The historicity of politics in Africa is here, very classically, a history of terroirs. (Bayart Reference Bayart1993: 265)
INTRODUCTION
The role of land in African agrarian societies is complex, contentious, and critical. Land continues to support the livelihoods of millions, power national economies, and hold deep symbolic power in rural as well as urban settings. One facet of land in Africa that is particularly sensitive is the process of geographic boundary formation. In West Africa, boundaries have historically been porous, vague, and of secondary political importance due to the region's sparse and shifting demographics, which rendered the control of people a more pressing task for authorities than control of territory (Raynaut Reference Raynaut1997).
As competition over land resources has become a defining characteristic of agrarian West Africa (Berry Reference Berry2009) and decentralized local governance has become widespread in the region (Idelman Reference Idelman2009), geographic boundary formation has taken on new importance within local political and social relations. Geographic boundaries in West Africa derive their importance from the role they play in defining the social groups through which people gain access to productive resources and participate in politics more generally (Berry Reference Berry1993; Kuba and Lentz Reference Kuba and Lentz2006). At stake are the critical issues of belonging and exclusion, as evocations of autochthony – rights that are based on settlement precedent and conquest – play an instrumental role in disputes over resource control (Breusers Reference Breusers2001; Chauveau Reference Chauveau, Kuba and Lentz2006; Lavigne Delville Reference Lavigne Delville2002). These are important matters as agriculture is now viewed as key to accelerating poverty reduction in sub-Saharan Africa and land in historically neglected areas is of growing interest to governments and private investors. Secondly, climate change-induced higher temperatures and more erratic rainfall in the region will likely increase migration within rural areas and heighten resource competition even further in the coming decades. Across West Africa, the responses to these challenges typically involve boundary-based planning efforts that are supported by national laws, international conventions and development agency programmes.
This paper focuses on the role of boundary formation within the relationship between land, territory and local political identity in agrarian West Africa. Certain scholars have focused on how the imposition of fixed boundaries on historically unbounded areas (Bassett et al. Reference Bassett, Blanc-Pamard and Boutrais2007; Le Meur Reference Le Meur2006) stokes intra-community resource competition and conflict (Gray Reference Gray2002). This paper follows these accounts of boundary formation, particularly in terms of their impacts on the politics between groups claiming autochthony and those who are categorized as ‘strangers’, such as migrant farmers and herders. However, this paper demonstrates that the political consequences of boundary formation depend most critically on how boundaries re-order the socio-spatial relationships between autochthonous hosts and strangers. Boundary processes affect these host–stranger relationships in contingent ways that contribute to divergent political outcomes that have important impacts on people's access to resources as well as on emerging local democratic institutions. Host–stranger relationships influence local political outcomes by structuring the covenants – customary agreements between farmers and herders (Djiré Reference Djiré2003; Traoré Reference Traoré, Toulmin, Lavigne Delville and Traoré2002) – that act as a critical link between the customary and formal spheres of governance.
The paper's argument is based on a case study of a recent development project that experienced success and failure in its attempt to facilitate a large-scale bounded livestock corridor in western Mali. Livestock corridors in dryland West Africa are delineated pathways between pastoral resources, such as water points and grazing areas, at scales ranging from a few to over 100 kilometres. Dryland ecosystems are characterized by variable rainfall, which necessitates such livestock movements, including seasonal migrations known as transhumance. Although corridors are increasingly necessary to sustain mobility-based livestock production in dryland West Africa, they represent controversial endeavours of boundary formation that disrupt local power relations between herders and farmers. Their importance derives as much from these political effects as from their intended purpose of facilitating livestock mobility.
The corridor project in question succeeded in one location and failed in another due to the specific socio-spatial characteristics of farmer–herder relationships in each place. Farmers in Dioumara rejected the proposed corridor because they perceived its boundaries as strengthening the land tenure of herders who had established rights outside the local host–stranger relationship structure. In that area, where farmers and herders coexist and compete in overlapping zones of contact (Lentz Reference Lentz, Kuba and Lentz2006), questions of political autonomy and primary resource rights are uncertain and highly contested. By contrast, farmers in the Fuladougou, an area where pastoralist herders are still seasonal guests, accepted the corridor because it provided them with a legitimate way of reinforcing their host–stranger relationship with the herders and therefore of maintaining political and land tenure supremacy. In doing so, farmers cunningly moved to pre-empt the sort of autonomous land rights and pastoral settlement that the project was seen to be strengthening further north in Dioumara.
