Gender, knowledge-sharing and management of shea (Vitellaria paradoxa) parklands in central-west Burkina Faso

Journal of Rural Studies 38 (2015) 27e38

Contents lists available at ScienceDirect

Journal of Rural Studies journal homepage: www.elsevier.com/locate/jrurstud

Gender, knowledge-sharing and management of shea (Vitellaria paradoxa) parklands in central-west Burkina Faso
ne Elias Marle Bioversity International, PO Box 236, UPM Post Office, Serdang, 43400 Selangor Darul Ehsan, Malaysia

a r t i c l e i n f o

a b s t r a c t

Article history: Received 25 September 2013 Received in revised form 12 January 2015 Accepted 26 January 2015 Available online

Research on agroforestry largely continues to ignore the significance of gender relations in shaping natural resource management processes. To contribute towards filling this gap, this study focuses on gender dynamics in Burkina Faso's centre-west region to examine how gendered knowledge and preferences affect the management and conservation of shea parklands. In-depth interviews and free-listing o, Lan and Prata reveal that despite a strongly exercises with Gurunsi and Moose women and men from Le gendered division of labour, women and men hold overlapping areas of knowledge about shea uses, yields and shea nut characteristics. Further, men and women farmers detailed the same management practices and factors guiding the selection and conservation of shea trees in cultivated fields. Similar fidelity levels (FLs) calculated from women's and men's responses show that top-cited uses, preferences and practices correspond across gender groups. This congruence is partly due to participants' personal experiences with the species, but also to knowledge sharing between the spouses that guides decisionmaking. Findings illustrate that the widely held assumption that men decide in matters of tree management overlooks the important contributions women may make to the process. The shea case suggests that intra-household knowledge sharing and collaboration may hold greater significance for achieving resilient resource management strategies than has been described in previous works on African agroforestry. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Shea Vitellaria paradoxa Burkina Faso Gender Indigenous knowledge Agroforestry

1. Introduction Native fruit trees provide crucial benefits to local ecosystems and livelihoods (Kalaba et al., 2009; Faye et al., 2010; Bayala et al., 2011). In sub-Saharan Africa, the shea tree (Vitellaria paradoxa C.F. Gaertn.) is indigenous to 18 countries across a 5000-km long and 500-km wide expanse of semi-arid savanna (Hall et al., 1996). Due to its myriad uses, farmers have selectively protected the species in their fields since 1000 A.D. (Neumann et al., 1998; Kahlheber, 1999). The result has been a discontinuous cover of scattered shea trees under which crops are grown (Pullan, 1974; Boffa, 1999). In these agroforestry parklands, which contain fewer trees than uncultivated lands, the species grows in nearly pure stands, illustrating the role of humans in shaping tree densities (Harlan et al., 1976; Boffa, 1999; Lovett and Haq, 2000a). Across the shea belt, shea trees are is prized for their nutritive fruit, medicinal properties, and hardwood, but also principally for

E-mail address: [email protected]. http://dx.doi.org/10.1016/j.jrurstud.2015.01.006 0743-0167/© 2015 Elsevier Ltd. All rights reserved.

their nuts, which women collect and transform into butter (Burkill, 2000). This butter represents the primary source of dietary fat for many agriculturalists living in the species' range and serves  important economic and cultural uses (Lykke et al., 2002; Pare et al., 2010). International demand for shea nuts and butter stems from their use in the agro-food and cosmetics industries, and provides significant export revenues to countries such as Burkina Faso, where shea nuts rank fourth among national export commodities (MEF, 2011). Yet, within West Africa, the importance of shea largely rests in the highly gendered context of its production and trade (Elias and Carney, 2007). The collection and transformation of shea nuts into butter, as well as the local sale of shea nuts and butter, are strongly associated with the female sphere of activity (Chalfin, 2004; Elias, 2010). Nonetheless, shea agroforestry is arguably guided by the specialized and interlocking knowledge repertoires and practices of women and men, who both value and use the species. To date, research on indigenous knowledge and management of shea trees has been limited to the species' eastern variety, V. paradoxa subspecies nilotica, and has not identified any

28

M. Elias / Journal of Rural Studies 38 (2015) 27e38

differences in women's and men's knowledge and management of the species (Gwali et al., 2011a, 2011b).1 Yet, the gendered dynamics surrounding the selective conservation and management of shea trees have not been systematically studied. Consequently, the aim of this paper is to examine the gendered ethnobotanical knowledge, management, and conservation of V. paradoxa subspecies paradoxa in the province of Sissili, Burkina Faso, where shea represents the most prevalent parkland tree species. As Howard (2003: 33) contends, “some researchers have attempted to develop conceptual frameworks to assess which factors motivate indigenous or peasant farmers to conserve biodiversity, but to date these have neglected to consider gender relations as potentially significant.” What is more, “examining gender helps us to understand how other forms of social difference influence rural environmental management, not just as ‘proxy’, but because other differences such as age, wealth or origins operate in genderdifferentiated ways” (Leach, 1994: 22). This paper draws attention not only to how women and men agriculturalists know, use, and manage shea trees, but also to how intra-conjugal knowledge sharing can inform decision-making and conservation strategies for the species. The discussion begins with a review of the literature on gendered use and management of trees, before turning to Amartya Sen's intra-household bargaining model of ‘cooperative-conflict’ that can foster an improved understanding of the gendered shea agroforestry system. After describing the study's context and methodology, I analyse the main findings related to the gendered dimensions of shea tree knowledge, management and conservation. I then examine the implications of these findings and the value of Sen's notion of ‘social connectedness’ for better conceptualizing natural resource management systems, before providing brief concluding remarks. 2. Gendered ecological knowledge and management of trees It is now widely acknowledged that gender is “a critical variable in shaping resource access and control, interacting with class, caste, race, ethnicity to shape processes of ecological change” (Rocheleau et al., 1996: 4). Throughout the world, use, knowledge, access, preferences and management of natural resources are primarily organized along gender lines. Historically-rooted and contextspecific norms and belief systems prescribe ‘appropriate’ behaviour for men and women and a gendered division of labour that guides resource use (Leach, 1994; Rocheleau et al., 1996). Largely due to their differentiated roles and responsibilities in production and reproduction, women generally collect forest products for food, fuel, fodder, medicine, and small-scale trade, whereas men contribute to these but also dominate the collection of animal products (e.g. through hunting) and the extraction of structural fibre such as timber for construction or sale (Jacobson, 1992; Gausset et al., 2005; Sunderland et al., 2014). In addition to gender specialization in the collection and processing of most types of forest products (Sunderland et al., 2014), gathering intensity differs as women collect a continuous flow of smaller quantities of forest products whereas men gather these sporadically and in larger quantities (Byers and Sainju, 1993). Customary and formal laws, which are also gendered, structure rights to forests and trees and influence incentives and capacity to manage tree resources (Fortmann and Bruce, 1988; Meizen-Dick et al., 1997; Quisumbing et al., 2001). In patrilineal systems, men

1

Gwali et al. (2011a, 2011b) have shown that Ugandan farmers across three farming systems recognized 44 ethnovarieties for shea, used the species for 36 different purposes, and managed it using various traditional practices.

have primary rights to land and natural resources whereas women's rights hinge upon their relationship with their male counterparts (Meizen-Dick et al., 1997). Men and women from the same household or community may have access to trees located in different spaces or even to different parts of the same tree. Hence, they often collect and use different products or gather the same products in different spaces (Wangari et al., 1996; Rocheleau and Edmunds, 1997; Howard, 2003). Gendered tenure regimes affect management strategies, as insecure access and control of forest and trees resources limits women's incentives to plant and manage trees to which their long-term rights are tenuous (Fortmann et al., 1997; Mukadasi and Nabalegwa, 2007; Howard, 2007). This gendered division of labour, use, and access to forest products and the multiple layers of institutions described above influence the knowledge and skills women and men acquire about tree resources (Wezel and Haigis, 2000; Howard, 2003; Ayantunde et al., 2008; Dovie et al., 2008). Such ecological knowledge is gleaned through first-hand experience but also through intergenerational and inter-group transmission in a “mostly unconscious, activity-situated, verbally and non-verbally communicated, observer-activated and learner-directed” process (Zent, 2009: 52). Factors such as kinship, age, specialization, and “motivation, ability and opportunities to learn” more generally intersect with gender to affect its acquisition (Boster, 1986: 434). Gendered patterns of ecological knowledge and resource use reflect and reinforce a gendered valuation of tree species, traits and products. Many authors have shown that preferences for tree species differ according to gender, with women often prioritizing species for household consumption and men those for sale (Kaur, 1991; Warner, 1993; Cavendish, 2000). Others suggest that this distinction has been overemphasized, but that gendered preferences do exist (Leach, 1994; Sunderland et al., 2014) and are conditioned by other factors of social differentiation such as marital status or age (Bonnard and Scherr, 1994). Gender also affects decision-making processes related to natural resources. Women are generally shown to be constrained in their ability to make decisions related to natural resource management. This may result from gender norms or technological biases, among others, that hinder their participation in formal decision-making forums such as forest user groups (FUGs) (Agarwal, 2001; Das, 2011) or in household-level decision-making processes (Abbas, 1997; Rocheleau and Edmunds, 1997; Chikoko, 2002). When women do participate in forest management committees, however, these institutions have tended to promote sustainable management practices as well as improved incomes from the forest (Upadhyay, 2005; Agarwal, 2009, 2010; Mwangi et al., 2011). Women's participation in decision-making processes thus shows promise for encouraging ecologically and economically sound resource management strategies. Drawing attention to these gendered specificities in natural resource systems is valuable for highlighting women's and men's respective contributions, opportunities and constraints in resource management processes. Yet, in this pursuit, the ways in which women and men often collaborate to help each other fulfil their gender-specific responsibilities remain poorly explored (Turner et al., 2000). This collaboration may represent a key feature of sustainable resource management strategies, and as such, requires further appreciation. 3. Sen's model Key notions from Amartya Sen's (1990) model of intrahousehold bargaining can improve our analysis of how women and men both specialize and cooperate in shea tree agroforestry. In his revised model of household economics, Sen conceives of the

