AGRICULTURE AND HUMAN VALUES - WINTER-SPRING 1991

Gender, Ecology, and the Science of Survival: Stories and Lessons from Kenya
Dianne E. Rocheleau

Dianne E. Rocheleau is an Assistant Professor in the Graduate School of Geography at Clark University in Worcester, MA. She holds a Ph.D. in Geography with a minor in System~ Ecology from the University of Florida. She teaches courses on social forestry, tropical ecology, political ecology, gender and development. Her research focuses on social and ecological dimensions of forestry and rural landscape change in East Africa and Central America. She conducted research on land use and watershed management in the Dominican Repubfic from 1979 to 1981, worked as a senior scientist at the International Council for Research in Agroforestry (ICRAF) in Nairobi from 1983 to 1986, and was a Forestry and Agricultural Program Officer for the Ford Foundation in Eastern and Southern Africa from 1986 to 1989. Dr. Rocheleau is senior author of Agroforestry in Dryland Africa and has authored several articles and book chapters on women, trees, tenure and land use. She serves on the advisory boards of the Land Tenure Center, the Wildlife and Human Needs Program of the Worm Wildlife Fund, and Society and Natural Resources Journal. She is also a member of the National Academy of Sciences (NRC/BOSTID) review panel on International Forestry Research.

ABSTRACT Sustainable development and biodiversity initiatives increasingly include ethnoscience, yet the gendered nature of rural people's knowledge goes largely unrecognized. The paper notes the current resurgence of ethnoscience research and states the case for including gendered knowledge and skills, supported by a brief review of relevant cultural ecology and ecofeminist field studies. The author argues the case from the point of view of better, more complete science as well asfrom the ethical imperative to serve women's interests as the "daily managers of the Hying environment". In the interests of both objectives the paper advocates an ethnoscience research approach based on empowerment of rural people, rather than simple extraction of their knowledge. The Kenyan case study of women's agroforestry work follows their response to the drought and famine of 1985 and chronicles the unfolding discovery of women's ecological, political, and social science as gendered survival skills. The case is re-counted as a story, in keeping with an explicit choice to learn through participation and to report through storytelling. The experience of rural women and researchers during the drought provides several lessons for both groups about their respective knowledge systems, their agroforestry work, and the relationship of both to local and national political economy.
Introduction Sustainability and participation may become the two most widely invoked and diluted words that issue from the mouths, pens, and other "word processors" of development planners and scientists during this decade. This new wave of enthusiasm for "sustainable" development has brought a concurrent upwelling of interest and activity in biodiversity and indigenous knowledge. Yet, researchers and practitioners have paid little attention to

gender and sustainability. This paper examines the place of rural women's ecological science in this purported sea change within the development process. It also reverses the question and considers the possible consequences of this new trend for rural women's futures and those of their communities (Redclifl, 1987). It questions the suppression, neglect, or even the extractive collection of women's ethno-scicntific information, and explores the comprehension and empowerment of their own science

156

Rocheleau: Stories and Lessons From Kenya of survival. 1983; Moore and Vaughan, 1987; Meyerhoff, 1982; Rocheleau, 1985, 1987, 1988, 1989; Merchant, 1980, 1989; Fieuret, 1979; Shostak, 1981; Murphy and Murphy, 1974; Fernandez, 1988; King, 1989; Shiva, 1988). Many of the same researchers have documented the changing fights, responsibilities, and tasks of poor rural women, and the imbalance of women's rights and responsibilities as resource managers. Yet very little of the literature on sustainability has treated gender as an issue. The implication is that half or more of indigenous ecological science has been obscured by the prevailing "invisibility" of women, their work, their interests and especially their knowledge to the international scientific community. In her aptly tiffedrecent book,StayingAlive: Women, Ecology and Development, Vandana Shiva (1988) posits survival as the ultimate criterion for verification of poor rural women's knowledge. Several other recent works on land use change, technology, power, and indigenous knowledge (Carney, 1988; Jiggins, 1986a, b; Richards, 1985 and 1986; Watts, 1983 and 1988; Mcxtimer, 1989; Stamp,1989; Talle, 1988; Meyerhoff, 1982) have also por~yed rural people's combined social, economic, and ecological strategies as a science of survival, whether in the ebb and flow of seasonal adversity and opportunity, or in the periodic crises (social, economic, political, ecological) that affect their fives and livelihoods. The understanding and protection of the environment as essential to rural livelihoods, work, and social welfare is a hallmark of poor peoples' and women's environmental movements in India (Shiva, 1988; Jain, 1984; Joshi,1982). Popular movements share similar practical ecological concerns in the Amazon Basin (Hecht and Cockburn, 1988; Mendes, 1989; Schmink and Wood, 1990; Posey, 1985), the savannas of East Africa (Matthai, 1985; Nyere~, 1989; Parkipuny, 1988) and in many other "fragile" or "marginal" rural environments throughout the world. From this perspective rural subsistence and health become environmental problems and the solutions involve questions of equity as well as economic and political redistribution (Guha, 1989). In contrast, the international scientific and development communities have tended to ignore rural people's science or have separated it from the larger context of daily life, labor, and livelihoods. Social scientists and ecologists alike often recast "indigenous knowledge" as an ethnographic artifact, as "unconscious ecological wisdom', or as part of the "environment" for the generation and introduction of new technology. Even these very partial and objectified views of rural peoples" science have not often been studied and understood as gendered knowledge and practice. We face the double task of reshaping the terms of discourse about popular, local ecological science (Thrupp, 1989; Bebbington, 1990) and introducing women's science and women's interests into the larger domain.

