Cultural survival and the geographic ordering of space are tightly interwined. Indigenous peoples, ethnic minorities, and other associations of peoples whose ways of life are under threat have long turned to the defense of specific territories as a means of increasing their control over the pace and direction of cultural change. Many have been assisted in their efforts by a growing willingness of government and multilateral and organizations to accept more participatory methods of social research, including sketch mapping and Participatory Rural Appraisals, ad by recent developments in mapping technologies, such as more user-friendly GIS software and GPS equipment. Indigenous peoples' organizations can now present proposals for the protection of their lands with the kind of technical sophistication and precision demanded by governments and aid organizations, without necessarily sacrificing the significance of the lands for their cultures.
WHY GENERED MAPPING?
While applauding these advances in protecting local self-determination, we would like to encourage the inclusion of a gender-based analysis of how spaces and places are used, valued and struggled over in specific cultures. Such gender based mapping is essential not only to protect women's independent sources of income and livelihood - maintaining a balance of power between men and women under changing social and ecological conditions - but also to preserve the diversity in local flora and fauna, since men and women can have dramatically different relationships to particular resources.
Mapping men's and women's access to and control over resources is not always easy to accomplish. The balance of rights, responsibilities, activities and knowledge between men and women in a particular culture is always subject to change, and is especially dynamic in situation where economies and ecologies are undergoing rapid transformation. Moreover, women's spaces are frequently nested between and within lands controlled by men, as thin strips of bush separating homesteads or even single trees scattered amidst a husband's cultivated fields, and can remain invited to the inattentive or untrained observer.
The difficulty in seeing and mapping women's rights in this context has in the past led to land tenure reforms and land use proposals that selectively disenfranchise women. Exclusive ownership rights become vested in male-headed households or male-dominated community organizations as a result of mapping and land reform initiatives uniformed by gender analysis. Husbands, brothers and sons can then restrict women's access to and use of resources in favor of their own, separate income-earning and subsistence activities, particularly in culture where the gender division of rights and responsibility within the household is relatively pronounced. Similarly, by classifying areas where women gather or extract uncultivated resources as either "pristine" wildlands or "unused" wastelands, expatriate planners or local men may inadvertently justify the conversion of these areas to land uses of little value to women. In the first instance, useful bush and scrub may be incorporated into protected areas where women are physically barred from access. In the second, lands may be cleared of a diversity of indigenous flora and fauna in favor "factory farms" with commercial monocrops. In both cases, women can lose effective access to a wide range of resources that were once readily available to them.
A number of "constituencies" have a stake in the elaboration of local peoples' territorial rights. In addition to those interested in the protection of cultures per se, both local and expatriater environmentalists have recognized the special role that many local peoples' organizations play in the protection of rare and biologically diverse habitats, "Green" commercial interests, local and foreign, have also supported indigenous land rights in the hopes of preserving potentially lucrative source of genetic material and tapping into local knowledge of its uses. Why should these groups take an interest in the way that space is gendered?
First, there is an ethical question; women are a legitimate constituency themselves, and their rights to resources are fundamentally human rights. Any erosion of women's access to resources, inadvertent or otherwise, is likely to have negative consequences for women's health and well-being. Land tenure reforms and land classification programs in Kenya, for example, have deprived women of access to common gathering areas where firewood, water, fodder, fiber, medicinal plants, and wild foods are found, increasing women's labor-burden and expense, while eliminating important sources of income and subsistence (Pala-Okeya, 1980; Wangari, 1991). Even when women are compensated for such losses in access with a reduction in responsibilities or alternative rights, there remains the ethical issue of women's rights to have a voice in how their environmental is changing.
Second, because women and men often have very different daily experiences, they frequently possess very different resource management skills and environmental knowledge (Jiggins, 1986). Eliminating women's spaces and places in the landscape may also eliminate this important source of ecological and cultural information, both from the local community and beyond.