POLITICAL ECOLOGY AND RESOURCE TENURE
This paper's concern with boundary formation as a socio-spatial practice is informed by political ecology through its attention to geographic scale and power relations that are influenced as much by discursive struggle as by material constraint (Bassett and Zimmerer Reference Bassett and Zimmerer2003). Boundary formation itself can be understood as a form of territoriality, which is a ‘historically sensitive strategy to affect, influence, and control’ (Sack Reference Sack1986: 1). Secondly, land use schemes such as livestock corridors, land titling initiatives and other zoning projects are being implemented within a broader West African political landscape that is being reformed through democratic decentralization, which, in many cases, has strengthened discourses of autochthony and other customary instruments of power (Chauveau Reference Chauveau2007; Hagmann and Péclard Reference Hagmann and Péclard2010; Nijenhuis Reference Nijenhuis2003). This complicates the implementation of national laws such as the Malian pastoral charter, which includes bounded livestock corridors as a central provision. The pastoral charter was first proposed as a policy measure several decades earlier (cf. Gallais and Boudet Reference Gallais and Boudet1980) but it did not become a law until 2001, which was after Mali's democratic revolution and coincided with a period of rapid development of the country's decentralization process. The charter's implementation became an integral part of this process and the ways in which local authorities attempted to gain legitimacy by mediating questions surrounding land tenure and group identity (Berry Reference Berry2009; Cleaver Reference Cleaver2007; Sikor and Lund Reference Sikor and Lund2009). It is therefore necessary to contextualize authority, territory and boundaries broadly within a framework of land tenure and specifically within the host–stranger relationship, which is a central organizing principle of customary tenure systems in West African agrarian political economies (Kopytoff Reference Kopytoff1987).
Land tenure and the internal frontier
Customary land tenure – defined as the social organization of agrarian space (Le Bris et al. Reference Le Bris, Le Roy and Leimdorfer1982: 14) – is based throughout rural West Africa on first occupancy, which provides later generations with autochthonous land tenure and political power over the territory claimed by their ancestors. Mobile pastoralists who pass through these areas have historically relied on secondary rights to water, grazing resources and passage, which they would acquire through host–stranger relationships with these land-controlling groups. A critical issue surrounding the livestock corridor in this case study is whether these groups perceived its establishment as falling within these traditional secondary rights or as a new form of land right that would exist outside their existing customary authority. As competition for land access in agrarian West Africa has increased and these systems of rights have come under strain, the political ramifications of territorial boundaries have also increased. As a result, host–stranger relations are now influenced by the actions of the state and other actors that are involved with interventions in land tenure relations and the formation of new boundaries (Jacob and Le Meur Reference Jacob and Le Meur2010).
Boundaries, territory and power
Physical boundaries have long been an integral part of agrarian socio-spatial organization in West Africa (Kuba and Lentz Reference Kuba and Lentz2006). In precolonial West Africa, vague and overlapping frontier zones typically separated political communities ranging from small villages to large kingdoms. Nonetheless, certain historically evident boundaries can be construed as fixed and their present-day effects are part of a historic continuity that is contingent on by whom, where and how they were formed.
The Maacina empire of Mali's inland Niger delta provides the best example of fixed boundary formation in precolonial West Africa. Cissé (Reference Cissé, Le Bris, Le Roy and Leimdorfer1982) describes how theocratic rulers territorialized clan lands and land use zones, including livestock corridors, in the form of leydi. Verkijika (Reference Verkijika1986) and Asiwaju (Reference Asiwaju1983: 46) provide several other examples where ‘concern for precisely demarcated territorial possession is not lacking’. Verkijika (Reference Verkijika1986: 61) notes the use of ‘geometric lines’ and ‘fixed points’ to define farmland boundaries in Cameroon, even in areas where political boundaries were vague. Africans were ‘aware of their territorial limits and when they crossed ethnic boundaries’ (Verkijika Reference Verkijika1986: 59), and an intimate connection continues to exist between spatial, social and political boundaries in all their forms. These divisions and boundaries have played a formative role in how Malians in rural areas and their customary leaders are responding to the country's decentralization process (Idelman Reference Idelman2009). Host–stranger relationships in particular, and the covenants they structure, are playing a prominent role in many of the disputes surrounding the boundaries and membership of local jurisdictions.
Livestock corridors, dryland ecology and governance in West Africa
Boundaries that delimit livestock corridors affect host–stranger relations in particular ways because they cut across the fundamental livelihood changes that pastoralist herders are currently experiencing. Mobility-based livestock production systems in West Africa have been eroded in recent decades by agricultural expansion and recurrent drought (Dorman et al. Reference Dorman, Hammett and Nugent2007). This has prompted many herders to settle and begin pursuing primary rights to land resources (Traoré Reference Traoré, Toulmin, Lavigne Delville and Traoré2002). Nonetheless, livestock mobility remains a critical adaptation to dryland ecological variability and corridors are taking on increased urgency to protect mobility in the current context of political decentralization and environmental change. This has created a major governance challenge in agrarian Africa: how to promote mobility and flexible resource access in the context of growing land pressure and heightened political competition over the control of resources (Cousins Reference Cousins, Toulmin and Quan2000).Footnote 1
Corridors with fixed boundaries are the conventional approach to protecting pastoral livestock mobility, yet, as the following case study will demonstrate, these corridors and the formation of their boundaries ultimately serve as negotiating arenas (Hagmann and Péclard Reference Hagmann and Péclard2010) for broader power struggles over resource tenure and access rights. Such interventions tend to produce normative categories that indirectly and unintentionally shape both the political arena and the relations between the actors themselves (Colin et al. Reference Colin, Le Meur and Leonard2009). As the two parts of this case illustrate, larger-scale boundary formation processes such as livestock corridors are inscribed and acquire meaning within local political landscapes and through the differing forms of autochthony that reproduce them. These differences have an impact not only on farmer–herder relations and their livelihoods but also on the ways in which their struggles are shaping new democratic institutions in rural areas.