"Sustainable" Development as Opportunity and Challenge The current concern for environment, development, and culture reflects a historical re-discovery of the connections between economies and ecosystems, between people and their environments, and between present decisions and futures possible. The "international scientific community" is poised uneasily at the turning point between the expanding world of Darwin's voyage on the Beagle and the shrinking biological community of Norman Myers' imperiled "ark" (Myers, 1979). While we arc still making entries in the catalog of life forms on earth, we are increasingly aware that wholepages, some yet unrecorded, are being erased even as we write. Rapid social, economic, and ecological change in many regions also threatens to undermine the foundations and unravel the fabric of indigenous science and practice under a wide range of ecological and economic conditions. Although less well-publicized than the extinction of species, this is no less a loss to our future lives and science. Among the dead and disappeared sciences already buried by careless development we have lost much insight that could inform our on-going search for humane and ecologically feasible futures (Brokensha et al., 1980). Environmental and social scientists alike now address "development" as residents of a shrinking planet faced with the erosion of genetic and cultural diversity. Ironically, the urgent concern over global environmental change and genetic erosion has spawned a serious attempt to deal with the complexity of local cultural and ecological realities in the farm and forest communities of Africa, Asia, and Latin America. Biological field scientists have for centuries depended on the skill and knowledge of indigenous "guides", collectors, hunters, and informants. Now, there is a growing awareness that indigenous science and practice may contribute far more to the cmrent understanding and to the future survival and use of a wide array of ecosystems. Likewise, some foresters and agronomists have discovered the value of local knowledge of species, environments, and production in the search for more productive, sustainable forests and agroecosystems (Warren, 1988). Conversely, many social scientists and representatives of farmers, herders, and forest dwellers have noted the importance of ecosystem integrity and biodiversity as the biological basis of lXX~ peoples" current livelihoods and future options.
Ethnoscienee, Gender, and Invisibility A seldom heard but growing community of concern has also noted the distinct roles and interests of women as the "daily manageas of the living environment" (Dankelman and Davidson, 1988). A steady stream of field research and informal reports as well as historical research indicates the gendered nature of ecological science and practice in most cultures (Chavangi, 1984; Hoskins,

157

AGRICULTURE AND HUMAN VALUES - WINTER-SPRING 1991 The Case for Gendered Etlmoscience in the Extractive Mode There is little doubt as to the utility of cataloging some discrete "bits" of rural women's ecological knowledge. An encyclopedic compendium of wild and cultivated plants, their uses, their habitats, and their environmental requirements in a given place can improve our collective ability to survive and flourish and our contributions to "sustainable" and equitable development elsewhere. Even extractive documentation and conservation ofsuchknowledge is better than the complete loss of that information. Yet the case for comprehending the whole system remains strong. There is clear instrumental value in contextual understanding and documentation of whole knowledge systems. This extends even to cases of "rescuing" residual knowledge from "dying" traditions. Whether the subject is alive or dead, context "makes sense" of individual "facts", a point equally well appreciated by archaeologists, medical practitioners, and paleoecologists as well as many post-modern social theorists. Even surviving remnants or memories of the larger scientific context will inform our understanding, as will the very story of its erosion and disintegration (Merchant, 1980). The rationale and conditions of indigenous scientific collapse and/or revolution would seem to warrant as careful an examination as the fact that a particular plant can survive drought or fertilize the soil or cure a common illness. In particular, the changing nature and status of women's science in a given place, under conditions of social and ecological change, may have much to tell us about the essential core and logic of their science, their own purpose and skill in using it, and the possible futures of similar knowledge systems elsewhere. However, whether we study the whole system or "just the facts" there is the danger that women's ecological science will be packaged as a product to be collected, owned, and sold in the "marketplace of ideas" of the scientific community. Beyond a Gendered Normal Professionalism The simple application of what Chambers (1986) has called "normal professionalism" ! to the study of women' s "indigenous knowledge" will most likely result in the appropriation and abridgement rather than the empowerment of women's social and ecological science. Normal professionalism in agricultural r e s e ~ h and extension has given us a one-way flow of "information" pushed from top to bottom, and gendered normal professionalism has extended that process to rural women. The same ethic applied to ethnoscience is likely to generate a oneway flow of "information" extracted from the bottom and pulled to rite top, with the possibility of adding women's information to the final product. Whether the process occurs within the context of social or biological inquiry, there is much to be gained from a more egalitarian treatment of rural women's science and popular science in general. I would pose the problem not as a simple need to prevent and control damage, but as an opportunity that requires careful and timely research and action. In most cases we are dealing with a fragile reality that could be, but need not be extinguished in our enthusiasm to "discover" and scientize ecological facts from the richness of women's lived science. Like all science, rural women's ecological knowledge derives and constantly recreates its meaning from a larger social and historical milieu. We can choose to understand and to encounter it whole, within that context, and to urge that others do the same. We can go further and shed the informant/'mterpreter dichotomy in order to engage in a more active exchange at the interface of knowledge systems, a move already advocated and demonstrated by both social and biological scientists operating beyond the confmes of their normal professional paradigms (Roe and Fortmann, 1981; Santkhan, 1985; Iiggins, 1986a~ Iuma, 1988; Wamalwa, 1989; Moore, 1986,1988; Toledo, 1989; Thrupp, 1989; Bebbington, 1990). While classical ethnographies often try to tell it like it was, albeit holistically, and biological science tends to tell it like it measures up, piece by piece, to external verification, perhaps we should be telling it like it is becoming of even telling its potential. The opportunity exists to share in the Cross-pollination of a multiplicity of sciences. The story that follows chronicles one such oplx~unity, the insights gained and the challenges posed. A tale of two agroforestry projects: "theirs" and "our~' In applying ethnographic and in general humanistic research methods much of what one comes away with is best distilled as a story. This applies equally well to our narration of the research experience and to the recording of peoples" answers to our questions and our listening presence. While records of this kind age often dismissed as "anecdotal" by "hard" scientists, these records may well distill the shared meaning of several distinct experiences that constitute a single class of event. Perhaps the critical difference between "mere anecdote" and significant story lies in the analytic content of the message and its synthesis of common content or process applicable to other circumstances. Both instructive stories and replicable experiments carry valid information for other times and places. 1 do not propose to offer here a solution nor an example to be replicated but rather to tell a story of a lesrningprocessat the interface of two worlds. I chronicle our own experience for two reasons. First, it serves as an analog for the more protracted and disparate experience in the larger research and development community. Second, it creates a shared point of departure toward future work at the soft edges of many sciences defined by gender, class,ethnicity,and locality. 2

