Gender Bias in Ethnobotany: Propositions and Evidence of a Distorted Science and Promises of a Brighter Future.

Gender Bias in Ethnobotany: Propositions and Evidence of a Distorted Science and Promises of a Brighter Future. Patricia Howard Distinguished Economic Botanist Lecture, kew royal botanical gardens, november 2006. This article argues that the knowledge and use of plant biodiversity is everywhere gender-differentiated. It explores the historical, conceptual, and methodological biases related to gender in much ethnobotanical research, draws out the consequences for ethnobotanical knowledge claims, and illustrates how a gender perspective changes our understanding of people-plant relationships in fundamental ways. The failure to deal with intracultural variation and especially gender differences in cognitive anthropological and ethnobotanical research has several associated errors: (1) the failure to research the knowledge and use of plants of both sexes, so that plant diversity and its uses are under-estimated; (2) the use of poorly-informed sources, leading to the improper identification of plants, their management, characteristics, uses or names; and (3) the misunderstanding of people-plant relationships, since a critical component of these relationships is not revealed. Consciousness of intracultural variation in people-plant relationships is increasing, but there is still a lack of methodological tools to deal with this. Key Words: ethnobotany, gender, scientific bias, intracultural differences, women, knowledge distribution, division of labor, cosmology, methodologies, history of ethnobotany, history of botany ────────────────────────────────────────────────── ‘Ethnobotanical and anthropological studies have frequently an unstated but critical male bias’. Miguel Alexiades, 1999, Ethnobotany of the Ese Eja: plants, health and change in an Amazonian society, p. 38. ‘Most of the ethnobotanical writings on female health issues were by foreign men, interpreting native men in turn interpreting native women’. James Duke, 1994, forward to the Amazonian ethnobotanical dictionary, p. iv. ‘The problem is that only by studying variation could the ethnographer be assured of having made the right choice [of key informant]’, James Boster, 1985a, Requiem for the omniscient informant: there's life in the old girl yet’, p. 194. Introduction Today and in the future, the way that we view women and gender relations in relation to the plant world will greatly influence our ability to halt the erosion of plant biodiversity across the globe, particularly of those plants that humans have found to be useful. I argue that, against most thinking on the topic, women collectively hold the majority (possibly the vast majority) of knowledge about the world’s plants. The simple explanation for this is that, throughout known history, women’s daily work has required more of this knowledge than men’s daily work. However, today, when it is perhaps more important than ever, women’s knowledge and management of plant biodiversity are underestimated and undervalued. The majority of the literature that directly deals with people’s management and knowledge of plants can still be termed ‘gender blind’ that is, unaware of the fact that women and men have different physical domains of work, knowledge, practices, interests and needs with respect to plants and their environments. ‘Gender blindness’, however, is not simply a product of ignorance – it is an historical legacy born out of a Western ‘cultural cognitive model’ that continues to distort science, particularly the science of ethnobotany. An example which is typical of such bias is found in an article recently published in Economic Botany about the use of wild plants by the small farming community of Las Pavas in Panama (Aguilar and Condit 2001). While the objective of the study was to determine ‘the importance of noncultivated plants’ to this community, the article’s title was ‘Use of native tree species’, since the researchers found that the vast majority of wild plants used by the informants were trees: ‘The community reported use of 119 non-cultivated plant species, including 108 tree species’ (Ibid: 223). This is a rather remarkable result since, even in forested areas, there is usually quite substantial use made of non-tree species. Such a result might lead one to ask whether the findings are exhaustive. Who, precisely, were the informants for the study whom the authors indicate represent the community? The section on methods indicated: ‘Information was obtained by consulting with 40 families of the area, mostly husbands, but women participated in some of the interviews at home’ (Ibid: 225). In a second phase, informants were accompanied on walks through pastures and farmland into forest areas where, it was noted, ‘Wives never joined these walks’ (Ibid: 226). The authors noted certain anomalies in their results when compared with the results of other studies from Costa Rica and Panama: Las Pavas, however, contrasts with other studies in the importance of construction and firewood uses, and the lack of medicinal uses. Sixty percent of the species at Las Pavas were mentioned for house construction, 34% for fuelwood, but only 8% as medicinal. In contrast, in a study in western Panama and Costa Rica, Hazlett (1986) identified 76 plant species used without cultivation by two villages, 14% of which for construction and 48% for medicine. . These other studies focused on indigenous groups, who ought to have longer traditions on medicinal and craft plants than Hispanic populations. But some of the differences must indicate researchers’ biases: Hazlett (1986) did not mention a single plant used for firewood, and surely Guaymi Indians burn wood (Ibid: 233). I argue that instead it is the results presented by Aguilar and Condit that are probably highly biased; that this bias is not easy for individual scientists to detect and is the result of the failure to consider women and gender relations in the research area and in particular the gender division of labor and gendered use of landscape spaces, which led the authors to seriously underestimate the use of wild plants in general and overestimate the proportion of trees, of certain uses and of certain landscapes. Aguilar and Condit are apparently unconscious of the presence of intra-cultural variation in wild plant knowledge and use in general, and of the activities and knowledge of women in their study area in particular. As is the case with most Central American Hispanic (mestizo) populations, it is usually women who make the greatest use of wild plants for medicinal purposes as well as for food, fibers, utensils, cosmetics and ornamentals, whereas men are primarily responsible for construction and are often responsible for firewood collection if this occurs far from home (e.g. Howard, pers. exp.; Kappelle et al. 2000; Ochoa et al. 1998; Stavrakis 1979). Further, women often use wild plants from spaces that are found closer to home in areas where secondary growth predominates, homegardens and borderlands. Finally, attempting to interview Las Pavas women at home in the presence of their husbands is not likely to elicit their knowledge, since mestizo women are renowned for being reticent to speak under these conditions. Aguilar and Condit, who failed to research women’s wild plant collection and use, assume that the grassland immediately surrounding the village is not a source of medicinals or food, and rather emphasize ‘the importance of forest to the community’ since this is where men access trees (2001: 233). Without conducting a study in Las Pavas on women’s wild plant use, it is impossible to demonstrate conclusively that Aguilar and Condit’s study would have had different results – but it is certainly valid to say that their results cannot be conclusive if they failed to systematically include women in their investigation. Singling out this piece of research is intended only to highlight what more generally may be considered as an Achilles heel in ethnobotanical research – the failure to consider intracultural variation and particularly the gender bias that leads to a distorted science. If it is of much comfort, in this, ethnobotanists are not alone. Gender research has clearly demonstrated the presence of gender bias in most social and natural sciences. This means that scientists take prevailing gender norms in a society to be ‘natural’ and often incorporate these norms into their theories as unquestioned assumptions. It also means that scientists assume male predominance and take men’s behavior and knowledge to be ‘standard’, whereas women are given little importance or their behavior is seen to be ‘deviant’ in comparison with men. Gender bias affects the theories, the questions formulated, the methods used, and the research outcomes. The repercussions go far beyond simply creating biased scientific knowledge: they extend into related practices, policies, and interventions that are intended to change the interactions between people, and between people and their environments. They can distort the outcomes in ways that are unanticipated and often not desirable. The ‘Who’ of Ethnobotany Before embarking on ethnobotanical research that involves investigating people’s knowledge and use of plants, probably the second most important decision to be made (ranking after the decision regarding ‘what’ the research is about) is ‘who’ will be selected as informants. This is usually presented in the literature in terms referring to groups of people such as ‘X tribe’, ‘farmers’, ‘forest users’, ‘healers’, etc. Within this, very often not all members of a specific group can be directly investigated, and a related question, no less important than the question of who ‘in general’, must be made regarding who ‘in particular’, that is who within the group will be selected to provide the information being sought. Given the scanty attention to this issue that is evident in much ethnobotanical literature, it appears that ethnobotanists regard it as a relatively minor problem of methodology: answers to the question of ‘who’ are seen as obvious or at least relatively straightforward, and typically descriptions don’t go much beyond rather general terms such as ‘key informant’, ‘experts’, ‘households’, ‘village sample’, ‘elderly’, etc. It appears that only rarely do ethnobotanists consider that the ‘who’ question is a major determinant of the replicability, validity or generalizability of research results, and also only rarely do they consider it as a theoretical problem. However, several decades have passed since it has been strongly argued (one might venture, demonstrated) that the question of ‘who’ has everything do with theoretical assumptions about how knowledge is distributed within cultures and about cognitive cultural models. Certainly work of cognitive anthropologists and ethnobiologists has not been ignored in ethnobotanical circles. Boster’s research on Aguaruna women manioc cultivators, which provide some of the most often-cited articles in the ethnobotanical literature (1985a; 1985b), provides a case in point. In another, albeit less cited article (1986), Boster drew out further the implications of his conclusions and highlighted four different assumptions about a cultural cognitive model and the associated distribution of knowledge about plant names: There is a single culturally shared model where all have access to the same knowledge There is no shared model and knowledge is random There is strong social differentiation, where each social group has its own shared model and knowledge There is a single culturally shared model where knowledge is learned to varying degrees His research showed that the last assumption is that which most closely fits with Aguaruna agreement on plant names, and he further demonstrated that certain people within that culture have knowledge that comes closest to what can be called that of the ‘omniscient informant’ – that is, it reflects the culturally shared model, and hence these people are worthy of being designated as ‘key informants’. Most ethnobotanists today are still like the ethnographers Boster described two decades ago: they ‘have been aware of variation for a long time, [but] they have often solved the problem by ignoring it, describing cultures as though they were uniformly shared’ (Ibid: 177). The problem for the researcher of selecting ‘who’ should be ‘key informants’ is predicated upon first understanding how knowledge is distributed: ‘. . only by studying variation could the ethnographer be assured of having made the right choice’ (Ibid: 194). Jumping across more than a decade to see what has happened since Boster made these critical observations, Medin and Atran pointed out in 1999 the significance of the problem of the ‘who’ question when they critique ethnobiologists for often failing to indicate the source of their data, and pointed out the consequences for scientific claims: Are the informants a representative sample or a few local wise people or experts? In some instances no mention is made at all of the informants as if the “facts” were free-floating entities in the culture. Without some more precise identification of the data, one cannot begin to assess basic requirements for science such as replicability (1999:5). Of all that Boster’s work is known for, I argue that perhaps his most important finding was that knowledge distribution in Aguaruna society reflects Aguaruna social structure, which determines the importance of particular plants to particular people within it. He determined that ‘a cultural model of manioc identification exists, and some people know it better than others.’ In general, ‘agreement is patterned in the way that one would expect such knowledge to be patterned . . . [where] variation between informants can be explained by factors that reflect differential knowledge: age, sex roles, and opportunities to learn’ (1985: 185). Boster’s research indicated that, among the Aguaruna, An individual can be considered as having a number of identities: a member of a society, an actor in a sex role in that society, a member of a household, and an individual. Different amounts of knowledge are shared at each of these layers. A certain amount can be presumed of any adult member of the society. That and more can be assumed as an actor in a sex role. Still more is shared between closely related people. Finally, certain knowledge is unique to the individual. As the difficulty or specialization of an identification task increases, agreement at all of the layers decreases but not at equal rates. Sharing at the most general layers falls off most quickly. Thus there is an apparent succession in the importance of these layers of identity depending on the difficulty of the task (Ibid: 191). Ethnobotanists still have a strong proclivity to overlook or underestimate the significance of these different identities and particularly to overlook women and gender differences, even though all contemporary societies are characterized by a gender division of labor, and the knowledge and use of plants is everywhere gender-differentiated. Like Boster, I argue that gender affects personal experience and motivation to learn, as well as opportunities for learning from others. But it also affects many other dimensions of the interaction between plants and people in ways that have been largely overlooked by ethnobotanists and even by cognitive anthropologists. Another proposition that I put forward is that the reasons for what is often termed ‘gender blindness’ are multiple and have much to do with shared beliefs or cognitive models that are prevalent in Western and other patriarchal societies. These models appear in scientific research as unquestioned assumptions, particularly about the way that knowledge is distributed, where men are seen as subsistence producers, knowledge holders and purveyors of culture, and women as caretakers. Hence it is a misnomer to say ‘gender blindness’, since ethnobotanists do pay attention to women when they consider that the realm of research is related to women’s ‘caretaking’ roles, or when ‘women’s problems’ (e.g. reproductive health) are the focus of the research. In fact, the problem is one of ‘gender bias’. Yet a third proposition is that gender bias in ethnobotany leads to a distorted science. In the rest of this paper, I explore and illustrate the historical roots of a conceptual and methodological bias regarding women and gender differences and relations in ethnobotanical research. Then I examine some general patterns in the gender division of labor regarding plant use and management that is the product of a long-term literature review, and discuss in some detail the gendered nature of indigenous or local knowledge. I then draw out the consequences of the neglect of gendered knowledge and use of plants for contemporary ethnobotanical knowledge claims. Further, I argue that the significance of attention to gender issues goes beyond correcting the above three errors; it also highlights the importance of other relatively neglected conceptual realms. Finally, I briefly review available texts on ethnobotanical methods to determine their value and lacunae when it comes to researching intracultural variation and gender. A Western Cognitive Model underpinning Ethnobotany: the ‘Cult of Domesticity’ Women manage and use plants largely within what is often socially defined as the ‘domestic’ realm. The ‘domestic’ realm is commonly portrayed as a reproductive sphere where women, as principle agents and managers, carry out unpaid, home-based activities that ensure the maintenance and functioning of people within households. The household, in turn, is characterized as the principle site of collective consumption. In Western cultures, this ‘reproductive domestic’ archetype is strongly embedded in what historians refer to as the ‘cult of domesticity’ that prevailed in Europe and its colonies from the 18th to the 20th centuries. Its influence on the development of the sciences – especially those relating to human evolution and to botany - has much to do with the contemporary failure to recognize and valorize women’s relations to subsistence and to the plant world. The ‘cult of domesticity’ and its corollary, the ‘reproductive domestic’ archetype, are features of a powerful system of ideas (a ‘cultural cognitive model’) that serves to obscure rather than to enlighten our understanding of people-plant relations. While I discuss the relationship between the ‘cult of domesticity’, botany, and ethnobotany further on, it is important to give attention to gender bias in the sciences dealing with human evolution, which for centuries obscured the longest standing and most important relationship that women have had with plants and with human subsistence. In her landmark review article, Fedigan (1986) showed that, up until at least the 1980s, sciences concerned with human evolution were dominated by a ‘Man the Hunter’ model, which portrayed women’s contributions to subsistence through plant gathering as part of the ‘domestic’ realm and as insignificant relative to men’s contributions through hunting. It was ironically at a conference entitled ‘Man the Hunter’, held in 1966, that it was first more widely recognized that the bulk of foodstuffs in hunting-gathering societies is provided by plant food gathered primarily by women. In 1968, Lee published his survey of 58 foraging societies demonstrating that hunted foods contributed only an average of 35% of total food supply, except in the Arctic, where meat is of primary importance (cited in Fedigan 1986: 47). Further reviews of a larger number of hunting-gathering subsistence based societies revealed that the ‘Man the Hunter’ model had no empirical substantiation. But other ideas closely associated with the ‘Man the Hunter’ model continued to have great influence. For example, due to their reproductive status, it was considered that women carried out lighter work and were tied to base camp. These assumptions were also refuted by empirical evidence: in foraging societies over most of the world “Sedentary women simply do not exist…. it has been found that women are away from basecamp for equivalent amounts of time and walk equivalent distances, carrying infants and heavier burdens than do the men” (Ibid: 49). In spite of the evidence that has amassed and the feminist theoretical contributions (including the emergence of a ‘Woman the Gatherer’ model of human evolution), the old assumptions continue to have great influence both among scientists and within society: ‘when Man the Hunter becomes Man the Provider, it is clear that powerful cultural sex role expectations inform these reconstructions’ (Ibid: 61). There are other ways in which the ‘cult of domesticity’ has affected Western scientific perspectives on women’s contributions to subsistence and their relationships with plants. In 18th century Europe, an explosion of interest in plants took place among the general population as well as among the scientifically and commercially inclined due to colonial plant-hunting expeditions. In Victorian England as well as on the continent, due to the associations in myth and literature between women, flowers, and gardens, and to the long tradition of women’s medicinal herbalism and homegardening, botanical work was considered to be part of the domestic sphere and much in line with feminine attributes, so women were actively encouraged to pursue the study and cultivation of plants (Bennett 2000; Hyde 1998; Shteir 1996). But these ideologies had an impact far beyond the domestic sphere, since the ‘cult of domesticity’ that prevailed in Europe in the 18th century also influenced the establishment of botany as an Enlightenment science. According to historian Ann Shteir (1996), who traced the influence of the ‘cult of domesticity’ on the development of botany, between 1760 and 1820, the ideas of Carolus Linneaus, widely acclaimed as the ‘father of Botany’, gained great popularity in Europe and his system of plant classification became widely accepted, which classified plants according to sex and then according to the number of pistils or stamens. The system was based on parallels between plant and human sexuality and on concepts of masculine and feminine that were prevalent in his day. He used anthropomorphic terms to characterize the sexuality of the plant world – such as ‘brides and bridegrooms’ ‘eunuchs’ and ‘clandestine marriages’. According to Shteir, He assigned a higher ranking to the class, a unit based on stamens (the male part), and a subsidiary ranking to the order, based on the pistil (the female part). He also represented the male part in plant reproduction as active and the female part as passive . . . he naturalized sex and gender ideologies of his day (Ibid: 16). The gender and anthropomorphic biases evident in Linneaus’s work went relatively unchallenged for nearly 70 years, in part since it fit with social concepts of how nature and societies should be organized. But, by the 1820’s, some botanists began to turn to plant physiology as a new area of inquiry, and developing ‘natural system’ approaches to classification based upon a series of characteristics rather than simply plant reproduction. Linnean botany was increasingly seen as the ‘lower rung of the ladder of botanical knowledge, associated with children, beginners, and women’ (Ibid: 31). Shteir further relates that, at the same time that Victorian England was romanticizing nature, botany was becoming professionalized, as symbolized by the inaugural speech of John Lindley as the first Professor of Botany at London University. In his inaugural address, Lindley strongly distanced himself from Linnean botany and allied himself with the continental thinkers. He insisted that botany should be concerned with plant structure rather than identification. But this is not all he did: as well, he insisted that ‘it has been very much the fashion of late years, in this country, to undervalue the importance of this science, and to consider it an amusement for ladies rather than an occupation for the serious thoughts of man’ (cited in Shteir: 156-157). So, Shteir relates, ‘During 1830-60, botany was increasingly shaped as a science for men, and the “botanist” became a standardized male individual . . . [women’s] botany was in the breakfast room’ (Ibid: 166). Not only were women largely excluded from training as botanists for almost the next hundred years; they were also to be given decreasing attention over time as plant knowledge-holders and managers, and their contributions to subsistence were to be largely ignored. During this period, however, there have been other trends evident. Early ethnobotanical research (eg sponsored by the Bureau of American Ethnology), often explicitly recognized indigenous women’s vast plant knowledge and expertise. It appears that the emergence of economic botany during the 1940s and 1950s, which often de-emphasized the ethnographic content of people-plant research, did a great deal to reduce attention to women; this may have been reinforced by the emphasis on hallucinogenics as ethnobotany reemerged during the 1960s and 1970s, which due to its subject matter also emphasized the importance of the shaman, typically a male expert. Today, a principle reason that scientists still ignore gender differences in intra-community knowledge distribution is that they overlook women’s contributions to subsistence through plant management and use. As an example, Nancy Turner’s (2003) extensive research on North American aboriginal people’s knowledge and knowledge transmission provides a holistic view of the range of women’s technical environmental and cultural knowledge that is entailed in plant-based subsistence. However, Turner (Pers. comm. 2002) argues that, whereas women’s work and knowledge were highly valued and considered as equal to men’s within these societies, scientific researchers often make wrong assumptions about the importance of women’s activities for subsistence. A common perception is that their traditional economies were based almost solely on the exploitation of salmon and other fish, and the hunting of marine and terrestrial mammals and large birds. In much literature, plant foods are categorized as secondary products subsumed under a general category of ‘berries and roots’. She observed that this marginalization of the significance of botanical food products and the processing of foods in general has the effect of devaluing women’s work and roles in these societies, since women were the main decision-makers, harvesters, and knowledge-holders in many aspects of food use. Because of this, there is still very little research that thoroughly and explicitly documents the full breadth of indigenous or peasant women’s technical environmental knowledge, although there is more that documents this in specific domains. What a wide-ranging and as yet ongoing literature review reveals is that, in all but the most highly industrialized regions of the world, what is characterized as the ‘reproductive’ domestic sphere is in reality a tremendously productive, albeit largely invisible realm. It contributes the majority of subsistence resources in many rural areas, and involves a highly demanding and holistic level of technical environmental knowledge and skills related to plants that can require at least a third of a lifetime to accrue, as well as frequent innovation. This is best seen by reviewing briefly women’s use of plants cross-culturally. Women and the Plant World: What Do We Know? This is a question that I have been trying to answer for the past several years in the context of a research programmed that I developed that is establishing a worldwide knowledge base on people-plant relations, partial results of which have been published (Howard 2003; Howard-Borjas 2001). The reason that this is difficult to answer is that the information is very disperse and hard to locate, being contained mainly in ethnographic research done on indigenous societies, and has been carried out mainly by anthropologists and ethnobotanists who in most cases were not specifically setting out to research women’s relations with plants but who, in any case, did collect and present data disaggregated by sex. Nor do the difficulties end once the manuscripts are collected. There is no ethnobotanical database equivalent to those anthropological databases that are constructed to classify and compare mainly case study evidence either across cultures, time periods or agroecosystems. Ethnobotanical studies of individual cultures and agroecosystems are generally focused on few resources (Ellen cite), and thus do not provide sufficient coverage to examine gender differences. Nevertheless, I have collected more than 1200 manuscripts, among with are more than 100 PhD dissertations, and am still in the process of reviewing these to understand women and gender in people-plant relations. Despite nuances, discrepancies, and incomplete information, I have been led to a conclusion that surprised even me: cross-culturally, women predominate in plant use and management as housewives, plant gatherers, homegardeners, herbalists, seed custodians and plant breeders, and gender relations universally have profound effects on how people and plants interrelate. Woman the housewife Only rarely is it considered that women, in their roles as housewives performing domestic tasks, sustain an intimate relationship with plants that has fundamental implications for the use and conservation of plant genetic diversity. ‘Domestic’ refers to unpaid, home-based activities that ensure the maintenance and functioning of people within households, contributing thereby to the reproduction of both generations of human beings as well as of entire societies (Madge 1994:280). These tasks include, among many others, food preparation, preservation, storage, and processing, and they are most often assigned to women and girls. What is associated with the kitchen (food processing, preparation and preservation) and with the pantry (storage) is in the majority of cultures conceived to be the domain of women. It is generally acknowledged that the majority of plant species used by humans are for medicines and food (Johns 1990), but an as-yet unrecognized but logical corollary of this that I put forth is that the kitchen is quite possibly the single most important site of plant use and conservation. Culinary traditions are a highly important aspect of cultural identity, and foods are clearly valued not only for their nutritional content, but also for their emotional, ritualistic, spiritual, and medicinal values. Food is, in most cultures, also a fundamental constituent of exchange and hospitality, which are in turn basic organizing principles of many traditional societies (see e.g. Brenton 1994; Cole 1983; Johnsson 1986; Meigs 1988; Pieroni 1999; Hamilton 2003; Trankell 1995; Weismantel 1991). While the idea of what constitutes an adequate meal or dish may be influenced by men (see, e.g., Counihan 1998; McIntosh and Zey 1998; Weismantel 1988), women are generally considered as the ‘gatekeepers’ of food flows in and out of the domestic sphere. Culinary traditions are perpetuated by the careful transmission of knowledge and skills, particularly from mother to daughter. Most importantly, culinary traditions and preferences, as well as the post-harvest processes that are required in order to provide edible and culturally acceptable food, have a marked influence on knowledge, selection, use, and conservation of plant diversity (see e.g. Barnett 1969; Defoer et al. 1997; Ferguson and