M. Elias / Journal of Rural Studies 38 (2015) 27e38

29

Fig. 1. Study sites.

household as the site of coinciding and competing interests where women's and men's spheres of production, consumption, and decision-making are separate but interrelated.2 Although women and men manage activities in a specialized manner, an integrated view of how their activities sustain and support each other must thus be taken to understand production processes. To fulfil their gender-specific responsibilities, household members will cooperate when they believe that doing so is more favourable to them than non-cooperation. Various cooperative arrangements may exist and will differentially favour some household members over others. Sen explains that the nature of the cooperative arrangements will influence the intra-household distribution of benefits and the household's response to conflicts of interest. The arrangement that is ultimately pursued is negotiated based upon on each member's relative bargaining power, which primarily depends on the person's ‘breakdown position’ or the alternatives awaiting him or her in the absence of cooperation (see also Folbre, 1986; Udry, 1996; Carter and Katz, 1997). Socioeconomic changes, such as new economic opportunities arising from new markets or technologies for shea, can alter the state of this breakdown position and shift the balance of power within the household, thereby affect collective and individual production and welfare (Agarwal, 1997; Doss, 2001). Of particular value here is the emphasis Sen places on the existence of separate domains of specialization that are articulated within the household and inform intra-household decision-making, and the ‘social connectedness’ of husband and wife who: live together under the same roofdsharing concerns and experiences and acting jointly. This aspect of ‘togetherness’ gives

2

This contrasts with Becker's (1981) household-welfare-function modeldthe best-known model of household resource allocation among economistsdthat sees household members as sharing a common set of preferences and a single decisionmaking logic. Becker's model is also referred to as the ‘common preference’ or ‘unitary’ model.

the gender conflict some very special characteristics. One of these characteristics is that many aspects of the conflict of interest between men and women have to be viewed against the background of pervasive cooperative behaviour (Sen, 1990: 147). This complex backdrop of gendered specialization, sharing, cooperation and conflict (or divergent interests) sets the stage for understanding the indigenous shea agroforestry system in Burkina Faso. 4. Setting the scene 4.1. Study sites o and the villages of This study was carried out in the town of Le Lan and Prata in Burkina Faso's province of Sissili, which occupies the country's central-west region (Fig. 1). Lan and Prata sit on the Ghanaian border and are situated nine kilometres south and 49 km o, respectively. The three sites were selected because of east of Le their women residents' membership in one of the largest unions of shea butter producers in the country, the Union des groupements de productrices de produits karit e de la Sissili et du Ziro (UGPPK-S/Z).3 Due to their location in a shea tree-rich zone and to their proximity to the Ghanaian border, the three sites have long been integrated in local and international shea butter markets. Local annual rainfall ranges between 800 mm and 1000 mm, with peak rainfall occurring between May and October (Guinko,

3 The UGPPK-S/Z, now renamed the Federation NUNUNA, was created within the context of international aid projects promoting women's empowerment through international shea butter sales. The union is composed of over 3000 women producers organised in 33 village women's groups. It is headquartered in the town of o and encompasses members from the surrounding villages of Lan and Prata, Le among others.

30

M. Elias / Journal of Rural Studies 38 (2015) 27e38

Table 1 Women and men's endogenous perceptions of the factors influencing shea yields. Factor

Climatic factors Biological factors Anthropogenic factors Tree traits

Strong winds High rainfall Semi-parasitic plants Predatory insects/fungus Farming practices Flowering behaviour (abundant, post-Harmattan) Leafing behaviour (extensive canopy, full leaf flush) Size ‘Gender’: male tree


s and Guinko, 1995). The region sits within the southern 1984; Fonte Sudanian zone and is species-rich compared to the northern parts of the country. Over the past 20 years, an influx of migrants from the country's northern and central areas has joined the province's indigenous Gurunsi inhabitants. These migrants, who mainly originate from the country's dominant Moose ethnic group, locally outnumber the Gurunsi (Howorth and O'Keefe, 1999a, 1999b). Nonetheless, Sissili's population density (21.5 people/square kilometre) remains low with respect to the rest of the country draogo, 2003; INSD, 2006). (Oue Contemporary Gurunsi and Moose agriculture is based on household production and silviculture, integrating some animal husbandry with subsistence and cash crop cultivation (Prudencio, draogo, 2003). Extensive agricul1993; Breusers et al., 1998; Oue ture is practiced with limited access to plough-animal traction, and soil fertility is maintained by rotational fallow, despite a shortening of fallow periods (Gray, 2003). Agroforestry is an integral component of local farming systems. Gathering of non-timber forest products (NTFPs), including shea fruit, shea nuts, leaves, medicinals, oils, and resins, occupies a central place in farmer livelihoods. Customary land ownership, which remains the norm in Burkina draogo, Faso, is lineage-based (Reenberg and Lund, 1998; Oue 2003). Descendants from the ‘first occupants’ who settled the land and brought it into productive use have privileged land rights draogo, 2003). They allocate tempo(Kevane and Gray, 1999; Oue rary or extended tenure rights to ‘latecomers’ or ‘strangers’ through a transfer transaction (Reenberg and Lund, 1998). In exchange for their labour on cooperatively farmed household fields, women and young adult men have usufruct rights to personal fields on which they individually retain rights to the products they grow (Marchal, 1983). Social institutions and taboos regulate access to tree species such as shea, the planting and felling of which are customarily restricted. Burkina Faso's forestry laws also forbid culling shea trees without a permit, but as in other parts of West Africa, the state lacks the resources to enforce these laws (Schreckenberg, 1999; Wezel and Haigis, 2000; Leach et al., 2011).

4.2. Methods Data for this study were collected in the three study sites between October 2006 and March 2007 with 90 women and 78 men of Gurunsi and Moose descent, the two dominant ethnic groups living in the study area.4 Specifically, 50 women and 45 men were of Gurunsi origin, whereas 40 women and 33 men were Moose. o, Lan and Prata) were Thirty women participants from each site (Le randomly selected from the roster of the UGPPK-S/Z, in which most

4 Slightly more women than men were interviewed because some of the female respondents were widows while the spouses of others were unable to participate.

Per cent of times cited

Effect on yields

Women (n ¼ 90)

Men (n ¼ 78)

13 5 38 3 10 28 15 10 3

8 8 31 2 39 10 20 22 4

Y [ Y Y [ [ [ [/Y Y

able-bodied adult, married Gurunsi and Moose women from local villages hold membership. This represents over one quarter of the women living in Lan and Prata, and follows Brewer et al.'s (2002) observation that twenty to thirty research participants generally suffice to identify a coherent set of items using free listing techniques, as was done in this study, when individuals share a reasonable degree of knowledge about the topic at hand. The number of participants selected proved to be sufficient to reach the point of data saturation in all three sites. The women respondents' husbands were also interviewed to enable cross-gender comparisons and assess the gender dynamics at play within the shea agroforestry system. Each participant was interviewed individually, privately in his or her compound or fields. Interviews were conducted using free listing techniques to collect data without initial prompts. This basic ethnographic tool consists of asking respondents to list as many ‘Xs’ as they can, where ‘X’ refers to a cultural domain, such as names of trees, uses for products, and so on (Weller and Romney, 1998; Brewer et al., 2002). Respondents were thus able to elicit and subsequently discuss any and as many points as they wanted for every question posed. Following Gatewood (1983, 1984) and Borgatti (1990), the number of responses participants gave in this way, or their “free listing capacity” (Brewer, 1995: 108), was then used as a loose indicator of their knowledge of the issues at hand. Questions primarily centred on participants' knowledge of the factors influencing shea yields, characteristics of quality shea nuts, and uses for the shea tree and its derivatives. As uses for shea butter are extensively documented and tend to outshine the species' other applications (Maranz et al., 2004), participants were asked to focus on the tree's non-shea butter related functions. Questions additionally revolved around shea tree management practices and the selective conservation of shea trees. Open-ended, follow-up questions were then asked to ‘thicken’ the discussion of the role the shea tree plays in the livelihoods and farming systems of participants, their shea tree management practices, skills and underlying knowledge repertoires. Men's and women's responses were open coded manually according to recurring themes (Stemler, 2001). A master list of key thematic categories was created and the frequency of responses that fell within each established category was calculated separately for women and men. Differences and overlaps across gendered responses were qualitatively analysed both in terms of these frequencies and in terms of the range of categories cited by women and men participants. Following Friedman et al. (1986), fidelity levels (FL) were calculated using the formula FL ¼ (Ip  100)/IU, where Ip is the number of participants who independently cited a specific variant (e.g. a specific use for shea) for a given domain of analysis (e.g. uses of shea) and IU is the total number of responses provided for all variants in a given domain of analysis (e.g. the total number of responses cited by participants for all the different uses