158

Rocheleau: Stories and Lessons From Kenya Women, agroforestry, and the gender division of labor In 1981 the International Council for Research in Agroforestry and Wageningen University began an exploratory On-farm agroforesU'y research project in a cluster of five villages in the semi-arid farm and rangelands of Machakos District Kenya. By 1983 a team of social and ecological scientists and students sought to broaden the scope of the ~ h design from agroforestry technology trials on ten farms to include the larger landscape and the community at large. They monitored the original trials of "alley cropping" (hedgerow intercropping of trees to produce mulch, nitrogen, and fuelwood) in plots of maize, beans, cowpeas, and pigeon peas, and soon confronted problems rooted in the gender division of rights, responsibilities and knowledge at household and community level (Rocheleau, 1985, 1988). In the ten on-farm trials, nine with men heads of household and one with a couple, the project team encountered two key problems that derived from the absence of women in design and management of new technologies. F'wst, some men failed in their efforts to grow their own seedlings in home nurseries due to lack of sufficient water. Although women were responsible for providing water, they had not been consulted about the decision to grow seedlings at the household level and were unwilHng to carry extra water from distant sources over a period of four months. Secondly, the proposed agroforestry practice that men were testing on farm (alley-cropping) responded primarily to women's problems with fuelwood supply and soil fertility for staple food Ixoduction. The women, who had not been directly involved in planning on most of the trials, were unimpressed with the amount and quality of fuewood produced by the seasonal pruning of the young tree rows of Leucaena leucocephalla and Cassia siamea. These compared poorly with pigeon peas (Cajunus cajan), a woody perennial that produced food and fodder as well as fuelwood and was already well incorporated into the existing system. Men's interests in timber and poles for construction or for sale were not addressed at all, which led them to trim the intended"green mulch" and fuelwood trees to produce poles, or to allow their goats to browse the trees, both at the expense of the researchers" intended products. Thus neither men nor women obtained what they could have from a resem~h design involving both as full participants (Rocheleau, 1985). This experience illustrated that women's labor and interests matter for technology design and testing, and that men's and women's domains are distinct at household level. "Public works" and the proper work of women's groups At the same time, researchers "discovered" the importance of women's self-help groups and their role in community level soil and water management work. In response to the limitations of the previous farm trials and the presence of the women's groups they initiated a small participant observation project to introduce agroforestry skills, species, and planting arrangements into the community-based soil conservation work of the women's groups. Researchers and Irainees joined the groups in their weekly soil conservation work sessions, mostly devoted to plugging of active gullies, and began to raise the possibility of using trees and grasses to stabilize the sites and to provide fuelwood and fodder cuttings (Rocheleau and Hoek, 1984). The groups agreed in principle but seemed uninterested. Researchers tried on numerous occasions to elicit women's groups" preferences for particular species or for particular tree products and services, without suggesting specific options, but with no success. As foresters had warned, people seemed to have little interest in trees and no strong preference for particular species or tree products. Finally, as the field team repeated the question for the fourth time in four weeks, some group members asked them to list exactly what they could offer. Once they heard that the range of options included fi~it and fodder trees, and were not limited to men's timber trees or foreign forester's woodfuel trees (as expected), they x~ked why anyone would want to waste good trees on such an obviously degraded site as the gully at which they were working. The participants in the soil conservation work requested trees to plant at home, and considered themselves to be earning the trees through the group work (Rochelean, 1985). Moreover, the groups gradually revealed themselves as associations of individuals and households, formed as reciprocal work groups and mutual aid networks rather than public works organizations. While five groups were at one point "mobilized" by local officials to work together at a site chosen by the project, the participating groups prevailed upon the field team to work with them separately in the next season's activities. Based on prior experience the women noted that the mulli-group situations stretched social networks too thin and over too large an area, while the individual groups had developed their own means of mutual accommo~tion and accountability among members and shared an interest in the same geographical space. Through a series of subsequent discussions participants explained that their groups each counted from 50 to 100 official members and n~mllydrew a work fot'ce of 20 to 30 women and sometimes a few men, for any given task. Their "real" work was reciprocal weeding and terrace repair labor on each other's cropland, which was then limited to one day per week, with two days devoted to road and gully repair. They pointed out that in most cases the road and gully work was an onerous duty performed to comply with public service demands from public officials. Finally, the project's entry into this work coincided with the onset of the drought and subsequent famine of 159