PREGESCO'S LONG-DISTANCE TRANSHUMANCE CORRIDOR
The livestock corridor that is the focus of this paper was part of a grass-roots development project initiated by the Swiss NGO Helvetas in response to a series of violent clashes between farmers and herders in western Mali that occurred during the late 1990s (Beeler Reference Beeler2006). Competition over resources and the erosion of livestock mobility were viewed as factors in farmer–herder conflict and the corridor was seen as a potentially effective way to improve relations through reforms in local resource governance. Although the project's agents ostensibly maintained advisory and coordinating roles, their work inevitably became politicized by acting as a technology of government (Cleaver Reference Cleaver2007) through their use of specific forms of participation and their framings of specific problems and solutions (such as mobility and corridors) in the evolving context of decentralization.
The establishment of an official livestock corridor along a traditional transhumance route became a major component of Pregesco's work. Although the project focused on a single route, its northern and southern portions are characterized by different histories, environments and political geographies. Pregesco intervened in a total of six communes: Dianguirdé and Dioumara to the north as well as Kotouba, Madina, Kassaro and Sebekoro to the south (Figure 1). The southern communes comprise an area known since the precolonial era as the Fuladougou. The project initially concentrated its livestock corridor efforts in the communes of Dioumara, Kotouba and Madina, and this paper focuses on the contrasting outcomes in these locations. Local participants in Kotouba and Madina worked collectively with Pregesco on the Fuladougou corridor, which forms the first part of this paper's case study.
Figure 1 The proposed livestock corridor and the six communes where Pregesco intervened
The Pregesco corridor in the Fuladougou: timing the transhumant arrival
Transhumant herders began using the Fuladougou as an annual dry season destination only following the devastating droughts of the early 1970s and 1980s. Initial forays into the Fuladougou had been relatively short due to exposure to bovine maladies such as trypanosomiasis and the fact that resources further north were still adequate. Herds travelling south could easily avoid cropped fields by waiting in the bush for farmers to harvest their crops, after which pastoralists would arrive in villages to exchange milk for grain. The droughts severely reduced the availability of fodder in traditional dry season grazing areas, and the Fuladougou, which had historically been too humid to support Sahelian livestock species, became an attractive dry season destination. Transhumant herders from the Sambourou clan of Dilly had established a host–stranger relationship with the autochthonous rights-holding Diakité clan during the first half of the twentieth century and it remained unchanged even as increasing numbers of herders began arriving in the area following the droughts.Footnote 2 The Sambourou–Diakité covenant had the broad contours of other host–stranger relationships in terms of its minimal conditions: transhumant herders were freely invited to come to the Fuladougou as seasonal guests on the condition that their animals did not damage crops or cause other problems for farmers.
Covenants between hosts and strangers, such as the one between the Sambourou and Diakité clans, represent an important channel for the reproduction of autochthony as it concerns access to productive resources. That particular covenant was a manifestation of the Fuladougou's geography and its recent precolonial past. Following the destruction of the area by jihadist El Hadji Omar Tall in the mid-nineteenth century, the Diakité clan's customary authority became hegemonic over nearly all of the Fuladougou, since Tall's forces did not occupy the area and no other group could exercise credible and antecedent claims of conquest or occupation (Brottem Reference Brottem2013). French colonial authorities later formalized that hegemony by appointing Diakité clan leaders to local positions in the colonial administration. The sparse population, due to isolation and endemic river blindness, further reinforced the Diakité clan's hegemonic rule over most of the Fuladougou and ensured that it would continue during the post-independence era. At the time when the Sambourou pastoralists approached the Diakité clan about becoming seasonal guests, its leaders spoke of the milk and meat that their presence would bring.Footnote 3 Herein lies an important underpinning of the original Diakité–Sambourou covenant: its design, reflecting Kopytoff's (Reference Kopytoff1987) frontier concept, had much more to do with the recruitment of new subjects than outright control of the transhumant pastoralists, which would later become its paramount objective.
The original covenant suited both groups' needs until ten to fifteen years ago, when early herd arrival started to become a problem. Herders began lengthening their seasonal stays in the Fuladougou from three to four months to upwards of ten months out of the year. According to herders, the main reason was that surface water resources began drying much more quickly along the transhumance route at the end of the rainy season and this accelerated their rate of descent. A second reason was the expansion of frontier agriculture in areas that were historically devoid of settlements, including along the boundaries of the Boucle du Baoulé biosphere reserve, an area that herders used while waiting for crops to be harvested in the Fuladougou. The disappearance of the uncultivated bush that served as a waiting area has compounded the problem of earlier, faster descents into the Fuladougou as animals have fewer places to graze than before and face greater risk of damaging fields. Moreover, cultivation contributes to the elimination of perennial grasses (Miehe et al. Reference Miehe, Kluge, von Wehrden and Retzer2010), which historically provided nutritious livestock fodder during the long, slow descent at the end of the rainy season. Bush fires now eliminate the annual grasses that grow in their place and remove the ecological basis of the traditional practice of transhumance (Brottem Reference Brottem2013).