Mkandawire 1993; Lope-Alzina 2003; Shellie-Dessert and Hosfield 1990). The last statement is illustrated well by the case of the Andes, the cradle of the world’s potato diversity. In an area where environment and agronomic conditions alone cannot explain the great diversity of potato and maize varieties produced, Zimmerer showed that specific varieties are cultivated in order to meet precise culinary requirements: in potato production, "..groups of species correspond to different uses, such as freeze-drying, soup-making, and boiling. In maize, on the other hand, different preparations rely on groups of cultivars. Agriculturalists utilize culinary distinctions as the basis for planting separate fields in different ecological habitats" (1991:301-302). Maize is secondary in the diet to potatoes, but this does "not preclude extensive genetic diversity, complicated folk taxonomic classification, and elaborate ritual significance" (Ibid: 322). Besides parching and boiling maize, maize beer, hominy, crushed maize, popcorn, mush, corn-on-the-cob, soup thickener, pudding and tamales are made. "Primarily women brew maize beer, a laborious, multi-step process lasting 2-3 days which includes several alternative techniques... " (Ibid: 324). Cabrera (1997) found in Tomoaya, in the State of Mexico, similar motivations for maize varietal maintenance where relatively isolated and diverse ethnic communities have conserved varieties that meet their specific culinary traditions and religious purposes. Among the Tukanoan peoples of the Colombian Amazon, cassava (Manihot esculenta Crantz) is the principle staple crop and only women produce it. Wilson (1997) researched cassava crop performance and women’s varietal preferences in order to learn why women plant more ‘bitter’ varieties that require a labor-intensive process to detoxify the tuber for human consumption, instead of the ‘sweet’ varieties that don’t require detoxification (see also Spijkers and Box 1971; Dole 1978; Chiwona-Karltun 2001). He concluded that: It was expected that interviews would demonstrate that yield and/or resistance to environmental stressors is the most important consideration a Tukanoan woman makes when selecting cultivars for her garden. The results, however, indicate that while yield and damage suffered by the plant may be factors which influence cultivar selection, it is the foods which can be made from each cultivar which is the most important consideration (Ibid: 152). Across the world, in Southwest China, hybrid varieties of maize have been promoted by the government for at least two decades, and have been widely adopted by farmers. However, in-depth research showed that farm women, who represent about 90% of the farmers in the region, continue to maintain local varieties of maize for culinary purposes: “Local sticky is a sticky waxy variety…taken as a special quality food for festivals. Almost every household maintains a small plot of Local Sticky 1 in its vegetable garden or back yard, despite its low yield...Local White is maintained by some farmers because of its sweet stalks, preferred by kids as a kind of sugarcane” (Song 1998:143). In central Malawi, the women in Dedza Hills gave two reasons for keeping large numbers of bean varieties. First, they reported that they needed many varieties because, if some failed during the growing season, others would survive to feed their families. Second, they held that each variety had its own characteristics and fulfilled different household needs. For example, some varieties produced superior pods or leaves, which are important items in the diet immediately preceding harvest when food supplies are often low. Other varieties are grown because they are fast-cooking, store well, or are easy to sell on the local market. Their reasons for selecting the four bean varieties that they plant in greatest quantity, in descending order of importance, are yield, taste, cooking quality, marketability, date of maturity, health-related issues, insect and disease resistance, and ability to withstand environmental stresses. Altogether, three-quarters of the reasons for selecting varieties had to do with factors other than environment and agronomy. “Chief among these were taste considerations, cooking quality, and health concerns, which accounted for nearly half the responses" (Ferguson et al. 1990: 276). Several other studies similarly indicate the importance, if not predominance, of culinary qualities in the maintenance of a diverse variety of cultivars (see, e.g., Cartledge, 1995; Defoer, et.al. 1996; Ashby & Herpen 1991). The interrelationships between the maintenance of cultural identity, culinary traditions, and biological diversity are clearly evident in the exchange and movement of plant genetic resources through migration (see, e.g., Niñez 1987; Esquivel & Hammer, 1992). The significance of culinary traditions and women’s domestic work for the preservation of traditional varieties upon migration is illustrated by the homegardens of households that migrated from a subsistence agricultural economy in the Yucatan Peninsula to a wage-labour, cash economy in Quintana Roo, Mexico. Greenberg’s (2003) in-depth research revealed that immigrant home lots are sites of in situ conservation not only of traditional crops, but also of elements of traditional Yucatec cuisine which helps to preserve the cultural identity of immigrants in their new environment. Some 140 plant species were found in 33 gardens, most of which were for culinary use. In the traditional gender division of labor in the Yucatan, men’s work is primarily in the field and the forest, and women's domain has historically been the homegarden and the kitchen. As immigrants, women create and maintain homegardens even when the household economy is based principally on cash and women engage in wage labour. There are three major benefits for immigrants who adhere to traditional practices after migrating: (1) traditional gardening and cooking make the transition less stressful; (2) some measure of control can be exerted over the new cultural elements that immigrants adopt; and (3) these practices made important contributions to the household economy in the form of food and income. In the Yucatan itself, crop biodiversity is eroding as traditional agriculture declines, which is threatening both plant genetic material and cultural continuity of the Yucatec Maya. Immigrant gardens are conservation sites for traditional crop species and varieties outside of the region and provide the only source of traditional foods in the new settlement area. Across much of the African continent, culinary traditions include the use of sauces, relishes, and soups that make the bland carbohydrate staples consumed in the region more palatable and nutritious. These preparations involve a great diversity of plant species, both wild and cultivated, including many ‘indigenous’ vegetables that must be produced or gathered on a very substantial scale. It is usually women’s task not only to prepare these accompaniments, but as well to produce or procure the plants that go into them (see e.g. Chastenet 1991; Chweya and Eyzaguirre 1999; Fleuret 1979; Malaza 2003; Nordeide et al. 1996; Ogle and Grivetti 1985; Ogoye-Ndegwa and Aagaard-Hansen 2003). For example, Akoroda (1990) reported that Telfairia occidentalis is a fluted pumpkin consumed as a ‘relish’ or soup ingredient accompanying yams, cassava, or cocoyam among the Igbos in Nigeria. Used by 30-35 million people, it is a leaf, shoots, and seed vegetable that women are culturally obliged to produce or procure for their households. The nutritional content of these staple accompaniments is essential and often superior to exotic fruit and vegetables, and they are also often less expensive. However, "Fruits and vegetables are not accorded the importance they deserve in the diet of West African people...and are hardly considered food. At best, they are prepared and accepted as a relish or a flavouring agent...it is the basic staple which is considered 'food'" (Akoroda, Ibid: 30, citing Sai, 1965). Local and foreign ethnobotanical literature rarely contains information on this type of food plant since the number of species is large, many are very localized since they are grown in small patches in homegardens, boundary lands, or between crops, or gathered wild or semi-wild, and they are mainly managed by women (Ibid.:31; Chweya and Eyzaguirre 1999). The significance of these ‘accompaniments’ for the preservation of folk varieties is only now being recognized. For example, with respect to multi-purpose species such as cowpea, cocoyams, sweet potatoes, cassava, and pumpkins, selection also occurs on the basis of the leaf as the primary character, for its use in sauces and relishes (Chweya and Eyzaguirre 1999). Even in Western Europe, substantial knowledge of wild food plants is held and passed down from mother to daughter in order to maintain culinary traditions, and knowledge of wild plants is maintained by women throughout generations in order to maintain culinary traditions. In Garfagnana, a village in Tuscany in central Italy, 700 people responded to questionnaires on wild plant gathering. “Most of the information about fungi and wild fruit gathering were obtained from men, while the gathering of green [plants] seems to be the domain of the women" (Pieroni 1999:329). “Traditional cooking with wild gathered plants in Garfagnana is very complex and includes dishes which are part of the social heritage" (Ibid: 330). Men have long sold mushrooms outside of the region. "The most common culinary preparation of vegetables (erbi) is in the form of a variety of vegetable soups…Never fewer than twenty, but often more than forty species are used. The number depend [sic] on the season..and the availability of plants" (Ibid.). “In these regions, even the term "kitchen" (cucina) is sometimes used to denote such vegetable soups or the essential plants used in them.." (Ibid: 339). The survival of this knowledge and use of plants is attributed to "the role that traditional cooking maintained in many families in the valley… traditional society is changing rapidly and many more women work outside the home, but they still preserve the female heritage of food traditions, especially the gathering of wild species" (Ibid: 340). Which plants are selected, managed, produced, and conserved for food depends on a wide range of criteria related to palatability, culinary qualities, and beliefs about health and nutrition. But domestic work entails more than cooking: it also entails processing, preserving, and storing plants, and plant varietal selection criteria are related to processing characteristics, storability, preservation methods, the technology available for these, and on local knowledge, labor, and fuel availability. Food processing and cooking are even more essential in most traditional societies because they make plants edible through detoxification, which requires in-depth knowledge of plant characteristics. In the Upper Arun Valley of Eastern Nepal, the jangal are common property areas including forest land, fallow swidden fields and grass and pasture surrounding villages up to the timberline. Although there are no distinct gender differences in the collection of jangal foodstuffs, in the household women assume the major responsibility for their management. "The extensive ethnobotanical knowledge possessed by Rai and Sherpa women is apparent in the elaborate processing techniques used to remove the toxic compounds…" (Daniggelis, 1997:238). Food processing, preparation and storage activities are inter-related in terms of labor and time (often representing a series of steps carried out sequentially), and in terms of techniques (the way plants are processed influences the way they can be stored and consumed). They are also conditioned by other factors such as humidity and incidence of pests and diseases. These tasks are often indivisible: the same person selects, separates, processes, and stores plant products simultaneously for the next crop, for home consumption, and for sale. The knowledge and skills required to develop, maintain, and innovate in this post-harvest food chain are complex, dynamic, and vital, and in most societies pertain to women (see e.g. Arenas 1999; Battock and Azam-Ali 1998; Bjoernsen Gurung 2002; Dole 1978; Longe 1988; Lope-Alzina 2003; Muhammad 2002; Trankell 1995; Yen 1975). In a case study in the Western Division of the Gambia, Madge (1994) researched the preparation, preservation and storage of wild food plants. "The recipes used for cooking collected foodstuffs are mostly to make sauces…cooked especially during labour bottlenecks in the farming calendar..since wild leaf sauces require less cooking time and, therefore, need less fuelwood than sauce recipes which use cultivated species" (Ibid.:286). Young women recall more recipes than older women since young women are more often responsible for cooking and processing, and poor women were also more knowledgeable than wealthier women because they rely more on collected food, and wealthier women tend to purchase this food or disparage some types of it. Women who lived through major droughts also had greater knowledge as did women who were born in comparison with women who had in-migrated. Local knowledge in food processing and storage often correlates with scientific knowledge. In fermentation, women stress the importance of heat which also reflects the degree of bacterial and enzymatic activity. Water in which fish has been cooked is saved and used to make a sauce because, women report, '..the water contains the goodness’ – that is, water-soluble B complex vitamins. Also during cooking, "..acid fruits reduce bacterial growth and shortens cooking time; this probably explains why velvet tamarind (bujaala), Dialium guineense is sometimes added to cooking water…" (Ibid.). Each surveyed household in the village stored between three and five wild plant and animal species, some in large quantities, even during periods of abundance. Sun drying is the most commonly used food preservation method: "Strict weight and colour controls are used to ensure throrough drying and successful preservation" (Ibid: 290). Smoking and salting are two other preservation methods used, and herbal preservatives sometimes added to repel insects. Some fruits are preserved in liquid in airtight containers, and root crops and oil are buried. "The storage location of particular foodstuffs is specific. African locust beans (bukombong), Parkia biglobosa, for example, are stored in the kitchen on the bench above a fire since constant smoking ensures that the food is protected from insect pest attack.. Bush teas are not stored in the house rice store because they attract termites which will eat the rice" (Ibid.:288). The integrity of these processes is essential to health and family well-being: not only do culinary traditions and knowledge directly affect household nutrition, but storage and preservation also considerably lengthen the ‘shelf-life’ of both gathered and cultivated food and are therefore essential to ensuring food security (e.g. Muhammad 2002; Battock and Azam-Ali 1998; Madge 1994). Historical research carried out on the indigenous women of the Northwest Pacific Coast of the United States showed also that most plant resources were seasoned and processed by methods which required special techniques as well as storage (Norton, 1985). Plant foods provided requisite nutrition for the Coastal people and, when harvested and stored in quantity, were dependable, all season staples. "The edible portions of plants have a truly limited season of harvest and without processing and storage vegetable foods would have been unavailable during a large part of the year…The primary relevance of plant foods (whether staple or supplemental) to