AGRICULTURE AND HUMAN VALUES - WINTER-SPRING 1991 1984, which began early in the drier parts of Machakos. The self-help groups had acquired greater importance due to the drought, as access to public relief food and/or "food for work" projects was expected to depend on individual wealth and influence or on group membership and participation. The flurry of group soil conservation work at sites chosen by local officials had more to do with drought and famine "insurance" than with specificcommunity priorities for soil and water conservation. The time, place, and public image of group work is not necessarily indicative of the group's larger purpose and women's interests. Re-shaping gendered knowledge and practice As the drought set in the project distributed the first set of 15 "sample" trees to members from five groups. They planted the seedlings at their respective homes, which for most women was a first. While a few influential men had planted numerous citrus and timber trees in the past, and many men had purchased or collected an occasional tree for plot boundaries or home compounds, this was a new activity for women. The project team monitored the placement and performance of the trees and found the women to be undaunted by the poor performance of the mostly exotic trees in the face of drought, hungry cattle, and termites. When they met to plan the next season's work the five participating groups requested that researchers cease and desist on the soft conservation front and instead devote their efforts to group tree nurseries, with provision of additional seedlings from outside to supplement their first efforts. In effect their primary concern was to get the process of tree domestication into their own hands and their own heads within the context of self-help groups. The project team monitored people's choice of species, their choice of nursery sites, their subsequent construction and management of the mwseries,and group members' planting and management of trees. Through their participation people taught the research team much about their ability and motivation not only to adopt innovations but to innovate independently and in groups. They developed alternatives both to their existing systems and to pre- packaged "scientific"agroforestry technologies. The same process of group tree propagation and individual tree planting also demonstrated the clear gender division of control over land and water, as well as the women's substantial skill in negotiating with men for rights of access and use. Each women's group had to prevail upon a wealthy head of household, usually a group leader's husband or relative and in one case the group leader herself, for secure long term access to a site with a reliable water somee. All but one of the five groups succeeded in securing a suitable tree nursery site. Likewise individual women negotiated lime and space at household level for the planting of their own tree seedlings. 160 After two full seasons of planting trees some group leaders raised questions about the newly introduced exotic species, especially their performance and management over time and in larger scale plantings. In spite of misgivings about the advisability of alley cropping, the researchers arranged a visit by group leaders and interested men and women to the most successful of the original 10 farm trials, to serve as a focus of discussion. Not only did the group members critique the technology as a package, they also chose specific elements (species, planting arrangement, tree managemenO that appealed to them. Men and women had somewhat different priorities: timber for men and soil fertility and crop production for women. However, elder women heads of household shared the men's interest in timber for home use and for sale, since they needed cash and building material for their households and would have control over both. Women without resident husbands also expressed strong interest in fodder trees, since they managed livestock. Participating group members questioned the field team at length about the rationale and principles of the alley cropping design. Later some women returned to the project extensionist with examples of local practice and individual innovations that constituted more appropriate application of the same.principles to serve the purpose of soil fertility improvement with plant biomass. They reported adding leaves and twigs to cattle pens to "bulk up" the manure supply to fertilize cropland. They collected most of the biomass from fencerow trees (includ-

ing Euphorbia terucaili) and dispersed trees in grazing land (Combretum and Terminalia species). Subsequent
discussions with groups revealed widespread experimentation and experience with these recently developed cattle-pen-composting techniques. Many women used leaves of both exotic and indigenous frees common in the area, all of which were resistant to termites and drought. Through their response to the project, women and men in this farm community revealed that they have ongoing experiments of their own, including women's exploration of formerly male domains of management. They both have much to gain as well as to contribute by participating in formal research with outsiders and they are more likely to do so effectively in activities on individual farms, with group participation. Farming and land use as if famine mattered As the drought wore on, the project team had expected a decreasing intexest in tw.es, since so many women faced the overwhelming reality of hunger, water shortages, and fodder shortage, threatening the well-being of their families and the the survival of their draught animals and even small stock. Instead the famine and fodder shortage spurred a resurgence of interest in a wide variety of indigenous trees as reserve fodder sources, in tree crops rather than livestock as assets, and in a diversity of wild fruits and vegetables that provide nutritious snacks, combat the effects of malnulrition, or serve as substitutes for

Rocheleau: Stories and Lessons From Kenya other foods (Rocheleau et a/.,1985; Wachira, 1987). Whole extended families took to the woodlands, bushlunds, and in-between places (fencerows, roadsides, and streambanks) in search of possible fodder sources. Some people tested leaf samples of several tree species on their cattle. The elders in the community made a concerted effort to recall which tree leaves had served in the past as drought reserve fodder. However, unlike the last drought and fodder shortage of similar magnitude in 1946, there was less grazing land, less flexibility of livestock movement within the region, and in over 60% of the households women, not men, were responsible for livestock management (as supervisors or directly, as herders). Women relied heavily on their prior knowledge of wild foods and acquired new knowledge about fodder plants through hearsay and widespread experimentation with trees and shrubs in range and woodlands. Given the role of wild foods and indigenous plants in general as poor people's drought and famine reserves, researchers approached women's groups with the possibility of protecting and managing some of these plants in-situ or domesticating them on-farm within agroforestry systems. While group members were at first incredulous of outsiders" interest in "primitive practices" and "poor people's food", they gradually rallied around the idea. Moreover, they insisted on including medicinal plants, an unexpected turn for two reasons. While the project team had associated traditional medical practice with men, there was a well developed practice by specialized women herbalists (mid-wives and general practitioners) as well as widespread knowledge and practice of basic herbal medicine among women over thirty years of age ("unschooled" and taught by mothers and elder women). The researchers encountered a widely shared concern over the local disappearance or scarcity of particular medicinal herbs as well as specific indigenous fruits and vegetables. The subsequent ethnobotanical survey of men and women in the five villages started with the "general public", proceeded to the specialists and went back again to the women's groups and their children (Rocheleau et al., 1985 and 1989; Wanjohi, 1987; Munyao, 1987; Wachira, 1987). Together they identified 118 indigenous or naturalized wild plant species used for medicine and 45 for food. Of these, participants selected five fruit trees, three vegetables, and three medicinal plants for potential domestication in agroforestry systems or small gardens. They also named several fruits, vegetables, and medicinal plants as candidates for special protection in place, although women were quite cynical about their ability to enforce management rules in public and shared lands. While men's and women's priorities varied they knew many of the same places, classes of ecosystems, and plant associations. They tended to know and use different species, or in some cases, different products from the same species. Whereas men's widely shared traditional knowledge of indigenous plants had been most developed in rangeland food and fodder, their outmigration, sedentarization, and formal schooling had all militated against the transmission of this gendered science and practice to the young. Some men knew a great deal about specific classes of wild plants for specialized purposes (charcoal, brick-making fuel, carvmg, local timber, bee fodder, and medicine), but the knowledge was unevenly dispersed and decreased markedly among the younger men in the community. Among women there persisted a widely shared, high level of general knowledge about wild food, craft, and medicinal plants, with an overall reduction in the scope and depth of proficiency among younger women. However, the knowledge gap between generations of women was not nearly as pronoanced as that for men. Some members of the community attributed the persistent decline in indigenous knowledge to formal schooling and a rejection of"primitive" tradition by the young. Perhaps mote important, men's outmigration had simultaneously removed adult men as tutors and created a labor shortage and double workload for women, leaving little time for traditional education in multi-generational groups of either sex. Moreover, women had different rights and responsibilities than in the past and had to acquire and maintain an ever broader range of new knowledge and skill. The differential erosion of local ecological science among men and women may also reflect their respective rights, responsibilities, and 013portunities in farm versus wage labor sectors. While women maintain livelihoods and retain rights of access to land through residence and agricultural production, young men aspire to leave home and to succeed as wage laborers in nearby cities and towns, without fear of sacrificing their long term access to land. The feminizatiou of famine and drought response and the requisite science of survival reflects the new spatial division of labor between men and women into rural and urban domains. These changes demonstrate that the boundaries of gendered knowledge are neither f'Lxed nor independent. Content and distribution of gendered knowledge influences and is influenced by the gender division of rights and responsibilities in national, regional and local context. One man's field becomes a women's group commons The importance of wild plants during the drought was obvious. The ethnohotany survey also confirmed that most women normally drew upon fodder, fuelwood, and sometimes wild food sources beyond the boundaries of household land, as did their children. However, those most reliant on resources outside their own land stated that their children were unlikely to enjoy the same facility of access to shared lands in the future. 3 They noted that community level land tenure, land use, and vegetation changes proceeded on their own momentum, outside the conlzol of individ~ml~ and small groups. 161