As herders adapt to these changes by adjusting their transhumance patterns, they face an added challenge in the Boucle du Baoulé biosphere reserve. Although the biosphere reserve includes a pastoral zone that enables herders to pass through without trespassing in the wildlife zone, relations between herders and park rangers are tense and unpredictable. As herders spend more time in the reserve, they clash with foresters with increasing frequency. The dynamics of herd movements through the reserve have recently had a significant influence on farmers' attitudes towards transhumance. In October 2009, dozens of herders fled from foresters and abandoned their animals, which caused them to move uncontrolled towards the Fuladougou's cultivated areas. These events resulted in a spike in crop damage cases that year and hardened the farmers' stance towards transhumant herders as ‘nomads’ and ‘outsiders’, despite the fact that many of the herders had begun spending the majority of their time in the Fuladougou and were making substantial investments in the community. The stance of local farmers was increasingly at odds with the livelihood aspirations of transhumant herders, many of whom wish to settle permanently in the Fuladougou, as others have done in Dioumara, where the second part of the case study takes place. This latent conflict over the terms of belonging for the herders has become a pivotal factor in negotiations over Pregesco's proposed livestock corridor. Even as herders began seeking to settle in the Fuladougou, many local farmers were calling for their systematic expulsion. This would represent a revocation of the Diakité–Sambourou covenant, which, in the eyes of farmers, has been violated repeatedly and flagrantly.
The Pregesco livestock corridor quickly became the arena of ‘contextualized negotiations’ (Hochet Reference Hochet2005: 7) through which the two groups would struggle to secure their own interests and ambitions. For farmers, the corridor itself became the means by which they could reproduce their autochthony under a new covenant in the guise of a modern land use plan that would satisfy the NGO, donors and the Malian government. This strategic move came as a direct result of threats to the Diakité's hegemony that had once characterized their customary autochthonous rule and their host–stranger relations with transhumant herders. In-migration had made the Diakité clan a minority group in their homeland, transhumant herders were increasingly assertive about their rights as local citizens, and their local governments were obliged to implement national laws – all of which would inevitably weaken the Diakité's authority even further.
The corridor, as the following section demonstrates, represented a fundamentally different kind of covenant that would potentially enable the clan to address their tenuous position vis-à-vis the herders. While the original Sambourou–Diakité covenant tacitly aimed to recruit transhumant herders as seasonal guests, the new covenant represented a way for the Diakité clan to reassert its vulnerable political power by enmeshing it within the new local government structure via the corridor project, which was an important component of official development strategy in the communes of Kotouba and Madina. Since the Diakité could not – and most would not want to – actually expel the herders, their overarching political goal would be to use the corridor to dictate the terms of legitimacy for transhumant herder presence in the Fuladougou. The process of the corridor's delimitation illustrates how, over the period of several years, the clan would achieve this goal.
When Pregesco began the delimitation process in 2003, autochthonous farmers and transhumant herders differed immediately over its position. Farmers wanted to formalize a corridor that would follow the western boundary of the biosphere wildlife zone (Figure 1). This would keep roaming livestock a safe distance from their fields, but herders complained that cropped fields had occupied most of the area along the reserve's boundary, which offered little in the way of water and held a greater risk of punishment by park rangers. Herders proposed instead to formalize their traditional route, which connected to a series of large wetlands that lay much closer to autochthonous settlements and fields (Figure 1). A decisive factor in the corridor planning process was the lack of autonomous settlements of sedentary herders. A small number of pastoralists had settled, but only as individual households within farming villages. This settlement pattern stands as a contrast to Dioumara in the way in which it left the Diakité clan's autochthonous land tenure rights uncontested and the process of corridor delimitation largely in their hands. Farmers unanimously opposed linking the corridor to the wetland areas, which, like the biosphere reserve, had become surrounded by agricultural hamlets over the past fifteen years. Aside from this common position, however, attitudes towards the corridor varied between agricultural villages depending on their geographic location and position within the local political hierarchy.
Representatives of Tourébougou, the northernmost village of the Fuladougou, immediately accepted a corridor that would pass through the centre of their territory for two reasons. The first is that very few herders stop in the area due to its lack of dry season resources, so the corridor would truly be a passage that would facilitate herd movements in and out of Tourébougou's territory. Secondly, Tourébougou's acceptance of the corridor legitimized their agricultural expansion into the biosphere reserve pastoral zone because the village had ceded arable land for its establishment. This situation was reflected in the village's attitude towards herders: the chief's son described ‘good relations with the herders’, shared social ceremonies and mutual respect. This position contrasted sharply with the attitudes of the migrant Bambara, themselves guests of the Diakité clan, who live in a cluster of hamlets located just 10 kilometres to the south of Tourébougou and who were excluded from the corridor planning process. Farmers in those settlements feel that the herders ‘do not respect’ their fields and ‘bring nothing of value’ to the area, and they claim to have little contact with the herders, who tend to move through the area at night.
