), including the labour of young autochthons. As early as this time, the scale of their farms was a lot less large than the rich Akan planters in the eastern region of Ivorian UGF, who employed migrant and family labour (a part of them from slave origin).

Increased mobility following the abolition of the Native Code (Code de l’Indigénat) and of forced labour in 1947 led to further tree-crop planting (dominated at that time by coffee), but also the arrival of further migrants, especially Baoulé. They asked local chiefs and ‘big men’ to provide access to land under the framework of the tutorat. However, beginning in the mid-1950s, when the first plantations began to yield, local tuteurs demanded that customary ‘gratitude’ become a substantial fee. Local conventions between tuteurs and Baoulé settlers resulted in tensions within the indigenous communities regarding the distribution of this compensation among the villages, lineages and family heads. This notwithstanding, Baoulé migrants obtained access to land by negotiating with heads of family groups. This resulted in serious disputes between villages and district headmen, as well as between lineage and family heads and, finally, between generations.

From 1947, the abolition of forced labour freed the Gban younger generation from the manpower demands of local notables. Moreover, freedom of movement granted to all African people gave youth new room for manoeuvre via emigration. Encouraged by Abidjan's rapid economic growth (out-migration accounted for 6 per cent of Gban population in 1951), the young could now acquire part of their bride wealth payments independent of elders and participation in family farming. Conversely, the fees settlers paid to family heads allowed the latter to counterbalance their labour losses due to the withdrawal of youth from family farms. But the corollary of this involvement of Gban youth in new urban sources of income via emigration was a degree of voluntary exclusion from local (land-linked) institutions. This is to be seen especially in the religious sphere. Intergenerational tensions reached such heights that administrative reports of the 1950s describe a real state of ideological–religious confrontation between the ‘young’ and the ‘elders’. Young Gban adhered massively to ‘new’ religions (Catholicism and a syncretic Deima cult) which opposed both the sorcery and fetishism of the elders, and conversion by certain notables to Islam. It was these young Gban who also adhered in large numbers to the newly-founded PDCI-RDA, an anti-colonial political party, despite colonial repression and the opposition of a majority of Gban chiefs, who objected to the alien origin of the local leaders of the PDCI (principally Baoulé and Dioula).

A massive shift of rural population to the pioneer forested lands of the west was encouraged by the PDCI-RDA party, backed by the African farmers’ union (SAA), and more particularly by their Baoulé leader, Félix Houphouët-Boigny. Baoulé claims against increasingly heavy land fees under cover of the tutorat were supported, under the banner of PDCI-RDA party, by the emerging Ivorian political elite, a good part of them having vested interests in large coffee or cocoa farms located outside their home region. Furthermore, the establishment in the colony of ‘internal autonomy’ (i.e. home rule) in 1956 favoured a perception by the political elite that control of the coffee and cocoa sector represented not only the key to the economic growth of the country but also a base to strengthen a clientelistic system of power brokerage. At full independence (1960), what was at stake was clearly the expansion of the western agricultural frontier (the plan for the ‘mise en valeur de l’Ouest’) free as much as possible from claims by autochthons over settlers. This was clearly the case in the Oumé region, at the time one of the most important places for new planting.

The Politics of the Tutorat as an Evolving Institution on an ‘Internal Frontier’

It is not in the implementation of official development plans or of legal mechanisms, however, that we find the really decisive interventions on the part of the government and the ruling elite in order to open the western forest belt to settlers,23 but in a range of other interventions of a fundamentally political and clientelistic nature (political pressures exercised on the villages’ authorities to welcome strangers, and protection of the latter in case of conflicts). The alleged under-settlement of the Gban forest and the weak development of the available forest resources was justification to welcome ‘brothers’ in need for land. The moral and cultural references to the tutorat made by the administrative authorities was accompanied by instructions, which aimed at closely framing its implementation. These instructions were grounded in the order issued by Houphouet-Boigny himself in 1963: ‘la terre appartient à celui qui la met en valeur’ (the land belongs to those who develop it). This dictum assumed the force of law, though it completely contradicted the provisions of legislation inherited from the colonial period, and formally forbade land fees to be levied on the settlers.

But the post-colonial state exercised hegemony on a shoestring. From the 1960s to the 1980s, it left the practical arrangements regarding the transfer of land rights to an increasingly large number of migrants largely in the hands of local political brokers, and of the micro politics prevailing within local multiethnic communities. Progressively, Gban tuteurs circumvented the state's prohibition on land fees by incorporating an increasingly large monetary component in the initial symbolic gifts sanctioning the stranger's access to land, and by increasing and monetizing demands regarding subsequent ‘duties of gratitude’.24

This inventive use of custom, however, gave rise to competing interpretations concerning the bundle of rights effectively transferred to the settlers. On the one hand, the generalization of transfers of land rights to settlers gradually transformed them into clients, henceforth subject to illegal and clandestine land fees. On the other hand, the tendency to increase and monetize social obligations in return for access to land encouraged the idea among settlers that they had engaged in a purchase–sale transaction. From the autochthons’ perspective (and, generally, for settlers also) an economic transaction in no way cancels the moral and political obligation of ‘gratefulness’ the settler owes to a Gban tuteur– even if the Gban often use the term ‘sale’ to qualify these transfers.

Despite tensions apparent even at the time, the benefits (even if very unequally distributed) from economic growth during the 1960s and 1970s (due to westward agricultural colonization of the country's forests) were sufficiently large for autochthons and other stakeholders to submit to government policy on agricultural colonization under the guise of the tutorat. The ‘Ivorian miracle’ gave the state the means to guarantee fixed prices to farmers, to subsidize imports and to improve the general standard of living. However, in return, the government expected that farmers submit politically and acknowledge the legitimacy of the state party and its local agents in interfering in local land matters. This informal pact also included specific compromises between the state party and the Ivorian and non-Ivorian immigrants, who gained protected access to land in exchange for electoral support (non-Ivorian residents were authorized to vote up to 1990).

The compromise concerned also autochthonous young people, whose land access was threatened by land transfers to settlers. Their support was gained through free schooling, access to urban employment and – at least in theory – assistance once they wished to establish themselves as ‘modern farmers’. In effect the state became their patron, and not the local elders (see Sierra Leone case study below). Given the difficulties involved in having their own farms, and in the context of an expansion of administrative and urban employment, Gban youth based their future plans on an urban model of livelihood.

Considerable investment was made in education, and the access to employment was partially guaranteed during the boom years between 1960 and 1970. Strategies of urban schooling and migration drew the support of lineage and family heads. Covering the expenses incurred in education and urban migration of sons was one of the reasons behind the pressure Gban tuteurs exerted on strangers, and was a factor in pushing elders to transfer more land to new strangers. Gban elders generalized use of labour and agrarian contracts (specially sharecropping) with migrant workers to compensate for loss of family labour. Urban migration by Gban youth facilitated land transfers to immigrants while effectively justifying, for a time, divestment of land opportunities by young people. When it became apparent that the civil service and the urban private employment market were not enough to absorb all rural youth migrants, a policy to turn autochthonous youth into ‘modern planters’ was implemented during the 1970s. This was to little avail. First, government projects to settle young Gban farmers clashed with the land prerogatives of lineage and family elders. The second and major reason was that young candidates had no strong inclination towards agricultural labour.

To use a term inspired by Kopytoff (1987), these developments changed the neo-traditional tutorat relationship into a multilayered ‘internal frontier institution’. The tutorat structured an interface between a plurality of social spheres: intra-communal relations, local political brokers and the state, the micro-politics of belonging in multi-ethnic communities, land relations between customary owners and settlers, and, last but not least, the intergenerational relations between autochthonous elders and youth. In particular, a growing part of youth expenditure for schooling and urban migration was supplied indirectly by strangers tied to elders. By and large, Gban communities, less transformed by mercantilism since the pre-colonial period, followed a trajectory of ‘non-class-based’ agrarian development, in which potential conflict between the generations was managed under the neo-customary institution of tutorat, a device for accumulating surpluses from migrant workers, and not from a lineage under-class of former slaves (as in the Sierra Leone case). At the same time, Gban elders were unable to develop a local class of ‘big planters’, in sharp contrast with the eastern Akan region. On the contrary, local big planters in Oumé district were found among Akan Baoulé and, progressively, Burkinabé migrants. The leading role of the ‘strangers’ in the process of socio-economic differentiation was going to be a major cause of Gban youth's hostility.

Intra-family and Autochthon–Settler tensions: The Politicization of the Tutorat

Since the mid-1980s, the urban economic crisis and the failure of the model for social progress based on education and urban migration has resulted in a significant return by townspeople, especially unemployed young people, to their home villages, at the very time decreasing cocoa prices and the effects of liberalization policy were making local elders more dependent on rents from the tutorat relationship with settlers. During these fifteen or twenty disturbed years, a range of grievances became stronger preoccupations for the local population, particularly the mobile youth. It is these claims that conflict has picked up. They concern mainly the ‘policy of the elders’ with regard to the management of family land and income derived from strangers.

