Abstract

This paper examines female hunting among a group of Aka forest foragers ("pygmies") of the Central African Republic where women net-hunt more frequently than men. The study aims to understand the contexts of female hunting and contribute to the paucity of descriptive and systematic studies of women hunters and gender task allocation among foragers. Contexts predicted from human behavioral ecology and cultural anthropology are considered and evaluated. Most of the contexts for female hunting predicted by the evolutionary and cultural theoretical orientations occurred among the this group of Aka: game were relatively abundant and women received relatively high caloric returns from hunting; game animals were acquired synchronously; hunting took place with other adults; Aka women had access to the means/technology of efficient hunting; Aka male ideological/political control of women was minimal; and, cultural precedents existed that enabled women to obtain knowledge and experience in hunting. Modifications to both evolutionary and cultural theories that deal with female hunting and gender task allocation among foragers are suggested and an integrated approach is described.

Key Words: foragers, Central Africa, sexual division of labor, women hunters

Andrew J. Noss1 and Barry S. Hewlett2

Casilla 2417

Santa Cruz, Bolivia

Washington State University

Vancouver, WA 98686

Corresponding author:

Barry S. Hewlett

hewlett@vancouver.wsu.edu

360-546-9449 Office

541-929-3144 Home

Running Head: Female Hunting in Central Africa

It is obvious that women can hunt, but they do not do so in all contexts. The primary aim of the paper is to describe the cultural and environmental contexts of female hunting among one group of Aka forest foragers ("pygmies") of the Central African Republic where women net-hunt more frequently than men. Larger theoretical issues regarding gender allocation of tasks among foragers are considered once the specific contexts have been described and analyzed.

Ethnographers sometimes mention that women hunt, but few studies have systematically evaluated the contributions of women hunters. For instance, Brightman (1996) indicates that many Woods Cree women trapped fur-bearing animals alone or in groups and some women, both married and single, hunted moose, caribou and bear. Romanoff (1983) describes Matses women in the Peruvian Amazon who accompany their husbands, helping to chase and kill animals, while Hurtado et al. (1985) indicate Ache women in Paraguay help men search and transport captured animals, occasionally killing animals themselves. Biesele and Barcaly (2001) describe excellent Ju/'hoan women trackers of large game who accompany their husbands and contribute substantially to their hunting success. While these are important ethnographic descriptions, the ethnographers provide few details about the contexts, frequency and significance of female hunting. An exception to the descriptive accounts of women hunters is the systematic study of Agta women hunters by Estioko-Griffin and her colleagues (1985, 1986, Estioko-Griffin and Griffin 1981, Goodman et al. 1985). Most Agta women hunt and a limited number of Agta women are proficient hunters with bows and arrows, machetes, knives, traps, and dogs. Their study was important because it demonstrated that: 1) women hunters with young children did not experience higher child mortality than women with children who did not hunt; and, 2) female hunters can be just as efficient/successful as male hunters. However, Agta women hunted substantially less than men and tended to hunt in groups and with dogs for smaller animals.

Researchers from two general theoretical orientations--evolutionary and cultural--have identified contexts under which women are likely to hunt. Enormous diversity exists within each of these theoretical orientations, but the assumptions that guide their hypotheses are distinct. A brief survey of the current literature in each orientation is provided before an ethnographic overview of the Aka is presented. We refer readers to Mukhopadhyay and Higgins (1988), Bird (1999) and Brightman (1996) for recent and more exhaustive literature reviews on the sexual division of labor among foragers.

Several, quite distinct, approaches to human evolutionary biology exist (Smith 2000), but researchers in human behavioral ecology (HBE) have contributed the greatest number of studies to the sexual division of labor among foragers (Bird 1999, Hawkes 1991, Hurtado and Hill 1990). HBE generates hypotheses about human behavior starting from the assumption that human activities should enhance reproductive fitness. Human behavior is flexible and few behaviors are hard-wired (i.e., genetic), but humans are predicted to try and maximize fitness within their demographic, ecological and cultural environments. Behavioral ecologists use Trivers' (1972) parental investment theory to explain why men and women have different reproductive strategies and trade-offs in a given environment. Since human females invest more in offspring than do males, the trade-offs of alternative subsistence tasks are different for females than for males.