Tourébougou's stance on the corridor also posed problems for the autochthonous hamlet of Moptiko, which is located directly to the south of Tourébougou on the banks of a key water source (Figure 1). The chief of this hamlet repeatedly invoked the previous corridor that passed along the biosphere reserve boundary and directly through the recently established Bambara settlements. This position proved untenable at the level of the commune, however, and Moptiko, along with several other hamlets, was obliged to accept that the corridor would pass through its territories. The village of Madina faced similar problems with the proposed corridor. Madina is situated at the base of a large rock formation and faces out towards the biosphere reserve, both of which limit the village's access to arable land. A wealthy and influential village leader, Bakary Diakité, used the village land shortage as a pretext for clearing a field in the middle of the proposed corridor, arguing that ‘the land adjacent to the village is exhausted’ and the corridor ‘runs through the village's fallow area’. Despite pleas from most other stakeholders involved, Bakary's field remained in place for several years and the corridor was eventually moved closer to the reserve boundary in order to accommodate it.
Despite these contestations, by the summer of 2011, the local governments in Madina and Kotouba were in the process of ratifying the boundaries of the corridor of varying width (500 metres to 2 kilometres) that would run nearly 100 kilometres north to south through the two communes. This was a remarkable feat considering that the process touched on highly sensitive land tenure relations and involved fifteen official villages as well as many more unofficial hamlets in their jurisdictions. Despite the corridor's remarkable success on some levels, it nonetheless continued to pose a number of unresolved challenges. The issue of herd movements and crop damage beyond the corridor boundaries remains a concern. As the secretary general of one local government commented: ‘The corridor could make things worse. Now farmers will look at problems that occur outside the corridor as evidence of the herders’ bad intentions.’ Additionally, the large scale of the corridor and its implication of movement led it to be designed as a linear feature that cuts through the landscape. This belies two important characteristics of transhumance. First, herders establish their seasonal camps at various points in proximity to, but not necessarily within, corridors. Secondly, herd movements are geographically variable, which means that livestock inevitably leave the corridors and traverse cultivated areas to access water and grazing resources, which are typically inadequate along or within corridors. This reflects a basic problem of pastoral resource availability and, as one herder put it, ‘I'm not sure the corridor will work as it will not provide our animals with enough to eat.’
The Fuladougou's corridor and the imperative of movement
A principal reason for the project's relative success in the Fuladougou was that the corridor's design suited the interests of farmers who perceived it as a way to facilitate and even ensure the seasonal movement of transhumant livestock herds out of their village territories. They logically argued that they were benevolently ceding territory and recognizing the seasonal access rights of herders, who then must reciprocate by acknowledging farmers' land rights and by staying out of the Fuladougou during the cropping and harvesting seasons. Since the events of 2009, farmers in the Fuladougou have unanimously tied the legitimacy of the corridor to a proposed schedule of herd movements that corresponds to their agricultural calendar. A common refrain is: ‘We have no problem with the nomads, as long as they show up after January.’ But this position is deeply out of step with transhumant herders' livelihood aspirations, perceptions of their own legitimate belonging in the Fuladougou, and, especially, the changing environmental conditions in the area that make a February arrival exceedingly difficult for their livestock.
Even as the corridor project steadily progressed towards completion and herders quietly continued to seek ways to settle in the Fuladougou, tensions remained over the corridor and what implications it would have for farmer–herder relations. The need to address lingering tensions from the events of 2009 led the Association pour le Développement de la Fouladougou (ADF) to call a meeting in March 2010 between farmers, herders, local authorities and local clan members residing in the capital city of Bamako. The official objective of the meeting was to settle differences over the corridor in order to solidify its implementation. However, the political impact of the meeting came because it served as the moment when the renegotiation of the Diakité–Sambourou covenant crystallized in the official policy process. The meeting allowed different key actors to weigh in, such as a prominent lawyer from Bamako, Mamadou Diakité, who reminded everyone that deliberations must be guided by the rule of law, which was a direct reference to the pastoral charter. This was a subtle way of stating that, no matter how angry farmers were, they could not stop the corridor from being established. Yet the farmers were nonetheless given a prominent role in the meeting, which was well attended and thus made a strong impression on those involved, including all the relevant local politicians – the mayors, the sous-préfet – and even several foresters from the biosphere reserve. As they listened to the farmers' complaints about recurrent crop damage, the meeting and its outcome would result in the blueprint for official local policy on farmer–herder relations in the years to come. This policy would carry the contours of a new covenant that would both redefine Diakité autochthony and imbricate it with the formal sphere of local governance.
Although the farmers had had their day, the ADF was also cognizant of the problems facing herders in the north and proposed a transhumance calendar based on annual rainfall, which was reasonable given its variability from year to year. However, this would not only mean that herders would continue to arrive early when rainfall was poor, but that this would be officially sanctioned. Farmers, by this point, were adamant that the corridor would be acceptable only if it were based on a calendar that suited their own needs as local autochthones and holders of customary land tenure. Transhumant herders would have their corridor, but before the end of January and after 1 June they would not be welcome in the Fuladougou. The meeting ended on an ambiguous note but it was a clear political victory for farmers who wanted to maintain political power through an updated host–stranger relationship with transhumant herders by ensuring that the corridor, once officially established, would be linked to livestock movements and a seasonal herder presence in the Fuladougou. Project workers recognized that the calendar posed problems but knew that they were in no position to contest the views of farmers, particularly in terms of land tenure. Local officials, aware that the outcome was well within the legal mandate of the pastoral charter, were happy to uphold the law while satisfying their main political constituents: local farmers, particularly members of the Diakité clan. Meanwhile, farmers themselves accomplished their larger political goal by using the corridor to pre-empt, or at least forestall, herders' establishment of autonomous land rights, which had been the cause of so many problems further north in Dioumara.