Once back in the village, many ‘home-coming youths’ are at a disadvantage in claiming family land occupied by the ‘old men’, by brothers remaining in the village, and by migrants on plots ceded through the tutorat. The difficulties they encounter back in their own villages lead them not to give up on their urban projects. An entire population of ‘rurbanized’ youth therefore has developed in Gban villages. They are neither completely excluded from the rural milieu, nor completely cut off from the urban network. Some of them take part in work ‘societies’ in employment with their elders. But they encounter competition from migrant labourers from the North and sometimes run up against the unwillingness of the ‘old men’ to remunerate the work of their ‘children’ (contrary to the situation in the Akan region in the south-east of Côte d’Ivoire, where employment of family members as share-croppers is quite common). Certain young people exposed to education and urban skills ‘get by’ with ‘petty civil service’ functions in the small village bureaucracies (serving co-operatives, or as secretaries to village chiefs, or parent-teacher associations) or working in casual jobs (as petty traders, touts for cocoa buyers and timber yards, etc.). But there are no major agricultural innovations. By and large, the production systems remain centred on traditional small-holder coffee and cocoa production. For the most part, innovations concern diversification of forms of agrarian contracts and transactions (share-cropping, rental, loan, the use of ‘small papers’ in land transactions) between autochthons and migrants – a clear reflection of the extent to which the local elders depend on tutorat income and on the labour force of migrant workers from the North.

The youth have acquired autonomy in only one particular case: access to marriage. Bride wealth payments and work in return for marriage – a point of control in the class-stratified variant of the LMP – have become marginal.25 But there is open tension in families surrounding the payment of school fees, the funding of urban migrations, family work and inheritance, and the distribution of tutorat income. Young people try to finance their urban migration and – a new phenomenon – their migration to Europe through clandestine ‘sale’ of portions of forest situated far from the village and even through the sale of family plantations.

Tensions within families and between generations in turn feed tensions between autochthons and settlers. The young people or townsmen who went ‘back to the village’ exercise strong pressure on their family heads to recover portions of land transferred to migrants, and some have even tried to intimidate settlers and regain parcels by force. Furthermore, the death of the initial tuteur or settler often favours a renegotiation by the tuteur's heir or a reaffirmation of the tutorat relation regarding the settler's heir. Throughout the western forest area, this crisis is manifested in the politicization of the land issue. Increasing land shortage and loss of confidence in government policy encourages the urban elite and local politicians to play the ethnic card in order to ensure their own political survival. The ideology of autochthony and the protection of ‘traditional’ land rights has become an increasingly effective electoral argument.

Young people, urban elite members originating in Gban villages and local politicians push for the renegotiation of the conditions under which land rights were transferred, and justify their demands by evoking a return to ‘autochthonous’ basic principles. Their efforts have been aimed in particular at restricting the bundle of rights transferred to Burkinabé farmers who, during this period of deep economic crisis and neo-liberal reform, had access to an abundant labour force and could continue purchasing plots. The complaints addressed towards Burkinabé and Dioula immigrants are the same that were addressed towards Baoulé settlers some decades ago, when the latter were perceived as beneficiaries of state protection: ‘they get richer and richer whereas local families are struggling with problems’, ‘they invest at home the money they earn locally’, ‘they show no interest in the village affairs, they no more respect their gratefulness duty towards the whole local community’, and so forth.

The announcement of the passing of the 1998 law on the domaine foncier rural, which has not yet been officially implemented, has bolstered the claims put forward by autochthons that they are the only legitimate holders of property rights. The law actually gives priority to ‘customary rights’ in the process of identifying and certifying land rights prior to their registration. It aims at converting the transfer of land rights between autochthonous tuteurs and foreign settlers, who cannot legally hold property, into a formal lease contract. However, it is quite likely that many settlers of Ivorian citizenship will not be acknowledged as the owners of their plantations by their tuteurs either, since ‘complete’ customary rights are based on autochthony.

The issue of land transfers to migrants is clearly noticeable in the political life of Gban villages. Since the beginning of the armed conflict, and even before the advent of violence, the questioning of village headmen and elected representatives considered too conciliatory vis-à-vis the migrants is indicative of the politicization of the institution of the tutorat. This politicization often also underpins deal-making in regard to village chieftaincies, where there is now a tendency to back urban elite members or young townsmen who went ‘back to the village’ in elections for headmen, because they are regarded as likely to be tougher than the former chiefs with respect to the migrants, and less likely to be corrupted in defence of village interests, in particular the defence of the interests of youth and non-residents. The conflict also particularly intensifies the victimization of the ‘strangers’, whether Ivorian or not. Burkinabés, northern settlers (often suspected to be foreigners, and supporters of the ‘rebels’), the supporters of A. D. Ouattara, the northern presidential opponent of Gbagbo, and Baoulé settler supporters of the PDCI, the former ruling party, are all likely candidates for suspicion.

Some Conclusions on the Côte d’Ivoire Case

The economic crisis, in both urban and rural environments, and the hijacking of people's frustrations by national leaders competing for control of the financial resources of the state contribute to explain both the vulnerability of rural Gban youth and their menacing attitude vis-à-vis the strangers.26 But in Gban country, as in central-western Côte d’Ivoire more widely, current economic and political crises feed on a particular historical trajectory, deep-rooted in issues pinpointed in the LMP model, and it is here that the effects of these crises have surfaced in intra-familial, intergenerational and intercommunity tensions. The involvement of Gban youth in the present conflict takes on its full significance in the light of principles of LMP1 social organization, the many profound mutations that have occurred since the pre-colonial period notwithstanding.

The model helps us understand the range of pre-existent claims entertained by mobile rural youth and nurturing violent conflict. During the conflict, as before the advent of violence, the main claims staked by most of the rural youth concern the ‘policy of the elders’– including local chiefs and politicians – with regard to the management of family land and income derived from strangers. The ‘strangers’, whether Ivorian or not, are victimized as the main beneficiaries of the past policy of settlement supported by the ‘PDCI state’ (allegedly) used by elders for their own ends. It is these same claims that the conflict picks up and amplifies in the course of collective action by road-blockers’ and young patriots’ groups.

But in contrast to eastern Sierra Leone, where ranked lineages and class-stratified communities have been shaped by mercantilism over a longer period, Gban communities are less transformed by mercantilism. An ‘egalitarian version’ of the LMP prevailing during agrarian transition in the pre-colonial period has provided a matrix of ideas and sentiment reworked in the agrarian developments of the colonial and post-colonial periods. In particular, the loss of young people to urban migration was made good by immigration of settlers, under the tutorat. It was the migrants, not the local elders, who were the leading actors of the penetration of the mercantilism. But this has created a dilemma, since the succession rights of the younger generation are now vested in two groups – the immigrants who met the shortfall of youth work, and the young autochthons that egalitarianism once benefited. When very adverse economic conditions happened from the 1990s, the prevalence of the ‘egalitarian version’ of the LMP then contributed to the twofold perception Gban young people have of the land question. On one hand, despite the growing tensions inside the families, the customary moral economy still retains the respect of youth. The claims staked by the young rural ‘patriots’ do not question the symbolic system of a gerontocratic order from which all members of the younger generation stand eventually to benefit. Unlike the deracinated underclass rebels in eastern Sierra Leone, for whom the local patrimonial hierarchy no longer has a place, Gban youth call for a renewal of ‘custom’ and of commitment to the primacy of membership in autochthonous communities as the basis for access to land. On the other hand, they dispute the elders’ monopoly of decision making over that land. It is the elders, not the institution, that appear to have failed. These claims are expressed especially as reproaches to the elders for their failure to ensure the means of a decent livelihood for youth, and focus in particular on the presumed economic collusion between elders and ‘strangers’. By linking intra-familial and intergenerational tensions on one side and intercommunity tensions on the other side, an answer seems to lie in purified tradition, leading to the victimization of the strangers. A political discourse of ‘autochthony’ ensues, seized upon by various national political interests for their own purposes.

However, increasingly difficult conditions for access to land resources is not the only issue to be addressed. There is also an issue of social advancement for rural youths who have often experienced failure in their attempts to make their living in urban contexts. Attention here should focus on the emergence of new settlement and livelihood systems not easily captured by a simple urban–rural dichotomy, where a growing proportion of income comes from non-agricultural sources. Also involved are claims staked by youth regarding local governance as practised by local and public authorities. One decisive element that transformed the feeling of land shortage into a feeling of land dispossession has been authoritative state intervention. As far as the settlement of migrants was the result, or considered by the local actors as the result, of past state policy, local tensions concerning the tutorat and landed property relations are entangled in both a local politics of belonging and in supra local debates about political brokerage and state legitimacy. These claims surface in a mixture of ‘autochthony’, nationalistic reference and mistrust vis-à-vis those in legal authority, a mixture that makes sense in the historical and political context of the agrarian pioneer front of western Côte d’Ivoire.


  1. Top of page
  2. Abstract

A Brief Account of the War

The rebel protagonist in the war in Sierra Leone called itself the Revolutionary United Front (RUF) and claimed Libyan support. The Benghazi-trained organizers first joined the Liberian uprising of Charles Taylor, before launching their own armed struggle in Sierra Leone in March–April 1991. Two small armed groups crossed into Kailahun and Pujehun Districts and formed bases in and around the Gola Forest. The leader of the northern group was an ex-soldier, Foday Sankoh, with links to the banned Sierra Leone People's Party (SLPP). Sankoh used widespread sympathy for the SLPP in Kailahun to recruit to the movement. The leader of the southern group – Rashid Mansaray, a Green Book student activist – rallied elements from an earlier but localized rural rebellion in Pujehun District in the 1980s.27

The RUF at first sought to capture territory, and administer it on Green Book lines, including establishment of ‘people's courts’.28 Government forces later pushed the RUF into forest recesses along the Liberian border, where it then formed training camps and launched far-reaching pin-prick raids. Uncertain how to deal with a persistent low-level guerrilla insurgency, the government (a military regime from 1992) offered a lucrative kimberlite (hard rock) diamond mining contract to a British (Canadian-registered) mining company with links to former South African army experts, to help pay for a counter-insurgency strategy devised by veterans of ‘dirty wars’ in Angola and Mozambique (Hooper 2003, appendix).29 Pressing for elections as the solution to the problems in Sierra Leone, the international community excluded the RUF from that process.30 After elections and change of regime in early 1996, the South Africans began to mobilize, train and arm a national civil defence force (henceforth CDF) to replace the ineffective government army. Harried by the CDF, the RUF entered into peace negotiations begun by the military government. But the South African security advisers persuaded the government to use a cease-fire as a screen (Hooper 2003, appendix) to attack rebel camps in the run-up to a peace agreement in Abidjan (30 November 1996). Foday Sankoh signed the deal only to cover his retreat.

