PhD, Graduate School of Geography, Clark University, Worcester MA, USA

October 1997

Supervised by
Douglas L Johnson, Billie Lee Turner II.

Reviewed by
Dianne Rocheleau, Robert E Ford (Loma Linda Univ, California)

The Central Plateau of Burkina Faso in semi-arid West Africa is well-known for a variety of soil and water conservation and environmental management programs initiated by the state, NGOs and international donors since the late 1970s. Development organizations have a strong presence in this region of high out-migration, rapid population growth and visible land degradation, and their presence has been encouraged by the government. A study of the impacts of 'village land use management' was carried out in cooperation with an environmental program in Bam Province from 1992-1993, in order to assess the environmental, social and economic impacts of these environmental conservation activities. A cultural ecological analysis was made of the livelihood systems of two remote Mossi and Yarse villages in Rollo departement, and the technical and social interventions carried out in these same communities were also monitored. The local Sudano-Sahelian ecology consists of diverse landscapes showing considerable modification, and is dominated by rain-fed agricultural production in nucleated Mossi settlements. Human adaptation to changeable resource endowments, technology, labor and ecology give rise to high variability in livelihood decisions, resulting in complex spatial patterns. Crop yields are variable, and are frequently insufficient to meet household and community needs. Intensification of agriculture is selective, and specific to certain individuals and sites. Off-farm incomes, animal sales and remittances from migrants are significant for many households.

Village land use management (gestion des terroirs villageois), engages sedentary 'communities' to overcome conflicts and to participate enthusiastically in communal environmental management of their 'village territories'. This approach has many positive benefits for social solidarities and for land rehabilitation, but it can overlook the complexities of land use, agricultural knowledge and the plural rationalities of Mossi and Yarse farmers. As this approach to land degradation spreads to other Sahelian countries and is widely adopted by development aid agencies for its 'participatory' elements, it first requires rigorous evaluation at the local level. The approach has nonetheless generated high levels of participation, new institutions, and buffered many individuals against food stress and hardship. By combining an analysis of agrarian and social systems with work on development interventions, the study contributes to a political ecology of environmental management. Using hybrid research methods, it bridges an all-too-common divide between theoretical understanding of nested scales of human-environment interventions, and the practical efforts of arid lands managers.

Order from University Microfilms International, Dissertation Number AAT 9813128, from here Price used to be $19.50 as PDF (web download to your printer), $29.50 Unbound Paper Copy.



  • Moral geographies and fieldwork
    Plan of action


  • Introduction
    The wider context for the study
    The local context for this study
    Organization of the study, and methods used
    Analytical Frameworks
    Cultural Ecology and the adaptational frame
    Political ecology
    Action-centered rural development
    Conclusion - the approach adopted


  • The Mossi: a people
    Political and social development of Burkina Faso
    Economy and Trade
    Agricultural production
    Social Organization
    Environmental Characteristics


  • Introduction
    The effects of land degradation on farming systems
    Land rehabilitation techniques on the Central Plateau
    Soil and Water Conservation - the project environment
    Gestion des terroirs - a second-best form of community development?
    Local Projects


  • Background
    Trends in land use
    Soil and water resources
    Development and history of village organizations
    Forestry and fuelwood
    Animal ownership and agro-pastoralism
    Wild foods and food stress


  • Introduction
    The Mossi Farming System
    Field positions and social changes
    The ring management system
    Pest Management
    Labor use and the farming cycle
    The Farmer Survey
    Farm sizes and soil type
    Productivity measures and labor inputs
    Discussion: agrarian change and agricultural intensification


  • Introduction
    Harvest data
    Soil and water conservation and agriculture in the study villages
    The benefits of SWC for crop yields
    Discussion and conclusion


  • Introduction - productive bricolage
    The major off-farm income sources
    Perceptions of expenditure and revenue sources
    The special role of migration
    Expenditure and income balances


  • Development messages and community needs
    A social actor perspective
    Supporting local institutions
    Communication problems in Gestion des Terroirs
    The village level
    Confusions at the village scale
    Confusions in the project
    A case study of the Training and Visit (T&V) extension system


  • Land degradation and "planned" resource management
    Agrarian change
    Village Organizations and Grass-Roots action
    Village Land
    Use Management and "community" resource management
    The implications for geographical practice