M. Elias / Journal of Rural Studies 38 (2015) 27e38

for shea). The FL indicates the degree of consensus among participants concerning the variants within a given domain of analysis. FLs were calculated separately for women and men. This data was interpreted in tandem with the rich interview data that complemented and explained the reasons behind the observed patterns. 5. Gendered knowledge repertoires 5.1. Shea yields and nut traits Participant women and men demonstrated a detailed understanding of the factors influencing shea yields, including those stemming from the environment in which the tree grows and those associated with specific tree characteristics. As shown in Table 1, male and female respondents correlated shea yields with the same climatic, biological, and anthropogenic factors, albeit in different proportions. There was high consensus among women and men that semi-parasitic plants are the most serious problem afflicting shea trees in the study sites (FL ¼ 30.6 and 21.8 for women and men, respectively) (Table 2). Women and men also recognized that the strength and timing of winds affects the ability of trees to maintain their flowers and eventually fruit. Nearly 40 per cent of the men and ten per cent of the women interviewed also noted that

31

anthropogenic practices such as cultivating, burning and felling closely clustered trees affect shea yields, as shea trees in farmed fields are larger, benefit from nutrient inputs from fertilizers, and are protected from fruit pillaging squirrels and bush fires (FL ¼ 27.7 for men and 8.2 for women). Men and women participants further recognised that the top three traits to influence an individual tree's yields were its flowering and leafing behaviour as well as its size (Table 2). A few women and men also reported the effect of tree ‘genders’ on yields, with one older Moose man explaining that “among all tree species, some individuals are non-productive by nature: they do not flower or yield fruit. They are called ‘male’ trees because they cannot bear children; they only change beauties by shedding their leaves with the seasons.” A younger male Moose respondent specified that some ‘male’ shea trees (Moore: Tam daaga; Gurunsi: Souanb
e) produce no flowers, whereas others flower but produce no fruit or very small fruit that bear no nuts. All female participants and 95 per cent of male participants claimed to know which shea trees provide the best yields in their fields, with some respondents ascribing names to these trees. When asked whether the shea trees offering good yields remain the same from year to year, 70 per cent of male participants answered in the affirmative, explaining that these trees also produce around the same period and the same number of times a year.

Table 2 Comparison in knowledge, practices and selection preferences related to shea across gender groups. Category

Knowledge

Variant

Factors affecting productivity

Environmental/anthropogenic

Tree characteristics

Uses for shea tree and derivatives

Management practices

Selection factors

Semi-parasitic plants Winds Farming practices Rainfall Predatory insects Flowering behaviour Leafing behaviour Size Gender

SF Medicine Shade Firewood Nutrition (fruit) Soap production (Shea nut residues) Fertilizer (leaves) Fodder (semi-parasitic plants) Fire ignition (Shea nut residues) Micro-climate improvement Construction Insect/fungal repellent (bark, burned) Honey production (beehives in tree) SF Cultivate/weed beneath tree Optimize tree spacing Parasite removal Pruning Build mound around tree Dig water retention ditch around tree Fertilizer application in field Refrain from bark removal Control field burning SF Yield Spacing Shading effects Size/age Vulnerability to parasites Fruit and nut characteristics SF

Women (n ¼ 90)

Men (n ¼ 78)

F

FL

F

FL

35 12 9 5 2 25 14 9 2 113 56 25 22 18 5 3 3 2 1 0 0 0 135 51 28 18 13 6 1 1 1 0 119 30 30 30 22 22 3 135

30.6 10.2 8.2 4.1 2.0 22.5 12.3 8.2 2.0 e 41.5 18.5 16.3 13.3 3.7 2.2 2.2 1.5 0.7 0.0 0.0 0.0 e 42.9 23.5 15.1 10.9 5.0 0.8 0.8 0.8 0.0 e 22.0 22.0 22.0 16.0 16.0 2.0 e

24 6 31 6 2 8 15 17 3 112 37 12 8 31 3 12 3 2 2 3 2 1 116 59 20 34 25 10 2 2 0 5 157 32 26 26 16 12 5 118

21.8 5.5 27.7 5.5 1.4 6.8 13.7 15.0 2.7 e 31.9 10.3 6.9 26.7 2.6 10.3 2.6 1.7 1.7 2.6 1.7 0.9 e 37.6 12.7 21.7 15.9 6.4 1.3 1.3 0.0 3.2 e 17.0 13.7 13.7 8.7 6.6 2.9 e

n ¼ number of participants; F ¼ number of times response was cited; SF ¼ total number of responses cited by gender group in given category; FL ¼ fidelity level.

32

M. Elias / Journal of Rural Studies 38 (2015) 27e38

In contrast, the same proportion of female interviewees believed that good producers vary inter-annually. This last observation lends better support to the existence of productivity cycles, reported by both male and female participants, in which trees (and individual branches) that produce well for one or two years show lower yields the following year or two and so on. Nearly all women participantsdwho are the ones who process shea nuts into butterdwere also attuned to the characteristics of shea nuts yielding superior quantities of quality shea butter; namely small-size, closed shell, hard, reddish colour and windfalls (nuts collected from the ground) that are commonly found in uncultivated, bush fields. Despite the fact that men, as per local gender roles, do not produce shea butter, two-thirds of male interviewees acknowledged that hard, reddish, fallen shea nuts are best for making shea butter, provided that they are properly transformed. A handful of women and men participants elucidated that male spouses acquired this knowledge by observing women's activities and discussing issues such as shea nut quality with their wives. Since shea butter is frequently prepared within the homestead, men may be present during certain processing stages, such as during shea nut sorting. While sorting shea nuts, which is among the least physically demanding steps in the butter making process, shea butter producers are seated on the ground and able to speak among themselves and with other household members. This creates a privileged setting for conversing with husbands, who are often seated in a nearby shaded area of the courtyard. Such times are ripe for discussions about shea, especially since development projects and economic trends have increased the monetary value of

the commodity and men's interest in the shea trade. The majority of women respondents (70 per cent) noted that quality shea nuts used for making butter consistently come from the same trees, whereas a handful of male participants explicitly felt that this is not a man's concern. One middle-aged Gurunsi man explained that, “sometimes I see women walk by certain shea trees without collecting their nuts so I know they must not be of good quality. However, I don't know which trees give good nuts, and this is not my problem.” Male respondents mostly showed interest in fruit taste, with all male participants claiming to know the trees producing sweet tasting fruit. Male and female respondents explained that each tree's potential to provide these fruit is unique and cannot be determined based on specific characteristics. Moreover, participants did not perceive any correlation between shea fruit and nut traits. In sum, female and male participants correlated the same range of factors with shea yields, but cited these in different frequencies. There was high consistency across genders in terms of the top factors believed to influence shea yields, although men were more attuned to the influence of anthropic practices on shea yields, likely because they are the ones to carry out many of the farming and burning practices reported. Overall, men and women felt differently about whether specific shea tree individuals consistently produce good yields. Nearly all women and most men claimed to recognize good quality shea nuts, but men could not identify the trees providing these nuts. As discussed below, these issues hold significance at the time of tree selection and can influence shea tree conservation decisions.

Fig. 2. Uses for the shea tree and its derivatives according to women and men participants.

M. Elias / Journal of Rural Studies 38 (2015) 27e38

33

Fig. 3. Women's and men's shea tree management practices in the study sites.