The terms of the ‘peace’ required the South Africans to leave the country, and the army, resentful at the CDF, mutinied (May 1997), displacing the elected government. The junta made an unexpected peace offer to the RUF, and the movement, regrouping in the bush, found itself junior partner in a short-lived military regime. This meant that the British–South African mining interest lost access to its concession, and urgent attempts were made to revive the earlier counter-insurgency strategy, but this time using a British security company (Sandline International) to support the CDF. This resulted in political scandal in the UK (Sierra Leone Arms Investigation 1998). Nigerian peace-keeping troops attacked the capital, removing the junta in early 1998, and the RUF returned to the interior, seizing a belt of land extending as far as the diamond-mining districts of the east (Phase 4).

A period of chaotic and uncontrolled violence followed, in which scattered army and RUF groups revenged themselves on communities supporting the CDF, and was ended by the Lome peace negotiations in 1999, and after further fighting, by the Abuja agreements of 2000 and 2001. Demobilization took place, and an official end to the war was declared in January 2002. Statistics revealed the majority of RUF fighters to be rural in background, with a predominance of Mende-speakers from the east, many recruited along the Liberian border.

Of areas taken over in the earlier phases of the insurgency – Kailahun (‘Burkina’) and Pujehun (‘Libya’) – only parts of Kailahun remained firmly in RUF hands at cessation of hostilities.31 In Kailahun – our case study – the war was less about the diamonds than serious (and still unresolved) social tensions of an agrarian character, skilfully exploited by the RUF.

The Agrarian History of the Liberian Border Under British Rule

A focal point of RUF control from beginning to end of the conflict was a block of territory adjacent to the Gola Forest in southern Kailahun District. The Gola Forest was itself a product of trade wars in the nineteenth century. Today it is an extensive boundary wilderness between Liberia and Sierra Leone which emerged from forest regrowth in an area devastated by trade rivalries between Gola and Mende warlords controlling portions of trade routes running from the Upper Niger basin to the Galinhas estuary and Cape Mount in the mid-nineteenth century (Richards 1996, 95–100). The separation of Mende and Gola speakers west and east of the Gola Forest served to define international spheres of influence in the colonial period. Mende-speaking communities belonged to British-ruled Sierra Leone and Gola-speaking communities to Liberia, a settler territory established for former American slaves (D’Azevedo 1969–71).

Wars in the pre-colonial period were typically ended by marriage alliances among ruling clans, and a dense network of such marriage-based links connected chiefly elites on either side of the frontier. Some local leaders continued to oppose British rule into the twentieth century and made good use of their connections in north-western Liberia and south-eastern Guinea to maintain an independent way of life. The British extended their control in eastern Sierra Leone by reconfirming the authority of warlord and mercantilist lineages running the regional trading routes through the western forest margins.

Under Indirect Rule British officers supervised a ‘customary law’ protecting the property rights of land-owning lineages, including rights over women at marriage (Fenton 1948). Land was leased or loaned under a non-monetary system similar to the tutorat, requiring recognition that the lessee was a ‘stranger’.32 Marriage involved customary exchanges including bride service and prestige goods. Domestic slavery was a tolerated feature of British rule until 1928. After formal emancipation, ex-slaves then formed a large underclass dependent on former owners for access to land and the means to marry.

Indirect Rule was favoured by the British in Sierra Leone mainly to keep administrative costs low. A major factor shaping developments in early colonial Sierra Leone was an uprising of Mende chiefs in 1898. The extension of British law into the interior (the Protectorate Ordinance of 1896) threatened abolition of slavery and slave dealing, a reason cited for the uprising.33 After the rebellion had been ended, the British appointed new chiefs from among elite families, and shelved plans for abolition. Fenton (1948) reports that recruits to British-supervised chieftaincy were only forthcoming after colonial restrictions placed on forced labour were withdrawn. Chiefs were permitted (under the Forced Labour ordinance, applicable until the 1950s) to command unpaid labour from subjects. Unpaid labour is still required by chiefs for community purposes (Mende, ta yenge– literally ‘town work’) and covered by a special exemption to the section banning forced labour in the 1991 constitution.

Close to Freetown, and along the line of the government railway to the Liberian border, the percentage of the population counted as slaves declined within a few decades to around 10–15 per cent from an estimated late nineteenth-century rate of 50 per cent (Grace 1977, 428). But percentages along the Liberian border remained stubbornly high. This forested and well-watered region was suitable for cocoa and coffee, and many chiefs invested in plantations, using their farm slaves. Farm access roads were also built to link with the railway with community forced labour. Unpaid labour extracted under custom was thus functional to a cash crop revolution.

Customary courts established under British supervision in fact actively maintained slavery. Courts heard cases for the reclaim of runaways into the 1920s.34 In 1927 the Supreme Court in Freetown found in favour of provincial owners using ‘reasonable force’ to recover slaves. This stirred up such a storm of protest – combined with pressure from the League of Nations – that abolition became inevitable (Ordinance 24 of 1927). Liberia followed suit in 1930.

The British emancipated slaves in West Africa according to the so-called Indian Model, i.e. slavery lost its legal status but did not become illegal (Miers and Klein 2004, 4). The intention was to keep most slaves working for former owners until the wage economy spread. The measure minimized pressure from owners for compensation, but it meant in practice that few steps were taken to inform slaves of their rights, or to help foster independence, e.g. through land grants or skills training. In short, emancipation fostered ingenious post-slavery adaptations, of which ‘woman damage’ will be cited below.

Slave origins were not readily forgotten, even if local convention forbids them to be discussed (Grace 1977, 429). Only children of chiefs were deemed to have full rights. When Richards (2005a, 588) asked, after the civil war, how villagers translated the new language of ‘human rights’ introduced by the humanitarian agencies, he was told the phrase used was ‘we are now all children of the chief’.

Why the Ranked LMP Endured in Sierra Leone

‘Customary’ institutional adaptations of the colonial period often survived into the post-colonial period. The reasons remain controversial. Hyden (1980) has argued that customary arrangements are functional to maintenance of an ‘uncaptured’ (self-sufficient) peasantry. Here we argue that they were functional to the spread of cash-cropping. But we note that the nationalist classes also helped shape colonial Indirect Rule, so it was a system they knew how to manage in otherwise uncharted political waters. A number of leaders worked with what they had inherited from this earlier period. This was the case in Sierra Leone.

In 1961 the SLPP under Sir Milton Margai formed the first independent government, having come to power with strong backing from rural chiefs. In response, customary privileges were preserved. One of the services of chiefs to the government was to deliver the vote of the large client underclass. Even in the 2007 general election the electoral commissioner had cause to complain that not all rural chiefs yet accepted that they had no right to tell their rural constituents how to vote.

This was more than natural conservatism. The reasons the ranked LMP lasted so long in Sierra Leone are linked to specific developments in the rural political economy at independence. Unlike in Côte d’Ivoire, cash crops were displaced in Sierra Leone by minerals as the motor of the economy. Multi-national mining capital withdrew from Sierra Leone during the 1970s and 1980s, and increasingly emphasis swung towards alluvial diamonds (Zack-Williams 1995). Alluvial workings are little more than hand-dug pits in swamps and river terraces, and thus are widely scattered in rice swamps and plantation land. Land-owning lineages are essential partners in alluvial operations. But diamonds are easily smuggled and hard to tax. Successive governments in need of resources to build their own patronage networks turned to agricultural exports instead, and thus depressed the agrarian sector. The cash cropping zone in eastern Sierra Leone, in consequence, was exploited by government monopsony marketing boards, and never became a target for widespread immigration. In consequence there was no large pool of outsiders from whom wealth could be extracted under the tutorat to smooth inter-generational relations among autochthones, as in Côte d’Ivoire.

In any case, the ranked lineages viewed clients as belonging to a social underclass deserving of only limited rights. The main concern of mercantile chiefs was to invest in their own children outside the rural sector, rather than find places for members of the ex-slave underclass within the local agrarian hierarchy.35 As one young ex-fighter sadly remarked concerning the number of children of the elite living safely overseas, ‘when war comes our leaders have wings to fly’.

In 1968 the SLPP government – the main bulwark for the protection of the ranked LMP – was replaced by the All People's Congress (APC) regime of Siaka Stevens. Stevens was a trade unionist from the industrial mining sector, and claimed much of his support in the largely diamond-free north of the country. At the outset he made use of socialist ideas, hoping to build a power base through state investment in education and jobs. For a time export taxes and overseas aid grants kept his one-party state (advised by the East Germans) afloat, but a combination of poor commodity prices, the flight of international mining capital, and beginnings of donor-imposed ‘structural adjustment’ reduced his sources of ‘state patronage’, and increasingly required more direct intervention in rural affairs, in part to head off SLPP opposition challenges, but also to tap increasingly important flows of wealth controlled by rural landowners in alluvial mining districts.