This dissertation has touched on the history and cultural ecology of the Mossi people, their transformation of the Central Plateau of Burkina Faso, and the failures and successes of environmental management initiatives in this region. The scale of analysis has shifted from the actions of individual farmers to the policies of aid donors and state agencies. The work is bound together by the need to understand changing human-environment relations, and the particular impacts of development organizations working in the domain of gestion des terroirs villageois (village land use management) and soil and water conservation. This Chapter summarizes the main findings, and brings the study back to its entry-point: the implications for sound environmental management, and geographical understanding of human-environment relations. Firstly, I discuss how local views and reactions to land degradation must be viewed as adaptive responses that incorporate external techniques and assistance where these accord with local social solidarities (solidarities that favor some communal activity at the community level) and individual needs. Secondly, I extend this argument to account for patterns of agrarian change. The case studies reveal that farmers are selectively intensifying and extensifying their agricultural systems under conditions of labor shortage and poor crop yields. Soil and water conservation is merely one component of adaptation to adverse conditions. Because the gestion des terroirs model of land rehabilitation is initiated by outside actors and there are still significant problems in its delivery to poorer or marginalized sectors of rural society, I term it a "second best" form of community development, albeit an exciting one, that can easily overlook inter-community differences and power structures unless it is very closely monitored or led almost entirely by inclusive village organizations. In section 9.4 I suggest that assistance to the farmers of the Central Plateau with environmental rehabilitation and the development of new agro-ecosystems is an important arena for the involvement of geographers, who are poorly represented in the region and in practical development work. I justify this presumptuous critique by pointing to the spatial skills required in the gestion des terroirs approach, and the frequent failure of political ecologists to actually work with the same development organizations that they are all to ready to criticize as power-laden and sometimes ineffective actors in rural development.

9.1 Land degradation and "planned" resource management
degradation continues to pose a threat to livelihood systems and the well-being of rural people on the Mossi Plateau. Its ecological, social and political complexities are now reasonably well understood in broad terms. The challenge to the development of sustainable agro-ecologies is less to design high-yielding farming systems (which, while beginning with simple erosion control, may involve agrochemical or other relatively costly inputs still scarce and expensive in this region), but rather to create sound management practices which local people can adapt and change to suit their own personal aspirations and those of their communities. This resource management agenda can adapt to a variety of demographic and political situations, and reflects current thinking in "farmer participatory research" (Reijntjes et al, 1992). It has laudable aims, but it has also been accused of being uncritically supportive of local community "groups", without defining the power structures and inequalities of those groups. In Burkina Faso it has been linked to a considerable confusion of donor interventions, projects, programs and government schemes in rural areas. The particular brand of development thinking represented by the German environmental project described in this study (PATECORE) accounts for local participation in environmental planning by means of research with groups of farmers, meetings, encouragements to institutions at the village level, and support to local government agencies. It also provides and supports the use of limited "external" technologies such as trucks and carts, to aid land rehabilitation activities largely carried out by farmers themselves. The state plays a role in this process as a bilateral partner at project level, and through providing agricultural extension agents who monitor and assist local level interventions.

This approach has done much to support environmental rehabilitation at the local level, and many of its training activities and interventions have allowed resource-poor farmers to participate in new activities. In many ways, the project has expanded the "range of choice" in everyday life (to adopt Gilbert White's famous phrase) and has assisted indigenous experimentation within locally adapted livelihood systems. The Central Plateau is now distinguished by its innovative approaches to the conservation of natural resources, notably soil and water, and is a true laboratory for such techniques.

The history of these interventions, as viewed through the eyes of farmers in two study villages, reveals much of interest about the process of environmental management, and shows how it has succeeded in some areas but failed in others. GTV methods build upon a sense of non-economic motivation and communal sharing in Mossi society (Fiske, 1990b). It is important, therefore, to identify how important these communal sharing motivations are in everyday decisions and actions. As Raynaut (1997:321) reminds us, Sahelian societies do not consist only of individualistic, rational economic actors. Nonetheless, Mossi farmers prefer (or are forced) to maximize their own market or economic opportunities at certain times, and I have shown in Chapter Seven that marketing and income generation are important to most households. Societies like the Mossi have shared values, traditions and world-views but this does not mean that they are composed of people with similar social relations and economic means. Environmental management is an activity where communal sharing motivations are vital, because of the large co-ordinated labor inputs required for diguette construction, tree-planting and so-on, and because of the need for inclusive village organizations to oversee and take decisions. But, simultaneously, there must be benefits to individuals if the efforts invested in communal activities are to be sustained, and room for their own experimentation so that new local technologies can be developed (Eyben & Ladbury, 1995). Even as early as 1993, farmers in Ibi and Toessin were not receiving benefits from their participation in group work. Data on the numbers of people involved in Toessin group activities (Figure 4.22) showed that the numbers participating in diguette construction, for example, were beginning to tail off towards the end of the study period. Women in particular were excluded from the direct benefits of soil and water conservation. Even if they manage to improve yields on their own land, it is doubtful they will be rewarded with security of tenure or more support or contributions from men. Therefore their interest in helping with SWC on communal household fields will wane. Following the logic of the macro-sociologists like Michael Thompson et al (1990) and Alan Fiske (1990b), "social solidarities" or "communal sharing" require constant re-reinforcement and negotiation - and a recognition of those left out of this process - if they are to overcome the countervailing tendencies of individual actors to seek different paths towards meeting their subsistence needs - through migration, trading or opting out of village institutions entirely.