Human behavioral ecologists have contributed to major revisions of anthropological views of the sexual division of labor. Hawkes et al. (1997) indicate that males have more to gain reproductively than do females by "showing off", and thereby gaining status and prestige in the community. A successful male hunter may substantially increase his reproductive fitness by attracting another wife or through extramarital relations. Since women have invested more time and energy in children than men and know the child is theirs, they are predicted to focus on providing for their children on a day-to-day basis. Among most foragers, prey that are widely shared with others are usually asynchronously acquired, provide large amounts of meat and are associated with high risk of failure. Male hunters who target these widely shared game animals do not gain by providing more for themselves or their family, but by gaining favorable attention or increased mating opportunities. In contrast, smaller-sized prey that are reliable and not widely shared are often consumed by family groups. Hunters who target these resources derive immediate benefits from provisioning their family. Hawkes et al. make the point that is that men's big game hunting, which was previously regarded as parenting effort (i.e., provisioning the family) in the hunting hypothesis, may often be mating effort. Fathers are showing off rather providing. Women, grandmothers in particular, according to Hawkes et al., are the consistent providers of forager families with young children. Hawkes et al. do not directly address the issue of female hunting, but if women are the day-to-day providers of their families and if women regularly hunt, then they should hunt for small and medium sized game that is regularly acquired and shared within the family rather than hunting for large game that is irregularly acquired and shared with the entire community.

The Hawkes et al. theory is useful for predicting the types of game animals women should hunt, but it does not help to predict the contexts in which women might hunt rather than gather or engage in another subsistence activity. Other human behavioral ecologists (Bailey and Aunger 1989) argue that women are likely to hunt in contexts where hunting produces more calories per unit time than other subsistence alternatives. Bailey and Aunger compared Mbuti forager women’s participation in net-hunting to Efe forager women's participation in horticultural labor (i.e., assisting Lese farmer women in their fields). Their research indicated that Mbuti women net-hunted because they obtained more calories per unit time from net-hunting, in large part due to meat markets in the Mbuti area, than they would have from providing labor to village women in exchange for village crops (e.g., manioc, corn). Efe forager women did not hunt because meat markets did not exist in their area and they obtained more calories per unit time from helping horticultural women than they could from hunting. Efe women received manioc and other village crops in exchange for their labor while Mbuti women sold/traded meat from the hunt for village crops. This finding is also consistent with the Agta study in that Estioko-Griffin (1986) study which indicated that Agta women hunted, in part, because of high return rates in comparison to horticulture--women hunted in the most remote parts of the forest where returns were high. Human behavioral ecologists predict that women are more likely to hunt in contexts where hunting is more productive than alternative subsistence activities.

The hypothesis that women do not hunt because they are constrained, burdened or immobilized by child care activities because they are concerned about the survival of their infants has a long history in anthropology (Mukhopadhyay and Higgins 1988, Brightman 1996, Kelly 1995). This hypothesis is consistent with HBE in that females are predicted to be more concerned than men about trade-offs with child care. Hurtado et al.'s (1992) time allocation data demonstrated that Hiwi and Aché women with infants gathered less than women without infants. These authors suggested that Hiwi and Ache women gathered rather than hunted because women were more likely than men to be concerned about health risks that might lead to increased maternal or child mortality. However, the authors do not present data to evaluate the hypothesis and the study of Agta women hunters suggests otherwise. Recent research among Australian foragers and a review of the HBE literature point out that women's trade-offs with child care "may explain variability in time allocation of food production according to reproductive status, but they do not explain differences in the resource choice of men and women (Bird 1999: 72)." That is, women with infants may work less than other women, but there is no evidence that infant care explains why women gather rather than hunt. The Agta data are consistent with this recent view in that Agta women who regularly hunted did not experience greater infant or child mortality than women who engaged in gathering or farming (i.e., hunting does not lead to "risks" and greater child mortality), but women with infants did not hunt as frequently as women without infants. Agta hunting requires stalking and stealth which may not be as compatible with infant care as other hunting techniques (e.g. driving or flushing animals into nets). It is important to mention that HBE does not focus on lack of female strength as a factor to explain why women do not hunt or the types of game women hunters may target.