By the 1980s state socialism was abandoned, and Stevens (and his successor Joseph Saidu Momoh, former head of the army) delved deeply into chieftaincy affairs. Under the British system of Indirect Rule, chiefs were elected by an electoral college representing the main land-owning lineages. It was relatively easy to buy votes in such a system, and Stevens replaced SLPP-leaning chiefs with party loyalists where he could. Broadly speaking, he did this within British-devised rules of succession, i.e. the candidate backed by State House might not be the popular choice but would still represent one of the land-owning lineages. In other words, Stevens changed the incumbent while keeping the ranked lineage system intact.

Chiefs are elected for life, and outside the diamond districts Stevens was often content to leave a hostile chief in place, though denying the incumbent State House patronage. This was his approach to the entire Kailahun District, a hotbed of opposition support. When strong-arm measures failed to influence political opinion – lorry-loads of ‘thugs’ are reputed to have ‘disappeared’ during the 1977 elections – Stevens cut off the entire area from state resources. Projects were withdrawn, road repairs were abandoned, and teachers lacked salaries for lengthy periods.

This had two results. Younger people found themselves without education. A cash-strapped President Momoh, in default with the IMF in 1987, handed the RUF a recruiting card when he declared that education was a privilege not a right. Meanwhile, the chiefly classes of Kailahun, short of revenue, sought income where they could, and increased pressure on the local underclass. Custom became more important than ever as a basis for extracting farm labour. Fines on young men for marriage cases and other breaches of chiefdom law (e.g. absence from unpaid community work, often for ‘projects’ from which chiefs pocketed benefits) intensified. Underclass resentment reached a boiling point. The RUF rounded up children from failed rural schools and a resentful underclass of young labourers, proclaiming its revolutionary slogan ‘no more master, no more slave’.36

The Crux of the Matter –‘Woman Damage’

As explained above, the crux of control in the ranked lineage mode of production is marriage. Dupré and Rey (1973) argued that the reproduction of this mode is managed by making it hard for underclass youths to marry without backing from elders. If post-colonial involution of this mode of production was a factor in the war in Sierra Leone, we should expect to find two sorts of data – evidence that rebel combatants saw problems over marriage as an issue, and evidence that lineage elders maintained or intensified customary sanctions over village marriage. We have in fact found both sorts of evidence.

First, we note that marriage was an explicit concern for a significant number of RUF rank-and-file. In a large random sample of ex-combatants, Humphreys and Weinstein (2004) discovered, apparently somewhat to their surprise, that few former RUF cadres rated diamonds an advantage of joining the war, but that as many as 20 per cent mentioned the movement had enabled them to find a marriage partner.

If control by elders over marriage remains a systemic source of local agrarian social control we should also expect ‘woman damage’ cases to remain important elements in customary court activity. In fieldwork in the 1940s, Little (1951, 150–3, 186) examined 200 cases in six chiefdoms and found debt, ‘woman damage’ and land issues to be the main issues before the courts. He notes that ‘woman damage’ cases were more frequent than land cases.

We examined records over five years from three representative chiefdom courts re-opened in 2001. One was located in a cocoa-growing chiefdom close to the Gola Forest. Another was in a rural rice-farming district. The third was a court in a peri-urban (non-agrarian) chiefdom.37 In conformity with Little's findings over half a century earlier, the three main classes of case were debt, marriage and land, and marriage cases were more frequent than land cases in the two agrarian chiefdoms.

Land cases mainly reflected disputes between land-owning groups over vague boundaries associated with family land in a shifting cultivation system. In the case of ‘woman damage’, the plaintiff is generally a polygynist seeking damages for an affair with one or other of his wives, though it might also involve other classes of elder to whom marriage prestations are due. The judges are invariably persons claiming such rights themselves. The tort addresses a presumed challenge to the right-holder's authority over a woman, not a dilemma of sexual morality; cases of incest (Mende –simongama), for example, are dealt with in a different way (e.g. by participation in a cleansing cult, Humoi).

A total of 542 land and ‘woman damage’ cases were brought before the three courts over the period September 2001–February 2006, together accounting for about one-third of all cases (‘woman damage’ cases 20 per cent, land cases 13 per cent). Land cases exceeded marriage cases (42 and 10 per cent of all cases) only in the peri-urban chiefdom (building land on the edge of a conurbation was much disputed). In the two exclusively rural chiefdoms, ‘woman damage’ cases were up to ten times more frequent than land cases (20 and 28 per cent compared to 2 and 4 per cent).

In marriage cases, plaintiffs are elders and defendants youths. What is the probability that the plaintiff might lose, when marriage and land cases are compared? Averaging land and marriage, judgement favoured the plaintiff in 45 per cent of cases, and the defendant in 8 per cent of cases (24 per cent of cases were settled out of court and 24 per cent by other means, e.g. through ‘ordeals’). In land cases defendants succeeded in 14 per cent of cases, and plaintiffs in 42 per cent of cases, i.e. the defendant had a one in four chance of succeeding in a case settled in court. But in marriage cases a defence succeeded in only 2 per cent of cases, compared to the plaintiff obtaining the verdict in 56 per cent of cases, with the balance settled out of court (42 per cent of cases). In other words, it is seven times more difficult for a defendant to win a marriage case than to win a land case, a measure of the odds stacked against an underclass youth (land cases, by contrast, are between social equals).

Village youths say that elders marry young girls polygynously, knowing they will form alliances, and that they can fine their paramours and force them to work on their farms and plantations. This is confirmed by Gibbs, for the Kpelle (a ranked lineage society of north-western Liberia), when he writes that a prominent person (toh nuu) might allow some of his polygynously married women ‘to become the consorts of poor men of the lower class who become his . . . workmen or clients . . . another source of labour to work on his farms’ (1965, 215). Little adds that among the Mende ‘the husband makes no objection to a wife having “friends”, provided the latter work for him’ (1951, 152). It is probably not too strong to claim that ‘woman damage’ is thus a means of re-imposing a kind of agrarian servitude (i.e. it forces young men into agricultural labour at very low rates). Labour is in shorter supply than land in rural Sierra Leone (Richards 1986). Hence it is no surprise to find that ‘woman damage’ cases are more numerous than land cases in the two farming chiefdoms.38

In Sierra Leonean customary courts there is no scale of fines. Inflation long ago eroded gazetted amounts. The court decides on a sum it thinks appropriate. This is often related to assessments of what the defendant can pay. Young men report that the fines will be much steeper if they have just harvested a crop. Village elders assessing one such case were heard to say ‘we will squeeze him until he is dry’. In eight cases for which we have data (all in the cocoa chiefdom), amounts ranged from Le 40,000 to Le 200,000, averaging Le 134,000 (c. US$50–60). This is equivalent to one third of a typical farm labourer's annual income. Inability to pay large fines was cited as among reasons for joining a militia group in the war (Richards 2005a, 577–80). The government promised (in 2001) to gazette a revised list of tariffs, but this had not yet taken effect at the time of our survey (2006).

What constraints are imposed on those who cannot pay? Village lock-ups and stocks are still common. In one post-war consultation a young man was brave enough to speak publicly about the threat of force. ‘You see this fine new court the British have built for us, with a lock-up attached; after you have gone the elders will stick me in it for speaking out of turn’ (Mende fitiyai– arrogance, rudeness).

Where no cash can be borrowed, fines are either commuted to labour on the plaintiff's farm or the case is ‘bought’ (i.e. a farmer looking for labour will pay the court, and the defendant works for his benefactor). Not paying, and fleeing the village, is favoured by some. This means becoming a ‘stranger’ in another chiefdom. There may be little advantage over working off the fine at home, since the ‘landlord’ has a shrewd idea that the new ‘stranger’ has been forced to relocate, and demands labour-intensive tests of loyalty as a result. It may be only after many years of reliable ‘service’ that the patron helps his client find a wife.

A second exit option is sweated labour in alluvial diamond pits. The dream of absconding with a large stone adds allure to an otherwise back-breaking option, though in truth few diggers ever find such stones, and wages barely cover subsistence. The real advantage of life in a diamond camp is freedom from close regulation by village elders. One former RUF fighter stated (Richards et al. 2004, 6) that if he worked hard – to make a drum of oil palm, say – his wife's family would demand most of it, but that if he tried to avoid such payments his wife's people would drag him to court and the chiefs would fine him. He described village marriage as synonymous with slavery. The advantage of life in the diamond camps was that he was free to form a relationship with any woman who would accept him.

In its early days the RUF formed bush camps and raided for young women, enrolled within a ‘combat wives unit’ (i.e. the women were married to the fighters).39 Accounts of camp life paint a picture of a rough-and-ready social order built on violence but also on egalitarian notions opposed to principles of village hierarchy (Coulter 2006; Peters and Richards 1998; Richards 2005c). Meillassoux (1991, 147–8) suggests this kind of violent meritocracy has challenged societies built on lines of descent over several centuries:

The powerful Segu State . . . emerged [in the 17th–18th centuries] from the political development of brigandage . . . it built a new society which tore up clan society . . . [The founder] Biton Kulibali [a man of non-aristocratic origins] drew to himself men of various origins thrown up by the disordered times: escaped slaves, rebellious cadets, [commoners] who had been humiliated . . . a new society was being formed [that] gave precedence to associations rather than clans . . .