AGRICULTURE AND HUMAN VALUES - WINTER-SPRING 1991 Since small-holders and legally landless women relied on the shared use of private lands to make use of wild plants (Rocbeleau,1989; Rocbeleau and Fomnann, 1988), they had every reason to focus on social strategies to secure and maintain access to wild plants or on alternative ecological strategies for meeting contingencies. Their future access to these resources on shared lands would depend on careful cultivation of social and political networks, as would their influence on soil, plant, and water management decisions taken by largeholders and male owners of family plots. Poor women's experience during the drought exemplified the careful interweaving of social and ecological knowledge to survive in the cross-currents of erratic environmental conditions with uncertain terms of resource use, access, and control. As the drought persisted peoples" terms of access to "off-farm lands" acquired increasing importance and their domains of use and access became clearer as individuals and groups relied more and more on shared resources as reserves. The map of actual land use and source areas that emerged from their activities bore little resemblance to the formal survey maps that denoted ownership, yet was in large part circumscribed by these legal boundaries established in 1972. Off farm lands as used here denotes any land outside the household property of the user;, in this case it included roadsides (public land), sUeam banks and riversides (a combination of public and private property), hillslope woodlands (mostly private land), the dry forest across the river (national government land) and most importantly, the grazing lands, woodlands, fencerows and gullies of other farmers (private holdings, small, medium, and large). The latter category became increasingly significant as grazing and browsing animals and individual collectorsdepleted the reserves in less protected areas. Moreover, the private holdings were usually much closer to the users' homes and farmlands, which reduced the long treks for both people and their livestock when both were already weakened by hunger and malnu~tion. When drought gave way to famine the women's groups emerged as a critical link to shared use of private lands. As the community tacitly declared a state of emergency, so did they call upon those with greater resource endowments to share an increasing proportion of those resources with others. However, this social pressme applied to the act of sharing, not to the naming of the beneficiaries; in fact, participants would be a more apt term, since those in need were recognized by largeholders according to longstanding relationships of reciprocity, most often and most predictably in the context of kinship or women's self help groups. In effect, the communitypeeled back the survey map to reveal another map of potential use and users, derived from traditional rules of reciprocity and mutual aid. Yet the power to determine exactly who could use exactly 162 which resources, and where, had shifted from the community and large kin groups to the individuals who controlled the legal boundaries. Thus a third map emerged which combined traditional norms with new loci of power, at both community and household level. The fact that men actually owned most of the private plots and formally controlled the public lands, set the stage for a gendered struggle for access to resources no less serious for its fmesse and skillful manipulation by individual women and self-help groups. Poor women's new and traditional knowledge of ethnobutany was necessary but not sufficient for coping with drought and famine. They also required access to resources controlled by men at household and community level. Women's groups and individuals had to mobilize substantial political skill to legitimize and tap their "social credit" at household, group, and community levels.
The name of the famine

All subsequent discussions of species preference, tree man~ement, and land use for the next two years were influenced by the experience of the drought. Not only bad people re-diacovered trees and wild plants as sources of food, fodder, and medicine, they had observed the diminished but still substantial yield and the survival of one man's citrus orchard in comparison to the death or distress sale of their livestock. Moreover, there was widely shared interest in planting fodder trees and in the introduction of fruit trees for both home use and sale. Many people had acquired a healthy skepticism about overreliance on cash income to offset the effects of famine. In fact, this last point illustrates an oR-neglected dimension of indigenous knowledge: the learning, storage, and transmission of knowledge about social political, economic, and environmental change in the form of oral history, particularly in the naming of events. In the case of Katham& people "reckon time in famines, and remember them by name". The name of the last famine captures the painful irony of the changing times: "I shall die with the money in my hand" (Alice, personal communication; Rocheleau and Jama, 1989). The codification of knowledge in the form of famine names records the central smlwise of the last famine, makes sense of the experience, pre-empts the surprise of similar incidents in subsequent years and informs practical, popular planning measures to prevent future famines altogether. Experience in Kathama demonstrates that indigenous knowledge extends well beyond the confines of botany and agriculture, and well into the domains of environmental history and practical political economy.
Conclusion Perlmps the most salient feature of ethno-science research with rural women is that it can "buy time" and "create space" for the women themselves to take stock of the larger scale processes working against ecological,