Cooperative female hunting is also predicted from an evolutionary perspective. Cooperative hunting may lead to greater sharing of captured animals and decrease the day-to-day variability in returns from hunting (Hames 1989). Cooperative hunting may also decrease the risk of rape, capture or violence to women and children by neighboring males (Hrdy 1999).

In summary, an evolutionary perspective predicts that frequent female hunting is most likely to occur when: risk of pursuit failure is relatively low; caloric return from hunting is relatively high by comparison to other subsistence options; animals are acquired synchronously (i.e., on a regular basis); women do not have an infant (unless the hunting technique is consistent with infant care); and, hunting takes place with other adults.

Considerable theoretical diversity also exists within the "cultural" orientation on the sexual division of labor, but in general cultural anthropologists tend to reject evolutionary perspectives and emphasize the cultural and historical construction of the sexual division of labor. For instance, Brightman (1996: 718) characterized evolutionary approaches as "hegemonic" and concluded that "physiological theories are irrelevant to the foraging division of labor". His literature review suggested that gendered classifications were completely arbitrary and the consequence of cultural taboos, gender identities and gender politics. Specifically, he hypothesized that women seldom hunted because they were excluded from the technological means of hunting (i.e., taboos utilized to exclude women's use of weapons--spears, arrows, harpoons and guns). He pointed out that even among the Agta, men hunted alone with bows or guns and use stalk or ambush techniques, while most women hunted in teams with dogs and machetes (Estioko-Griffin 1985: 23-24). He indicated that female hunting may be less efficient than male hunting but for cultural rather than physiological reasons. He concludes that "the sexual division of foraging labor derives from the interested appropriation by men of hunting labor and the social capital accruing to its products." Brightman's review predicts that women will hunt more frequently when male control ideologies/politics are minimal and taboos do not exclude women from hunting technology.

Ethnographic decision-making models of the sexual division of labor (Mukhopadhyay 1984, Gladwin1989) also have important implications for understanding female hunting among foragers, especially in culture change situations. This perspective is important because the group of Aka in this study are experiencing relatively rapid economic change. The ethnographic decision-making approach indicates that every culture has "precedents" or schema for deciding gender task allocation. "Cultural precedents" refer to role-based obligations/responsibilities (Mukhopadhyay 1980) as well as knowledge, experience and feelings about specific subsistence activities/tasks (D'Andrade and Strauss 1992). Cultural precedents have histories, are both dynamic and conservative through time, and are utilized by individuals to make task allocation decisions. For instance, Hewlett (1996) utilized a "cultural" approach to critique the Bailey and Aunger evolutionary hypothesis described above. The critique emphasized that Mbuti women net-hunted because of a long culture history with Bantu-speaking farmers that net-hunted. Congo Basin foragers that net-hunted (e.g., Mbuti, Aka, Bakola/Bagyeli, Bongo) had long social-economic histories with Bantu-speaking horticulturalists who hunted with nets, whereas forager groups that did not use nets (e.g., Efe and Baka) had long social-economic histories relations with Oubanguian or Sudanic speakers who hunted with bows-and-arrows, traps or spears. The critique also demonstrated that Mbuti women were net-hunting long before a market for meat existed.

Feminist and praxis-based cultural anthropologists also argue against a "homogenous" view of men or women and argue for a more contextual analysis of the social roles of "sisters", "wives", "daughters" or "grandmothers". This is consistent with the evolutionary perspective where costs and benefits of an individual's reproductive strategies are predicted to vary according to different natural and social environments encountered during the life course. But the perspectives diverge as follows: evolutionists hypothesize that all task allocation decisions maximize an individual's reproductive fitness; whereas cultural anthropologists are more likely to hypothesize that the variation in task decisions is patterned by how the individual negotiates the culturally and socially constructed aspects of daily life, having little to do with reproductive fitness.

In summary, "cultural" approaches suggest that female hunting is most likely to occur when: women have access to the means/technology of efficient hunting; male ideological/political control of women is minimal; and, cultural precedents exist that enable women to obtain knowledge and experience in hunting.