Finally, women have proven themselves quite adept at resisting change inimical to their interests, where as allies or foes of environmentalists. In many cases, they have taken the lead in environmental movements, as in the environmental movements, as in the Chipko Movement's work to protect forest commons in Indian (Shiva, 1988). In other cases, however, women have proven formidable opponents of "green" initiatives, ripping up tree seedings in a reforestation project, for example, in order to protect their own rights of access to a parcel of land. Demarcating boundaries without their input may provoke them to undermine even the most carefully constructed territorial agreements.
GENERAL PRINCIPLES
How then should researches and advocate take account of gendered space and place in the creation of our geomatic images? Three general principles seem to us to be of particular importance: starting at the scale of everyday use of the landscape; incorporating the multiple rights and responsibility of resources user groups and the social relations that shape them; and actively seeking our multiple perspective on the use, value and meaning of landscape features. The scale at which initial research is conducted can be critical initial research is conducted can be critical because, as we have already mentioned, women's spaces are often nested between and within lands controlled by men. They may, in fact, be limited to particular resources, or even particular products of those resources, certainly much smaller than a single pixel in most land use or property images in use today. Starting at the scale of regions, communities, or even single villages, therefore, may render invisible many of the places and specific resources most crucial to women as they struggle to meet their daily responsibilities.
Yet the choice of scale is only one element in the analysis of gendered space and place. Who has what rights in resources is clearly an issue of social relations, and cannot be read from even a life-sized diorama without an understanding of how those relations work.
We mute being by carefully analyzing difference in men's and women's household and community responsibilities, labor obligations, rights of access to and control over specific resources and products, knowledge and decision-making authority together in complex patterns invisible to outsiders, at least initially.
A single tree in Kenya, for example, may have a male "owner", be cared for by a woman borrowing the land on which the tree is found, provide fruits to her and to another woman who lived on that plot when the tree was planted, and furnish small sticks and other fuel to all in good standing in the community. This calls for a multipurpose representation that can account for the complexity of the kinds of multiple, overlapping rights and obligations likely to be associated with any one spot on a conventional map.
Finally, the meaning ascribed to the landscape, the values attributed to its various elements, and the normative visions of that landscape's future can generate very different cognitive maps for men and women. A bushy roadside may not appear as a resource of any value to the men of a community, but may be the place where women most often graze their goats. Swamps valued by women primarily for the provision of materials for weaving mats and baskets may be identified by men as the ideal location for a eucalyptus plantation. Even the best multipurpose maps reflect just one of these perspectives, and it is women's perspective that have often been left unrepresented of distorted in maps-as-usual. We must therefore actively seek to construct "countermaps" with women, maps developed explicitly to present the views of women and other groups excluded both from scientific discourse and from political negotiations concerning the use and representation of land-scapes. Such countermaps can help women resist such geomatic acts of exclusion, and create alternative visions of local landscapes that take better account of women's rights and responsibilities. (See Nancy Paluso on the application of contermaps in broader context).
Once recognized, men's and women's separate perspectives on shared livelihoods and landscapes can be brought together in a multidimensional medium that does not obscure the claims of either. Countermaps can be used in opposition to or in combination with other maps to evaluate the likely effects of any changes in land use, property regimes, and land management practices or the terms under which men and women gain access to resources. Areas of complementary and conflicting interests between men and women can be highlighted, and more equitable and practical land use and management plans can be developed. Using a combination of maps, for example, men and women could agree that although the swamp is the best location for eucalyptus trees from the men's point of view, a nearby hillside is also suitable, and does not jeopardize women's handicraft production. Where only single maps are practical, they should reflect these kinds of compromises among multiple perspectives, rather than solely the male perspective.