Native peoples lay not in their environmental distribution patterns but in their ability to meet quantitative and qualitative nutritional requirements year-round” (Ibid.:112). While much ethnobotanical research investigates the ways in which plants are consumed as foodstuffs, there is little research that investigates the specialized ethnobotanical knowledge and related skills underpinning domestic practices in food storage, preservation, and preparation techniques. These practices require in-depth knowledge of plant characteristics and of environmental factors affecting processed plants. It is still infrequent that the specific ways in which plants are transformed and consumed are linked to the associated practices and knowledge, or to the domestic sphere, or to women. The preservation of women’s knowledge and skills in post-harvest tasks is just as vital to the conservation of PGR as the preservation of crops in situ for, without this, these plants are no longer useful to people. Woman the Gatherer It has become widely recognized that, in most foraging societies (those dependent mainly on hunting, fishing, and gathering), both historically and today, the bulk of foodstuffs is provided by gathering which is carried out primarily by women. According to one statistical analysis of 135 different societies around the world with various subsistence bases (e.g., agriculture, animal production, hunting, fishing, and gathering), women provide 79% percent of total vegetal food collected. Estimates from other databases are close to this score (Barry and Schlegel 1982). The plants or plant parts gathered by men and women often reflect the division of labor in other spheres. Women gather plants that they are ‘responsible’ for, such as those needed to make sauces and relishes or those that serve as inputs for their own production such as basket and cloth making. Men and women have different needs and responsibilities for gathered plants, and different knowledge and preferences with respect to them. Flickinger (1997) researched gender differences in local knowledge and use of forest plants in Utter Pradesh, India. In general, women have greater knowledge of the usefulness of plants than men and perceive their usefulness differently. Men’s priority uses of plants are for agriculture (fodder and mulch) and women uses are more related to the household - medicines, tonics, cleansers, fiber, food and tools. Further, much research shows that men often collect plants from ‘men’s spaces’ and women collect from ‘women’s spaces’. It may be that only men are allowed to enter ‘sacred groves’ or highland forests, whereas ‘women’s spaces’ are ‘disturbed’ environments such as field margins, irrigation canals, roadsides, and fallows. But in many societies, contemporary women, like their historic counterparts, also venture far from home to gather plants in relatively ‘wild’ places such as forests and savannah, and it can be that men are not be permitted to gather in ‘women’s wild spaces’. The idea that plants growing in natural environments are ‘wild’ is also often mistaken: many are not strictly either ‘gathered’ or ‘wild’ but are selectively managed and harvested. An example of how ‘wild plants’ are managed by women in their natural environments is provided by Native American basket producers in California. Basket making historically was based on the collection of white root where 250-750 plants were needed to make a single basket (Stevens 1999). While harvesting, women left the plants and removed the weeds, thereby cultivating the bed and enhancing the habitat for the production of new plants (Dick Bissonnette 2003). Women also cultivated the roots with digging sticks, encouraging the growth of long straight rhizomes. Upon harvesting, women left sufficient rhizomes in place to keep the patch viable for future use. This system was sustainable for hundreds of years. Women’s ethnobotanical knowledge of ‘wild plant management’ was essential for the survival of these tribes for at least several centuries. Nowadays, since the available gathering sites have disappeared, modern basket weavers are growing their own materials in back-yard gardens (Stevens 1999). Across the globe today, gathering provides a substantial contribution to rural livelihoods, particularly in areas where there is abundant genetic diversity and where populations are resource-poor and food supplies are short seasonally or during crises. However, foraging resources are declining rapidly. Population growth, market expansion, and environmental degradation are increasing the time and labor invested in foraging activities, particularly by women. The reduction of foraged foods in the diet is leading to poorer nutrition and is reducing emergency food supplies, thus increasing reliance on food purchases and decreasing knowledge and use of local plant biodiversity. Woman the Gardener Homegardens are the oldest and most widely used cultivation systems on the planet. They tend to have greater species diversity than cultivated fields. Tropical gardens are the most complex agroforestry systems known. Most definitions of homegardens refer to their location near the home, their function as a secondary source of food and income for households, the predominance of family labor, and their multi-functionality as aesthetic, social and recreational spaces, as well as for provisioning of medicines, herbs and spices, fodder, building materials, and fuel. While the gender division of labor in homegardening varies across regions and cultures, the close link between gardens and the domestic sphere everywhere ensures that women tend gardens. It is clearly women who manage homegardens across the developed world as well as in tropical Africa and Latin America, and they make strong contributions to homegardening in Asia, so that, globally, women hold the majority of knowledge, skills and responsibilities in homegardening. Like much of women’s work, homegardening is relatively ‘invisible’ and is often disparaged as ‘minor’ or ‘supplemental’ to agricultural production. The fact that the majority of garden produce does not enter into the market, that many of the plants cultivated are traditional varieties known mainly to local people, and that the land areas involved are generally small and near the home, all contribute to the continuing invisibility and devaluation of homegardens, which in turn contributes to the invisibility and devaluation of women’s contributions to plant biodiversity conservation. The importance of homegardens for plant conservation has been under-estimated. For example, when Alexiades investigated medicinal plant use among forest dwellers in the Venezuelan Amazon, he found that most medicinal plants are collected from fallow land and homegardens rather than from forests, which most researchers assume supply the largest proportion of medicinals. Gardens ‘. . .represent a ‘genetic backstop’, preserving species and varieties which are not economical in field production and are planted small-scale. . ‘ (Niñez 1987). In swidden cultivation systems, useful varieties that would be lost due to clearing and burning are transplanted to homegardens where they may thrive (Okigbo 1990). One of the most important reasons to conserve plants in situ rather than in gene banks is to permit their continued evolution, and it is in homegardens where much of this evolution takes place. Many authors have noted that farmers first experiment with new crop varieties in homegardens to determine their productivity before they are planted in fields. The migration of the potato from South America to other parts of the globe occurred through homegardens, and the diffusion of maize began when Incan women settled newly conquered territories and brought maize seed with them to plant in their new homes (Niñez 1987). Among the Maya in highland Guatemala, ‘Women educate children through the chores of the garden. They teach how to use farm tools, what plants need to thrive, and how to manage crops, especially through weeding and harvesting’ (Keys 1999: 89). Homegardens are a vital resource particularly for poor women since they permit them to provide additional food and income for their families. Many studies show that a woman’s garden provides basic nutrition in periods of food scarcity and food supplies year-round. Homegarden food production is not necessarily supplemental and the amount of labor used may be large in certain parts of the year. Over much of sub-Saharan Africa and Latin America, women are the predominant managers of urban homegardens that provide a substantial source of the total livelihood for low-income households through sales of produce and supplemental food supplies. Niñez showed that this holds even in developed countries such as the United States, where a community garden can produce an US$5000 of output with US$500 in input. During the Great Depression and World War II, over 40% of all fresh produce in the US came from homegardens and they were even more important in Europe. In former Soviet countries today, homegardens provide a very substantial proportion of total household food supply and studies show that these are managed predominantly by women. Woman the Herbalist The World Health Organization estimates that 80% of the world's population use plant medicines for their primary health care needs. Between 25 and 40% of all modern pharmaceuticals are derived from plants. Research on folk medicine and medicinal plants is booming, but this has tended to focus on the knowledge of folk medicinal specialists: shamans, midwives and herbalists (McClain 1989). Shamans and ‘medicine men’ usually have great power and status and the majority of these specialists are male. However, female priestesses are prevalent particularly in Africa and Asia. Herbalists, on the other hand, are specialists in treating illnesses through the use of plants and are frequently women; midwives are also herbal specialists and are usually women, although men can also be midwives. Women’s ethnobotanical knowledge and medicinal roles are often unexplored by ethnobotanists who tend to make a beeline for the ‘shaman’ or ‘medicine men’. The focus of ethnobotanists on healers and medical specialists as ‘key’ informants also means that they have often ignored lay persons. Awareness is growing that the ‘common’ knowledge of lay women is actually that which predominates in traditional health care systems. Most illnesses are not life threatening and expert medical advice is only sought when home remedies do not work (McClain, 1989: 21; Goody 1987). The medical role and knowledge of women is essential to the health of household members, and in several societies lay women have a greater role in the knowledge and use of medicinal plant resources than their male partners, as is demonstrated by many studies. Several researchers interpret the healing activities of women as an extension of childcare duties and their responsibility for family health and caring for the ill. Knowledge of herbal remedies is often passed along the female line as daughters take care of ill siblings. Ensuring local women’s continued access to and control over these plants is crucial both for rural health care and for genetic conservation. Men and women not only have different knowledge of medicinal plants: their knowledge is also structured in a different way, which is related not only to the division of labor, but as well to social power. Ethnobotanical research has often introduced a double bias: on the one hand it, has relied on a limited sample of predominantly male informants and, on the other, it has structurally neglected female healers and the realm of domestic curing and herbalism. Woman the Breeder and Seed Custodian Traditionally, plant breeders and seed custodians are small farmers, and often if not predominantly women. Women in sub-Saharan Africa as well as in indigenous societies in Latin America and the Pacific are usually directly responsible for crop production. As crop producers, they consider all of those selection factors that are critical to farmers who produce in marginal environments and manage many varieties for many purposes. For example, in Rwanda, women produce more than 600 varieties of beans (Sperling and Berkowitz 1994); while in Peru, in one small village, Aguaruna women plant more than 60 varieties of manioc (Boster 1985b). While both men and women are involved in crop selection and have highly specific knowledge and use a variety of criteria, these differ substantially between them, and women’s criteria and knowledge are more often overlooked by formal plant breeders and conservationists. Women often have a broader set of varietal selection criteria in comparison with men since they use plant materials in more diverse ways: for example, rice not only provides food, but also straw for thatching, mat-making and fodder, husks for fuel, and leaves for relishes. Women’s responsibilities for post-harvest processing and family food supplies means that women try to ensure that varieties are in line with culinary traditions, are palatable and nutritious, and meet processing and storage requirements. Several studies show that, even when women do not produce crops, men take their wives’ preferences and criteria into account when selecting varieties, but researchers mostly neglect this since they are not directly related to yield and pest and disease resistance. Very frequently, women are responsible for tasks related to seed management including seed selection, storage, preservation and exchange. Informal seed exchange systems are often female domains, and include mechanisms such as the bride price, gift giving, and kinship obligations, as well as market and barter transactions. Women’s predominance in seed management activities is often explained by the close relation that this has with post-harvest and domestic work. Others suggest a more cosmological explanation that may be found to hold across many traditional societies. In the Peruvian Andes, Zimmerer (1991) relates that women almost exclusively manage potato and maize seed. Men are forbidden to handle seed or enter seed storage areas. The explanation for women’s control of seed is to be found in Andean cosmology. In Quechua, plants that are useful to humans are all worshipped under the name of mother: Mama sara (maize), Mama acxo (potato), Mama oca (Mama cocoa). Andean thinking contains a dual concept of reality based on masculine and feminine principles. ‘Seed’ also refers to semen, providing a metaphor between the ‘seed’ that the male deposits in the womb and that which is sewn in the field, collected, and later deposited in the home (Tapia and de la Torre 1993). Throughout human history and across most societies, women and fertility, and seed and fertility, are equated. However, cultural beliefs also sometimes exclude women from certain aspects of seed management. Vedavalli and Kumar (1998) report that, among the Kurichiyas in Kerala, India, women’s involvement in rice seed selection and storage is considered to be polluting, a phenomena seen in other societies in relationship particularly to taboos around menstruation and childbirth that also affect plant management. What is Gendered Plant Knowledge, and Why is Plant Knowledge Gendered? In order to more fully explore the issues raised at the beginning of this paper regarding intracultural knowledge variation and the significance of gender therein, it is important to discuss the gendered nature of knowledge and its underpinnings. The simplest definition of gendered knowledge is that which is held either by men or by women, but not by both. Using this definition, appearances would tell us that gendered knowledge exists because men and women do different things: gendered knowledge is therefore a direct reflection of the gender