Historical records indicate a long tradition of women hunters in central Africa. Andrew Battel, an English sailor who was taken prisoner of war in Brazil by the Portuguese and then sent by them to present-day Angola around 1590, described pygmy hunters to a friend: "The women carry bow and arrow, as well as men, and one of these will walk in the woods alone, and kill the Pongos (gorillas) with their poisoned arrows (Battel, in Ravenstein 1901:59)."

Joiris reports that BaKola/BaGyeli women of Western Cameroon hunt with nets,

spears and dogs, and hunt with other women or their husbands.

Baka women of Eastern Cameroon, on the other hand, seldom hunt, but women will carry spears on their husbands' snare hunts and some women are known to be more efficient spear hunters than men. Women do not own spears but will use one belonging to their brother, husband or father (Joiris, personal communication).

Tanno (1976:114) observed 15 net-hunts among the Mbuti of the Ituri, noting that the net-hunt with the fewest women had the lowest return rate. On that day, only one woman and 12 men participated because the Mbuti had recently moved to a new camp and most women remained in camp to build the new houses. He also reported that net-hunts did not take place on certain days because the women were too tired.

Observations of women net-hunting without men have been noted by several Aka ethnographers. Aka men in most areas go spear hunting at least part of the year and during this period women do a sexually explicit dance called sapa enticing the men to return. In the meantime women do not hesitate to go net-hunting on their own. Kitanishi (1995:81) reported that Aka "women also handle the nets and beat the bush, when there are not enough adult and adolescent men. Occasionally, women lead net-hunting when no adult man participates in the hunt." He also observed a situation in which women spear hunted: "It took place when the men had gone spear hunting for several days (njango). A woman encountered a bush pig near the camp, and the women at the camp chased it with spears (Kitanishi 1995:89)." McCreedy (1994:15) also describes the importance of women in the net-hunt when she quotes an Aka man "Women are the arms of the dibouka [throw of the nets]." The Aka with whom she lived conducted a bobanda ritual when net-hunting was not successful. Men organized the ritual and pleaded with women to participate because women's enthusiasm and energy were viewed as crucial to the success of the ritual and future net-hunting.

While considerable variability exists in the frequency of central African forager women hunting, none of the groups seem to have sanctions against women hunters. It is important for both men and women to know a diversity of subsistence techniques so that men can take over general female activities and vice versa in case of necessity. As anthropologists have tried to explain the sexual division of labor they have probably also contributed to a perception of fixed activities by sex when in fact considerable flexibility exists in these activities, especially among foraging peoples.

About 20,000 Aka foragers live in the tropical forests of southern Central African Republic and northern Congo-Brazzaville. While enormous diversity exists between Aka groups in their reliance on farming, levels of acculturation, relationships with farmers, frequency of net-hunting, impact of commercial meat trade, etc., the basic organization of net-hunts is remarkably similar. Researchers have described Aka net-hunts in several locations, and there are several ways to organize a net-hunt, but the following description is common to all Aka and is the type most frequently practiced (Bahuchet 1975, 1979, 1985, Delobeau 1989, Demesse 1978, 1980, Hewlett 1977, 1991, Hudson 1990, Kitanishi 1996, Motte 1982, Noss 1985, Takeuchi 1995). Hunters form a circle or semi-circle with the nets, hanging the nets on small trees or vines, pegging the bottom of the net to the ground. Upon completing the circle or semi-circle, the net carriers (usually males) begin searching the interior, beating leaves and shouting to drive animals from their hiding places. Often one or more men with spears also search the interior. The other participants (usually females) remain hidden near the nets to spot flushed animals, to frighten them into the nets, and then to seize them before they escape. Animals captured in nets are killed with machetes or logs lying nearby. The hunters quickly determine that no animals are inside the circle, or that what is there has escaped or been captured. They then gather up the nets and move on to the next spot only a couple hundred meters away, and begin again.