In trying to understand why certain efforts at land rehabilitation or certain agricultural techniques presented in the dissertation are carried out, it is inadequate to assume that farmers are simply "adapting" to difficult circumstances by choosing a particular set of actions, as populist writers, and some forms of ecological reasoning, have led us to believe. They are applying different technical practices, and developing knowledge that draws only in part from development projects and extension agents. Supporters of farmer's knowledge now emphasize peasant logic and forward thinking, the marshalling of resources, and the overcoming of - not just the working with - natural ecological constraints and environmental problems (see Batterbury, 1996a, Netting, 1993:28). Indigenous agricultural knowledge, long sidelined by development programs, has consistently influenced (and evolved with) these programs to reflect the new demands placed upon dryland farmers. Farmers have incorporated useable techniques such as diguettes that have been developed in parallel environments, and have largely rejected inappropriate packages including chemical fertilizers (which are expensive, and poorly adapted to Sahelian soil fertility problems), "improved" cereal varieties (appropriate, perhaps, to better soils and valley-bottoms only), and expensive ox-drawn plows. This process of selection and adaptation is very much part of the transformation of Sahelian production systems and societies to meet changing opportunities and changing constraints, as suggested in Chapter 1 (Raynaut, 1997).

The "planning" involved in technical assistance to these farmers, of the sort described in this dissertation, cannot be equated with the dubious "social engineering" agendas that Richards (1993:71-2) has identified in much externally-imposed development work and agricultural research in West Africa, even if we must acknowledge that some of its activities can be misguided or partial. Farmers¼ skills and curiosity can guide these external efforts, as can an understanding of the livelihood system itself (Batterbury, 1994a). It is important to work with the best of these organizations, which are increasingly adopting hybrid organizational frameworks drawn from traditional resource management practices, the knowledge of migrants, traders, and extension services, and modern conceptions of development and community self-help (see Edmunds, 1997).

9.2 Agrarian systems and their change
Farming remains firmly established as the economic base, and satisfies the majority of community food needs even in years of poor rainfall. Part of the goal of SWC must be to aid food production. While the logic of traditional farming methods is well adjusted to ecological conditions and labor supply, the ability of these methods to provide for all household needs is increasingly called into question, and particularly in higher population areas of increased land pressures where crop yields have declined considerably over the years and large production units have "atomized" down to the household or individual level. Where production units are small, labor shortages and the effects of migration are keenly felt. In the early 1990s, neither study village managed to meet its subsistence needs from subsistence farming. Although the necessity of further micro-scale analysis on this issue is clear, it is simply impossible to say from this study what the future holds for the Central Plateau in terms of agricultural intensification trajectories. In Chapter Five, intensification was defined as an increase in the effort expended per unit area, to increase total output per unit area and over time. Soil and water conservation is a form of land use intensification because it involves investment of human and material resources in the land, and it is improving crop yields (Adams & Mortimore, 1997). It was shown to have wider benefits on moisture retention, soil quality and depth, and the resistance of the soil to further erosion, despite the difficulties encountered in measuring these variables in the field. It has arrived on the scene because it has been promoted by external actors, thus changing local technical know-how and farming practice. Almost every village in Bam Province has a patchwork of fields criss-crossed by stone lines and many now have several kilometers of diguettes as well as tree plantations and protected bush areas. The majority of these activities are in central, visible parts of the village territory. Despite these positive effects, individual farmers, and notably the poor or disadvantaged (those with insecure land tenure, some women, and those farming far from the center) may still not receive on-farm benefits from SWC carried out by village groups in central locations. The study therefore shows that farmers maintain their own risk adverse farming strategies, and adopt SWC selectively.