Rocheleau: Stories and Lessons From Kenya economic, and cultural diversity in their rural landscapes. If research results in documentation and discussion of gendered ethnoscience at community level, then rural women may make more informed choices about which species, which skills, and which visions of nature and society to carry into the shaping of their emerging ecological and economic futures. An action research program might also facilitate the discussion and transfer of knowledge between men, women, and children, as their roles, responsibilities, and interests change. 4 This process coald result in re-negotiation of the division of rights and responsibilities as well as domains of knowledge and sin, though not without substantial smaggle over conflicting interests at household, community, and larger units of organization. The mere recognition and documentation of survival as a gendered science in harsh and unpredictable environments (political, economic, ecological) may effect change at local and national level. At best it may even serve to re-establish the legitimacy and strengthen the dynamism and innovative capacity of rural women's and men's separate, shared, and interlocking knowledge as tools to shape their own futures. Acknowledgements I am indebted to the people of Kathama, Mbiuni Location, Machakos District Kenya, for their continuing and patient collaboration over a period of seven years. Thek efforts have supported research as well as training of young professionals from Africa, Europe, and North America. The leaders and members of several women's groups (Syokimau, Panda ukethe, Athi River, Kanzalu Range, and Katitu groups) have contributed their time, knowledge and insights to surveys, informal trials, student training" and publications. Mnsyoki has shared his land, water source, tree seed, sk/ll~, experimental results, and his honest judgement of agroforestx and related technolo7 gies. Alice, Ndungwa, Puninah and Winfrend (now deceased), led the women in the community in exploring new agrofores~y practices and in documenting their own experience and innovations. Veronica Ndunge and Lawrence Kyongo have measured tree growth and survival, conducted meetings, interviews and surveys, translated for outside researchers, and acted as extension agents. Japheth Kyengo has committed his own farm, and his community work to the testing of ideas, sl~s, and ixactices that he has developed as a field research assistant. He has modified many of the methods employed in the cotwse of field work, has informed otr analyses of results,, and has freely shared information with people in the community as well as outsiders. All of the work summarized in this paper also owes much to prior research conducted at Kathama by John Raintree and Remko Vonk, to collaboration with Ingrid Duchart, Kamoji Wachira, Luis Malaret, Muhamud Jema, and Alison Field, and to student researchers (A. van Hoek; B. Muchiri; E. van Djeil; P. Maundu; M. Mutiso; L. Pope; A. Morrill; I. Cantor). Field work was conducted under the auspices of ICRAF with funding from the Rockefeller Foundation and The Ford Foundation. Notes 1.Chambersdefines"normallxofessionalism" as careeristbehavior thatoftenplacestechnocra~ a n d b ~ imperativesabovethe needs ofpeople andfl~edemandsofthepanioAsrsitmtion. 2. This choiee to tell astory derivesfxomthe often unique circumo stancesofethnoscienti~ expertise.It alsoreflects the impot'am~of nmmiveaccountsin feministscholarship.Sto~lling" as opposed to experhnentalreporting"is alsoin keeping with the tradition aud spiritoflhohistm'icaland(simultaneousiy)sclentificrecordof peopleinKathama. 3.The termsharudlandsisusedas m alternativeto common lauds~:e the formal definition of the latter (Bmmley, 1986) excludes the complexpattemofus~ access,andcontroldescribed here. 4.For a separatediscussionof the methods usedin the field worksud thoseappropriateforfurtheracfionresearchsee Chambersa ai, 1989, ILEIA,1988;Jiggim, 1986;Rocheleanetd, 1988,R0cheleau, 1991. References Bebbington,Tony.1 9 9 0 . ' ~ F m n ~ K n o w k d g e , ~ m t i o n a l ~ and SustainableAgriculturalStrategies:ACase Studyfromthe EasternSlopes of lhoPemv ianAndes."Bulletin ofL ~ inAmeric~mRese~'ch.9 (2).Forthcoming. Brokensha. David, D.M. Warren and 0. Wemer. (eds.) 1980.Indigmous KnowledgeSystons andDevelopment. Unive~ity Ptessof Am~ic~ Lanham,Mefy]and. Brondey, D. 1986. "The common property challenge." InNatkmal Acadonc~ofScienceProceedingsoflheConferencesonCommonPropertyResourceM~magement.NaliunalAcademyPress. Camey,Judlth. 1988."SUugghnoverlandandcmpsinaninigatedrke schem~ TheGambia."In J.Davisun. Wom~andLoutTenure /~A~-~_a.Westview.Boulder. Chambers,Robert. 1986."NormalProfessionalism,NewParadigms andDevelopment."IDSDiscussionPaper 227.IDS Sussex. Chavangi,N.A. 1984."Culturalaspectsof fuelwoodprocurementin Kakameg~ KWDP"WotkingPaperNo.4.KenyaWood FuelDevelopmentPmject.Nairobi. Dankelman,IreneandJ0mDavidsm. 1988.Wom~ andEnv/ro,vnent in the Third World:A llicm~efor theF uture. EarlhscanPubllcafiunsLtd.Londen. Fernsudez,Maria.1988."Techoologicaldomainsofwomeninmixed fanning systemsofAndeanpeasant communities.'InS.Poats, M. Schmink and A. Spring. Eds. Gender Issues in Farming SystemsReseorch andF~ms/on. WestviewPress. Boulder. Fleuret, Ann. 1979."Methods for evaluation of the role of fruits and wild geens in S iuwabea diet: a case study." MedicalAnthropology.Spring,I979. G u h a , ~ 1989."Radical~envkonmentalismand wiklenesspreservafiun:aThinlWcfkicrit/que."Em/ro~mnta/ Eth/cs. 11 (1) 71 - 83. Hecht, Susanna and Alexander Cockbum. 1988. The Fate of the Forest:Developers, Destroyers andDefenders of theAmazon. Varso.NewYork. Hoskins,Madlyn. 1983.Rura/Womm, ForestOutputsandForestry Projects.F~A.O.Rome. Jain,Shobita.1984.'~tandingupforUees:Women'sm l e i n t h e ~ movement." Unasy/va36 (146): 12-20. Jiggins, Janice. 1986."ProblemsofUnderstanding andCommunication at the Interface of Knowledge Systems."In S. Poats, M. andA.SIxinyrEds.Gender ls,mesinFarming Systems 163