Aka gender roles on the net-hunt are distinct from those among the Mbuti net-hunters in the eastern Congo Basin (Ituri Forest) where women always beat and flush out game whereas men guard the nets and capture animals (Harako 1976, Ichikawa 1983, Tanno 1976, Turnbull 1961, Wilkie and Curran 1991). Aka women are generally responsible for carrying the meat, tending children, and gathering additional food items, but men are active in all of these areas as well. Women assist with setting up and removing the nets, and play the essential role of guarding the nets while the men are beating. Many animals outside the net circle are sighted first by a woman, who calls those nearest her to set up their nets to encircle the animal, or to attempt to spear it. However, Aka gender roles are interchangeable as some women own nets (26 percent in this study) and women that do not own nets often borrow their husband’s or another man’s net when that person does not participate in the hunt. In addition, nets often change hands during a hunt. For example, after carrying the net for several sets, a man will pass the net on to his wife for the remainder of the hunt. Women also participate in beating, though they tend to be quieter and less active than men.

Learning how to net-hunt comes early and studies have demonstrated that boys and girls by age 10 know how to chase various game animals into the net, butcher animals and divide animals according to formal rules (Hewlett and Cavalli Sforza 1988).

While not a central part of this study it is important to briefly describe sharing of game animals captured on an Aka net-hunt (Bahuchet 1991, Hewlett 1991, Kitanishi 1996, 1998). The owner of a captured animal (konja) is the owner of the net. Konja are usually men, but as mentioned above, one-fourth of the nets in Mossapoula are owned by women. Sharing of the game occurs in three stages: 1) the animal is usually butchered by the wife of the konja and it is formally distributed to people with specific roles on the hunt--e.g., head of the animal goes to the person who set up the net, rib and belly go to person who first seized animal; 2) the woman who butchers the animal informally distributes pieces of meat to 1-3 (more if it is a larger animal) adults on the hunt; and, 3) the men with meat from the first or second distribution give the meat to their wife or female relative, and she prepares a stew which is then divided into bowls for each of the other huts in the camp or camp sub-group if the camp is large. Kitanishi (1998:18) systematically evaluated all stages of sharing and reported that only 20% of a stew cooked by a women was eaten by her family. Most of the stew women and their children ate was from other women unless the amount of stew was very small, in which case sharing was minimal. Because net owners (usually male) were not involved in the sharing and receivers of meat were women, the game hunted was shared only among women. Meat brought back to camp was redistributed among women who were present. Kitanishi demonstrated that kinship distance was not a good predictor of the frequency of Aka sharing in stages one and two, but it was a better predictor in stage three--i.e., women cooking the stews were more likely to share with women that were biological kin. When camps were small and returns good all women shared with all other women, but as camp sized increased, women were more likely to share with kin and those in their sub-group. Both meat and vegetables were widely shared between women (Kitanishi 1998:19-20).

Aka sharing is somewhat different from that among Mbuti: an Aka woman frequently shares the pot of stew with women in all other huts (or sub-group in large camps) whereas Mbuti women share with men in a central hut and a few other women.

Finally, it is important to mention a few aspects of gender relations among the Aka. The Aka are probably one of the most gender egalitarian cultures in the ethnographic record. Husband and wife share subsistence, childcare, leisure and ritual activities. Women participate in several types of hunts and are not forbidden from carrying or touching hunting implements such as nets, spears, machetes, cross-bows, guns or small traps. In terms of childcare, Aka fathers provide more direct care of infants and young children than fathers in any other culture. While men usually hold the named positions of kombeti (leader), tuma (elephant hunter) and nganga (healer), women challenge men's authority on a regular basis and are influential actors in all kinds of decision-making (e.g., camp movement, extramarital affairs, bad luck on the hunt). Divorce is a matter of one partner moving out of the house. Physical violence in general is infrequent and violence against women is especially rare. Hitting a child or a spouse is cause for divorce (Hewlett 1991).

Most of the ethnographic studies of Aka net-hunting mentioned above were conducted in small communities, generally of 50 inhabitants or less, with hunting-gathering subsistence economies. This paper examines a much larger Aka community, Mossapoula (300 inhabitants) in an economy where hunting and gathering has declined in importance relative to a variety of agricultural and wage-labor opportunities.