The study highlights the process of selective intensification on the Central Plateau. In some cases farmers are selectively intensifying all or part of their land, adopting inexpensive agronomic improvements such as manure additions where appropriate, on their existing farmland. Selective intensification also involves the recuperation of degraded land, by means of progressive enlargements to field boundaries with stone cordons, zai, diguettes, and paillage. Elsewhere in the farming system, some farmers are holding constant their labor inputs but extending the cultivated area using plows. There are cases where soil and water conservation takes place on the same fields as agronomic improvements and plowing, and examples of diversification away from farming where migration and off-farm income generation proves more lucrative or more satisfying. These techniques are applied on sandy and gravelly mid-slope soils, valley bottoms, and at varying distances from settlements depending on the knowledge, means, and land access of the individual cultivator. These decisions are not random, but reflect the contingent circumstances and outlook of individuals. These circumstances may best by understood by empirical analysis of farmer praxis, and with reference to actor rationales, constraints, and models of behavior. They sum to certain "pathways of intensification", that are visible at the terroir scale as distinct changes in land use patterns over time.

The process of selective intensification indicates that the Central Plateau is very far from the "virtuous circle" that links population increase to a widespread intensification of technical practices in agriculture, as identified for zones of higher population (Raynaut, 1997:320; Tiffen et al, 1994; Turner et al, 1993). Indeed Marchal's work in Yatenga (1983a) contests the privileged position of population growth as a driving force toward sustainable high-yield agriculture, arguing against Boserupian logics in this region of median population densities. Although based on very limited and localized information, the data from the study villages tend to confirm Marchal's findings, although I am less pessimistic than Marchal about the future prospects of the region's environment and people. There has not been a historical conjunction of forces on the Plateau that has led to rapid agricultural intensification; this intensification has been opportunistic and selective, varying with land rights, the historical and social situation, and by locality. Under conditions of labor and land shortage, some farmers choose to take on more farm land to increase their gross yields; others may choose to intensify existing farms by investing in soil and water conservation as well as other improvements. Should a development project provide free assistance for natural resource improvements, it is profitable for farmers to accept this assistance and begin to intensify their farming strategies around the new erosion control structures, thus forming a new "pathway" to intensification. Without this assistance, other avenues - especially off-farm income diversification, and the accumulation of animal herds - will be favored.

9.3 Village Organizations and Grass-Roots action
Moving to the organizational level, the case study highlighted the dynamism of village institutions in the study villages and elsewhere on the Central Plateau. I noted that a variety of institutional forms can exist at this level, and I found that the re-invented naam work-groups and solidarity networks, discussed in the first Chapter, were much less well developed in the study region than in Yatenga Province. Not all local institutions will be appropriate contact-points for external support organizations like PATECORE, and further exploration of intra-community dynamics is required. The template of actor rationales provided by Fiske (1990b) is helpful here; it explains some of the motivations for participation in different institutional arrangements, and the shifting identities that the Mossi take on in everyday life. The relative inexperience of women's village groups gives them less bargaining power with external actors, and this contrasts with the large and active groupement villageois that owe something to the legacy of socialism in Burkina Faso.

A second point to emerge was that building anything - including stone lines, new dwellings and communal buildings - can elevate community status. Projects involved in environmental improvements and basic rural development activities need to recognize that such activities are undertaken by communities for reasons of prestige and as symbols of power, as well as for their more obvious benefits to crops and material welfare (Edmunds, 1997). A well-managed village center gives an image of superiority, or wealth, to neighboring communities and to local government officials, as does a tidy woodlot or a new well built with external assistance. Perversely, the act of working together in order to present this image to outsiders can lead to internal village conflict and power struggles being ignored, overcome, or at least temporarily set aside, as at Toessin. Completed "works" can be consciously undertaken to present a particular image of diligence to potential donors and projects - "see how organized and hard working we are." The literature on indigenous knowledge systems often skims over this aspect of conscious, goals-oriented forward planning, and it is particularly important that development projects and extension agents (and researchers concerned with environment and development issues) probe deep enough to uncover its workings.

Thirdly, I have showed that Mossi and YarsČ farmers organize their labor schedules not just to "farm" but to undertake a host of activities requiring deadlines and forward thinking. These "non-passive" activities (Raynaut 1997:211) included building protection works, improving tracks leading to the village, raising and planting saplings in visible locations, welcoming visitors, and engaging in a host of activities requiring sometimes considerable and costly planning. These operations vary by season and between years.