AGRICULTURE AND HUMAN VALUES - WINTER-SPRING 1991 R ese~ch and Extension. W estview Press. Boulder. Jiggins,Jardce.1986."WomenandSeasonality:CopingwithCrisisand Calamity."In R.LonghurstEd. Seasonal/ty andPoverty. IDS Bulletin 17 (3): 9-18. Johnson, Douglas H. and D.M. Anderson. 1988. The Ecology of Survi mL"Casea . d i e s j ~ m N o n ~ A~can ~wry. W estview Press.Boulder. Joshi,G. 1982.Menpropose, womenopp~e the destructionoffore.~s. Information Service on Science and Society-Related Issues. CentreforScimceandEnvirenmenLDelhi. Juma, Calestous. 1988.TheGeneHunters:Biotechnologyandthe Scramble forS eeds.pr:a~tm U m v ersity Press.Princeton. Juma,~ 1989.BiologicalDiversityandlnnovation:Conservingand Utilizing Go~ic Resourcesi n K e n y a . ~ Centrefor TechnologyStudies.Nahobi. King, YnesUa.1989."The eoologyoffeminism andthe feminismof ecology."h JudithPlant.Ed. HealingtheWounds. New Soaety Kirim, Amos aad C.Juma.Eds. 1989. Gaining Ground: Institutional Irmovatlons inland useManagement in Kenya. AfricanCentre forTechnologyStndles.Nmhebi. Matth~d,Wangari.1985.TheGreenBeltMovement.WangariMatthai, P.O.Box 14832, Naimbi. Mendes, C. 1989. Fight for theForest: ChicoMendesinHisOwn Words.LatinAmetica Bmem~Londom Merchant, Camlyn. 1980. The Death qtNature :Women, Ecology and theScientificRevolution.HarperandRow.SanFrancisco. Merchant, Carolyn. 1989.Eco/o$/calRevolutions:Nature, Gender and Science inNew England. The U nive~ty of North Carolina P r ~ Ch~p~lX-m Meyedmff~F-liTal-eth.1982.TheSocio economic and RitualRolesof Pok~Women.AnthropologyPh.D.Thesis.CambndgeUnivermeritand SustainableDevelopment.University of Groningen, The Netherlands.Dec.6-9,1988. Posey, Daryl. 1985. "Indigenous management of Iropical forest ecosystems: the case of the KayapoIndians of the Brazilian Am~zon."AgroforestrySystems3 (2): 139-158. Redclift' Michael. 1987.Sustainab/eDevelopment: Exploring the Contr_,M "_~k~ns. MethuenandCo.New York. Richards,Paul. 1985.IndigenousAgriculturalRevolution.'Ecology and Food Production in WestA f r i c a . H ~ n : Londom Ricluaxls,PanL 1986. C o p i n g w i t h H u n & e r : H a z a r d a n d ~ inanAfriccmRice.FarmingSystem.AllenandUnwin.London. Rochelean,D. 1991."Parfdpatn~researchin agroforesU~,Lean~g from experience and exl~mdingo~repetnire."Agroforestry Systems:9 (1). Rocheleau, D. 1989."The gender division of work, resources and rewardsin agrof~estry systems."In A.E.Kilewe,K.M. Kealey andK.IC Kebaaraeds.AgroforestryDevelopment in Kenya. ICRAF.Naimbi.pp. 228-7.A5. Rocheleau, D. 1987."ALand User Perspective for Agroforestry Research and Action."In K Gholz ed.A&roforeso~ :Realities, Poss/b///u'esandPotent/a/s.Marfnus Nij~off.Det'drecht. Rochelean,D. 1988."Women,trees andtenm~."InLFomnann and J.Bruce eds. Whose Trees? Proprietary dimenslons offorestry. W estview. Bouldex. Fp. 254 - 27 2. Rochelean,D.E. 1985 "criteriafon~,-apprsisalandre-design:inlxahousehold andbetween-household aspects of FSREin three Kenyanagmfores~pmje~."ICRAFWoddngPaperNo.37. ICRAF, Nsirobi. Rocheleau. D~.., K.K. Wachira, L Malaret and B. Wanjohi, 1989. "F,thnoecologicalmethodsto complementlocalknowledgeand fann~innovatimsin agmforestw."hAFocey,R.Chambe~and L. Thml~eds.Farmer Fir~.Intermediate Technology.London. Rocheleau, D.,P.Khasiala, M.Munym, M. Mutiso, E. Opala, B. WanjchimdA.Wmjuma.1985.Womm" suseqfoff-f~mland~: Implicationsfor agroforestryrese~ch.Pmjectretx~to theFord Foendation.Mimoo.ICRAF.Nah~bi. Rocbe~D.andA.Hoe~ 1984. ' T h e A p p ~ a t i e n ~ and LandscapeAnalysis in AgmforeslxyDiagnosisendDesign: A Case Study from KathamaSublocation, Machakos District, Kenya."WerkingPaperNo. ll.ICRAF.Naimbi. Rochelem,D. a n d L ~ 1988. 'AVcmen'sslm:~ andwomen's placesin ruralfoodix~__,__,,~ion systems:Thespatialdism'bution of women'srights, respomibilities and acfivities."Paperpresentedto the 7thWoddCongress ofR~al Sociology.25 lune1July.Bologna,Italy. Rocheleau,D.,F.Weber, andA.Field-Juma. 1988. Agrofor~ryin Dry/andA~/c,, ICRAF.NSIrobi. Rochdean, D. andM.Jama. 1989.AnnualReportofl:armingSystems and ~ Research Projoct.Sur,mmry o : ~ v i e w w i t h eiders at Kathama. manuso(q~ IDS. Nakobi. Roe, EmeryendLouiseFortmann. 1981. Semen and Slrategy:The changing orgenizatlon ofihe r ural water sector in B atswana. RuralDevelopmmtCommittee.ComellUnivenity.Ithaca. Samkhan,Jose, 1985. "Ecologicalandsodaloverviews ofethnobotmical research."Economic Botany.39: 431.43 5. Schmink,MarianneandChadesWood. 1990. "CentestedF~ntie~in Amazonia." Manuscript. Center forLatin American Studies.

s~y.
Moore,Henrietta. 1988.Fern/n/sinandAnthropo/ogy.Universityof bfumesotaPress,lVfirtm~polis. Moore,~ 1986. Space, Text and Gender: AnAnthro~logical StudyoftheM ara~a ofgenya.Camt,~Ig~ Cmnlxidge Univ er-