This is unnervingly close to being a picture of the RUF. The RUF attempted to replace ranked lineage society with new forms of association in consequence of which the world of the village was ripped apart. It is a worry to note, therefore, that the data discussed above come from post-war courts restored with donor assistance, suggesting that if the present analysis is correct the causes of conflict have been reconstituted.

Support for the Argument from the Other Side of the Fence

Cramer (2006) argues that attempts to restore pre-war conditions are characteristic of post-war reconstruction activities in general. But there were specific reasons behind British support for restoration of chiefly rule in rural Sierra Leone. This was a perception that the Sierra Leone government was losing control over its own forces. CDF grievances began to resemble RUF grievances, and CDF units began to act accordingly. In some areas CDF fighters established courts resembling the people's courts of the RUF. In a case described by Archibald and Richards (2002), young CDF volunteers, dissatisfied with the actions of chiefs over relief supplies, left the village and set up a bush camp similar to those of the rebels.40

The CDF was formed under the patronage of rural chiefs, and recruits were initiated into a Mende guild of special hunters (kamajoisia).41 The earliest volunteers said they were fighting to reclaim their land (interviews, Sugar Bowl Camp, Bo, June 1996). But as CDF numbers increased (eventually more than 35,000 civil defence volunteers were demobilized), recruitment became much more diverse. Even at the outset recruits stated that ‘we know RUF fighters are seeking to join the movement’. They implied these were infiltrators, but many young people joined whichever fighting force was accessible because they lacked other means of support.

Depending on area and circumstances, the CDF thus incorporated as many young people with underclass backgrounds as the RUF. As the needs and grievances of this underclass began to assert themselves, RUF and CDF fighters began to converge towards a common understanding of the crisis. In complex conditions following the collapse of the Abidjan peace process, formerly opposed units sometimes even merged, or swapped affiliation. A notable case was uncovered at Makali, where an entire unit of CDF fighters ‘converted’ to the RUF in the dying days of the war (Richards et al. 2003).

Shared perceptions are also evident in interviews reported by Peters (2006), Archibald and Richards (2002), and Richards (2005a). After the war, groups of CDF fighters encountered in an ex-combatant tracker study (Richards et al. 2003) were convinced they had been ‘cheated’ by chiefs and initiators. Some commented that ‘next time, they would fight for the rebel side’. In short, tensions between the underclass and chiefs affected the CDF almost as strongly as the RUF, throwing light on the claim that the war lacked clearly defined sides (Keen 2005). Convergence of motivation, and absence of ethnic polarization – the majority of CDF and RUF fighters were Mende – supports a conclusion that the target for violence in Sierra Leone was the system of social reproduction itself.

Some Conclusions on Sierra Leone (and on the Comparison with Côte d’Ivoire)

Under the conditions of national development as described – the difficulty of taxing alluvial diamond wealth, agrarian depression and structural adjustment – ranked lineage society in rural Sierra Leone underwent a process of involution. Unlike the Akan societies of eastern Côte d’Ivoire there was no agrarian ‘vent for surplus’, and this adversely affected the chances of social and economic integration of a large underclass. Specific conditions on the Liberian border generated by the opposition of the local rural elite to the regime of Siaka Stevens during the 1970s and 1980s led to pressure on the underclass and thus to intensification of ‘class oppositions’ within the ranked lineage mode of production.

The Mende of eastern Sierra Leone and the Gban of central-western Côte d’Ivoire share certain institutional features, including the tutorat (or landlord–stranger system). In both cases indigenous institutional capital has taken on a new life under modern agrarian conditions. But custom has had different consequences in an area of pioneer forest agriculture and in an area on the forest margins long transformed by merchant capital.

In conditions of frontier expansion Gban elders were able to extract wealth from a large influx of migrant strangers (both planters and labourers) using traditional means, and then use this wealth to help reproduce a younger generation of Gban urban migrants. It was only with the breakdown of the urban economy that this strategy caused trouble, since young people were unable to return and claim land leased to strangers. Intergenerational solidarity remains an important ideal; it is the elders (supposedly) that have betrayed the system. Violence targets the large number of migrant strangers admitted by these elders.

In eastern Sierra Leone, by contrast, there has been little long-distance labour migration into a (depressed) cash crop sector.42‘Children of chiefs’ have moved to town, and intensified labour burdens have fallen on the underclass elements remaining behind. Movement of ‘strangers’ occurs, but it amounts to little more than shifting from chiefdom to chiefdom – or into alluvial diamond mining – in the vain hope of escaping involutionary pressures. But there is no release; neighbouring chiefdoms have the same social and institutional rules. Even alluvial mining is controlled at local level by the high-ranking lineages. With the burden of reproducing the system falling not on migrant workers but on the local underclass, at a time when state resources dried up, pressure built not to attack migrants but to destroy the system. Customary hierarchy no longer offered a place for all (Douglas 2005). Anger at class oppression rather than ethnic hatred became the focus of the war in Sierra Leone, and gave it a non-ethnic character unusual by African standards.

Taken together our two cases suggest an important conclusion. Both versions of the lineage mode of production appear to have reached some kind of limit, and the search for agrarian alternatives may be more rapid and profound as a result (cf. Jackson 2007). For policy makers an important if so-far neglected possibility becomes apparent – that community failure is as plausible a cause of African conflicts as failure of the state.


  1. Top of page
  2. Abstract

To many commentators on recent African wars, armed conflict is driven by (vaguely defined) cultural or ethnic divisions. Equally vaguely, development institutions concerned with post-conflict recovery (led by the World Bank) pin their hopes (and money) on catch-all notions of ‘community-based’ reconstruction. This shift towards ‘culturalism’ in development thinking has been powerfully critiqued by Douglas (2004). Culture by itself causes nothing. Ethnic identity claims (and other cultural epiphenomena) are always products of organization, and competition among organizations. What is needed is an analysis linking organizational rivalries to the material struggles in which competing agrarian organizations engage. This we have attempted in this paper.

Reassembling the variant models of the LMP (once seen as competing explanations) into a regional, resource-based agrarian framework has allowed us to differentiate distinct kinds of organizational clashes, and to trace how these have become incorporated within two war-affected localities. We have looked closely at intergenerational tensions resulting from two differently configured crises of social reproduction. In the one case (class-stratified agrarian communities of the western flank of the Upper Guinea Forest), failure fully to incorporate a social underclass has resulted in iconoclastic violence targeting customary rural institutions. In the specific circumstances of economic decline described, rural chiefs continued to invest in their own children but neglected a large underclass, undermining societal cohesion. In the other case – the egalitarian communities at the core of the UGF – room for expansion on an extensive forest frontier gave lineage elders scope to adapt custom to their financial requirements for reproducing a younger generation. Urban economic failure then forced this younger generation back home, and a crisis of reincorporation resulted. Violence offers scope for adjustment to pressure when other means fail. Violence in central western Côte d’Ivoire is aimed not at the lineage system, as in Sierra Leone, but at incomers now sitting on the land. In one case war seeks to overthrow the system; in the other it is undertaken in defence of custom and tribe. A fundamental contrast between the ethnic violence associated with the war in Côte d’Ivoire and the class-based violence targeted against chiefly families in Sierra Leone, perpetrated by two groups of young men otherwise similar in their poverty and hyper-mobility, thus comes into focus.

What (if any) is the practical significance of our argument? Basically, it is to add weight to warnings already issued against a ‘cultural turn’ in development policy. ‘Community failure’ is as much a factor as ‘state failure’ in the wars we have described. The restoration of (failed) chieftaincy (a focus of British aid in post-war rural Sierra Leone) and ‘community-driven reconstruction’ (a World Bank mantra in Liberia and Côte d’Ivoire) are likely to have perverse effects unless solutions to basic conflicts of interest in rural communities are figured into the analysis. We would argue that conflict management seeking to strengthen (a mythically cohesive) ‘community’ at the expense of (an allegedly over-weaning) ‘state’ may foster the very kinds of conflicts it hopes to abate. Recent voices have called for land reform to shape the path to recovery in the West African conflict zone, and to return land to ‘community management’ (Alden Wily 2007), without adequate regard to the sociology of inter- and intra-communal tensions contributing to the crisis. If our analysis above is correct, this will be a dangerous move. What is needed, we suggest, is to recognize the variety of interests competing over land and other rural resources, and to incorporate sound analysis of such competing interests within national frameworks for social, economic and political reform. Addressing the complexity and intensity of struggles over agrarian resources is (we believe) basic to the attainment of peace.

  • 1

     On agricultural origins in West Africa, see Fuller (2005), Klee et al. (2000), Klein (1996) and Porteres (1976).

  • 2

     Mende-speakers from the forests of eastern Sierra Leone were much involved in regional trade networks and their ranked lineage communities were once highly dependent on slaves. Livingstone (1958, Table 1) cites a figure of 29.36 per cent for HbS frequency in Mende populations from Sierra Leone (one of the highest in the region). Although some slaves were incorporated into communities in the forest core, the degree of dependence on slave labour was much less. Livingstone (ibid.) lists HbS rates from zero to under 2 per cent for groups such as the Bakwe, Bete and Dan living in western Côte d’Ivoire.

  • 3

     In Côte d’Ivoire these Mande traders (tracing links to the Upper Niger and ancient Mali empire) are known as Dioula [Jula], and in Sierra Leone as Mandingo.