9.4 Village Land Use Management and "community" resource management
The dissertation has concentrated on the application of a territorial approach to local resource management, gestion des terroirs, in a period of rapid expansion, uptake, and experimentation with this approach across West Africa. The coordination of land rehabilitation exercises by PATECORE has had major, positive impacts on farmers' own efforts at environmental protection. Something never thought possible ten years ago - the transformation of many landscapes and the return to production of acres of barren land - has been carried out. An examination of aerial photographs of parts of the Central Plateau shows SWC structures to be highly visible features of the contemporary rural landscape, traversing fields, bush areas, watersheds and gullies. Most lines and dams are less than fifteen years old. That such a recent innovation has brought about such important changes to Plateau environments is particularly challenging to "crisis" models of rural Africa (Watts, 1989).

While this approach rejects standard technological packages and has been enthusiastically embraced by the Mossi, it is fundamentally a "community" approach to the problem of environmental degradation. It requires participation, a degree of inclusiveness, and a level of commitment by farmers to a territorial and social unit, the terroir. Although project managers are well aware of the compromises required of the approach to function efficiently, it is still true that gestion des terroirs approaches to resource management are guilty of lumping together diverse actors under the heading of "community" resource managers, and can sometimes downplay intra-village conflicts (Mearns, Leach & Scoones, 1997). By focussing on community-level processes, they can lose sight of intra-community differentiation and the requirement of many production units to have at least one active worker exploiting income-generating opportunities elsewhere in West Africa (Raynaut, 1997: 316). Allen & Hoekstra (1990:9; 1992) remind us that ecological communities are little more than "temporary assemblages", of organisms relating to one another through "competition, adaptation, mutualism and other community mechanisms." The networks of power within human communities - while these may be concealed by villagers - need to be understood, in terms of gendered access to land and other resources, availability of labor and capital, and social and political status in the community (Rocheleau et al, 1996), because communities, unlike populations, "involve many species" (Allen & Hoekstra, 1990:11).

It is still not clear from the study whether the development project, PATECORE, meets the majority of criteria expected of a "sustainable" environmental intervention, as discussed in the first Chapter. The central question is whether the resilience and stability of the human and ecological system - defined as the "action space" centered on the village terroir of agricultural societies - has been improved by these interventions. The answer is a qualified "yes". Soil and water conservation structures are certainly responsible for improving the resilience of farmers' plots to moisture stress in periods of poor rainfall, and have some positive effect on crop yields and biomass. PATECORE's soil and water conservation improvements have been of high quality and will, with minimal repairs, last for several decades. Plantations, compost pits, and the provision of donkey carts and basic tools are investments that will yield continuous returns, and a momentum of training on the use of tube levels and contour bund construction has already been built up which will allow present "agro-formateurs" to train their successors. Only time will tell, however, if these diguette structures will continue to be built and, importantly, be maintained by farmers. Only after several years will it be possible to state with precision whether yields have been consistently sustained.

In terms of the institutional sustainability of these interventions, however, few in the region are convinced that local government departments will be able to assume the capital cost of vehicle and truck maintenance should the project finish. A decision, taken at the outset of work (in 1986-1987), to aid the delivery of stones to the village by truck rather than by donkey carts, probably reflected the high level of bilateral support available, and this has proven extremely expensive by local standards (from 1994-1997, the cost of PATECORE's transport was around 3 million US dollars, financed by a German bank). Villages, grown used to the availability of trucks at minimal cost, would have to look elsewhere for transportation or fall back on local means if this form of assistance was curtailed. The prospect of creating a cadre of trained administrators and field agents within government ranks, capable of assuming some of the responsibilities currently being carried out by project staff, is more encouraging.