~yPre~
Moore, Henrietta, Megen Vanghan. 1987.'~uttingdown trees: and women, nulritien agricultmal and change intheNorthernProvinceo~bi~ 1920 - 1986."Afr/canAffa/rs (345):523-40. 86 Mortimer,MichaeL 1989.AdaptingtoDrought.CambridgeUniversity.Cembridge. Munym, P. 1987. "Theimpot~anceof gathered foodandmedicinal plant species in Kakuyuni and Kathama areas in Machakos." Annex 1 in ILK. W achira (ed.) Women's Use ofOff-Farm and BoundaryLands:AgroforestryPatentials.FmalProjectRepo~ ICRAF.NSIrob Murphy, Yolanda endRobert Murphy. 1974. Womenofihe Forest. ColurabiaUnivemityPre~.New Ymk. Myers,Norman. 1979.TheSinkingArk.Oxford.PergamonPress. N y ~ J u l i u s . 1989.Tes&nonybefmeWodd CommissimenEnvironmentandDevdopmmt'October, 1986.Harare,~mmbawe. Parkipuny,M,S. Ole. 1988. "TheNgnmngomCraterIssue:ThePoint of ViewoftheIndigenom MaasaiCommunityofNgoro~gom." P~pertm:sentedto theInternafionalCongresson NaaueManage164

Rocheleau: Stories and Lessons From Kenya UniversityofFlorida.Gainesville. Shiva,Vandana.1988.StayingA~ve:Women,EcologyandDevelopment.ZedPress.London. Shostal~Marjtxie,1981.Nisa:TheL ~feandWordsofa !Knm Woman. g VintageBooks.NewYork. Stamp, Patricia. 1989. Technology, Gender andPower inAft ica. IntmmtiooalDevelopmentResearchCentre(IDRC).Ottawa. Talle,Aude. 1988.Womenata[.z~ss:ChangesinMaasaiPastoralism andTheirEff~tsonGender Relations.Unive~sityof Stockholm. Stock~lm. Toledo, Victor M. 1989. "The ecological rationality of peasant produetion."In M.Altieriand $.HeehtEds.Agroecologyand SmallFarmDevelopment.CRCPress.BoeaRa~n, Florida. Thm~,L.A. 1989. "Legitimiz~ Lo~Knowledge:FromDisp~:e,meattoEmpowermenffcxThirdWorklPeople."Agr/cu/t~eand HmmmVa/ues.6 (3): 13 -24. Waehira,K.IC 1987.Women'suseofoff-farmandboundary lands: agrofore~rypotent/a/s. Projeetreportto theFord Foundation. Wamalwa,BettyNafun~ 1989. '~digenous knowledgeandnatural resources."InKiriro,AmosandC.Iuma.Eds. GainingGround: Institutional InnovationsinLand Use Management inKenya. Cent~forTeelmologyStudies.Nairobi. Wanjohi, B. 1987. Women's groups' gathered plants and their potenfals in theKathama~ Annex 1in ICK.Waehir~ (eeL) Women'sUseofOff-Farm andBoundatyLands:Agroforeslry

Potential~ ProjectReporICRAF.Naimbi. F'mal t

Wan-en,D.M. 1988. "Linkingseienlifieandindigenousagrieultoral systems."/nJ.I,inComl~onEd.TheTransformatianoflnternationalAgriculturalResearchandDeveiopment.Westview.Bouldef. Warm,MichaeL1983.Sile~Violence:FaodFamineandPeasan~in NorthamNigeria.Uniwrsity o f ~ P r e s ~ Bea~167. Waits,MichaeL1988."Semgglesoverland,smagglesovermeaining: of somethoughtsonnaming,peasantresistsnceandthepolitics place."InR.Golledge,H.CoucelisandP.GouldEds.AGround forCommonSearch. GeographicalPress. Santa Barbara.

bfnneo.Naimbi.ICRAF. 137pp.

INTERNATIONAL

DEVELOPMENT

ETHICS

ASSOCIATION

THIRD

INTERNATIONAL CONFERENCE ETHICS AND DEVELOPMENT

ON

"THE ETHICS OF ECODEVELOPMENT: CULTURE, THE ENVIRONMENT, AND DEPENDENCY"

The International Development Ethics Association (IDEA) announces a call for papers to be presented at its Third International Conference on Ethics and Development, to be held at the Universidad Nacional Autonoma de Honduras, June 21-27, 1992. The theme of the meeting is "The Ethics of Ecodevelopment: Culture, the Environment, and Dependency.' IDEA is a cross-cultural, multidisciplinary group of philosophers, as well as development theorists, policymakers, and representatives of grass-roots groups, who apply ethical reflection to development goals and strategies and to the relations between rich and poor countries. Established in Costa Rica in 1987, IDEA held its first conferences in Costa Rica (1987), Mexico (1989), and a workshop, 'Ethical Principles for Development: Needs, Capacities, or Rights?,' at Montclair State College (1991). IDEA's Third Conference aims (1) to promote a fruitful convergence of international development ethics and environmental ethics, (2) to forge an ethic of ecodevelopment that takes into account cultural, environmental, and economic values and constraints, (3) to influence the formation and implementation of ethical policies and practices concerning development and the environment. The International Conference Advisory Committee includes Luis Camacho (Costa Rica), Peter Penz (Canada), Ramon Romero (Honduras), Horatio Cerutti Guldberg and Laura Mues (Mexico), Nigel Dower (Scotland), Ken Aman, David A. Crocker, J. Ron Engel, Denis Goulet, Rachel McCleary, and Paul Streeten (USA). The deadline for submission of paper proposals is November 30, 1991. The deadline for finished papers/abstracts and advanced registration Is April 30, 1992. For further information about IDEA and its Third International Conference, contact David A. Crocker, IDEA, Department of Philosophy, Colorado State University, Ft. Collins, CO. USA 80523. Fax: (303) 491-0528. Telephone: (303) 484-5764.
165