5.2. Shea uses Shea nuts and fruit are only two of the shea tree's non-timber forest products of local importance. Agriculturalists also rely upon the species' bark, roots, latex, and leaves to meet their daily needs. Aside from shea butter, which all participants praised for its nutritional and economic value, male and female participants reported similar uses for the species, which require an intimate knowledge of the properties and processing of its derivatives. As

Fig. 4. Factors guiding shea tree selection among women and men participants.

shown in Fig. 2 and Table 2, both women and men most frequently cited the species' role in traditional pharmacopeia (FL ¼ 41.5 and 31.9 for men and women, respectively). Shea-based medicine is administered to persons of all ages to treat a variety of lifethreatening diseases such as dysentery and malaria. As expected, women, who are the ones to process the cures prepared with shea tree leaves, bark, latex, and roots, were more articulate about the tree's medicinal properties and the preparation and administration of shea-based remedies. Men were also aware of the diseases shea derivatives could cure and valued the species' medicinal properties, particularly given the poor availability and prohibitive cost of Western medicine and hospitalization. Yet, they were unable to describe how these cures are prepared. Aside from medicine, men and women emphasized the species' role in the provision of shade and nutrition. The high FL value (26.7 and 13.3. for men and women, respectively) illustrates the nutritional importance of shea fruit, which ripen during the agricultural period and quell the hunger of women, men and children while in the fields. In fact, one third of male respondents cited shea fruit consumption as the primary reason they conserve shea trees in their fields. Women also stressed the importance of shea as firewood, which logically follows the local gendered division of labour that ascribes gathering firewood to the female sphere of activities. Neither participating women nor men emphasized the species' importance in providing ecological services aside from the soil fertilizing effects of its leaf litter, which was mentioned by 15 per cent of male respondents and only three per cent of the women.5 This may be because the shea tree's other benefits are more immediately evident for ensuring the subsistence of male and female farmers than the ecological services it offers. Many

5 Shea leaf litter serves as compost (Bambara, 1993; Bayala et al., 2003, 2005); shea tree roots preserve soil integrity (Bonkoungou, 1987); the trees serve as wind barriers (Kessler, 1992; Kessler and Breman, 1991; Kainkwa and Stigter, 1994); and they improve soil water retention and infiltration (Boffa et al., 2000), as well as providing other ecological benefits.

34

M. Elias / Journal of Rural Studies 38 (2015) 27e38

respondents were adamant that shea is the most important local species, and nearly all men and women felt that its conservation is of critical importance. 6. Shea tree management Guided by the repertoires of ecological knowledge outlined above, the women and men interviewed have adopted agroforestry practices that deliberately or inadvertently improve the species' vigour and productivity. These practices represent a form of ‘protoculture’dor management of valued trees that are not deliberately planted (Boffa, 1995). Consequently, shea parklands reflect not only the shea tree's natural regeneration patterns, but also the management and conservation practises of female and male agriculturalists who value the species' myriad purposes. Although men are the visible managers of shea trees in West Africa, carrying out field burning and tree culling (Maranz and Wiesman, 2003; Chalfin, 2004), most women respondents demonstrated a detailed understanding of shea tree management practices. As shown in Fig. 3 and Table 2, women described the same management techniques as their spouses, and there was a high degree of consensus across genders around the top four practices, namely cultivating and weeding beneath shea trees (FL ¼ 42.9 for women, 37.6 for men), optimizing the spacing of shea specimens (FL ¼ 23.5 and 12.7 for women and men), removing semi-parasitic plants (FL ¼ 15.1 to 21.7 for women and men) and pruning (FL ¼ 10.9 for women and 15.9 for men). On average, male interviewees named more management practices than their female counterparts (2.1 versus 1.5 responses per participant, respectively), likely because they are generally the ones to perform some of these practices, such as pruning, removing parasites or field burning. 7. Shea tree conservation Coupled with the fact that it is not a planted species, does not depend on costly fertilizers for growth, and is well adapted to local ecological conditions, the shea tree offers relatively high returns on labour. According to interviewees, these returns justify the species' conservation in cultivated fields. All female and male interviewees stated that they conserve shea trees in their fields, selectively retaining specific individuals with favourable characteristics and eliminating unwanted specimens. Fig. 4 and Table 2 demonstrate that women and men cited similar, interrelated factorsdnamely yield, spacing, and shading effectsdas the primary variables influencing the selection process. The proportion of men and women reporting each of these factors closely matched, as did the FLs calculated for each gender group. In new fields, male interviewees stated that they might assess shea yields for as many as four years before eliminating the weakest producers. This makes sense given that 70 per cent of women and 30 per cent of men respondents believe that the best producing trees vary across years. Alternatively, participants rely on visible tree characteristics, including shape and size/age, to guide their choices. Obtrusive trees and/or branches are removed, and given the ambiguous relationship between tree size/age and productivity, some participants retain older trees while others conserve younger individuals. A specimen's vigour or resistance to parasitism also affects its chances of being retained. Surprisingly, neither women nor men gave fruit and nut characteristics much importance in the selection process, with the quantity of nuts produced largely trumping quality considerations.6

6 In contrast, Ghanaian farmers considered fruit taste a primary criterion for shea tree selection (Lovett and Haq, 2000b).

When asked whether they participate in the shea tree selection process, just over half of female participants stated that their husband single-handedly selects shea specimens for conservation. The others claimed to discuss the selection process with their husband, stating that women recognise the best producers. Hence, although men may ultimately choose which specimens to eliminate, female spouses guide their decision. In a woman's personal field, she alone decides which shea trees to conserve. In turn, nearly three quarters of male participants maintained that they discuss the selective conservation of shea trees with their wives before eliminating any specimen, as they claimed that women are more knowledgeable about shea tree yields and shea nut quality. The remaining quarter of respondents explained that women are not always in the fields to identify shea trees at the time of land clearanceda man's taskdand that this hinders their participation in the selection process. Women's input in shea tree selection may also occur in a less formal manner. For instance, a middle-aged Gurunsi woman recounted arriving in one of her household's cooperatively farmed fields just as her husband was preparing to fell the branch of a productive shea tree that was shading his cotton crop. She promptly requested that he conserve the branch due to the tree's good quality and generous yields. The husband followed her advise and adapted his management strategy accordingly. In sum, female and male participants citeddin closely matching frequenciesdidentical factors guiding the selection of shea trees in farmed fields. The majority of participants reported selectively conserving shea trees in a process guided by female and male knowledge of tree yields, as well as by their preferences for specific shea tree productivity, spacing, and shading patterns. The implications of these findings for the ways we conceptualize gendered experiences with natural resources are discussed below. 8. Discussion: gendered knowledge, intra-conjugal consultations and shea tree management decisions Five important points arise from an analysis of the gendered shea agroforestry system. First, both local women and men hold rich, if at times covert, ecological knowledge that can inform locally-relevant research on the species. Overall, the botanical
women and men corresponds knowledge described by burkinabe with the ways of using and processing shea tree derivatives described in other regions of the species' range (Gwali et al., 2011a, 2011b; Mapongmetsem et al., 2011; Poudyal, 2011; Okiror et al., 2012). Participant knowledge of shea is congruent with scientific findings about the species, although participants also recognized tree traits affecting productivitydsuch as the tree's leafing behaviour or its ‘gender’dthat had yet to be identified in other studies on shea agroforestry.7 In particular, the so-

7 Previous studies have reported vast fluctuations in annual shea yields (Hall et al., 1996; UNCTAD, 2006), with tenfold variations occurring across consecutive lerot, 1995), and five-fold differences arising between the best and worse years (Cre producing trees (Boffa, 1995). The scientific community has ascribed these fluctuations to abiotic factors such as the winds (Okullo et al., 2004b), soil moisture (Hall et al., 1996), fires (Abbiw, 1990; Hall et al., 1996), as well as the tree's location in fields versus in natural formations (Lamien et al., 2004). Moreover, researchers attribute high shea tree mortality rates to the semi-parasitism of several vascular plant species belonging to the Loranthaceae family (Bonkoungou, 1987; Boussim et al., 1993a, 1993b, 2004). Shea tree characteristics such as the tree's flowering period (Okullo et al., 2004a; Kelly et al., 2007) and size (Ruyssen, 1957; Boffa, 1995; Lamien et al. 2007) also affect productivity, although the relationship between shea tree size and yields is ambiguous. Three to five year productivity cycles are also widely reported among both participants and researchers (Hall et al., 1996; Chalfin, 2000; Bayala et al., 2008).