I showed in Chapter 8 that insufficient attention had been devoted to the village meeting, the "interface encounter" (Long, 1989) where outside and inside meet. Village meetings are the "transmission media" used in many GTV and most conservation activities. In some cases these meetings are not the ideal method to reach the poorest individuals, as the data on the Toessin meeting participation showed. Individuals may stay indoors, remain at their fields, misunderstand, or otherwise remain isolated from important sessions. Experiments with participatory appraisal exercises (PRA) have shown that large and small groups can be brought together to debate and explore key issues, and this technique may form an alternative to large group meetings, where it is relatively easy for visiting personnel to dominate the proceedings. Participatory research techniques suffer problems when used on a repeated basis and, without careful monitoring, can still leave certain voices unheard (Mosse, 1994; Olivier de Sardan, 1995; Jackson, 1997). It is true in most cases that, as Fowler (1992:17) points out, "to realize people's empowerment, participation requires terms of engagement between the intervening entity and a community that in practice the project paradigm does not allow." Women's organizations in particular are poorly addressed in many extension activities, and support to them was not consistent in the study villages despite womens' clear role in natural resource management work and agricultural production. Despite these problems, it can be concluded from the evidence in Chapter Four that isolated soil conservation works carried out without real communal support or participation - as is the case where a rich farmer decides to treat his or her own land alone - would have been less successful than the community-wide interventions that projects have supported. It is later in the cycle of intervention, as enthusiasm for project ventures begins to diminish and farmers turn to work on their individual bush plots, that individualized support to farmers for their conservation activities will be required.

With these caveats in mind, perhaps the most exciting aspects of the GTV approach, as referred to in Section 9.1, are its promotion of local territorial control; the positive effects on intra-community solidarity and communal sharing; and the use of GTV activities as a vehicle to expand the range of choices enjoyed by rural people on their lands. As Allen & Hoekstra (1995:9) and Chambers (1997) remind us, sustainable resource stewardship is meaningless without social justice. The range of social benefits came across strongly from the farmer interviews, and in two seasons of close contact with male and female farmers as they slowly developed links with project staff and the government extension services. Particularly notable were the activities of Toessin farmers, who have designed their own land rehabilitation measures from scratch, discussed these in village meetings with a high percentage of men and women from different lineages present, and continue to adjudicate between multiple criteria (material, and "symbolic") when deciding how to best get the work done. Consideration is given to land tenure in areas where land rehabilitation is programmed, the complaints of non-beneficiaries are discussed, local environmental considerations such as existing ravines and gullies are examined, and responsibilities for the work are divided up. In Ibi, however, facing more severe food supply problems, less income opportunities, and a much abbreviated period of contact with development agencies, the work of the project had less evidently "taken off" under local control.

To conclude my observations, I have tried here to present village land use management (GTV) as a "second-best" form of rural community development. This is because it is initially managed "from above", yet it retains a populist framework and a strong local input (Engberg-Pedersen, 1995; Schorlemer, 1996). The combined efforts of farmers, the project, and the local government services, are striving towards reducing the vulnerability of local systems to drought and food shortage, and there have been significant successes, as well as failures, in this regard. Firstly, I have shown that "land-poor" villages would need major changes to tenure laws and land access to plan effectively at the terroir level, so there will always be situations where the terroir remains an unsuitable scale of action. Secondly, high out-migration can evacuate communities beyond the point where land-based investment in environmental improvements remains the primary concern of the community, so farmers accept these interventions "on the road" to demanding improvements in social welfare or assistance with broadening their income sources. These two points are actually quite controversial, given the widespread and rather uncritical support offered to GTV by the World Bank and other donors. We should be wary of generalizing from these few project experiences or even a few communities; dryland West Africa is too complex for simplistic forms of community-led development.

9.5 The implications for geographical practice
Pioneering work by geographers working in Africa was critical in exposing the links between underdevelopment, poverty and land degradation (eg Knight, 1974; Porter, 1979; Marchal, 1977; Wisner, 1977), and has more recently been incorporated in a regional political ecology perspective that is sensitive to the wider political and economic environments in which land-users operate, take decisions, and carve out livelihood systems (Blaikie & Brookfield, 1987; Blaikie, 1991; Bassett, 1988; Bryant & Bailey, 1997). The central planks of regional political ecology were highly influential in this study, although ultimately the case studies were primarily focussed on local processes - not the whole range of process and scales demanded by a "nested" political ecology analysis. I have retained what I considered to be the most useful arguments from cultural and political ecology: adaptive frames, nested scales of analysis, and a critical perspective on explanatory variables and relationships - rather than attempting to overturn previous orthodoxies and to develop new theoretical propositions. This dissertation has - in a way that is modest in its aims - concentrated on local systems, and taken a hybrid approach to arrive at adequate descriptions of agrarian space, using a range of research techniques to explore the natural and social variables that structure landscapes and development interventions. There is growing interest in such hybrid studies that try to combine scientific, critical, and local knowledge sources, and quite often follow an inductive reasoning process (Batterbury, Forsyth & Thomson, 1997; Robbins, 1995; Rocheleau et al, 1996).