SPECIFIC GENDERED MAPPING TECHNIQUES
Beyond general principles, we would also like to discuss specific techniques that can help us to see the ways in which landscape are gendered in particular cultural contexts, record this information in images, and finally "scale up" or generalize from these detailed images without losing track of the spaces, resources, and associated rights of importance to women. A large body of literature discusses Participatory Rural Appraisal and related gender-sensitive research methods, including key informant interviews, focus group interviews, transect walks with key informants, and participant observation (Thomas-Slayter, Esser, and Shields, 1993). Each technique can be used either to gather general information about a particular landscape, in which case informants largely determine the content and direction of the research, or to elicit information on specific issues of concern to the researcher. In the former case, care must be taken that informants have not made inappropriate assumptions about the interests or knowledge of the researcher; in the latter case, researchers, should be way of overlooking or misinterpreting important data that do not fit neatly into their preconceived categories of analysis. Researchers may also need to decide whether they will interview men and women separately or together. Separate interviews have often proven useful in obtaining information on the rights and responsibilities specific to women. Joint interviews, in contrast, can reveal much about the process of how men and women negotiate resources use and access, high-lighting areas of gender complementarity and conflict.
Finally, focusing on different geographic scales can have an impact on the types of information obtained. Men and women may have interests in very different spaces, resources and tenure arrangements within their own households than in the context of the wider community landscape. As with all research, if time allows, the triangulation of techniques and use of follow-up interviews can help reduce the chances of misinter-pretation and improve the quality of the maps produced.
Once the information is collected, there are several options for converting it into useful and representative images. Researchers may try to do this on their own, generating diagrams, sketches, and maps based n the data collected from narrative sources and observations. The advantage of doing so is that it may represent the best match of media and skills with informants and researchers. Particularly when informants are elderly or without formal school experience, they may be much more eloquent, and feel more at ease, expressing their ideas verbally rather than visually. This can be true whether they are asked to use paper and pen to create images of their landscapes, or "found" materials such as leaves and pebbles.
Many researchers working in rural areas, however, have preferred to encourage representatives of various resources user groups to develop images themselves, thus facilitating the expression of the local meanings of landscape features and increasing informants' stake in the research process (Bruce, Fortmann, and Nhira 1993; Chambers, Pucey, and Thrupp, 1989; Scoones and Thompson 1992; Thomas-Slayter, Esser, and Shields, 1993).
Again, there are decision to be made as to whether or not the researcher should guide the sketching and mapping exercises based on his or her specific interests; whether and when women should be asked to make maps separately from men; and at which scales local images should be created. No matter who is actually doing the drawing, the process can be more or less interactive, depending on the time available for gathering data and the patience of both researchers and local informants. Maps drawn by researchers on the basis of local interviews and observations, for example, can be brought back of informants for comment, editing, and enhancement. Images produced by various user groups can also serve as a basis for further interviewing by the researcher, leading to the production of new and richer images.
The kind of sketch maps that can result from such an iterative process are illustrated in Boxes 1 and 2. Box 1 represents gendered resource use, access and control at the household scale in a village in Machakos District, Kenya. Box 2 situates a similar household in a community context in Ghuse, Lalitpur District, Nepal. Both maps were developed to synthesize information, group interview, household interviews, and key informant interview over a long period of collaboration. Though researchers may construct such images at home, based on field notes, they often choose to develop their notes, they often choose to develop their images while in the field, within the context of group meetings or key informant interviews. Images may be sketched by researchers from narrative accounts and participant observation or drawn by local participants and labeled or elaborated by researchers based on further discussion. In fact, combing narrative accounts and sketching simultaneously, with an active exchange of information between the narrator and the "artist," often has a synergistic effect, resulting in more detailed and accurate images. (Not that "accuracy" in this context may include effectively representing the ambiguity and complexity of multiple perspectives or the delineation of points and regions of conflict). Similar resource tenure maps, from individual plants and plots to whole watersheds and villages have been published in case study and policy papers under the auspices of the International Development Program's SARSA/ECOGEN Project and the Marsh Institute at Clark University.