M. Elias / Journal of Rural Studies 38 (2015) 27e38

called ‘gender’ trait, also reported by West African farmers with respect to the baobab tree (Assogbadjo et al., 2008), exemplifies the value of exploring the local perceptions of those who use and manage natural resources. Studying the ‘male’ trees identified by research participants, for instance, can enhance our understanding of the factors stunting shea yields and guide research for development efforts promoting the use and sale of shea tree derivatives. The management methods reported by the female and male respondents are also in line with those recommended by the scientific community (Kelly et al., 2004; Akais Okia et al., 2005). As above, however, male and female interviewees described additional practices such as the input of fertilizer and the creation of earth mounds or water retention ditches that have yet to receive research attention. Given the rapid expansion of fertilizer-intensive cotton cultivation in the province of Sissili and elsewhere in the country (Kaminski et al., 2009), the use of fertilizers was of particular concern to interviewees who sometimes correlated it with favourable shea yields and other times noted that it promotes the growth of semi-parasitic plants that stunt yields. The net effect of fertilizer additions on tree productivity requires further investigation. Despite the many management practices associated with the species, the shea tree continues to be considered ‘wild’ by local people because it is not a planted species. Many participants originally failed to recognize the techniques they use to enhance the species' productivity, partly because many of these are not performed explicitly for tree management purposes. For instance, semi-parasitic plants are primarily removed to provide fodder for domestic animals, and synthetic fertilizer is applied to improve cotton and maize yields. This, as well as the free-listing method used to collect datadwherein participants were asked to discuss ideas that came to mind without guiding promptsdlikely accounts for the 22 per cent of women and 12 per cent of men participants who claimed not to manage their shea trees in any way (Fig. 3). As mentioned earlier, the fact that fewer women than men identified ways in which they manage shea trees may be because some of the management techniques reported are specifically carried out by men. Yet, it is also possibly a function of the general lack of value local women attribute to their work and knowledge as compared to their male counterparts. In-depth interviews and observation are useful ways of encouraging such undervalued or ‘covert’ knowledge and practices to surface during the research process and to promote self-awareness of this knowledge among participants. A second significant point is that, although gender differences in ethnobotanical knowledge have been widely reported in Africa (Cotton, 1996; Wezel and Haigis, 2000; Ayantunde et al., 2008), such differences were not generalized in this study. Consistent with the local gendered division of labour, men and women listed the tree's various uses in distinct frequencies. Nonetheless, the range of their responses and the top-ranking responses they provided for each domain of analysis showed strong overlap. This can be explained by the central and constant importance the tree occupies for all village residents, who draw upon it for daily subsistence. Boster (1986) has indeed demonstrated that the greater the shared cultural importance of a species, the more knowledge about the species will be shared intra-culturally. Similar observations have been made in Uganda, where women's and men's classification knowledge of V. paradoxa (subspecies nilotica) ethno-varieties strongly matched (Gwali et al., 2011a) and in south-central Burkina Faso, where no consistent gender differences were found in Gurunsi identification of locally important species (Kristensen and Balsev, 2003; Kristensen and Lykke, 2003). Although women's knowledge of shea nut quality exceeded men's, two-thirds of male interviewees were surprisingly knowledgeable about the topic,

35

citing the same desirable shea nut traits as their female counterparts. Given that men do not process shea nuts into butter, this area of male knowledge cannot be attributed to first-hand experience. The third point, related to the second, is that although rural African men are generally perceived to be associated with tree management and conservation, both women and men may be knowledgeable about and involved in these processes. Again, the large areas of overlap in their responses and the similarly high FLs for a given variant across gender groups suggest that women and men value common factors with respect to shea tree selection and that they collaborate in the species' management and conservation. This is also confirmed by participant reports of intra-conjugal consultations during shea tree selection activities. Hence, men may play a preponderant role in the process, but in the case of shea, which is strongly associated with the female sphere of activities, women contribute their knowledge and preferences to the process in a more or less formal manner. The fourth point, arising from the last two, is the need for a more fluid conception of how traditional ecological knowledge develops and circulates within the household. This knowledge is related to a person's social identity features such as gender (Wangari et al., 1996; Rocheleau and Edmunds, 1997; Camou-Guerrero et al., 2008), yet it is not bound by these. Despite the gendered nature of resource use, local knowledge of valued resources is shared through daily interactions among spouses. This may occur to meet practical goals, such as developing resource management strategies, or simply for the sake of conversation between spouses. It is, in fact, difficult to imagine that in subsistence-oriented communities husband and wife would not discuss matters of importance to their daily subsistence, such as resource use. In the shea case, some of the congruence between male and female responses can likely be attributed to such knowledge sharing among spouses. It is in this respect that drawing upon Amartya Sen's model of intra-household bargaining can broaden our understanding of the gendered shea tree agroforestry system. As mentioned earlier, the notion of ‘social connectedness’ advanced by Sen recognizes that women and men have different spheres of activities and decision making, but challenges the assumption that these gendered spheres are separable. Summing up this idea, Razavi and Miller (1995: 15e16) assert that, “although the gender division of labour involves men and women undertaking different activities, it also entails an intricate and changing system of cooperation and exchange.” Bonnard and Scherr (1994: 89) further state: the existence of joint decisions, multiple objectives, and mutual dependence within the household makes it difficult to construct generalized rigorous and pragmatic models of household or individual behavior. Even if men are the only ones planting trees, women may still provide the critical watering and pruning labor, and their relevant opportunity costs will greatly influence the survival rates of project trees. While men use a specific species for poles, women might use them for fodder, green manure or special herbs and medicine depending on the species selected. There may be differentiated preferences for secondary tree products as well. Joint management and decision making naturally arise out of these scenarios. Likewise, Lope-Alzina (2007: 34) describes the selection of crop varieties in Mexico as “a process of negotiation: decision-making is identified as a prerogative exercised by both men and women and based on the knowledge and skills that each hold, and the trust and recognition of these by their partners.” This aspect of ‘social connectedness’ is frequently overlooked in studies describing the relationship between African spouses primarily as one of separation or conflict, and spheres of female knowledge and male

36

M. Elias / Journal of Rural Studies 38 (2015) 27e38

management of tree resources as chiefly independent. Results from this study suggest the need to more readily recognize the knowledge transfers occurring within the household that allow for informed decision making in the natural resource management and conservation arenas. This information sharing is  and Lassoie (2011: 255) have termed a step towards what Asse ‘gender-inclusive decision making’ on parkland issues, wherein “female adult members are explicit joint partners in choosing natural resource use and conservation actions.” In Sudano-Sahelian Mali, this type of decision-making culture was associated with innovative and adaptive parkland soil and tree management prac and Lassoie, 2011). Also of importance will be attributing tices (Asse attention to factors such as age or seniority within the household that influence the knowledge held and ability to make or contribute to resource management decisions. Finally, the shea case hints at the rapidly evolving nature of gendered knowledge and resource management practices over time. Shea butter is currently gaining value due to international market trends and development projects that seek to promote women's empowerment through shea sales. As this happens, male spouses are showing interest in shea nut processing, and seeking what may be broader knowledge about the resource from their shea producer wives. As the local and international political economy is in continuous flux, so too is intra-household knowledge sharing that influences the collaborative and competing resource use and management decisions of husbands and wives. Stringing these points together illuminates the ways the different layersdknowledge, decision-making and practicesdthat comprise resource management systems are articulated.8 Gendered knowledge is acquired and shareddor not, depending on the cooperative or competing interests of household membersdto inform decisions about resource management. Members choose which information to share and which to withhold based on their collaborative or competing interests.9 When a set of resource management outcomes is strongly desirable to both husband and wife, spouses have high incentives to collaborate in and thus exchange relevant knowledge. Decisions, which may ultimately be made jointly or not, then partly hinge on this layer of knowledge acquisition and exchange. The layer of practice, which is most visible and often conveys the impression that men are the primary resource managers, cannot be detached from the more covert knowledge and decision-making dimensions upon which it rests. The gender dynamics that shape these layers are in constant evolution as intra- and extra-household factors that bear upon them fluctuate. Each of these gendered layers comprises the cooperative and conflictive elements described by Sen; and all must be considered together to understand natural resource management strategies.