Environmental issues in developing countries are perceived differently by diverse actors and it is often a failure to communicate between actors that leads to a "messy" and conflictual policy environment, and poorly thought out transfers of technology or knowledge (Blaikie, 1995a; Batterbury, 1996a). Under such conditions, projects are likely to fail, or miss a "moving target" - the land users. In the West African Sahel, we need to study the important social and technical changes that development projects are setting in motion, as well as beginning to understand how these projects "think" about the people and environments with which they are working. Geographers are well placed to assist in this through the range of critical skills they gain in their training, especially where this involves study of "nature-society" or "people-environment" interactions, and long-term fieldwork (Bebbington & Carney, 1990; Bennett, 1976; Binns, 1995; Turner, 1989). It should not compromise our social consciences to work with those in positions of influence and power to improve their interventions and to help ensure positive benefits to rural people and their livelihood systems, and to help translate the wishes of rural dwellers into "project language" (Chambers, 1992; Ndione et al, 1995; Rocheleau, 1994; Scoones & Thompson, 1994).

Some of the study presented the results of my effort to apply the techniques of cultural ecology in field conditions. My initial conviction that ecological variables would structure many aspects of farming systems and cultural practices was only partially borne out in the case study. While the imprint of soil and water conditions are seen clearly in many aspects of Mossi systems, the social structure itself is a complex amalgam of hierarchical and egalitarian relationships that appears to have developed autonomously from the natural milieu. These systems are unlike the stratified or segmented lineage systems of neighboring groups, as cultural materialist arguments would predict (Harris, 1979). Rather, the "splendid diversity" of technical practices (Netting, 1993), particularly in the agricultural domain, emerged from this limited examination of two villages. The reverse impacts of people upon the agro-ecological system were evident, and these were mapped through land use change analysis and consultations with farmers about the extent and duration of erosion processes and the changes wrought by the extensification of agriculture under population growth and technical change. It would be presumptuous to base a critique of cultural ecology on the limited data obtained. Indeed Rappoport claims all holistic academic efforts at understanding, in which we may include cultural ecology, are no more than institutionalized in relatively obscure academic disciplines in which they are isolated from the conduct of everyday affairs. Whether or not their findings and perspectives are "accurate, "true", "correct", or "adaptive" is almost beside the point because they are powerless. (Rappoport, 1984:329)

I have some sympathy with this damning claim. I understand the worry that ecological ideas cannot provide an adequate account of nature, or adequately guide the work of geographers (Demeritt, 1994). But recent advances in ecology, and environmental history, are considerate of scale-dependent processes, resilience, and complex human impacts, and the links are made to concerns about social justice and the practicalities of environmental action (Simmons, 1997). Rather than move human-environment work onward to a position where our primary task as geographers is to "...situate competing knowledges about the world", and "...make visible the material and discursive effects of different environmental narratives" (Demeritt, 1994) I feel it is more productive to first obtain practical findings, and policy implications, beginning with a cultural-ecological analysis. The real challenge is to ensure that the theoretical apparatus we use is well balanced between the natural and social sciences. In the dissertation, the much-derided concept of human adaptation to environment, as employed in many cultural ecology monographs, was useful as a heuristic concept for explanation. But it was the diversity of adaptive response by Mossi and YarsČ farmers at multiple and overlapping scales, that proved most striking, and of most relevance for informing the community-based land use management process.