  • 4

     Patrilineages are commonest but there are important matrilineal exceptions in the eastern UGF. It matters (here) only that land access and other rights of community membership are reckoned by descent. Assigning rights through the lines of both mother and father (cognatic reckoning) is widespread, often linked to (alliance-building) avunculate marriage (Currens 1972; Murphy and Bledsoe 1987; Leopold 1991).

  • 5

     For ethnographic documentation see Meillassoux (1964). For a comparable picture of pioneer agriculture in thickly-forested south-eastern Liberia see Massing (1980).

  • 6

     We define mercantilism as monopoly control of commodity trade.

  • 7

     Little (1951, 141) points out that among the Mende (male) eldership is defined by polygyny (a person who has no more than one or two wives is spoken of as a ‘small boy’).

  • 8

     A proverb in Mende translates as ‘no one is for him [or her] self, everyone is behind someone’, thus even patrons have patrons – i.e. hierarchy is all-encompassing.

  • 9
  • 10

     Wilks (1993, 80–2) centres his discussion of Akan societies on (i) the agrarian system before and after the forest conversion, (ii) agricultural manpower needs corresponding to these two phases, (iii) the strong social integration of non-Akan and slaves into the lineages in the second phase, (iv) resulting decline of matri-clans.

  • 11

     Among egalitarian groups from western Côte d’Ivoire, those achieving the best integration into trade were also able to integrate larger numbers of domestic slaves, thus extending their productive (and reproductive) capacities (Chauveau et al. 1981; Léonard and Vimard 2005). H. Memel-Fotë (2007) underlines that slaving in Ivorian forests was not exclusive to class-stratified communities.

  • 12

     Rebels in the war in Sierra Leone were sometimes referred to as ‘raray’ (rare, i.e. street youth, cf. Abdullah 1997). The Krio dictionary (Fyle and Jones 1980) gives a complicated etymology concerning street entertainers, but popular understanding implies an undisciplined ‘vagrant’ (i.e. a runaway).

  • 13

     For opposite tendencies due to the impact on youth of recent structural changes within the agricultural economy in eastern Ghana see Amanor (2005).

  • 14

     Congrès Panafricain des Jeunes Patriotes (Panafrican Congress of Young Patriots) led by Charles Blé Goudé.

  • 15

     The most important incident was an attack by loyalist aircraft on the positions of the Forces Nouvelles in Bouaké and Korhogo in November 2004. The consequences of this attack included bombardment of a French position in Bouaké by loyalist forces, organized violence of ‘young patriots’ against French expatriates in Abidjan, and reactions by the French Army.

  • 16

     The (non-pejorative) term ‘stranger’ corresponds to local usage and designates any group of settlers originating outside the local community, regardless of nationality.

  • 17

     From the (French) term ‘tuteur’ commonly used locally.

  • 18

     For a general discussion of the issue of the first comer–late comer relationship and its structuring role in African political history, see Kopytoff (1987) and Chauveau et al. (2004).

  • 19

     Dioula [Jula] is a generic term to refer to Muslims from northern Côte d’Ivoire or the Sahel. It has the same sense as ‘Mandingo’ in Sierra Leone, referring to someone of Mande origins, a Muslim trader, or both.

  • 20

     Fieldwork is currently in progress in Oumé and other rural districts.

  • 21

     Parti Démocratique de Côte d’Ivoire, Ivoirian section of the Rassemblement Démocratique Africain.

  • 22

     The following sections draw on Chauveau (2000, 2005b, 2006, 2007), Bobo (2002, 2005), Chauveau and Bobo (2003), Koné (2002). Background information on the Gban region was found in the archives of the Oumésous-préfecture (political reports 1913–1963).

  • 23

     For instance, a draft law strengthening the prerogatives of the state over land tenure was stopped in 1962 after discontent by the customary authorities.

  • 24

     Land transfers via the tutorat provided tuteurs and their family groups with another advantage: the ‘placement’ of settlers on areas bordering family, lineage or village land reinforced customary property rights vis-à-vis nearby families, lineages or villages, or even contributed arguments to justify customary property rights where these were not clearly recognized.

  • 25

     The only case where social constraint is very heavy for the younger generation concerns funeral payments for the father.

  • 26

     Unlike what is happening in far western regions, young patriots seldom directly threaten settlers, except during road blocking at the beginning of the conflict.

  • 27

     RUF aims proved hard to establish. There was no media contact, and prisoners-of-war were killed (Richards 1996, 28). At the end of the war the movement's main political activists (a group of about 400) were promptly imprisoned (May 2000) and only released (without charge) in 2006–7 (Da Silva 2007). The explanatory agenda meanwhile was dominated by former youth radicals opposed to the war (Abdullah 1997; Kandeh 2001). The thesis by Peters (2006) is an important source for RUF perspectives.

  • 28

     For the movement's early campaign see RUF/SL (1995), a pamphlet prepared with assistance from a conflict resolution agency at the time of the Abidjan peace negotiations.

  • 29

     Hooper (2003, appendix) deems the operation in Sierra Leone an application of the most successful counter-insurgency doctrine in recent African history. Its origins lie in Rhodesian special operations in the 1970s (Cilliers 1985).

  • 30

     Inconveniently, the RUF was pressing for accountability in diamond mining as its main political demand. An influential figure in the democratic transition was a retired British officer of the secret intelligence service (MI6) managing the company holding the kimberlite concession in Sierra Leone from 1995. He was identified in a UK parliamentary report as also active in the ‘Arms to Africa’ affair (Sierra Leone Arms Investigation 1998).

  • 31

     The Makeni–Magburaka–Koidu axis in the north was known to the movement as ‘Togo’.

  • 32

     In Mende the tuteur (patron) is hota-kee, i.e. father of the stranger (hota).

  • 33

     ‘It is probable . . . that British interference with certain native practices, such as slave-dealing, in particular, had already aroused a good deal of resentment among the chiefs, whose source of income and prestige was largely bound up with the commerce’ (Little 1951, 58).

  • 34

     In Pujehun District in 1919–20, 23 per cent of cases in customary courts concerned runaway slaves (Grace 1977, 427).

  • 35

    This is most obvious in diamond chiefdoms, where revenues from alluvial mining have helped create overseas diaspora groups. The sizeable Kono community in south-east London would be an example. The Kono are a Mande group occupying the largest alluvial diamond-mining area in eastern Sierra Leone.

  • 36

     Elders in an RUF-controlled village interviewed in 2001 said, of the Stevens and Momoh eras, the roads were impassable, ‘but we could head-load our cocoa over the [adjacent] border and sell it in Monrovia’; young people without plantations, they added, ‘joined the RUF’ (Richards et al. 2003).

  • 37

     We thank Alfred Mokuwa for collecting these data.

  • 38

     One chief (in the cocoa chiefdom) took five different young men to court for ‘woman damage’ in 2004.

  • 39

     Some female captives preferred to train as fighters.

  • 40

     The word used was fakai (= slave village).

  • 41

     A Temne chief in the north sent for ritual experts from the Mende south to have his own fighters initiated into the CDF, explaining the advice came from ‘Branch Energy, the mining company’ (Richards et al. 2003).

  • 42

     A survey in 13 villages around the Gola North reserve on the eve of the war showed that 89.1 per cent of strangers came from other Mende-speaking communities and only 6.3 per cent from elsewhere in Sierra Leone and 4.7 per cent from neighbouring countries (Richards 1996, 135).