9. Conclusion This study has shown that women and men have distinct but strongly overlapping areas of knowledge about the shea tree, its

8 Similarly, Berkes (1999: 8) defines traditional ecological knowledge as having dimensions of knowledge, practice, and belief. The gendered norms clearly that underlie the acquisition of gendered knowledge and practice were beyond the scope of this paper, but comprise yet another layer of resource management systems. 9 This knowledge plays a role in strengthening (or weakening, in its absence) the relative bargaining position of household members. Although the relationship between knowledge and power has been extensively demonstrated by Foucault, among many others, the role knowledge plays in determining the relative bargaining position of household members is rarely discussed in intra-household bargaining models.

uses, as well as its management and conservation. This can be partly attributed to their personal experiences with the species, but also to knowledge sharing between spouses. In particular, the notion that men decide in matters of tree management and the visibility they receive as primary resource managers overlooks the fact that women's contributions to the process may be significant. The case of shea suggests that such intra-household sharing of knowledge about valuable natural resources may hold greater significance for achieving resilient resource management strategies than has been described in previous works on African agroforestry. In particular, recognizing that knowledge, decision-making and practices are interlaced and shaped by gendered relations of cooperation and conflict can improve our understanding of resource management processes. Shea parkland conservation programs are being widely implemented throughout the shea belt. These initiatives, as well as studies on shea, must enlist the cooperation of local agriculturalists whose endogenous knowledge of the species' ecology, management and particular challenges can guide the creation of culturally sensitive and ecologically rational arboreal management programs. In these undertakings, working with women and men farmers will be essential given the roles both spouses play in shea tree management and conservation processes. Acknowledgements The author gratefully acknowledges the Social Sciences and Humanities Research Council (SSHRC) of Canada, and the International Development and Research Centre (IDRC) for funding this o, research. Heartfelt thanks are due to the men and women from Le Lan and Prata who generously contributed their time and knowl, edge to this study; to Reid Cooper, Azizou Yago, Fatimata Traore draogo and Nassiratou Ne bie  for field assistance; and Pamoussa Oue to Sarah Turner, Stephanie Coen and two anonymous reviewers for their valuable comments on the manuscript. References Abbas, J.D., 1997. Gender asymmetries in intra-household resource allocation in sub-Saharan Africa: some policy implications for land and labour productivity. In: Haddad, L., Hoddinott, J., Alderman, H. (Eds.), Intrahousehold Resource Allocation in Developing Countries: Models, Methods and Policy. Johns Hopkins University Press for the International Food Policy Research Institute, Baltimore. Abbiw, D.K., 1990. Useful Plants of Ghana: West African Uses of Wild and Cultivated Plants. Intermediate Technology Publications and The Royal Botanic Gardens, Kew, London. Agarwal, B., 1997. ‘Bargaining’ and gender relations: within and beyond the household. Fem. Econ. 3 (1), 1e51. Agarwal, B., 2001. Participatory exclusions, community forestry and gender: an analysis for South Asia and a conceptual framework. World Dev. 29 (10), 1623e1648. Agarwal, B., 2009. Gender and forest conservation: the impact of women's participation in community forest governance. Ecol. Econ. 68, 2785e2799. Agarwal, B., 2010. Gender and Green Governance: the Political Economy of Women's Presence within and beyond Community Forestry. Oxford University Press, New Delhi and Oxford. Akais Okia, C., Obua, J., Agea, J.G., Agaro, E., 2005. Natural regeneration, population structure and traditional management of Vitellaria paradoxa subspecies nilotica in the parklands of northern and eastern Uganda. Afr. Crop Sci. Conf. Proc. 7, 1187e1191. , R., Lassoie, J.P., 2011. Household decision-making in agroforestry parklands of Asse Sudano-Sahelian Mali. Agrofor. Syst. 82, 247e261.
le
Kakaï, R., Chadare, F.J., Thomson, L., Kyndt, T., Sinsin, B., Van Assogbadjo, A.E., Gle Damme, P., 2008. Folk classification, perception and preferences of baobab products in West Africa: consequences for species conservation and improvement. Econ. Bot. 62, 74e84. Ayantunde, A.A., Briejer, M., Hiernaux, P., Udo, H.M.J., Tabo, R., 2008. Botanical knowledge and its differentiation by age, gender and ethnicity in Southwestern Niger. Hum. Ecol. 36, 881e889.
re organique selon les syste
mes de culBambara, D., 1993. Dynamique de la matie ture dans les sols agricoles du finage de Thiougou. Approche quantitative (Ph.D.  de Ouagadougou, Ouagadougou. thesis). Universite draogo, S.J., Eyog-Matig, O., 2011. Editorial for Bayala, J., Lamien, N., Butare, I., Oue

M. Elias / Journal of Rural Studies 38 (2015) 27e38 special section: food tree species and poverty alleviation. New. For. 41, 277e279. Bayala, J., Ouedraogo, S.J., Teklehaimanot, Z., 2008. Rejuvenating indigenous trees in agroforestry parkland systems for better fruit production using crown pruning. Agrofor. Syst. 72, 187e194. Bayala, J., Mando, A., Ouedraogo, S.J., Teklehaimanot, Z., 2003. Managing Parkia biglobosa and Vitellaria paradoxa prunings for crop production and improved soil properties in the sub-Sudanian zone of Burkina Faso. Arid Land Res. Manag. 17 (3), 283e296. Bayala, J., Mando, A., Teklehaimanot, Z., Ouedraogo, S.J., 2005. Nutrient release from  (Vitellaria paradoxa) and Ne re  (Parkia decomposing leaf Mulches of karite biglobosa) under semi-arid conditions in Burkina Faso, West Africa. Soil Biol. Biochem. 37, 533e539. Becker, G.S., 1981. A Treatise on the Family. Harvard University Press, Cambridge, Mass. Berkes, F., 1999. Sacred Ecology. Traditional Ecological Knowledge and Resource Management. Taylor and Francis, Philadelphia and London, UK. Boffa, J.M., 1995. Productivity and Management of Agroforestry Parklands in the Sudan Zone of Burkina Faso, West Africa (Ph.D. thesis). Purdue University, Purdue. Boffa, J.M., 1999. Agroforestry Parklands in Sub-saharan Africa. In: FAO Conservation Guide. FAO, Rome. No. 34. Boffa, J.M., Taonda, S.J.-B., Dickey, J.B., Knudson, D.M., 2000. Field-scale influence of  (Vitellaria paradoxa) on Sorghum production in the Sudan zone of Burkarite kina Faso. Agrofor. Syst. 49, 153e175. , Butyrospermum paradoxum Bonkoungou, E.G., 1987. Monographie du karite
ce agroforestie
re a
usages multiples. Institut de (Gaertn. F.) Hepper, espe  Recherche en Biologie et Ecologie Tropicale, Ouagadougou. Bonnard, P., Scherr, S., 1994. Within gender differences in tree management: is gender distinction a reliable concept? Agrofor. Syst. 25 (2), 71e93. Borgatti, S., 1990. Using ANTHROPAC to investigate a cultural domain. Cult. Anthropol. Methods Newsl. 2 (1), 8. Boster, James, 1986. Exchange of varieties and information between Aguaruna manioc cultivators. Am. Anthropol. 88 (2), 428e436. , G., 2004. Mistletoes of the agroforestry Boussim, I.J., Guinko, S., Tuquet, C., Salle parklands of Burkina Faso. Agrofor. Syst. 60, 39e49. , G., Guinko, S., 1993a. Tapinantus parasite du karite  au Burkina Boussim, I.J., Salle
re partie. Bois Fore ^ts des Tropiques 238, 45e47. Faso, premie , G., Guinko, S., 1993b. Tapinanthus parasite du karite  au Burkina Boussim, I.J., Salle
me partie. Bois Fore ^ts des Tropiques 238, 53e64. Faso, deuxie Breusers, M., Nederlof, S., Van Rheenen, T., 1998. Conflict or symbiosis? Disentangling farmer-Herdsman relations: the Mossi and Fulbe of the Central Plateau, Burkina Faso. J. Mod. Afr. Stud. 36 (3), 357e380. Brewer, D.D., 1995. Cognitive indicators of knowledge in semantic domains. J. Quant. Anthropol. 5, 107e128. Brewer, D.D., Garrett, S.B., Rinaldi, G., 2002. Free-listed items are effective cues for eliciting additional items in semantic domains. Appl. Cogn. Psychol. 16, 343e358. Burkill, H.M., 2000 The Useful Plants of West Tropical Africa, second ed., vols. 4 and 5. Royal Botanic Gardens, Kew. Byers, E., Sainju, M., 1993. Mountain ecosystems and women: opportunities for sustainable development and conservation. In: Paper Presented at the United Nations International Consultation on Advancing Women in Ecosystem Management, Washington, D.C. Camou-Guerrero, A., Reyes-Garcia, V., Martinez-Ramo, M., Casas, A., 2008. Knowledge and use of plant species in a Raramuri community: a gender perspective for conservation. Hum. Ecol. 36, 259e272. Carter, M., Katz, E., 1997. Separate spheres and the conjugal contract: understanding the impact of gender-biased development. In: Haddad, L., Hoddinott, J., Alderman, H. (Eds.), Intrahousehold Resource Allocation in Developing Countries: Methods, Models and Policies. The Johns Hopkins University Press, Baltimore, MD, pp. 95e111. Cavendish, W., 2000. Empirical regularities in the povertyeenvironment relationship of rural households: evidence from Zimbabwe. World Dev. 28, 1979e2003. Chalfin, B., 2000. Risky business: economic uncertainty, market reforms and female livelihoods in Northeast Ghana. Dev. Change 31, 987e1008. Chalfin, B., 2004. Shea Butter Republic. Routledge, New York. Chikoko, M.G., 2002. A Comparative Analysis of Household Owned Woodlots and Fuelwood Sufficiency between Female and Male Headed Households: a Pilot Study in Rural Malawi, Africa (Ph.D. dissertation). Oregon State University, USA. Cotton, C.M., 1996. Ethnobotany: Principles and Applications. Wiley, Chichester, UK. lerot, F., 1995. Importance of Shea Nuts for Women's Activities and Young Child Cre Nutrition in Burkina Faso: the Case of the Lobi (Ph.D. thesis). University of Wisconsin-Madison, Madison. Das, N., 2011. Women's dependence of forest and participation in forestry: a case study of joint forest management in West Bengal. J. For. Econ. 17, 67e89. Doss, C.R., 2001. Designing agricultural technology for women: lessons from twenty efive years of experience. World Dev. 29 (5), 2075e2092. Dovie, B.K., Witkowski, E.T.F., Shackleton, C.M., 2008. Knowledge of plant resource use based on location, gender and generation. Appl. Geogr. 28, 311e322. Elias, M., Carney, J., 2007. African shea butter: a feminized subsidy from nature. Africa 77 (1), 37e62.
Women Elias, M., 2010. Transforming Nature's Subsidy: Global Markets, Burkinabe and African Shea Butter (Ph.D. dissertation). McGill University, Montreal. Faye, M.D., Weber, J.C., Mounkoro, B., Dakouo, J.-M., 2010. Contribution of parkland