Similarly, deficiencies in the political ecology framework have emerged in my own efforts to apply a loose version of the "regional political ecology" model. Theorizing in political ecology is fraught with disagreement, as students of the Blaikie and Brookfield model have now begun to both refine and attack it, and the methodological ambiguities of the approach have continued to generate confusion and difficulties in forming comparative explanations. Recent work in this field has tended to overlook the nuanced ecological relationships of "continuously scaled fluxes" that ecologists are now directing us toward (Allen & Hoekstra, 1992), swinging too far towards a view that social actors determine most ecological outcomes, including land degradation (see discussion in Batterbury, Forsyth & Thomson, 1997, and in Zimmerer, 1996). There are good reasons why this has happened; the ignorance of gender, resource struggles, and social movements in earlier work has required comment and further conceptual development (Rocheleau et al, 1996). My primarily interest in this study has been in developing in an applied political ecology - a political ecology of environmental management - that fully recognizes the power of biophysical processes in shaping human responses in drylands. An applied political ecology will not only identify the powers of political actors linked to environmental issues (Bryant & Bailey, 1997) or the social constructions of nature by these groups and the power of social movements (Demeritt, 1994; Peet & Watts, 1996b). While it is true that new developments in political ecology open up a flexible analytical terrain suitable to the explanation of problems like land degradation and resource access in Africa, it is through critical support to environmental management - or arid lands management - that practical goals can be achieved through well-targeted environmental research with, and within, institutions. I have long believed that the identification of power dynamics, unequal access to resources, and globalization processes is a vital component of political ecology, but I also feel that explanation is a necessary but not a sufficient component of this exciting field (Batterbury, 1997c). The challenge is to turn complex explanations towards solving complex problems, whether this be through advocacy, research, or practical advice (Bryant & Bailey, 1997: 196; Price & Thompson, 1997). When beginning this work, I felt there was a need for the geographical community to become more closely involved in the network of organizations that are conserving natural resources and tackling persistent rural poverty in West Africa (Batterbury, 1997a). This does not mean taking sides with one set of actors against another. Farmers, grass-roots movements, NGOs, the state, and donors all play a role in social and ecological change, and none of their actions should be accepted without critical evaluation (Olivier de Sardan, 1995).

I have tried to show that institutions involved in land rehabilitation of the marginal environments of Burkina Faso are confronting issues central to political ecology models and human-environment studies across more than one "western" paradigm or disciplinary frontier. Much of the basic groundwork on village land use management is conducted by local geographers (as in PATECORE and the Burkina PNGT, for example); skills such as resource mapping, survey work of all sorts, air photo interpretation and the identification of social and ecological change are in everyday use. The whole notion of gestion des terroirs owes much to the French school of tropical agrarian systems research inaugurated by the geographers Sautter and Pelissier (Pelissier & Sautter, 1970; Imbs, 1982; Marchal, 1983a; Sautter, 1962) - a school of research almost completely ignored in anglophone writing on these same regions, despite its strong conceptual development and quite advanced methodologies. This work provided the tools to understand the village terroir in French West Africa. By helping to apply it, geographers - and their work - becomes implicated in current struggles to reclaim not only degraded land but also rural cultural identity; particularly where the latter is strongly linked to the quality of the local resource base.

9.6 Conclusion
The Central Plateau of Burkina Faso, one of the poorest rainfed farming environments in West Africa, now hosts a multitude of non-governmental, bilateral and other initiatives moving forward slowly and with the relatively modest aim of assisting farmers to achieve marginal increments in food security through locally managed environmental management initiatives (notably soil and water conservation). This is a strategy for rural development, and agrarian change, that is more fully cognizant of ecological and social diversity than past efforts, and frequently involves local land users and their organizations in setting research and program agendas, at least in part. It is time that these fruits of conscious planning and experimentation were recognized (Reij, 1994a:13). The outsider's role in this complex process may ultimately be to step back and provide a level of support necessary to allow this process of planning and inventiveness to take place, not to introduce exogenous blueprints (Richards, 1985; Chambers, 1983).

What building materials, and what tools, and what outside help will be needed to build and maintain the new agro-ecologies that the Mossi, and farmers in other dryland situations, are slowly developing? This analysis, which has focussed on many interlocking aspects of agrarian change and natural resource management, has shown that there is a need for modest external support to village organizations, in the form of both technical assistance and practical help and advice (Dumont, 1986). This particular form of low-level, flexible support is even more urgent in marginal farming environments like the Central Plateau, despite the adaptive capabilities of its farmers. Intensive and permanent cropping systems are still rare, out-migration is helping to transform production relations, and land pressures are intense in some areas. A diversity of users occupy the same geographical space and, for any interventions to be sustainable, the plurality of local relations between nature and society must be understood and respected (Raynaut, 1997; 315). The challenge is to provide a form of support that deals with land pressures, resource degradation, and social conflict. As currently formulated, the gestion des terroirs approach is best applied to communities with a strong degree of social cohesion, and relatively straightforward agrarian systems and land rights (as at Ibi). Where rights to land are contested, overlapping, and shared between multiple users (as at Toessin) successful interventions have only resulted because the local population has played along with gestion des terroirs actors for their own purposes, and have used them to assist in their struggles for better community resources and secure land rights. In both situations, there have been winners, and there have been losers. The challenge is to allow all local people to identify and create space for their own interests (Long & Villareal, 1994), and to leave decision making where it really belongs - with individuals, their social networks, and their community institutions.

Return to Simon Batterbury's homepage

1021 hits from July 1998- June 2000