  1. Top of page
  2. Abstract
  • Abdullah, Ibrahim, 1997. ‘Bush Path to Destruction: The Origin and Character of the Revolutionary United Front (RUF/SL)’. Africa Development 22 (3 and 4): 4576. (Special Issue: Lumpen Culture and Political Violence: the Sierra Leone Civil War).
  • Alden Wily, Liz, 2007. So Who Owns the Forest?’ An Investigation into Forest Ownership and Customary Land Rights in Liberia. Monrovia and Brussels: Sustainable Development Institute and FERN.
  • Amanor, K.S., 2005. ‘Jeunes, Migrants et Marchandisation de l’Agriculture au Ghana’. Afrique Contemporaine, 214: 85101.
  • Archibald, Steven and Paul Richards, 2002. ‘Conversion to Human Rights? Popular Debate About War and Justice in Central Sierra Leone’. Africa 72 (3): 33967.
  • Banégas, Richard, 2006. ‘Côte d’Ivoire: Patriotism, Ethnonationalism and Other African Modes of Self-writing’. African Affairs, 105 (421): 53552.
  • Banégas, Richard, 2007. Côte d’Ivoire: Les Jeunes ‘se Lèvent en Hommes’. Anticolonialisme et Ultranationalisme chez les Jeunes Patriotes d’Abidjan. Les Etudes du CERI, No. 137, juillet 2007.
  • Banégas, Richard and Bruno Losch, 2002. ‘La Côte d’Ivoire au bord de l’Implosion’. Politique Africaine, 87: 13961.
  • Bledsoe, Caroline, 1980. Women and Marriage in Kpelle Society. Stanford: Stanford University Press.
  • Bobo, K.S., 2002. La Question de l’Accès à la Terre des Jeunes et des Citadins de Retour au Village: Cas de Donsohouo dans la Sous-préfecture d’Oumé. Mémoire de maîtrise, Université de Bouaké, Département de Sociologie et d’Anthropologie.
  • Bobo, K.S., 2005. Les Rapports Fonciers Intrafamiliaux et Leurs Dynamiques dans le Centre-ouest Ivoirien: Cas de Bodiba dans la Sous-préfecture d’Oumé. Mémoire de DEA, Université de Bouaké, Département de Sociologie et d’Anthropologie.
  • Boone, Catherine, 2003. Political Topographies of the African State. Territorial Authority and Institutional Choice. Cambridge: Cambridge University Press.
  • Bouquet, Christian, 2005. Géopolitique de la Côte d’Ivoire. Paris: Armand Colin.
  • Chauveau, J.-P., 2000. ‘Question Foncière et Construction Nationale en Côte d’Ivoire. Les Enjeux Silencieux d’un Coup d’État’. Politique Africaine, 78: 94125. (English version: The Land Question in Côte d’Ivoire: A Lesson in History. International Institute for Environment and Development, Programmes Zones Arides, Dossier No. 95, août 2000).
  • Chauveau, J.-P., 2002. ‘La Loi Ivoirienne de 1998 sur le Domaine Foncier Rural et l’Agriculture de Plantation Villageoise: Une Mise en Perspective Historique et Sociologique’. Réforme Agraire (FAO) 2002 (1): 6278.
  • Chauveau, J.-P., 2005a. ‘Introduction Thématique: Les Jeunes Ruraux à la Croisée des Chemins’. Afrique Contemporaine, Dossier ‘Les Jeunes Ruraux en Afrique’, 214: 1535.
  • Chauveau, J.-P., 2005b. ‘Les Rapports entre Générations ont une Histoire. Accès à la Terre et Gouvernementalité Locale en Pays Gban (Centre-Ouest de la Côte d’Ivoire)’. Afrique Contemporaine, 214: 5983.
  • Chauveau, J.-P., 2006. ‘How Does an Institution Evolve? Land, Politics, Intergenerational Relations and the Institution of the Tutorat between Autochthons and Migrant Farmers in the Gban Region (Côte d’Ivoire)’. In Landrights and the Politics of Belonging in West Africa, eds R.Kuba and C.Lentz, 21340. Leiden: Brill Academic Publishers.
  • Chauveau, J.-P., 2007. ‘La Réforme Foncière de 1998 en Côte d’Ivoire à la Lumière de l’Histoire des Dispositifs de Sécurisation des Droits Coutumiers. Une économie politique de la question des transferts de droits entre autochtones et ‘étrangers’ en Côte d’Ivoire forestière. In Law, Land Use and the Environment: Afro-Indian Dialogues, ed. Christoph Eberhard. Pondichéry: French Institute of Pondicherry.
  • Chauveau, J.-P. and K.S. Bobo, 2003. ‘La Situation de Guerre dans l’Arène Villageoise. Un exemple dans le Centre-Ouest ivoirien’. Politique Africaine, 89: 1232.
  • Chauveau, J.-P. and J.-P. Dozon, 1985. ‘Colonisation, Économie de Plantation et Société Civile en Côte d’Ivoire’. Cahiers ORSTOM des Sciences Humaines, 21 (1): 6880.
  • Chauveau, J.-P. and E. Léonard, 1996. ‘Côte d’Ivoire's Pioneer Fronts: Historical and Political Determinants of the Spread of Cocoa Cultivation’. In Cocoa Pioneer Fronts since 1800: The Role of Small Holder, Planters and Merchants, ed. W.G.Clarence-Smith, 17694. London: Macmillan.
  • Chauveau, J.-P., J.-P. Dozon and J. Richard, 1981. ‘Histoires de Riz, Histoires d’Ignames: Le Cas de la Moyenne Côte d’Ivoire’. Africa, 51: 62157.
  • Chauveau, J.-P., J.-P. Jacob and P.-Y. Le Meur, 2004. ‘L’Organisation de la Mobilité dans les Sociétés Rurales’. Autrepart, 30: 323.
  • Cilliers, J.K., 1985. Counter-insurgency in Rhodesia. London: Croom Helm.
  • Colin, J.-P., 2005. ‘Le Développement d’un Marché Foncier? Une Perspective Ivoirienne’. Afrique Contemporaine, 213: 17996.
  • Collier, P., 2000. ‘Doing Well Out of War: An Economic Perspective’. In Greed and Grievance: Economic Agendas in Civil Wars, eds M.Berdal and D.M.Malone. Boulder, CO: Lynne Rienner.
  • Coquery-Vidrovitch, C., 1969. ‘Recherches sur un Mode de Production Africain’. La Pensée, 144: 6178.
  • Coulter, Christine, 2006. ‘Being a Bush Wife: Women's Lives Through War and Peace in Northern Sierra Leone’. PhD Thesis, Uppsala University, Department of Cultural Anthropology and Ethnology.
  • Cramer, Christopher, 2006. Civil War is not a Stupid Thing. London: C. Hurst.
  • Currens, G.E., 1972. ‘The Loma Avunculate: An Exercise in the Utility of Two Models’. Ethnology, 11: 11121.
  • D’Azevedo, Warren L., 1962. ‘Some Historical Problems in the Delineation of a Central West Atlantic Region’. Annals, New York Academy of Sciences, 96 (2): 51238.
  • D’Azevedo, Warren L., 196971. ‘A Tribal Reaction to Nationalism, Parts 1–4’. Liberian Studies Journal, 1 (2): 121; 2 (1): 43–63; 2 (2): 99–115; 3 (1): 1–19.
  • Da Silva, Clare, 2007. ‘Justice for Sierra Leone?’ Washington Times, 1 June.
  • Douglas, Mary, 1970. Natural Symbols: Explorations in Cosmology. London: Barrie and Rockcliff (2nd edn, Routledge 1996).
  • Douglas, Mary, 1993. In the Wilderness: the Doctrine of Defilement in the Book of Numbers. Sheffield: Sheffield University Press.
  • Douglas, Mary, 2004. ‘Traditional Culture – Let's Hear No More About It’. In Culture and Public Action, eds VijayendraRao and MichaelWalton, 85109. Stanford: Stanford University Press.
  • Douglas, Mary, 2005. ‘A Feeling for Hierarchy’. In Believing Scholars, ed. James L.Heft, 94120. New York: Fordham University Press.
  • Dupire, M., 1960. ‘Planteurs Autochtones et Étrangers en Basse Côte d’Ivoire’. Etudes Eburnéennes, 8: 7237.
  • Dupré, Georges and Pierre-Philippe Rey, 1973. ‘Reflections on the Relevance of a Theory of the History of Exchange’. Economy and Society, 2 (2): 13163.
  • Fenton, J.S., 1948. Outline of Native Law in Sierra Leone. Freetown: Government Printer (Revised Edition).
  • Ford, M., 1992. ‘Kola Production and Settlement Mobility Among the Dan of Nimba, Liberia’. African Economic History, 20: 5163.
  • Fuller, Dorian, 2005. ‘Crop Cultivation: The Evidence’, ‘Farming: Stone Age Farmers of the Savanna’ and ‘Farming: Tropical Forest Zones’. In The Encyclopedia of African History, ed. K.Shillington. New York: Fitzroy Dearborn.
  • Fyle, Clifford and Eldred Jones, 1980. A Krio–English Dictionary. Oxford and Freetown: Oxford University Press and University of Sierra Leone Press.
  • Gastellu, J.-M., 1989. Riches Paysans de Côte d’Ivoire. Paris: L’Harmattan.
  • Gastellu, J.-M. and S. Affou Yapi, 1982. ‘Un Mythe à Décomposer: La Bourgeoisie de Planteurs’. In État et Bourgeoisie en Côte d’Ivoire, eds Y.-A.Fauré and J.-F.Médard. Paris: Karthala.
  • Gibbs, James, 1965. ‘The Kpelle of Liberia’. In Peoples of Africa, ed. JamesGibbsJr. New York: Holt, Rinehart and Winston.
  • Grace, John, 1977. ‘Slavery and Emancipation among the Mende in Sierra Leone’. In Slavery in Africa: Historical and Anthropological Perspectives, eds SuzanneMiers and IgorKopytoff. Madison: University of Wisconsin Press.
  • Hair, Paul, ed., 2000. Hawkins in Guinea 1667–1568. History and Culture Series No. 5, University of Leipzig Papers on Africa (Instituet fuer Afrikanistik).
  • Hecht, R.M., 1985. ‘Immigration, Land Transfer and Tenure Changes in Divo, Ivory Coast, 1940–80’. Africa, 55 (3): 31935.
  • Holsoe, Svend, 1977. ‘Slavery and Economic Response among the Vai (Liberia and Sierra Leone)’. In Slavery in Africa: Historical and Anthropological Perspectives, eds SuzanneMiers and IgorKopytoff. Madison: University of Wisconsin Press.