Before a Yugoslavian lumber company opened in 1971, the Mossapoula Aka lived next to the Kaka fishing village of Mbinjo, which is now Bayanga. By the early 1980s most of the Aka had moved to the small dirt airstrip and worked for the logging company. They moved 5 km north of Bayanga to their present location at Mossapoula in the mid-1980s because of problems with the increasing population of non-Aka residents. The non-Aka population increased because of perceptions of job opportunities with the logging company and conservation agencies (non-governmental organizations such as World Wildlife Fund). As the population increased, competition over land for manioc fields increased and the demand for Aka labor and forest products increased. The 5 km distance meant less harassment from villagers. Some of their Kaka patrons moved with them and Mossapoula included some 80 Kaka residents across the road from the Aka, but traditional ties with Kaka were weak and Aka were no longer obliged to give Kaka a proportion of their game.

The first author collected the data from September 1993 to December 1994 in the village of Mossapoula, within the Dzanga-Sangha Special Reserve of southwestern Central African Republic (Figure 1). Data were collected to evaluate the impact of Aka subsistence activities on the new reserve.

[INSERT FIGURE 1HERE]

During week-long periods (Monday-Friday) for a total of 90 observation days the principal daily activity was recorded for each adult and adolescent resident: 126 females, and 116 males. In analyzing individual activity patterns a sub-sample of those whose activities were tracked for at least 60 of the 90 observation days was defined: 113 females, and 99 males. Households were defined as all adults and children sleeping in a single house.

In order to observe hunting activities, the first author accompanied 76 net-hunts: 58 during the 90 observation days, 10 on other days, and 8 net-hunts from forest camps. All hunts on observation days were accompanied, except in a few instances when more than one group went hunting in which case the larger group was observed. Daily activities for each individual were recorded during the day as the person was observed, or in the evening upon returning to the community. Aka ages were determined by A. Kretsinger (pers. comm.), director of the conservation agency's community health program that had served Mossapoula for five years.

All captured animals were weighed. Potential yields in kilograms of edible meat assumed that 60 percent of the live weight was edible, and that all meat was consumed in Mossapoula. Approximately 80 percent of meat from net-hunts was consumed in Mossapoula, and 20 percent sold or traded to merchants from Bayanga (Noss 1995). It was impossible to determine actual sharing and consumption by individuals because meat was divided and distributed during the course of each hunt and again on returning to Mossapoula before and after cooking.

Five kilometers to the south of Mossapoula was the town of Bayanga which houses 2,500 residents, a French logging company sawmill, and the headquarters of a US based conservation agency that administered the Dzanga-Sangha Special Reserve and Dzanga-Ndoki National Park. Aka tend to view logging and conservation facilities as alternative clusters of "resources" and consequently the number of Aka in Mossapoula has slowly increased over time as Aka from Congo move in next to their Aka relatives in Mossapoula. Aka were drawn to rubber and coffee plantations during the colonial period and more recently they were attracted to conservation facilities or other neocolonial structures (e.g., logging companies, missions, NGOs).

DATA

Traditionally hunter-gatherers, the Mossapoula Aka during the study period engaged in a variety of subsistence and economic activities including hunting, gathering, fishing, farming, mining, day labor and formal employment. Despite numerous alternatives, hunting remains economically important in providing food and income for the Mossapoula Aka (Noss 1997, 2000). Mossapoula Aka employ several hunting methods: community net hunts, staking with crossbow or gun, and the setting of snares. Of these methods, communal net-hunting is the most important in terms of yields and participation rates, and the only hunting method practiced by Mossapoula Aka women (Noss 1995). Unlike Aka women in other locations, Mossapoula women did not accompany men for crossbow and small trap hunting.

Net-hunts included an average of 45.6 participants but participation ranged from 5 to 99 individuals. Net-hunts began anywhere from 5 minutes (200 m) to over 2 hours (10 km) from Mossapoula. Net-hunts took place on average one hour's walk and 3.8 km from the village. Mossapoula hunters formed a circle with 4 to 20 nets (14 nets on average) and nets measured 5-40 meters in length, averaging 18.3 meters, and 1-1.5 meters in height. With an average of 16 casts per hunt and 14 nets per hunt and an average of 18 m per net, the diameter of the net circle of nets was estimated to be 90 m. Mossapoula Aka covered approximately 11-13 km on an average net-hunt. The hunters completed the entire cycle of setting up the nets, beating, and moving to the next site in only 10-15 minutes. Thus each net-hunt comprised 4-6 casts per hour and hunts lasted about 6 hours (including time walking to and from hunt), for a total of 10-30 casts in one day (average 16).