37

trees to farmers' livelihoods: a case study from Mali. Dev. Pract. 20 (3), 428e434. Folbre, N., 1986. Hearts and spades: paradigms of household economics. World Dev. 14, 244e255.
s, J., Guinko, S., 1995. Carte de la ve ge tation et de l'occupation du sol au Fonte veloppement Rural (IDR), UniBurkina Faso (Notice explicative). Institut du De  de Ouagadougou, Ouagadougou. versite Fortmann, L., Bruce, J.W., 1988. Whose Trees? Proprietary Dimensions of Forestry. Westview Press, Boulder, CO. Fortmann, L., Antinori, C., Nabane, N., 1997. Fruits of their labors: gender, property rights, and tree planting in two Zimbabwe villages. Rural. Sociol. 62 (3), 295e314. Friedman, J., Yaniv, Z., Dafni, A., Palewitch, D., 1986. A preliminary classification of the healing potential of medicinal plants, based on a rational analysis of an ethnopharmacological field survey among Bedouins in the Negev desert, Israel. J. Ethnopharmacol. 16 (2e3), 275e287. Gatewood, J.B., 1983. Loose talk: linguistic competence and recognition ability. Am. Anthropol. 85, 378e387. Gatewood, J.B., 1984. Familiarity, vocabulary, size, and recognition ability in four semantic domains. Am. Ethnol. 11, 507e527. ni, SouthGausset, Q., Yago-Ouattara, E.L., Belem, B., 2005. Gender and trees in Pe Western Burkina Faso. Women's needs, strategies and challenges. Geografisk Tidsskrift-Danish J. Geogr. 105 (1), 67e76. Gray, L., 2003. Investing in soil quality: farmer responses to land scarcity in Southwestern Burkina Faso. In: Bassett, T., Crummey, D. (Eds.), African Savannas: Global Narratives & Local Knowledge of Environmental Change. James Currey, Oxford, pp. 72e90. ge tation de la Haute-Volta (Ph.D. thesis). Natural Sciences Guinko, S., 1984. Ve  de Bordeaux II, Bordeaux. Department, Universite Gwali, S., Okullo, J.B.L., Eilu, G., Nakabonge, G., Nyeko, P., Vuzi, P., 2011a. Folk classification of shea butter tree (Vitellaria paradoxa subsp. nilotica) ethno-varieties in Uganda. Ethnobot. Res. Appl. 9, 243e256. Gwali, S., Okullo, J.B.L., Eilu, G., Nakabonge, G., Nyeko, P., Vuzi, P., 2011b. Traditional management and conservation of shea trees (Vitellaria paradoxa subsp. nilotica) in Uganda. Environ. Dev. Sustain. 14 (3), 347e363. Hall, J.B., Aebischer, D.P., Tomlinson, H.F., Osei-Amaning, E., Hindle, J.R., 1996. Vitellaria paradoxa: a Monograph. Publication Number 8. School of Agricultural and Forest Sciences, University of Wales, Bangor. Harlan, J.R., De Wet, J.M.J., Stemler, A., 1976. Plant domestication and indigenous African agriculture. In: Harlan, J.R., De Wet, J.M.J., Stemler, A. (Eds.), Origins of African Plant Domestication. Aldine, Chicago, pp. 3e19. Howard, P.L., 2003. Women and the plant world: an exploration. In: Howard, P.L. (Ed.), Women and Plants: Gender Relations in Biodiversity Management and Conservation. Zed Books, London, pp. 1e47. Howard, P.L., 2007. Are there customary rights to plants? An inquiry among the Baganda (Uganda), with special attention to gender. World Dev. 35 (9), 1542e1563. Howorth, C., O'Keefe, P., 1999a. Farmers do it better: local management of change in southern Burkina Faso. Land Degrad. Dev. 10, 91e109. Howorth, C., O'Keefe, P., 1999b. Drought-induced resettlement: a case study from Burkina Faso. Reg. Environ. Change 1 (1), 15e23. mographie et de la Statistique), 2006. Recensement INSD (Institut National de la De ne ral de la population et de l’habitation de 2006. Retrieved online July 8 ge 2009, at: http://www.insd.bf/. Jacobson, J.L., 1992. Gender Bias: Roadblock to Sustainable Development. In: Worldwatch Paper, vol. 110. Worldwatch Institute, Washington, D.C. Kahlheber, S., 1999. Indications for agroforestry: archaeobotanical remains of crops and woody plants from Medieval Saouga, Burkina Faso. In: Veen, M.V.D. (Ed.), The Exploitation of Plant Resources in Ancient Africa. Kluwer Academic/ Plenum, New York, pp. 89e100. Kainkwa, R.M.R., Stigter, C.J., 1994. Wind reduction downwind from a Savanna woodland edge. Neth. J. Agric. Sci. 42 (2), 145e157. Kalaba, F.K., Chirwa, P.W., Prozesky, H., 2009. The contribution of indigenous fruit trees in sustaining rural livelihoods and conservation of natural resources. J. Hortic. For. 1 (1), 1e6.
Kaminski, J., Headey, D., Bernard, T., 2009. Institutional Reform in the Burkinabe Cotton Sector and its Impacts on Incomes and Food Security: 1996e2006. IFPRI Discussion Paper. International Food Policy Research Institute, Washington, D.C. Kaur, R., 1991. Women in Forestry in India. Working Paper No 714. Women in Development Division, Population and Human Resources Department, World Bank, Washington, D.C. Kelly, B.A., Bouvet, J.M., Picard, N., 2004. Size class distribution and spatial pattern of Vitellaria paradoxa in relation to farmers' practices in Mali. Agrofor. Syst. 60, 3e11. Kelly, B.A., Fleury, S.G., Bouvet, J.-M., 2007. Impact of agroforestry practices on the flowering phenology of Vitellaria paradoxa in parklands in southern Mali. Agrofor. Syst. 71, 67e75. Kessler, J.J., Breman, H., 1991. The potential of agroforestry to increase primary production in the Sahelian and Sudanian zones of west Africa. Agrofor. Syst. 13, 41e62.  (Vitellaria paradoxa) and Ne re  (Parkia Kessler, J.J., 1992. The influence of Karite biglobosa) trees on Sorghum production in Burkina Faso. Agrofor. Syst. 17, 97e118. Kevane, M., Gray, L., 1999. A Woman's field is made at night: gendered land rights and norms in Burkina Faso. Fem. Econ. 5 (3), 1e26.