division of labor, which is an assumption that, for example, Boster appears to support. This emphasizes experience and practice as sources of knowledge, as well as the transmission of practical knowledge and skills that are required to carry out tasks. According to this line of reasoning, plant knowledge is gendered to the extent that a gender division of labor exists with respect to the use, management, and conservation of plants. But is the gender division of labor sufficient to explain gendered knowledge? In the first place, gender divisions of labor are variable over time, as well as within and between cultures and economies. Since they are variable, they must themselves be subject to explanation. In the second place, not all knowledge is experiential and practical. In all societies, the gender division of labor is related to religious and other values and belief systems, and to cultural conceptual models of masculinity and femininity and norms about behavior that is appropriate for each sex. These conceptual models and social norms prescribe the type of activities and responsibilities that are appropriate for men and women, as well as with whom men and women of different social positions can appropriately interact. Very importantly for plant and other environmental knowledge, these beliefs and norms extend to men’s and women’s relations to different physical spaces and environments. As demonstrated by ethnographic research on women foragers who travel very long distances while carrying infants (Fedigan 1986), there is no physical impediment to women’s mobility, but there are many cultural restrictions. For example, Hays (1974) reported that, among the Mauna in the Eastern Highlands of Papua New Guinea, wild vegetable collection (mushrooms, vegetables, fruits, and ferns) is carried out by women unless it involves tree-climbing. However, ‘the forest is primarily man's domain. Myths tell of the tragedies that occurred when women went high into the bush. Women do go in the forest to collect . . . However, there is an upper limit of approximately 7,500 feet beyond which they are not supposed to go’ (1974: 59). Hays noted, however, that surprisingly for him, he discovered that women were able to accurately identify species of plants that were available only above these altitudes and that were exclusively used by men. Apparently, women did venture into these areas, but they did so ‘in secret’ (Pers. Com. 1997). Descola (1994) reported that women’s gardens are exclusively women’s spaces, where men rarely venture to go. If men and women have access to different physical spaces and environments, it stands to reason that their environmental knowledge will also differ. Further, cultural norms and different responsibilities means that men and women relate differently to different groups of people (e.g., ‘mothers-in-law’, ‘male outsiders’), that is, they have different social networks. Therefore, their ‘knowledge networks’ also differ, which affects the type and quality of knowledge and knowledge transmission. Knowledge transmission is increasingly the subject of research because the mechanisms are poorly understood, and plant and environment-related knowledge transmission is eroding rapidly within many indigenous and peasant societies. Turner (2003), for example, shows how crucial women are as educators of children of both sexes for the perpetuation of the understanding of food production systems. Her extensive research shows that knowledge transmission is not confined to instruction, but as well occurs through participation, observation, and conversation at times of harvesting and processing of food and related resources. Nor is this transmission confined to practical knowledge: the values and philosophies held and taught by women are an integral component of food production, and are as well key to cultural integrity and traditional ecological knowledge systems. These are transmitted through stories, songs, games and many other cultural media, very often through the female line. Another example is provided by the Kel Ewey Tuareg people in Niger, West Africa, where women traditionally are herders and gatherers and enjoy significant prestige. Rasmussen shows that the inheritance of herbal healing knowledge occurs along the female line and is traced back to Tagurmat, an ancestress who passed down the knowledge of trees and healing to women. The knowledge is secret and ‘. . it is transmitted to, belongs to, and is practiced and managed by women, like property . .’ (1998: 148). For other cases dealing with knowledge transmission among women see e.g. Dick Bissonnette 2003, Ertuğ 2003, Hoffmann 2003, Keys 1999, Pieroni 2003. Women and men also have different access to formal and exogenous knowledge. In many societies, women have less access since they receive less formal education, have less contact with extensionists and other government agents, are less able to leave their communities, and are less able to speak languages or dialects other than their own (see, e.g. FAO 1993). FAO (1983) conducted a survey of extension services in 115 countries that showed that women only received between two and ten percent of all extension contacts and a mere five percent of extension resources worldwide. However, the opposite can also be the case: for example, in many hunting-gathering societies, women are often reported to venture further than men on food gathering forays (see e.g. Anderson 1983; Brightman 1996; Carlstein 1982; Fedigan 1986); the same has been reported for women herbalists who travel extensively to procure medicinal plants (Rasmussen 1998). Marketing and barter of seeds and plant materials may be women’s prerogative, and women may also maintain more contact with kin in their villages of origin, all of which means they may often in fact have more contact with outsiders’ knowledge and genetic resources than do men (see e.g. Longley 2000; Coughenour and Nazhat 1986). The gendered nature of knowledge, then, is not simply a function the gender division of labor, but rather is embedded in cosmologies, beliefs, and norms about appropriate behaviors. Descola (1994) captured the intimate relationship between gender divisions of labor, knowledge, cosmology, and subsistence in his work on the Achuar, a sub-group of the Jívaro who live in Amazonian Ecuador. Manioc is accorded a very special status in Achuar society, and women are the exclusive cultivators. The tutelary female spirit of women’s gardens, Nunkui, resides in the topsoil, and women gardeners must have ‘direct, harmonious, and constant commerce with Nunkui’ in order to have successful gardens (Ibid: 192). Nunkui is the creator and mother of all cultivated plants in Achuar gardens. Her deeds are recorded in a myth that has a similar structure among all Jívaro peoples. This myth holds that women were allowed to harvest manioc from Little Grandmother’s (Nunkui-Uyush) garden but, one day a foolish woman ridiculed her and Nunkui-Uyush prohibited her from harvesting manioc. The foolish woman then took care of Little Grandmother’s infant and, when the child started magically pronouncing the names of the cultivated plants, they appeared in the garden, and therefore are the offspring of Nunkui. However, the child was mistreated and all of the cultivated plants shrank until they were tiny. It was only through Nunkui’s compassion that women were given seeds and cuttings to plant new gardens. It is therefore imperative that a woman secure Nunkui’s constant presence in the garden and that she take every precaution not to offend her . . . a garden’s life span and productivity depend as much on the magical skills of the woman who works it as on local ecological constraints. These skills are designated by the term anentin, which, applied to an individual, denotes at once the scope of magical knowledge, the capacity to manipulate the symbolic fields specific to his or her sex, and the particularly fruitful relations entertained with the guardian spirits that govern the spheres of activity in which the individual engages . . . To be anentin one needs to know a great number of anent, magical songs, since it is basically by means of these incantations that a woman can hope to communicate with Nunkui and with the plants in her garden (Ibid:198). Besides the anent, a woman gardener has to use gardening charms (nantar) in order to be successful. To be effective, the charms have to be activated by the proper anent: otherwise they might become uncontrollable and dangerous. There are no nantar for the species that men cultivate, and the use of these charms is the exclusive privilege of women. The power of nantar resulting in an especially fine garden is, however, proportional to their potential for harming, threatening both the offspring of the woman gardener or any other being entering her garden. ‘Ownership of nantar is exclusive and a closely guarded secret . . . Like gardening anent, nantar are inherited through the female line and are probably the most precious possession a mother can transmit to her daughter’ (Ibid: 206-207), together with the knowledge of how to control the nantar’s magical powers: ‘nantar are a device that permits every woman not only to maintain the individualized autonomy of her symbolic practice, but also to control concretely the very access to the exclusive domain where this symbolic practice is exercised and reproduced’ (Ibid). This clearly illustrates that knowledge is also not equally distributed within communities because often certain types of knowledge are purposefully not shared, kept secret, or shared only with very specific groups, often of the same sex (see, e.g. Abbink 1993; Alexiades 1999; Browner and Perdue 1988). Just as in post-industrial societies, knowledge in indigenous and peasant societies is used to confer status, manipulate social relations, gain material advantage, and maintain control over certain aspects of one’s life. As is discussed later in this paper, it is also clear that knowledge is an integral part of power relations. Gender relations are also power relations, and the distribution of ethnobotanical knowledge within societies cannot be understood without reference to belief systems that legitimize and mediate relations of power between men and women, as Shteir’s analysis of the history of Western botany so clearly reveals. These power relations and the ways in which they are manifest may be relatively transparent and clearly demarcated by men’s exclusive occupation of prestigious social positions. These relations may also be much more subtle, embedded in language and in the control over the production and exchange of goods. Kothari critically examines the treatment of gender issues in medical ethnobotany, and particularly emphasizes the significance of gender power relations for understanding ethnomedical systems and the distribution of medicinal knowledge between the sexes. He argues that ‘ethnobotanists . . . tend to make a beeline for the “shaman” or “magico-expert” healer, an office that in many cultures is reserved exclusively for men.’ ‘While women may hold some if not most medical ethnobotanical knowledge, by virtue of their sex they may be denied the power and status of the “expert”’ (2003: 151). This suggests that ‘gender must be explored as an integral aspect of the local power or “prestige structure” in which knowledge is held, generated, and expressed’ (Ibid). In contrast, ethnobotanists have paid an inordinate amount of attention to women's plant knowledge with respect to reproductive health, which they assume is exclusively the domain of women, while overlooking the breadth of their curative knowledge. Redressing this inattention to women’s ethnomedicinal knowledge and the knowledge-power interface, he argues, is a necessary point of departure if the relevance of gender issues in ethnobotanical knowledge is to be grasped. He discusses three theoretical models that have the potential to spark a deeper conceptualization of gender power relations in ethnobotanical research. In addition, language has been determined to be a principle vehicle for the conveyance of normative rules regarding gender relations within cultures. Nevertheless, it appears that little ethnobotanical research; particularly that which deals with folk taxonomic systems, addresses the question of how the gendered nature of language may affect the ways in which people name and classify plants, and how this in turn may reflect gender power relations. Sillitoe (2003), in his pioneering work on the Wola in the highlands of Papua New Guinea, was perhaps the first ethnobotanist to look at this in depth when he explored the Wola ‘gender classification of crops.’ In Wola society, crops are divided into male and female categories that tend to have a correspondence with the gender division of labor in plant production, which in turn is related to anthropomorphic concepts of male and female sexuality. In practice, crops are ‘male only’, ‘predominately male’, ‘both sexes’, ‘predominantly female’, or ‘female only’, and sanctions are in place to ensure that ‘male only’ and ‘female only’ crops are not produced by the inappropriate sex. When discussing crops, ‘the Wola distinguish between them by using two forms of the verb “to be”. There is a connection between this and the sexual categorization of these plants’ (Ibid: 169). One verb is used for things that are horizontal to the ground and the other is used for things in an erect state. ‘Things that stand erect are strong and masculine for them, whereas recumbent things are weaker and female’ (Ibid). Women’s crops are generally associated with the ‘horizontal’ ‘female’ plants, whereas male’s crops are associated with the ‘erect’ ‘male’ plants. Those that do not fit within this system (e.g., crops designated with a male verb but cultivated by women) are generally those that have been recently introduced to the region. The explanation for this system, according to Sillitoe, is ‘the overwhelming importance of female crops in the Wola diet . . . Without women and their crops the Wola could not exist, whereas they could if men's crops disappeared’ (Ibid: 178). Women produce, but men exchange wealth. ‘The danger is that men, whose reputations depend on handling wealth, may be tempted to produce wealth . . . such behavior would be antithetical to the ceremonial exchange system, which requires them to obtain valuables through transaction, not production, and then to give the items received away again’ (Ibid). Men are therefore prohibited from producing ‘women’s’ crops, and this prohibition is embedded in Wola language and folk taxonomy. Gender Bias and Ethnobotany: Evidence of a Distorted Science I and others have argued that a serious shortcoming of much ethnobotanical research is that it often takes the plant knowledge of a few people, particularly of men, to be representative of the knowledge of entire cultures, in spite of the fact that the knowledge and use of plant biodiversity is everywhere gender-differentiated. There are three associated errors. The first is related to the failure to research women’s knowledge and use of plants, which is an error of omission - the species and varieties that only women know are omitted, and therefore biological diversity is under-estimated. This error is only rarely acknowledged by researchers. After reviewing hundreds of scientific publications, I have found only one where the omission and its subsequent consequences for scientific validity was acknowledged, in relation to research into the indigenous knowledge and vegetation use of Bedouin tribes in Egypt: It should be noted that these discussions and the subsequent work reported here were entirely with male respondents. The results, therefore, very much reflect a male world and there is, consequently, a need to carry out