The Pregesco corridor in Dioumara: contested tenure and failure
The Pregesco livestock corridor was successful, albeit with controversy, in the Fuladougou but it failed entirely in Dioumara. This is largely attributable to the permanent settlement of herders and their tenure relations with farmers in that commune. This section describes how certain geographic and historic factors in Dioumara led farmers to perceive the corridor project in very different terms from those in the Fuladougou: namely, as a transfer of land rights to settled herders. Dioumara, located in the semi-arid Sudano-Sahelian zone, has agro-ecological and historical characteristics that differ drastically from the more humid Fuladougou. Once the dry season destination for Malian herders who would pass the rainy season in Mauritania, Dioumara has, since the droughts of the early 1970s, hosted sedentary herders as well as those who leave for the Fuladougou during the dry season. Dioumara is more densely populated and ethnically diverse than the Fuladougou: sizeable communities of Bambara, Sarakolé, Kakolo, Moorish and Fulani inhabit the area. In the Fuladougou, a single clan had established autochthonous hegemony over a relatively contiguous geographic area, but, in Dioumara, autochthony is characterized by a patchwork of village territories with Bambara and Sarakolé groups maintaining first-occupancy rights within much smaller areas. These characteristics are due to Dioumara's location in the population arc that runs through the Sudano-Sahelian portion of western Mali, where settlement densities are higher compared with those further south, due to the legacies of the precolonial Kaarta and Ghana empires (Raynaut Reference Raynaut1997). Additionally, Dioumara's geographic proximity to areas such as Dilly, which is under the customary control of Fulani pastoralists, was a critical factor in how the corridor project evolved.
Dioumara also has a more complex and contentious history of corridor projects, and they have had a much greater impact on the outcome of the Pregesco corridor than in the Fuladougou. The area has long been an important zone of agro-pastoralism and, during the 1970s and 1980s, the Opération de Développement Intégré du Kaarta (ODIK) was the first project to implement a system of livestock corridors and grazing zones. ODIK was one of the top-down, integrated rural development projects that were active throughout the country before democratization, and, although it was part of an effort by the state to exert control over natural resources, the operation was also driven by the concern that agriculture mechanization was putting pastoral resources at risk. The project established livestock corridors on two occasions – in 1982 and 1986 (Letheve and Dainro-Tadion Reference Letheve and Dainro-Tadion1996) – and delimited the corridors each time with concrete markers.
The first attempt failed, according to local accounts, due to a lack of stakeholder involvement. The second attempt involved pastoralists who nonetheless complained that they were ‘not given responsibility’ and therefore ‘did not complain’ when farmers tore out the markers and began cultivating within the corridors created by ODIK (Letheve and Dainro-Tadion Reference Letheve and Dainro-Tadion1996: 29). Once the project was terminated, farmers denied that the corridors ever even existed. At that point, it was possible for farmers to use such a tactic because Mali lacked accountable local institutions that would ensure the protection of the corridors and negotiate any conflicts surrounding their use. The denial was a strategic move on the part of farmers who correctly sensed that pastoralists would try to use the corridors as part of their post-drought strategy to secure land rights in Dioumara, where they could cultivate crops more dependably than in their customary territories further north.
A second component of ODIK's work in the area that greatly affected farmer–herder relations was the construction of four pastoral wells near the autochthonous villages of Mangara and Zambougou, located along historic transhumance pathways that originate in the Sahelian rangelands further north (Figure 2). The wells were intended to improve the management of livestock movements by providing animals with water at the end of the rainy season, which would ostensibly slow the movements down during a time of year when crops further south are highly vulnerable to damage. Instead, herders used the wells as an opportunity to establish permanent settlements near two of them in the territory of Zambougou. This enabled herders to establish autonomous land rights, which initially were not enough of a concern for neighbouring farmers to contest and potentially cause a conflict to erupt. However, in the 1990s, when farmers began to rely on the wells for their newly acquired livestock as well as their own drinking water, competition with the herders started to become a problem. By this time, however, not only were the herders staying in the area for the entire year, but they also had made substantial investments in maintaining the wells, which conveyed user rights over them. When conflicts over the wells worsened relations between the farmers and herders, a stakeholder committee implicated both groups in their management, which further strengthened the herders' rights to reside and participate in politics in Dioumara.