38

M. Elias / Journal of Rural Studies 38 (2015) 27e38

Kristensen, M., Balslev, H., 2003. Perceptions, use and availability of woody plants among the Gourounsi in Burkina Faso. Biodivers. Conserv. 12 (8), 1715e1739. Kristensen, M., Lykke, A.M., 2003. Informant-based valuation of use and conservation preferences of Savanna trees in Burkina Faso. Econ. Bot. 57 (2), 203e217. draogo, S.J., Diallo, O.B., Guinko, S., 2004. Productivite  fruitie
re du Lamien, N., Oue  (Vitellaria paradoxa Gaertn. C. F., Sapotaceae) dans les parcs agroforestiers karite traditionnels au Burkina Faso. Fruits 59, 423e429. Lamien, N., Tigabu, M., Guinko, S., Oden, P.C., 2007. Variations in dendrometric and fruiting characters of Vitellaria paradoxa populations and multivariate models for estimation of fruit yield. Agroforestry Systems 69, 1e11. Leach, H.B., Van der Stege, C., Vogl, C.R., 2011. Baobab (Adansonia digitata L.) and Tamarind (Tamarindus indica L.) management strategies in the midst of conflict and change: a Dogon case study from Mali. Hum. Ecol. 39, 597e612. Leach, M., 1994. Rainforest Relations. Edinburgh University Press, London. Lope-Alzina, D., 2007. Gendered production spaces and crop varietal selection: case study in Yucatan, Mexico. Singap. J. Trop. Geogr. 28, 21e38. Lovett, P.N., Haq, N., 2000a. Diversity of the sheanut tree (Vitellaria paradoxa D.F. Gaertn.) in Ghana. Genet. Resour. Crop Evol. 47, 293e304. Lovett, P.N., Haq, N., 2000b. Evidence for anthropic selection of the sheanut tree (Vitellaria paradoxa). Agrofor. Syst. 48 (3), 273e288. Lykke, A.M., Mertz, O., Ganaba, S., 2002. Food consumption in rural Burkina Faso. Ecol. Food Nutr. 41 (2), 119e153. Mapongmetsem, P.M., Nkongmenek, B.A., Rongoumi, G., Nguemo Dongock, D., mes d’utilisation des terres sur la conserDongmo, B., 2011. Impact des syste gion des savvation de Vitellaria paradoxa Gaerten. F. (Sapotaceae) dans la re ennes. Int. J. Environ. Stud. 68 (6), 851e872. anes soudano guine Maranz, S., Wiesman, Z., 2003. Evidence for indigenous selection and distribution of the shea tree, Vitellaria paradoxa, and its potential significance to prevailing parkland tree patterns in sub-Saharan Africa North of the Equator. J. Biogeogr. 30 (10), 1505e1516. Maranz, S., Kpikpi, W., Wiesman, Z., De Saint-Sauveur, A., Chapagain, B., 2004. Nutritional values and indigenous preferences for shea fruits (Vitellaria paradoxa C.F. Gaertn F.) in African agroforestry parklands. Economic Botany 58 (4), 588e600. Marchal, J.Y., 1983. Yatenga, nord Haute Volta: la dynamique d'un espace soudanolien. ORSTOM, Paris. sahe MEF (Ministry of Economy and Finance of Burkina Faso), 2011. Investing in the Future: Accelerated Growth and Sustainable Development. Business Environment, Investment Opportunities and Public-private Partnership. MEF, Ouagadougou. Meizen-Dick, R.S., Brown, L.R., Feldstein, H.S.M., Quisumbing, A.R., 1997. Gender, property rights, and natural resources. World Dev. 25 (8), 1303e1315. Mukadasi, B., Nabalegwa, M., 2007. Gender mainstreaming and community participation in plant resource conservation in Buzaya county, Kamuli district, Uganda. Afr. J. Ecol. 45 (S1), 7e12. Mwangi, E., Meinzen-Dick, R., Sun, Y., 2011. Gender and sustainable forest management in East Africa and Latin America. Ecol. Soc. 16 (1), 17. Neumann, K., Kahlheber, S., Uebel, D., 1998. Remains of woody plants from Saouga, a Medieval west African village. Veg. Hist. Archaeobotany 7, 55e77. Okiror, P., Agea, J.G., Akais Okia, C., Okullo, J.B.L., 2012. On-farm management of Vitellaria paradoxa C.F. Gaertn. in Amuria district, Eastern Uganda. Int. J. For. Res. 8. Article ID 768946. Okullo, J.B.L., Hall, J.B., Obua, J., 2004a. Leafing, flowering and fruiting of Vitellaria paradoxa subsp. nilotica (shea butter tree) in Savanna parklands in Uganda. Agrofor. Syst. 60, 77e91. Okullo, J.B.L., Obua, J., Okello, G., 2004b. Use of indigenous knowledge in predicting fruit production of shea butter tree in agroforestry parklands of North-eastern Uganda. Uganda J. Agric. Sci. 9, 360e366. draogo, M., 2003. New Stakeholders and the Promotion of Agro-silvo-pastoral Oue Activities in Southern Burkina Faso: False Start or Inexperience?. Drylands Issue Paper No. E118 International Institute for Environment and Development (IIED), Edinburgh. , S., Savadogo, P., Tigabu, M., Ouadba, J.M., Ode n, P.C., 2010. Consumptive values Pare

and local perception of dry foresty decline in Burkina Faso, West Africa. Environ. Dev. Sustain. 12, 277e295. Poudhyal, M., 2011. Chiefs and trees: tenures and incentives in the management and use of two multipurpose tree species in agroforestry parklands in northern Ghana. Soc. Nat. Resour. 24 (10), 1063e1077. Prudencio, C., 1993. Ring management of soils and crops in the west African semiarid tropics: the case of Mossi farming season in Burkina Faso. Agric. Ecosyst. Environ. 47, 237e264. Pullan, R.A., 1974. Farmed parkland in west Africa. Savanna 3 (2), 119e151. Quisumbing, A.R., Payongayong, E., Aidoo, J.B., Otsuka, K., 2001. Women's land rights in the transition to individualized ownership: implications for treeresource management in western Ghana. Econ. Dev. Cult. Change 50, 157e181. Razavi, S., Miller, C., 1995. From WID to GAD: Conceptual Shifts in the Women and Development Discourse. In: UNRISD Occasional Paper Series OBP 1. UNRISD, Geneva. Reenberg, A., Lund, C., 1998. Land use and land right dynamics-determinants for resource management options in eastern Burkina Faso. Hum. Ecol. 26 (4), 599e620. Rocheleau, D., Edmunds, D., 1997. Women, men and trees: gender, power and property in forest and Agrarian landscapes. World Dev. 25 (8), 1351e1371. Rocheleau, D., Thomas-Slayter, B., Wangari, E., 1996. Gender and environment: a feminist political ecology Perspective. In: Rocheleau, D., Thomas-Slayter, B., Wangari, E. (Eds.), Feminist Political Ecology: Global Issues and Local Experiences. Routledge, New York, pp. 3e23.  au Soudan. Agron. Trop. 12, 143-172, pp. 279e306, pp. Ruyssen, B., 1957. Le karite 415e440. Schreckenberg, K., 1999. Products of a managed landscape: non-timber forest products in the parklands of the Bassila region, Benin. Glob. Ecol. Biogeogr. 8 (3e4), 279e289. Sen, A., 1990. Gender and cooperative conflicts. In: Tinker, I. (Ed.), Persistent Inequalities: Women and World Development. Oxford University Press, New York and Oxford, pp. 123e149. Stemler, S.E., 2001. An overview of content analysis. Pract. Assess. Res. Eval. 7 (17). Retrieved online April 15, 2010 at: http://PAREonline.net/getvn.asp?v¼7&n¼17. Sunderland, T., Achdiawan, R., Angelsen, A., Babigumira, R., Ickowitz, A., Paumgarten, F., Reyes-Garciåla, V., Shively, G., 2014. Challenging perceptions about men, women, and forest product use: a global comparative study. World Dev. http://dx.doi.org/10.1016/j.worlddev.2014.03.003. Turner, N.J., Ignace, M.B., Ignace, R., 2000. Traditional ecological knowledge and wisdom of aboriginal peoples in British Columbia. Ecol. Appl. 10 (5), 1275e1287. Udry, C., 1996. Gender, agricultural production, and the theory of the household. J. Polit. Econ. 104 (5), 1010e1046.  dans le secteur des produits de UNCTAD, 2006. InfoCOMM: Information de marche base. Retrieved online May 15, 2007 at: http://www.unctad.org/infocomm/ francais/karite/culture.htm http://www.unctad.org/infocomm/francais/karite/ societes.htm http://r0.unctad.org/infocomm/francais/karite/marche.htm. Upadhyay, B., 2005. Women and natural resource management: illustrations from India and Nepal. Nat. Resour. Forum 29 (3), 224e232. Wangari, E., Thomas-Slayter, B., Rocheleau, D., 1996. Gendered visions for survival: semi-arid regions in Kenya. In: Rocheleau, D., Thomas-Slayter, B., Wangari, E. (Eds.), Feminist Political Ecology: Global Issues and Local Experiences. Routledge, New York, pp. 127e154. Warner, K., 1993. Patterns of Farmer Tree Growing in Eastern Africa: a Socioeconomic Analysis. Tropical Forestry Papers No 27. International Centre for Research in Agroforestry, Nairobi. Weller, S.C., Romney, A.K., 1988. Systematic Data Collection. Sage, Newbury Park, CA. Wezel, A., Haigis, J., 2000. Farmers' perception of vegetation changes in semi-arid Niger. Land Degrad. Dev. 11 (6), 523e534. Zent, S., 2009. Traditional ecological knowledge (TEK) and biocultural diversity: a close-up look at linkages, delearning trends and changing patterns of transmission. In: Bates, P., Chiba, M., Kube, S., Nakashima, D. (Eds.), Learning and Knowing in Indigenous Societies Today. UNESCO, Paris, pp. 39e57.