equivalent work with Allaqi women before a full understanding can be gained (Briggs et al. 1999: 92). The second error is one of unreliability. It is related to using sources that are not well informed, which leads to the improper identification of plants, their management, characteristics, uses, and names. Numerous studies have shown that women are often more able to correctly identify these parameters in comparison with men, particularly with regard to plants that fall directly into their domains (Howard-Borjas 2002). Men and women may have some knowledge regarding the plants that are primarily used or managed by the opposite sex, but generally their knowledge is more limited or imprecise with regard to various characteristics of those plants. Such is the case in the Peruvian Andes, where Zimmerer investigated gender differences in ethnobotanical knowledge of maize and potato cultivars that he associated with the gender division of labor: Male cultivators tend to be less accurate and less specific in naming [different varieties] . . . They apply fewer names overall than women and tend to mislabel uncommon taxa . . . Not only the extent of cultivar knowledge but also its focus varies between sexes. Men know especially little about culinary properties such as mealiness or taste which are key conceptual markers of cultivars . . . Variations between men and women in the identification and naming of maize cultivars exceed the gap which characterizes sex-based differences in the knowledge of potato types. This contrast parallels the division of labor, which is more pronounced in maize than in potato agriculture . . . women alone provide most of the labor for maize cultivation in many households. The classification of uncommon cultivars as intermediate or diseased forms of more common taxa reflects the lack of knowledge among male cultivators concerning identification and naming (1991: 314). It is usually impossible to determine whether the first two errors have been committed. Research results are presented in such a way that it is not known whether women were included among informants since references are to gender-neutral descriptors such as ‘farmers’, ‘rural dwellers’, ‘peasants’, ‘experts’, ‘informants’, ‘healers’, etc.. In the majority of those cases where it is made explicit that women were included in the research (as interviewers and/or as interviewees), which is the case in quite a number of publications, data are nevertheless most often not sex-disaggregated and there is no analysis of sex differences nor of gender relations. It is as though researchers are aware that they must include women in their research, but yet do not consider gender differences as significant for their results. Only rarely are any of the methodological difficulties of researching women mentioned (see final section of this paper), which leads a sensitive reviewer to be concerned that, even though women might have been included, their participation was not necessarily fully secured (as was the case with Aguilar and Conduit’s study, for example). In fact, it is in those cases where researchers report that there were ‘no gender differences’ in knowledge or use of plants that explanations should be forthcoming as to why this might be the case – as Boster pointed out, the lack of gender distinctions in knowledge and use generally applies only in those cases where one is referring to ‘the most culturally important plants’ (1985a: 186), that is, those that are likely to be known to and used in similar ways by virtually all adults. The third type of error is one of interpretation, leading to a misunderstanding of people-plant relations since a critical component of these relationships –- gender relations -- is not revealed. An example is provided by contrasting research findings regarding genetic erosion among potato and maize landraces in the Andes, where male outmigration, both temporary and permanent, is high. First, Brush et al. (1992) hypothesized that, among other factors, if (male) farmers migrate to earn income off-farm, this . . may encourage diversity by providing capital to finance the cost of maintaining diverse landraces. However, it may also create disincentives for diversity. Typically it requires the farmers’ selection criteria and expertise, and hired labor is likely to be an imperfect substitute for the farmer's own labor time in this regard (Ibid: 371). They tested this hypothesis in the Paucartambo and Tulumayo valleys of Peru and found that on-farm diversity was negatively affected by off-farm employment. This they attributed to ‘a high opportunity cost of efforts to maintain diversity where off-farm income opportunities exist’ (Ibid: 380). But, Zimmerer (1991) found in one of the same valleys that cultivar loss was not due to the absence of the male farmer who has the principle expertise. Male off-farm labor does not decrease the expertise available for cultivar selection since it is women farmers who mainly hold this expertise in the first place. While there are many complex transformations occurring in agriculture due to increasing commodity production, contract farming, and temporary male out-migration, one of the most important outcomes is the ‘feminization of agriculture’ with the concomitant increasing pressure on women farmers’ labor, which is leading to the loss of cultivar diversity (Ibid: 307-315). In a later article, Brush (1995) proposed a conceptual framework for understanding farmers’ in situ conservation of landraces, where four factors were particularly important: Land holding fragmentation means that farmers manage several fields and cultivate folk varieties in at least one or more of these fields; Marginal agronomic conditions mean that folk varieties are competitive with improved varieties since they perform as well or better in these environments; The relative isolation of many traditional farming systems means that market imperfections are created so that improved varieties lose their commercial advantages; and Farmers’ cultural diversity and their preferences for maintaining genetic diversity mean that they maintain folk varieties. Whereas all of the points that Brush makes may be considered to be generally valid, there are vital points that would not be overlooked if women or gender relations were considered. With regard to the first point, Brush did not mention that men and women often manage different fields, with different responsibilities for providing plant resources, different access to technology, labor, credit, knowledge, and markets. The pressures on plant biodiversity in one field may therefore be quite different from those on another field, and for different reasons (Malaza 2003; Wilson 2003; Wooten 2003). With respect to Brush’s second point, it has frequently been shown that the land to which women have access for field crop production is more marginal in agronomic terms than that to which men have access. For example, in The Gambia, studies have shown that women have access to lowland swamp areas and men to upland areas, and the species and varieties that each plant correspond to these different agro-ecological conditions (Dey 1983; Schroeder 1999). In Cameroon, Nchang Ntumngia (1997) showed that women traditionally produce cassava on marginal land whereas men produce cash crops on higher quality land but, with the decrease in prices for men’s export cash crops and the introduction of improved cassava varieties, men have begun to plant modern cassava varieties on higher quality land, leaving women to continue to produce the traditional varieties that are adapted to marginal land. With respect to Brush’s third point, men and women often have access to different markets, where women are mainly able to access local markets where the demand for local varieties is often greater, and men have greater access to urban and national markets, where the demand for modern varieties is higher. Brush does not mention the fact that production for subsistence is more oriented toward varieties and species that are traditionally consumed in the local diet, and is also often in the hands of women and performed in homegardens. See, e.g. Nchang Ntumngia 1997, Defoer et al 1996, Chewya & Eyzaguirre 1999, and Wooten 2003, Wilson 2003, Greenberg 2003, and Malaza 2003. Finally, with regard to Brush’s fourth point, this paper shows clearly how cultural identity, plant diversity, women and the domestic sphere are clearly inter-related and highly important, and yet are very frequently overlooked. Including women in ethnobotanical research is not only a matter of ensuring that a full accounting of plant knowledge and use is accomplished. Understanding people-plant relations is far more encompassing, having to do as well with the influence of plants on human culture, and today has very great importance considering the need to understand the driving forces behind genetic erosion and discover means to preserve both plants and those cultures that have served as their custodians. Promises of a Brighter Future: Toward Gender-Sensitive Frameworks and Methods There are reasons for optimism and promises of a brighter scientific (as well as applied) future. In spite of the clear presence and persistence of gender bias within the discipline of ethnobotany, it is quite evident that changes in researchers’ sensitivity to women and gender relations are occurring especially over the past decade, as is evident in the statements found at the beginning of this paper and in many other ethnobotanical contributions. It is interesting to note that gender blindness was not so obvious in people-plant research prior to the emergence of ‘economic botany’ around the 1940s. This is at least in part attributable to the fact that much historical ethnobotanical research was more ‘anthropological’ and ‘ethnographic’ in comparison to ‘economic botany’, which focuses much more on plants and their potential economic uses than on people. Much ethnobotanical research still focuses exclusively on plants. Many researchers certainly have come to realize that knowledge is not evenly distributed throughout indigenous and peasant societies; conceptual and methodological frameworks for research that take into account social structures and intracultural variations are becoming more sophisticated as botanists and social scientists begin to cross disciplinary lines in earnest. I suggest this is due in part due to the increasing demands for applied knowledge in conservation efforts. This may be seen to follow a trend present in other applied disciplines such as forestry, agricultural development and soils conservation, where the cognizance of the importance of women and gender relations grew principally due to the repeated failure of development efforts due to mis-targeting and misunderstanding of local conditions and needs. Some ethnobotanists today are aware that gender matters; enough gender-sensitive research has been done to permit the development of a state-of-the-art assessment of the field, and the panel presented here today provides yet further testimony to the fact that women and gender relations in ethnobotany are becoming an integral part of the field. There are very many topics that can and need to be explored to arrive at a reasonable conceptual framework for ethnobotanical research in its many guises that have to do with intracultural variation. I select one example where we can quite legitimately ask, what at this time do ethnobotanists have ‘on-the-shelf’ that can guide us toward, at minimum an adequate selection of informants, and is it gender sensitive? Few of the articles or books dealing specifically with methods in ethnobotany that I have reviewed Articles and books reviewed dealing with methods and informant selection in ethnobotany include, among others: Alexiades 1996, Cotton 1996, Cunningham 2001, Fowler 2001, Given & Harris, 1994, Guarino 1995, IDRC 1998., Martin 1995, Maundu 1995, Medin & Atran 1999, Rao 1987, Richardson & Stubbs 1978, and Robbins et al 1916. deal at with intracultural variability or with gender relations, although most do discuss informants in one respect or another. Going back as far as 1916, when discussing the emerging science of ethnobotany and its methods as an introduction to their research findings on the ethnobotany of the Tewa, Robbins et al. gave the following advice: ‘A prime necessity is a good native informant; indeed it is better to have several informants preferably older men or women . . As a means of checking the accuracy of information obtained it is also well to work with different individuals or groups of individuals separately and to compare the results’ (1916: 2). The emphasis is on key informants and the assumption is that older persons of either sex are the best representatives of the collective knowledge of their cultures, which is an assumption that many ethnobotanists seem to share today. In 1978, Richardson and Stubbs, in their book on ethnobotany that included a chapter on methods, were quite concerned with the problem of establishing good relations with the community that is being researched, stressing constantly the need to avoid offending people. In their discussion on informant selection, they only stressed the avoidance of associations with local elites that may bias ‘more helpful’ informants against them. They mentioned a problem related to gender, however, that is rarely tackled in ethnobotanical literature today, which refers both to the gender division of labor and gendered social norms, as well as the need to consider the sex of the interviewer: Few cultures are acquainted with the rhetoric of sexual equality. In many cultures the possession and transmission of shamanistic and related ethnobotanical information may be privy to a particular sex. It may be considered taboo to provide this knowledge to a member of the opposite sex. If, as a potential investigator, you are of the 'wrong' sex in terms of the information desired, you would do better to leave the proposed work to someone else. The probability of obtaining accurate data is low (1978: 224). Apparently such advice was not taken seriously even after they wrote, since Duke admitted in 1994 that ‘Most of the ethnobotanical writings on female health issues were by foreign men, interpreting native men in turn interpreting native women’ (Duke and Vasquez 1994). Rao (1987), in his recommendations about informant selection in a paper on ethnobotanical methods, expressed the gender bias that is elsewhere so prevalent but less explicitly stated: ‘Valuable data can be collected with the help of knowledgeable local informants. . . Locating proper knowledgeable informants is not easy. In most tribes or villages, normally there will be one or two elders, who are familiar with the herbal medicines. These 'medicine men' are very resourceful . . . These people are to be contacted for collecting information about medicinal herbs’ (1987: 35). There was apparently little progress as late as 1994, when Given and Harris, in their handbook on ethnobotanical methods, made no mention at all of intracultural variation and were principally concerned with the issue of establishing trust with informants, and they did not mention any concern with establishing rapport with women informants. Whereas individual researchers such as Boster had taken up the issue of intracultural variation and gender fairly seriously and discussed them in relation to informant selection, these issues do not appear in discussions explicitly addressing ethnobotanical methods until the mid-1990s, when several key texts were published: two on ethnobotanical methods (Alexiades 1996; Martin 1995) and one general review of ethnobotany that contained a chapter on methods (Cotton 1996). These authors began to recognize and critique the assumptions made by most ethnobotanists to date. For example, Alexiades pointed out: A common mistake when interacting with unfamiliar cultures is assuming that one is dealing with a culturally homogeneous group and that culture is a monolithic entity that can be studied through interaction with the most ‘authoritative’ informant. Although much ethnobotanical information is widely shared, there is often a considerable degree of intracultural variation and specialization in ethnobotanical knowledge, even within small communities (1996: 10-11). Martin (1995) in turn asked, ‘Whom do you ask about plants, animals and other elements of the natural environment?’ and indicated that many researchers are casual about this choice. ‘They speak to the first person encountered, ask for the individual who knows the most about plants in the community or interview anyone who is willing to talk with them’ (1995: 97). The means to overcome this problem is to gain an appreciation of intracultural variation in knowledge distribution. To do this, Cotton suggested that, when ‘the overall range and distribution of information is unknown, it is advisable to start with a random sample which provides a representative cross-section of the information held by a community as a whole’ (1996: 101), which Alexiades and Martin also recommend. Alexiades added a specific reference in this context to gender: . . .researchers need to make decisions relating to the type and size of sample of their informants. . . Whereas some studies may include all the people living within a region or habitat, others may be based on a community or a smaller sub-section of the community. Gender, age, and social roles, for example, may also be used to define the sampling universe (1996: 77). However, he implied that gender is an optional variable, rather than a structural factor that must be considered in all analysis of intracultural variability. This echoes the position of both Martin and Cotton at the time: Cotton wrote that ‘researchers must ensure that their method takes account of sociological variables such as age, gender, occupation, education and class, to ensure that the data collected are not biased in favour of any one social group’ (1996: 102); Martin (1995) also referred (albeit obliquely) to gender when he emphasized that such a random sample is intended to exclude bias: . . . for example, more men than women or more old people than young people . . .This is particularly important in situations where the dominant social group - men, elders, wealthy or highly educated people - tend to be the first to come forward to be interviewed. Picking a random sample, or at least one that represents the diversity of the community, will ensure that you speak with the silent majority which often includes women and children, the poor and other groups who have often been ignored by fieldworkers (Ibid: 97). However, in the case where special informants are required, Cotton argued that, ‘where more specific information is required, specialist informants may be chosen in consultation with community leaders and other community members’ (1996: 106), without cautioning, as Martin did, that that men will often not openly acknowledge women’s specialized knowledge and that women are also often not community leaders. Martin further insisted that informants be identified by sex in research results: ‘you should characterize the local people who work with you and show how they represent the overall community by using sociological variables, including gender, place and date of birth, education, literacy, languages, years of formal education, work, etc.’ (Ibid: 101-102). By 1999, Alexiades had changed the position he expressed in 1995, where gender was an optional variable for social stratification when it comes to understanding intracultural variation in knowledge. Much like Boster argued in 1985, he now considered that individual competence, motivation, experimentation, and access to knowledge are important to explaining intracultural variation, but noted that: . . . these in turn are determined by age, gender, kin relations, social roles and degree of contact with other social actors and sources of knowledge. In addition, knowledge differences between women are correlated to kinship, as both plant varieties and knowledge are mainly exchanged between members of family and extended kin (1999: 335). The most recent book discussing ethnobotanical methods is that of Cunningham (2001), which refers to applied ethnobotany. It could be considered to present an interesting contrast if it is true that the pressure to deal with gender and women is a result of the need to find pragmatic solutions to development and conservation problems, as I argue earlier in this paper. In it, he explicitly mentioned the issue of power, and indicated that rural communities are ‘not homogeneous, but are complex networks, divided on the basis of power, gender and specialist interest groups’ (2001: 11) that have a stake in maintaining specific resources. The reasons for being concerned with these networks and divisions are for him utilitarian rather than academic. The involvement of resource users in applied ethnobotanical efforts, he argued, accomplishes numerous objectives simultaneously: it ‘provides valuable insights into the scarcity of useful plant species’ and ‘for identifying possible key species’; 2) permits the development of conservation plans with rather than for resource users; and 3) ‘enables specialist user groups to be identified’ (Ibid). When asking ‘who are the resource users’ he indicates that one should refer clearly to age, sex and specialization e.g. ‘herbalists or weavers’ (Ibid: 15). However, he makes no further attempt to systematically treat the question of informant selection or gender. There is one methods text that much more explicitly addressed gender, although it is not explicitly directed at ethnobotanists. Guarino, when discussing the use of secondary sources in germplasm collection, provided a rationale for particularly stressing the roles of women and emphasized power relations: They are . . . the main custodians of crop-related knowledge, though there is often a discrepancy between female workloads and experience on the one side and their social status (and power) on the other. Even where women’s direct production tasks are more limited, they will probably be responsible for pre-sowing, processing, storage or cooking activities, and thus hold key information on germplasm performance and quality. Further, their general lack of decision-making power on land use means women often cultivate the main field crops at marginal sites, and thus have specialist knowledge of landrace performance with respect to problem soils or environments. In addition, women’s trading networks and kin relationships are often the main channels for acquisition and exchange of germplasm (1995: 196-197). Further, he explicitly addressed the need to research women as a matter of efficiency: ‘At a time of shortages of staff and resources, the formal sector cannot afford to ignore the accumulated experience of local people, and in particular local women, in solving their own conservation and development problems’ (Ibid: 198). Selecting the right informants also is a matter of validity and comprehensiveness and has great relevance to germplasm collection. For example, in reference to locating and accessing target areas and materials, he pointed out that specialized, marginal environments are often missed because of their small size and dispersion and attributes this in part to bias. ‘They are also often neglected by agricultural professionals, who tend to be male, have shorter time horizons than farmers and concentrate on staple and cash crops. These environments are often tended by women, may take years to develop, often feature “unimportant” crops and do not fit in easily with conventional station-based research’ (Ibid: 200). To date, only the IDRC (1998) has provided a methods guideline for its own Sustainable Use of Biodiversity programmed on how to integrate gender analysis into biodiversity research, which is the first systematic treatment of the topic that could act as a starting point for ethnobotanical research more generally. Besides providing key concepts (which may also be seen as its principle limitation, since these are derived essentially from gender and agriculture research, rather than being based on ethnobotany), it discusses data collection, interpretation and analysis, and provides a series of research tools. It is explicitly directed toward the project cycle and hence incorporates methods that are based essentially on rapid rural appraisal that again can only be taken as a starting-point for scientific research, but it also provides useful discussions of how instruments more generally applied in ethnobotany such as matrix ranking can be used to detect gender differences. There are more issues entailed in carrying out research that takes into account intracultural variability and particularly gender differences, one of which has been given somewhat more attention, although sporadically, in the ethnobotanical literature. Sex and the single male researcher Plotkin, in Tales of a Shaman’s Apprentice (1993), related a very interesting confession and an anecdote that is highly relevant to this discussion: Although I have little knowledge of the plants involved, I am convinced that there exists a wealth of ethnobotanical treatments for menstrual problems, birth control, difficult childbirth, and so on, which is simply unavailable to the male ethnobotanist. Take, for example, the famous piripiri of the western Amazon. In the 1940s, Nicole Maxwell, a New York socialite, packed her bags and headed to the Amazon. . . lived for many years among tribes like the Witoto and the Bora, carefully collecting as many of their plants as she could. A plant called piripiri was used by the women of the tribe as a contraceptive. They brewed it as a tea, and anyone drinking only a few cups reportedly would be infertile for a period of up to three years. But Maxwell was never able to get fresh plant material into the laboratory for testing. It was not until 1986 that piripiri was collected and tested, by an American woman named Karen Lowell, who was living with the Shuar Indians . . Lowell was taken into the women’s gardens and taught about the plants they cultivated. Most, like cassava and papaya, were grown as food. From some of the plants, like cotton and u-shuh berry, the women produced trade goods like hammocks or dyes. A few were medicinal. The famous piripiri, which Lowell had heard about but never seen, turned out to be a sedge. Laboratory analysis is currently under way . . . (1993: 105). The anecdote is interesting not only in that it draws attention to the importance of the sex of the researcher and access to indigenous women’s knowledge regarding reproductive health, but also because it reveals that Shuar women’s more general plant knowledge, use and management also apparently went unexplored until a woman researcher stepped onto the scene. In spite of the lack of specific theoretical or methodological attention to gender and intracultural variation in people-plant relations, setting out to research women’s knowledge and use of plant resources and of the relations that surround these is a goal that many ethnobotanists have at one time or another, and for one reason or another, actively pursued. Several ethnobotanists have noted the importance of reaching women but at the same time have noted some of the difficulties of doing so, precisely due to gender relationships – in fact, this appears to be the moment when they actually encounter the fact that gender norms are highly significant arbitrators of human behavior. Besides Richardsons’ and Stubbs’ concern mentioned earlier, Cotton noted that ‘…factors such as cultural restraints on single women must also be considered, as members of different sociological groups [sic] may respond differently to researchers of a certain gender or social status’ (1996: 106). Alexiades (1999) spoke from personal experience when he pointed out that the collaboration with his wife in his research presented great advantages since interactions between male ‘outsiders’ and native women are culturally proscribed because single adults may have illicit sexual relations. It would have been difficult for him to interview women and information would likely have been distorted. He considered that the male bias that he encountered in ethnobotanical and anthropological research is due essentially to these problems. There are yet other obstacles to incorporating women into ethnobotanical research. Kothari (2003) reported that, for a participatory ethnobotanical research project in which he was involved, having an equal number of women interviewers was crucial to obtaining access to women’s knowledge, but that getting women to participate as interviewers was fraught with difficulties and only really became possible when a literacy requirement was dropped. Other gender-related problems have been mentioned by researchers, such as the insistence of men that they represent the household, women’s ‘shyness’ and inexperience in working with researchers, and restrictions on women’s mobility that present difficulties when using them as interviewers (see Brodt 1998; Martin 1995; Rusten 1989). Seldom, however, do ethnobotanical researchers offer greater methodological insights into ways to deal with such issues, although references can be found dispersed throughout the literature. For example, Maundu, when referring to group interviews, indicated: The quality and quantity of the information generated over a given period will be affected by such factors as size, composition, psychological state of the group, and social, economic and cultural factors. The sex and age ratios will determine the degree of involvement and freedom of expression. . . In most communities the output of a group will be influenced by the representation of the various sex and age segments. In mixed groups, men will tend to dominate the talks. Women generally feel most free among other women, and in some communities women will not even speak in the presence of men. Younger women usually feel free only when there are no men at all present, while in the absence of women, men are often impatient and argumentative (1995: 2). A Requiem for Ethnobotany – There’s Life in the Old Girl Yet At this time, ethnobotanists must refer to literature that provides frameworks and methods for researching intracultural variation and gender relations that are largely outside of their disciplines, although the amount of material that can be found within the discipline is fortunately growing daily. Until that time when ethnobotanists attribute enough importance to intracultural variation, to women and to gender relations in local plant management, the theoretical frameworks that inform them, the methods employed and the resulting empirical data will be insufficiently robust to permit answers to be found for the most basic questions posed. Fortunately, ethnobotanists may take heart - that time appears to be upon us. Literature Cited Abbink, J. 1993. Me'en Medicinal, ritual and other plants: a contribution to Southwest Ethiopian ethnobotany. Journal of Ethiopian Studies 26(2): 1-19. Aguilar, S. and R. Condit. 2001. Use of native tree species by an Hispanic community in Panama. Economic Botany 55(2): 223-235. Alexiades, M. (ed.). 