Several related methods for eliciting and summarizing information about gendered domains of space, place and power in the landscape can also be used to good effect. Among those we have found most useful are: realistic and detailed landscape sketches with species labels and gender symbols; larger-than-women's resources; and felt boards that allow participants to simulate changes in land use and various options for farm and village "development". Using the felt boards and symbols of plants, animals, buildings, and other landscapes features, both built and natural, small groups of men and/or women can recreate the past, explain the present, or create new landscapes. This activity is nested within "what it?" discussions of issues including gendered access to, valuation and use of resources (Rocheleau et al., 1994).
By using such an interative and collaborative approach, and by combining narrative accounts with visual representations, we have been able to obtain varied and complex information while maintaining an awareness of its gendered qualities. Yet we are constantly confronted with another difficulty: how to maintain that awareness while genderalizing from the information in such a way that it is useful to policy-makers, project planners, and community leaders. Standards practice in resource mapping and the delineation of territories tends to smooth over the small spatial anomalies on large-scale maps. These generalization have often been based on technocratic imperatives, to the detriment of women's interest.
This need not be the case, however, Take as an example the production of a land use/land cover map at a scale of 1/50,000. We could use the existing land use categories, such as pasture, lowland rice fields, and swamplands, and relabel them to reflect the presence of resources important to women. This isn't a matter of inserting every detail of the landscape into these new categories, but rather of preserving pervasive but microscale features. Where women gather a significant portion of their fuelwood from scrub scattered amidst pasture, the "pasture" category could be relabeled "pasture with scattered scrub for fuelwood." "Lowland rice fields," might be relabeled "lowland rice fields with hedges for goat fodder" where hedges are a vital source of seasonal fodder for women's goats.
We may also wish to quantify the presence of a resource valued by women Swamps in Uganda often contain palm trees used by women in the production of handicrafts; a map's classes or categories could indicate the relative densities of palms per unit area.
Maps of property rights can be treated in the same way. This "small-holder farms" can be redefined as "annual crop lands with communal gathering areas"; tribal hunting grounds" may be relabeled "women's forest gardens with men's hunting grounds"; and "tribal grazing commons" might become "tribal grazing commons with lineage-based rights for women to gather fruits from specific trees". Though some of these titles may be less than elegant, they reflect actual land use much more accurately that the pervious ones.
We can also mitigate the problems of generalizing our information by keeping in mind that no single map is "correct" or "final". Maps heretofore considered "overview" or "summary" maps should be viewed instead as "negotiated settlements" arrived at by a host of user groups reflecting multiple perspectives and uses of resource. The features and categories included in these maps may be the result of compromises which reflect the priorities of each user group, as well as each group's power to realize its own priorities in cooperation or competition with others.
As social and ecological conditions change, the terms on resource use and management may be renegotiated, necessitating the redrawing of our maps. Following the eviction of settlers from a game reserve in western Uganda, for example, elephant grass became an increasingly important source of fuel for many women cut off from their previous supplies of fuelwood. Any map of this region based on local interviews should reflect the struggle of these women to gain free access to grasslands. Single "overview" maps can effectively represent the "negotiated settlements" which allow for the shared use and management of particular landscapes by distinct user groups. To do so, however, they must be based on the full range of local perspectives.
Although,we have focused on the ways in which land use is gender-based, similar divisions of interest and opportunity between any land user groups can be represented as we have described. What is important is that we learn to see the rights, responsibilities, knowledge and perspectives of the respective user groups; map them in a way which recognizes the multiple uses of multiple users; and generalize without losing sight of the types of negotiated settlements reached in the everyday practice of living in and sharing landscapes. Though the form and method of creating geomatic images may vary significantly from situation to situation, the graphic illustration of the division and sharing of resources, work and knowledge can help fieldworkers to understand existing land use systems and tenure regimes and to design realistic and equitable interventions in partnership with local users.
Article copyright Cultural Survival, Inc.