  • Hooper, J., 2003. ‘Appendix: Sierra Leone’. Bloodsong! An Account of Executive Outcomes in Angola. London: HarperCollins (2nd edn).
  • Humphreys, Macartan and Jeremy Weinstein, 2004. What the Fighters Say: A Survey of Ex-combatants in Sierra Leone, June–August 2003. CGSD Working Paper No. 20. New York: Columbia University.
  • Hyden, Goran, 1980. Beyond Ujamaa in Tanzania: Underdevelopment and an Uncaptured Peasantry. London: Heinemann.
  • International Crisis Group, 2003. Côte d’Ivoire: ‘The War is Not Yet Over’. 28 November 2003. ICG Africa Reports No. 72, Freetown/Brussels.
  • International Crisis Group, 2004. Côte d’Ivoire: No Peace in Sight. 12 July 2004. ICG Africa Reports No. 82, Dakar/Brussels.
  • International Crisis Group, 2005. Côte d’Ivoire: Le Pire est Peut-être à Venir. 24 mars 2005. ICG Rapport Afrique No. 92.
  • Jackson, Paul, 2007. ‘Reshuffling an Old Deck of Cards? The Politics of Local Government Reform in Sierra Leone’. African Affairs, 106 (422): 95111.
  • Jones, Adam, 1983. From Slaves to Palm Kernels: A History of the Galinhas Country (West Africa) 1730–1890. Wiesbaden: Adam Steiner.
  • Kandeh, Jimmy, 2001. ‘Subaltern Terror in Sierra Leone’. In Africa in Crisis: New Challenges and Possibilities, eds Alfred TundeZack-Williams, DianeFrost and AlexThomson, 17995. London: Pluto Press.
  • Keen, David, 2005. Conflict and Collusion in Sierra Leone. Oxford: James Currey.
  • Klee, M., Barbara Zach and Katharina Neumann, 2000. ‘Four Thousand Years of Plant Exploitation in the Chad Basin of Northeast Nigeria I: The Archaeobotany of Kursakata’. Vegetation History and Archaeobotany, 9: 22337.
  • Klein, A., 1996. ‘Toward a New Understanding of Akan Origins’. Africa, 66 (2):
  • Köbben, A., 1963. ‘Land as Object of Gain in a Non-literate Society. Land-tenure among the Bete and Dida (Ivory Coast, West Africa)’. In African Agrarian Systems, ed. D.Biebuyck, 24566. London: Oxford University Press.
  • Koné, M., 2002. Gaining Rights of Access to Land in West-Central Côte d’Ivoire: Bodiba (Oumé) et Zahia (Gboguhé). London: International Institute for Environment and Development.
  • Kopytoff, Igor, 1987. ‘The Internal African Frontier: The Making of African Political Culture’. In The African Frontier. The Reproduction of Traditional African Societies, ed. I.Kopytoff, 384. Bloomington: Indiana University Press.
  • Léonard, É. and P.Vimard, eds, 2005. Crises et Recompositions d’une Agriculture Pionnière en Côte d’Ivoire. Dynamique Démographiques et Changements Economiques dans le Bas-Sassandra. Paris: IRD-Karthala.
  • Leopold, Robert S., 1991. ‘Prescriptive Alliance and Ritual Collaboration in Loma Society’. PhD Thesis, Indiana University.
  • Little, Kenneth, 1951. The Mende of Sierra Leone: A West African People in Transition. London: Routledge and Kegan Paul.
  • Little, Kenneth, 1965. ‘The Political Function of the Poro, Parts 1 and 2’. Africa, 35 (4): 34965 and 36 (1): 69–71.
  • Livingstone, Frank B., 1958. ‘Anthropological Implications of Sickle Cell Gene Distribution in West Africa’. American Anthropologist, 60 (3): 53362.
  • Livingstone, Frank B., 1989. ‘Who Gave Whom Hemoglobin S? The Use of Restriction Site Haplotype Variation for the Interpretation of the Evolution of the B S–Globin Gene.’ American Journal of Human Biology, 1 (3): 289302.
  • Losch, Bruno, ed., 2000. ‘Côte d’Ivoire. La Tentation Éthnonationaliste’. Politique Africaine, Special issue, Vol. 78.
  • Massing, Andreas, 1980. The Economic Anthropology of the Kru (West Africa) . Wiesbaden: Franz Steiner Verlag.
  • Meillassoux, Claude, 1964. Anthropologie Economique des Gouro de Côte d’Ivoire. Paris and The Hague: Mouton.
  • Meillassoux, Claude, 1973. ‘The Social Organization of the Peasantry: The Economic Basis of Kinship’. Journal of Peasant Studies, 1 (1): 8190.
  • Meillassoux, Claude, 1991. The Anthropology of Slavery: The Womb of Iron and Gold. London and Chicago: Athlone Press and University of Chicago Press.
  • Memel-Fotê, Harris, 2007. L’Ésclavage dans les Sociétés Lignagères de la Forêt Ivoirienne (XVIIe-Xxe siècle). Abidjan-Paris: CERAP-IRD.
  • Miers, Suzanne and Martin Klein, 2004 [1999]. ‘Introduction’. In Slavery and Colonial Rule in Africa, eds SuzanneMiers and MartinKlein. London: Routledge.
  • Mikell, Gwendolyn, 1992. Cocoa and Chaos in Ghana. Washington: Howard University Press.
  • Mkandawire, Thandika, 2002. ‘The Terrible Toll of Post-colonial Rebel Movements in Africa: Towards an Explanation of Violence Against the Peasantry’. Journal of Modern African Studies, 40 (2): 181215.
  • Moran, Mary, 2006. Liberia: The Violence of Democracy. Philadelphia: University of Pennsylvania Press.
  • Murphy, William and Caroline Bledsoe, 1987. ‘Kinship and Territory in the History of a Kpelle Chiefdom (Liberia)’. In The African Frontier: The Reproduction of Traditional African Societies, ed. IgorKopytoff. Indiana University Press: Bloomington.
  • Peters, Krijn, 2006. ‘Footpaths to Reintegration: War, Youth and Rural Crisis in Sierra Leone’. PhD Thesis, Technology and Agrarian Development Group, Wageningen University, Netherlands.
  • Peters, Krijn and Paul Richards, 1998. ‘Why We Fight: Voices of Youth Ex-combatants in Sierra Leone’. Africa, 68 (1): 183210.
  • Porteres, Roland, 1976. ‘Domestication of African Cereals: Eleusine, Fonio, Black Fonio, Teff, Brachiara, Paspalum, Pennisetum and African Rice’. In The Origins of African Plant Domestication, eds JanDe Wet, AnnStemler and JackHarlan. The Hague: Mouton.
  • Raulin, H., 1957. Mission d’Étude des Groupements Immigrés en Côte d’Ivoire. Paris: ORSTOM.
  • Richards, Paul, 1986. Coping with Hunger: Hazard and Experiment in an African Rice-farming System. London: Allen and Unwin.
  • Richards, Paul, 1996. Fighting for the Rain Forest: War, Youth and Resources in Sierra Leone. Oxford: Currey (additional material 1998).
  • Richards, Paul, 2005a. ‘To Fight or to Farm? Agrarian Dimensions of the Mano River Conflicts (Liberia and Sierra Leone)’. African Affairs, 106 (417): 57190.
  • Richards, Paul, 2005b. ‘New War. An Ethnographic Approach’. In No Peace, No War: An Anthropology of Contemporary Armed Conflicts, ed. PaulRichards, 121. Oxford and Athens: James Currey and Ohio University Press.
  • Richards, P., 2005c. ‘Green Book Millenarians? The Sierra Leone War from the Perspective of an Anthropology of Religion’. In Religion and Civil War in Africa, ed. NielsKastfelt. London: C. Hurst.
  • Richards, Paul, Steven Archibald, Khadija Bah and James Vincent, 2003. ‘Where have all the young people gone? Transitioning ex-combatants towards community reconstruction after the war in Sierra Leone’. Unpublished report for the National Commission for Disarmament, Demobilization and Reintegration, Freetown.
  • Richards, Paul, Khadija Bah and James Vincent, 2004. Social Capital and Survival: Prospects for Community-driven Development in Post-conflict Sierra Leone. Social Development Papers: Community Driven Development/Conflict Prevention and Reconstruction, Paper No. 12, April 2004. Washington, DC: The World Bank.
  • Richards, Paul, Steven Archibald, Beverlee Bruce, Watta Modad, Edward Mulbah, Tornorlah Varpilah and James Vincent, 2005. Community Cohesion in Liberia: A Post-war Rapid Rural Assessment. Social Development Papers: Conflict Prevention and Reconstruction, Paper No. 21, January 2005. Washington, DC: The World Bank.
  • Rougerie, Gabriel, 1957. ‘Les Pays Agni du Sud-Est de la Côte d’Ivoire Forestière’. Études Éburnéennes, 6: 7212.
  • RUF/SL, 1995. Footpaths to Democracy: Toward a New Sierra Leone. No stated place of publication: The Revolutionary United Front of Sierra Leone.
  • Seddon, David, ed., 1978. Relations of Production: Marxist Approaches to Economic Anthropology. London: Frank Cass.
  • Sierra Leone Arms Investigation, 1998. A report into the Sandline affair by Sir Thomas Legg and Sir Robin Ibbs (ordered to be printed by The House of Commons, 27 July 1998). London: The Stationery Office (available at, follow links to official documents).
  • Smillie, Ian, Lansana Gberie and Ralph Hazleton, 2000. The Heart of the Matter: Sierra Leone, Diamonds and Human Security. Ottawa: Partnership Africa Canada.
  • Stavenhagen, Rodolfo, 1969. Les Classes Sociales dans les Sociétés Agraires. Paris: Éditions Anthropos.
  • Terray, E., 1995. Une Histoire du Royaume Abron du Gyaman. Des Origines à la Conquête Coloniale. Paris: Karthala.
  • Wilks, I., 1993. Forests of Gold: Essays on the Akan and the Kingdom of Asante. Athens: Ohio University Press.
  • Zack-Williams, Alfred B., 1995. Tributors, Supporters and Merchant Capital: Mining and Under-development in Sierra Leone. Aldershot: Avebury Press.