1996. Selected guidelines for ethnobotanical research: a field manual. New York Botanical Garden, Bronx, New York. Alexiades, M. 1999. Ethnobotany of the Ese Eja: plants, health and change in an Amazonian society (plant medicinals). Ph.D. thesis, City University of New York. University Microfilms International, Ann Arbor, Michigan. Anderson, M. 1983. Woman as generalist, as specialist, and as diversifier in Saami subsistence activities. Humboldt Journal of Social Relations 10(2): 175-197. Barry, H., Schlegel, A. 1982. Cross-cultural codes on contributions by women to subsistence. Ethnology 21(2):165-188. Battcock, M, and S. Azam-Ali. 1998. Fermented fruits and vegetables : a global perspective. FAO Agricultural Services Bulletin No. 134 Food and Agriculture Organization of the United Nations (FAO), Rome. Bennett, S. 2000. Five centuries of women and gardens. National Portrait Gallery Publications, London. Boster, J. 1985a. Requiem for the omniscient informant: there's life in the old girl yet. Pages 177-198 in J. Dougherty, ed., Directions in cognitive anthropology. University of Illinois Press, Champaign. Boster, J. 1985b. Selection for perceptual distinctiveness: evidence from Aguaruna cultivars of Manihot esculenta. Economic Botany 39(3): 310-325. Boster, J. 1986. Exchange of varieties and information between Aguaruna manioc cultivators. American Anthropologist 88: 428-436. Briggs, J., M. Badri, and A. Mekki. 1999. Indigenous knowledges and vegetation use among Bedouin in the eastern desert of Egypt. Applied Geography 19(2): 87-103. Brightman, R. 1996. The sexual division of foraging labor: biology, taboo, and gender politics. Comparative Studies in Society and History 38(4): 687-729. Brodt, S. 1998. Learning from the land: local knowledge systems of tree management in Central India, Ph.D. thesis, University of Hawaii. University Microfilms International, Ann Arbor, Michigan. Browner, C. 1989. The management of reproduction in an egalitarian society’. Pages 58-71 in C. Shepherd McClain, ed., Women as healers: cross-cultural perspectives. Rutgers University Press, New Brunswick and London. Browner, C., and S. T. Perdue. 1988. Women’s secrets: bases for reproductive and social autonomy in a Mexican community. American Ethnologist 15(1): 84-97. Brownrigg, L. 1985. Home gardening in international development: what the literature shows. League for International Food Education, Washington, DC. Brush, S., E. Taylor, and M. Bellon. 1992. Technology adoption and biological diversity in Andean potato agriculture. Journal of Development Economics 39: 365-387. Brush, S. 1995. ‘In situ conservation of landraces in centers of crop diversity’. Crop Science 35(2): 346-354 Chewya, J., and P. Eyzaguirre (eds.). 1999. The biodiversity of traditional leafy vegetables. International Plant Genetic Resources Institute (IPGRI), Rome. Cotton, C. 1996. Ethnobotany: principles and applications. John Wiley and Sons Ltd., Chichester, West Sussex, UK. Coughenour, C., and S. Nazhat. 1986. The process of agricultural change among women farmers of North Kordofan, Sudan. Ahfad Journal: Women and Change (Sudan) 3(1): 19-28. Cunningham, A. 2001. Applied ethnobotany: people, wild plant use and conservation. Earthscan, London. Defoer, T., Kamara, A., and H. de Groote. 1997. Gender and variety selection: farmers' assessment of local maize varieties in southern Mali. African Crop Science Journal 5(1): 65-76. Denevan, W., and C. Padoch (eds.). 1987. Swidden-fallow agroforestry in the Peruvian Amazon. Advances in Economic Botany 5. New York Botanical Garden Scientific Publications Dept., New York. Descola, P. 1994. In the society of nature: a native ecology in Amazonia. Cambridge University Press, Cambridge. Dey, J. 1983. Rice farming systems: case studies of current developments and future alternatives in upland rice and inland swamp rice. Expert Consultation on Women in Food Production, Rome, Italy, 7-14 December 1983, ESH: WIFP/83/4. Food and Agriculture Organization of the United Nations (FAO), Rome. Dick Bissonnette, L. 2003. The basket makers of the Central California Interior. Pages 197-210 in P. Howard, ed., Women and plants. Gender relations in biodiversity management and conservation. Zed Press and Palgrave Macmillan, London & New York. Dirar, H. 1992. Traditional fermentation technologies and food policy in Africa. Appropriate Technology 19(3):21-23. Duke, J., and R. Vasquez. 1994. Amazonian Ethnobotanical Dictionary. CRC Press, Boca Raton. Ertuğ, F. 2003. Gendering the tradition of plant gathering in Central Anatolia (Turkey). Pages 183-196 in P. Howard, ed., Women and plants. Gender relations in biodiversity management and conservation. Zed Press and Palgrave Macmillan, London & New York. FAO (Food and Agriculture Organization of the United Nations). 1993. Agricultural extension and farm women in the 1980s. FAO, Rome. _____ 1989. Household food security and forestry: an analysis of socio-economic issues. FAO, Rome. Fedigan, L. 1986. The changing role of women in models of human evolution. Annual Review of Anthropology 15, pp. 25-66. Flickinger, D. 1997. Rehabilitation of degraded tropical forests in India's Western Ghats: silvicultural and socio-economic implications of multiple species plantations. Ph.D. thesis, University of Florida. University Microfilms International, Ann Arbor, Michigan. Fowler, C. 2001. In the field with people, plants and animals. A look at methods. Pages 149-162 in R. Ford, ed., Ethnobiology at the millennium: past promises and future prospects. Anthropological Papers, No. 91. A collection of papers from the Ethnobiology 2000 millennium conference in Ann Arbor. University of Michigan Museum of Anthropology, Ann Arbor. Given, D., and W. Harris. 1994. Techniques and methods of ethnobotany: as an aid to the study, evaluation, conservation and sustainable use of biodiversity. Manaaki Whenua Landcare Research, Lincoln, N.Z.. Good, C. 1987. Ethnomedical systems in Africa. Guilford Press, Hove (UK). Greenberg, L. 2003. Women in the garden and kitchen: the role of cuisine in the conservation of traditional house lot crops among Yucatec Mayan immigrants. Pages 51-65 in P. Howard, ed., Women and plants. Gender relations in biodiversity management and conservation. Zed Press and Palgrave Macmillan, London & New York. Guarino, L. 1995. Secondary sources on cultures and indigenous knowledge systems. Pages 195-215 in L. Guarino, R. Rao, R., and P. Reid, eds., Collecting plant genetic diversity: technical guidelines. CAB International, Wallingford, UK. Hays, T. 1974. Mauna: explorations in Ndumba ethnobotany. Ph.D. thesis, University of Washington. University Microfilms International, Ann Arbor, Michigan. _____ 1997. Personal communication. Hoffmann, S. 2003. Arawakan women and the erosion of traditional food production in Amazonas Venezuela. Pages 258-272 in P. Howard, ed., Women and plants. Gender relations in biodiversity management and conservation. Zed Press and Palgrave Macmillan, London & New York. Howard, P. 1982-1990. Personal experience as gender expert for the United Nations system in Central America. _______ 2003. Women and the plant world: an exploration. Pages 1-48 in P. Howard, ed., Women and plants. Gender relations in biodiversity management and conservation. Zed Press and Palgrave Macmillan, London & New York. Howard-Borjas, P. with W. Cuijpers. 2002. Gender and the management and conservation of plant biodiversity,’ in H. Doelle and E. DaSilva, eds., Biotechnology, in Encyclopaedia of Life Support Systems (EOLSS), Oxford, UK. ERL: http://www.eolss.net. Hyde, E. 1998. Cultivated power: flowers, culture, and politics in early modern France. Ph.D. thesis, Harvard University. University Microfilms International, Ann Arbor, Michigan. International Development Research Centre (IDRC). 1998. Guidelines for integrating gender analysis into biodiversity research: sustainable use of biodiversity program initiative. IDRC, Ottawa. Johns, T. 1990. With bitter herbs they shall eat it. Chemical ecology and the origins of human diet and medicine. The University of Arizona Press, Tucson. Kappelle, M., Avertin, G., Juarez M., and N. Zamora. 2000. Useful plants within a campesino community in a Costa Rican montane cloud forest. Mountain Research and Development 20: 2, 162-171. Keys, E. 1999. Kaqchikel gardens: women, children, and multiple roles of gardens among the Maya of Highland Guatemala. Yearbook Conference of Latin Americanist Geographers 25: 89-100. Kothari, B. 2003. The invisible queen in the plant kingdom: gender perspectives in medical ethnobotany. Pages 150-164 in P. Howard, ed., Women and plants. Gender relations in biodiversity management and conservation. Zed Press and Palgrave Macmillan, London & New York. Kumar, S. 1994. Adoption of hybrid maize in Zambia: effects on gender roles, food consumption, and nutrition. Research Report No. 100m. International Food Policy Research Institute, Washington, DC. Longley, C. 2000. A social life of seeds: local management of crop variability in North-western Sierra Leone. Ph.D. thesis, University College London. Malaza, M. 2003. Modernization and gender dynamics in the loss of agrobiodiversity in Swaziland's food system. Pages 243-257 in P. Howard, ed., Women and plants. Gender relations in biodiversity management and conservation. Zed Press and Palgrave Macmillan, London & New York. Martin, G. 1995. Ethnobotany. A methods manual. Chapman and Hall, London. Maundu, P. 1995. Methodology for collecting and sharing indigenous knowledge: a case study. Indigenous Knowledge and Development Monitor 3(2). Internet version. McClain, C. 1989. Reinterpreting women in healing roles. Pages 1-19 in C. McClain, ed., Women as healers. Cross cultural perspectives. Rutgers University Press, New Brunswick & London. Medin, D., and S. Atran. 1999. Introduction. Pages 1-16 in D. Medin and S. Atran, eds., Folkbiology. Massachusetts Institute of Technology, Cambridge, MA. Nchang Ntumngia, R. 1997. Gender power dynamics and knowledge systems in household food security: a case study of Malende and Mautu cassava farmers, Muyuka – Cameroon. M.Sc. thesis, Wageningen Agricultural University, the Netherlands. Niñez, V. 1987. Household gardens: theoretical and policy considerations. Part I. Agricultural Systems 23: 167-186. Norton, H. 1985. Women and resources of the Northwest Coast: documentation from the 18th and early 19th centuries. Ph.D. thesis, University of Washington. University Microfilms International, Ann Arbor, Michigan. Ochoa, L., Fassaert, C., Somarriba, E., and A. Schlonvoight. 1998. Conocimiento de mujeres y hombres sobre las especies de uso medicinal y alimenticio en huertos caseros de Nicoya, Costa Rica. [Men and women's knowledge about medicinal and food species in home gardens in Nicoya, Costa Rica]. Edición especial: tésis de maestría en agroforestería 1997 [Special issue: master's theses in agroforestry 1997]. Agroforestería en las Américas 5: 17-18, 7-11. Okigbo, B. 1990. Homegardens in tropical Africa. Pages 21-40 in K. Landauer and M. Brazil, eds., Tropical home gardens. Selected papers from an International Workshop held at the Institute of Ecology, Padjadjaran University, Banderung, Indonesia, 2-9 Dec. 1985. United Nations University Press, Tokyo. Pieroni, A. 2003. Wild food plants and Arabëresh women in Lucania, Southern Italy. Pages 66-82 in P. Howard, ed., Women and plants. Gender relations in biodiversity management and conservation. Zed Press and Palgrave Macmillan, London & New York. Plotkin, M. 1993. Tales of a Shaman's Apprentice. Viking, New York. Rao, R., and P. Hajra. 1987. Methods of research in ethnobotany. Pages 33-41 in S. Jain, ed., A manual of ethnobotany : proceedings of the training course and workshop of ethnobotany, held at Lucknow, 10 - 15 march, 1986. Scientific Publishers, Jodhpur. Rasmussen, S. 1998. Only women know trees: medicine women and the role of herbal healing in Taureg culture. Journal of Anthropological Research 54(2)148-171. Richardson, W., and T. Stubbs. 1978. Plants, agriculture, and human society. W. A. Benjamin, Menlo Park, California. Robbins, W., Harrington, J., and B. Freire-Marreco. 1916. Ethnobotany of the Tewa Indians. Smithsonian Institution, Bureau of American Ethnology Bulletin 55. Part I. Washington DC: Government Printing Office. Rocheleau, D. 1989. Gender division of work, resources and rewards in agroforestry systems. Second Kenya National Seminar on Agroforestry. International Center for Research in Agroforestry (ICRAF), Nairobi, Kenya. Rusten, E. 1989. An investigation of an indigenous knowledge system and management of tree fodder resources in the Middle Hills of Western Nepal. Ph.D. thesis, Michigan State University. University Microfilms International, Ann Arbor, Michigan. Schroeder, R. 1999. Shady practices. Agroforestry and gender politics in The Gambia. University of California Press, Berkeley. Scoones, I., M. Melnyk, and J. Pretty, eds. 1992. The hidden harvest: wild foods and agricultural systems. A literature review and annotated bibliography. Sustainable Agriculture Programmed, International Institute for Environment and Development, London. Shteir, A. 1996. Cultivating women, cultivating science. Flora's daughters and botany in England, 1760 to 1860. Johns Hopkins University Press, Baltimore. Sillitoe, P. 2003. The gender of crops in the Papua New Guinea Highlands. Pages 165-180 in P. Howard, ed., Women and plants. Gender relations in biodiversity management and conservation. Zed Press and Palgrave Macmillan, . London & New York. Song, Y. 1998. 'New' seed in 'old' China. Impact of CIMMYT Collaborative Programmed on Maize Breeding in South-western China. Ph.D. thesis, Wageningen Agricultural University, the Netherlands. Sperling, L., and P. Berkowitz. 1994. Partners in selection: bean breeders and women bean experts in Rwanda. CGIAR, Washington, DC. Stavrakis, O. 1979. The effects of agricultural change upon social relations in a village in Northern Belize, Central America. Ph.D. thesis, University of Minnesota. University Microfilms International, Ann Arbor, Michigan. Steinkraus, K. 1992. African alkaline fermented foods and their relation to similar foods in other parts of the world. Pages 87-92 in A. Westby and P. Reilly, eds., Traditional African foods: quality and nutrition. Proceedings of Regional IFS Workshop, Dar es Salaam, Nov. 1991. International Foundation for Science, Stockholm. Stevens, M. 1999. The ethnoecology and autecology of White Root (Carex barbarae Dewey): implications for restoration. Ph.D. thesis, University of California at Davis. University Microfilms International, Ann Arbor, Michigan. Tapia, M., and A. de la Torre. 1998. Women farmers and Andean seeds. Food and Agriculture Organization of the United Nations (FAO) and International Board for Plant Genetic Resources (IPGRI), Rome. Turner, N. 2002. Personal communication. _____ 2003. 'Passing on the news': women's work, traditional knowledge and plant resource management in indigenous societies of North-western North America. Pages 133-149 in P. Howard, ed., Women and plants. Gender relations in biodiversity management and conservation. Zed Press and Palgrave Macmillan, London & New York. Vedavalli, L., and N. Anil Kumar. 1998. Wayanad, Kerala. Pages 96-106 in M. Swaminathan, ed., Gender dimensions in biodiversity management. Konark Publishers Pvt. Ltd., Delhi. Wilson, M. 2003. Exchange, patriarchy and status: women's homegardens in Bangladesh. Pages 211-226 in P. Howard, ed., Women and plants. Gender relations in biodiversity management and conservation. Zed Press and Palgrave Macmillan, London & New York. Wooten, S. 2003. Losing ground: gender relations, commercial horticulture and threats to local plant diversity in rural Mali. Pages 229-242 in P. Howard, ed., Women and plants. Gender relations in biodiversity management and conservation. Zed Press and Palgrave Macmillan, London & New York. Zimmerer, K. 1991. Seeds of peasant subsistence: agrarian structure, crop ecology and Quechua agriculture in reference to the loss of biological biodiversity in the Southern Peruvian Andes. Ph.D. thesis, University of California Berkeley. University Microfilms International, Ann Arbor, Michigan. PAGE 38 PAGE