Figure 2 The traditional transhumance pathway and key sites of pastoral settlement in Dioumara
Pastoralist herders had begun settling in the commune of Dioumara following the 1970s droughts, particularly in areas where they could establish defensible claims that their new camps were located within territories controlled by their clans from Dilly, located to the north-east (Figure 2). Lacking the hegemonic autochthony of their counterparts in the Fuladougou, sedentary farmers were not able to effectively counter these claims, even as the stakes of land tenure rights were steadily increasing in the area. The herders' claims were strengthened by the presence in Dilly of a powerful Islamic marabout and ethnic Fulani, Baba Hama Kane, who was able to exert moral authority concerning issues between farmers and herders. Meanwhile, as farmers increasingly invested in livestock and settled herders began farming (Toulmin Reference Toulmin1985), competition for land resources grew rapidly.
Largely sedentary and facing diminishing areas of natural vegetation, livestock owners in Dioumara began relying on feed supplements, which incited them to further expand their participation in markets, strengthen their level of organization, and increase their level of autonomy from local farmers (Hochet Reference Hochet2005). Expanding fields also meant that crop damage became a growing problem that was easily resolved between farmers from related clans but highly politicized between farmers and herders. This prompted new calls for initiatives to protect pastoral resource access and improve farmer–herder relations, which led to a third wave of livestock corridor delimitation through gestion des terroirs villageois (GTV) projects. However, since these focused on sedentary villages, the corridors were inadequate for transhumance because they did not connect rainy and dry season grazing areas across large enough geographic scales. Furthermore, they were not formally linked to the local governments emerging through political decentralization so they did not gain adequate political legitimacy.
The corridor project initiated by Pregesco carried more political weight than the GTV projects as it was supported by the pastoral charter and slated by local governments to become official policy. The Pregesco process in the north involved participants from two communes, including eighteen villages and thirty hamlets, as well as a new inter-communal commission to coordinate the work. Among the project objectives was achieving formal recognition for the traditional corridor, which had been informally accepted by farmers for more than a century. The justification was that agricultural expansion had reached the point where the existing system, which required herds to navigate the shifting fallow areas between fields, was no longer viable. During the planning process, farmers acknowledged this challenge and the presence of pastoral resources in their village territories.
By this time, agricultural and sedentary pastoral villages in Dioumara coexisted geographically and politically in complex and conflict-ridden relationships within which herders were not the strangers, nor the guests, of local farmers. The corridor would not replace a farmer–herder covenant grounded in strongly autochthonous claims or otherwise represent a way for farmers to control the movements of herders as it did in the Fuladougou. The annual departure of livestock from the Fuladougou, which created a geographic pretext for farmers' autochthonous relationship to herders, had been lost decades earlier in Dioumara. Farmers there could not create a discourse of herders as ‘nomads’ as they had done in the Fuladougou. By the time Pregesco began, herders in Dioumara had established multi-stranded rights through various channels – both formal and informal – that resulted in a situation of direct competition with farmers for access to productive land resources. Farmers therefore perceived the corridor in zero-sum terms of land tenure and Pregesco as a pro-herder project because the project was attempting to codify new territorial areas for use by livestock, which would prohibit farmers from cultivating where they still possessed customary land tenure rights. Despite the commune's official support for the Pregesco project, the personal view of Mayor Batou Sissoko during the summer of 2010 was indicative of the general sentiment in Dioumara:
We need a corridor. But it is false to say that the corridor is for everyone just because everyone has livestock. There are peasants and there are herders. It is the peasants who will lose land because of the corridor and that is unacceptable.Footnote 4
The mayor had taken a political position, at least vis-à-vis his local constituents. He is a member of an autochthonous lineage and was voted into office by, and collects taxes largely from, local farmers. Further, as local farmers have framed the corridor as a transfer of tenure, the mayor was also quick to emphasize that land tenure is strictly a village affair, stating that ‘The commune doesn't really meddle in tenure issues,’ despite the fact that the local governments in Mali possess the legal powers to make decisions that would affect land tenure in order to implement the pastoral charter under national law. By contrast, although tenure is a village-level concern in the Fuladougou as well, it played a much less significant role in discourse surrounding the corridor there. In fact, by discursively linking the corridor to a calendar of herd movements, farmers in the Fuladougou with autochthonous rights had ensured that it would not become a tenure issue because any eventual settlement by herders would come with strict conditions and without claims to land.
Dioumara residents had become accustomed to land tenure interventions, expressed by an oft-repeated view among local farmers: ‘We do not want it but we know that you will make the corridor anyway.’ Although local residents initially cooperated with the Pregesco planning process, certain individuals expressed their opposition through provocation. For example, during the cropping season of 2009, one farmer cleared a field in the middle of a portion of the proposed corridor. While the farmer invoked the standard narratives of autochthonous domain and land scarcity, the act was an intentional gesture towards the corridor and any political actor who might support it. The view of Samba Samoura, a Dioumara elder, captured the project's political undertones:
Before, herders used to go up to Mauritania and then down to the Fuladougou. We want them to continue doing this. It is the Fulani with few animals who want the corridor. They like conflict, they don't pay taxes, and the corridor becomes a domain for them. They want to establish fields but most of all maximize their time here.
Although the Pregesco staff had pushed the corridor project forward with some success, when the consultations with each of the eighteen villages were complete and the time came to formalize the corridor, village leaders did a political about-face and declared unanimous opposition to it. On 29 November 2011, three delegates from each of the villages and hamlets were summoned to the Dioumara mayor's office for a final meeting, at which, presumably, the corridor would become official. After the president of the inter-communal commission and deputy mayor Samou Diarra reiterated the importance of ‘farmer–herder cohabitation’ and the need for a ‘shared recognition of the corridor’, delegates were given the floor. Among those who spoke was Michelle Traoré, who stated:
I think that the corridor[s] are recognized by everyone. Each year, the Mauritanian transhumant herders pass through here. By defining corridor[s] recognized by everyone, there will be fewer farmer–herder conflicts.
However, other delegates did not share Traoré's view. Baboutiné Samoura from the village of Touba Madina declared:
I think we need to let the herders continue finding their own path [through our territories]. It would be detrimental to take away land to serve as corridors. What will we do with the fields that are currently situated within the proposed corridors?
Néké Sissoko added: ‘We can cohabit without being placed in a straitjacket.’ He then proposed a break in the meeting for the village delegates to convene together. Following a thirty-minute break, delegate Haidara Diarisso from Dioumara took the floor to express their collective position that:
Based on the decisions of the village chiefs, we express our disagreement with the establishment of a fixed corridor. We wish that the herds pass as usual without other requirements. Our fields may need to move at any time so [a fixed corridor] is a problem. No single village is in agreement with the corridor even if certain ones previously accepted it.
The stage was set for a complete rejection of the corridor, which was the cornerstone of Pregesco's efforts to improve resource governance, particularly access to pastoral resources, within the legal framework of decentralization. Before the meeting ended, the representative of the area's herders, Hamala Sow, expressed his frustration in a statement that reflected the project's original justification: the traditional system of herd mobility was no longer effective.
I participated in the delegate meeting and I do not support their resolution but my lone voice was unable to influence the discussion. Look at how the corridors and wetlands are obstructed by fields. I think times have changed and we must not stay stuck on one position.
However, the sentiment among the farmers in Dioumara was that the corridor would become yet another avenue, like the wells before it, for herders to establish land rights outside the host–stranger relationships that generally provide farmers with a modicum of control over their herding neighbours. The weakness of host–stranger relations and fragmented autochthony in Dioumara are the direct result of its geography as a diverse agro-pastoral zone of contact where land rights and political power are more diffuse than in the Fuladougou. Although customary authorities in Dioumara effectively deployed the power of autochthony in their rejection of the Pregesco corridor, this particular outcome belied the weakness of their position vis-à-vis pastoralist herders in the area.
CONCLUSION
This paper has argued that the Pregesco livestock corridor demonstrates how boundary-making processes that accompany resource governance initiatives enter into host–stranger relationships in geographically and historically contingent ways that shape their outcomes. In these two cases, whether the corridor was perceived as a tenure claim or as a way to pre-empt potential tenure claims depended on the differing geographies of farmer–herder political relationships as well as on herders' changing livelihood strategies.
The divergent trajectories of the Fuladougou and Dioumara reflect two important long-term trends that are relevant to local democratic institutions across agrarian West Africa. The first trend concerns changing regimes of autochthonous rights, which will continue to play a fundamental role in political discourse and practice in the region. In terms of this paper's case studies, the Diakité clan is well positioned to maintain the strength of its claim to customary power in the Fuladougou. This may allow it to continue influencing policies that emanate from the local government, as it did when the clan redefined its covenant with transhumant herders. By contrast, in Dioumara, farmers used their autochthonous authority in their resistance to a legally sanctioned livestock corridor but they are unlikely to continue winning political battles through an antagonistic stance towards official policies. Such a stance will most likely weaken their customary power over the long term. In the Fuladougou, more deeply entrenched autochthonous rights will affect resource access regimes in important ways, as corridors and, eventually, grazing areas increasingly become aspects of everyday life. Farmers seeking new farmland and herders seeking new pasture will modify their strategies accordingly. In Dioumara, the erosion of customary authority will affect these decisions in different and perhaps surprising ways.
The second and more general trend concerns democratic decentralization, which is poised to take on greater importance in Mali as a proposed solution to the country's ongoing political crisis in its northern regions. This paper has demonstrated the specific ways in which democratic decentralization represents a moment when norms, rules and institutions themselves are reworked and eventually stabilized in geographically nuanced ways. Scholars have long appreciated the legally plural nature of rural institutions in West Africa without paying adequate attention to these geographic dimensions and how they create important divergences in the trajectories of local institutional change. Rather than focusing on the weakness of the central state in rural areas and the parallel systems of governance that result, this paper suggests that more focus is needed on how national initiatives such as the pastoral charter find expression within local autochthonous relationships and influence underlying institutions as a result. As decentralization deepens in West Africa, these dynamics will become a more prevalent and important part of local governance.
ACKNOWLEDGEMENTS
This research was funded through an IIE Fulbright Fellowship and a National Science Foundation Doctoral Dissertation Improvement Grant (BSC # 1130081). Special thanks to Mamadou Diakité, Ibrahim Sow, Bakary Coulibaly, and many unnamed Malians who contributed to this research.