The net-hunt was primarily a hunt for the blue duiker Cephalophus monticola (<= 6kg), which comprised 75 percent of captures. The three other primary prey species were the brush-tailed porcupine Atherurus africanus (<= 4kg), the bay duiker Cephalophus dorsalis (<= 25 kg), and the Peters' duiker Cephalophus callipygus (<= 22kg). Together these four species comprised 97 percent of net-hunt captures. Larger animals were rarely encountered, in part because they were driven away by the noise of the net-hunting group moving through the forest. In any case, large animals including bay and Peters' duikers often tear through the nets or jump over them to escape (Noss 1998, 2000).

Only 7% (5/76) of hunts were unsuccessful. The average number of animals captured was seven. With 14 nets on average on the hunts this means that approximately 50% of families captured something every net-hunt.

Table 1 compares some of the features of Mossapoula Aka net-hunting with

[INSERT TABLE 1 HERE]

features of net-hunting among other Aka groups in northern Congo-Brazzaville and Mbuti groups in the Ituri (Congo-Kinshasa). Mossapoula Aka net-hunts were distinct in several ways: 1) nets were very small; 2) hunters made substantially more casts of the nets during the relatively short net-hunt; 3) more people participated on the hunt; 4) most of the participants were women; 5) net-hunts were relatively infrequent; and 6) return rates were relatively high. The Mossapoula return rates in Table 1 are averages of all Mossapoula net-hunts; the returns were higher in the forest camps (2.23kg/hunter) in comparison to net-hunts from the village of Mossapoula (0.92kg/hunter).

The relatively high return rates were, in part, due to relatively high abundance of game animals. There were 10-20 blue duikers, 1.2-2.0 bay duikers, 0.9-1.2 Peters' duikers and 2.7-5.3 brush-tailed porcupines per square km in the Mossapoula area (Noss 2000). Elephants (Loxodonta africana cyclotis) were observed on 14% of net-hunts and gorillas (Gorilla gorilla gorilla) were observed on 6% of hunts. No other researchers have noted encountering large mammals on net-hunts on such a regular basis.

While Mossapoula Aka net-hunts were distinct in several ways, they were similar to descriptions of net-hunts in other areas of the Congo Basin as follows: small to medium sized duikers represented more than 90 percent of all captures, and someone on the hunt captured at least one animal on 90 percent or more of all hunts (i.e., the success rate of the group was close to 100 percent).

Women and men of all ages actively participated in net-hunting, with girls and boys beginning to contribute as young as five years of age. One woman hunted through her eighth month of pregnancy, and resumed hunting one month after giving birth. Infants accompanied their mothers. The oldest active net-hunter was a great-grandmother more than 60 years old who frequently hunted with her own net.

Overall, women were much more likely to net-hunt than were men. Only 4 of 126 females (3 percent) in Mossapoula did not net-hunt during the entire observation period, compared to 12 of 116 males (10 percent) who never hunted (chi-sq.=5.17, p< .05). Women also outnumbered men on net-hunts (Table 1) even though the adult male-female sex ratio in the village was near equity (92:100). Figure 2 demonstrates that women across the life-span net-hunted significantly more frequently than men. On

[INSERT FIGURE 2 HERE]

occasion (2 hunts), the net-hunting groups comprised only women. The only net-hunts in which women did not participate were the net-hunts staged for tourists visiting the Dzanga-Sangha Reserve. "Tourist hunts" were opportunities to earn cash, as tourists paid a $10 fee to the group whether or not anything was caught. Tourists hired a small group of 10-15 hunters, and young men more interested in cash than in obtaining food items were the first to volunteer for this opportunity. It is also possible that the tourists and their park guides assumed that young men were the best or even the only net-hunters.

Women’s participation in net-hunting also determined whether net-hunting took place at all. No net-hunts took place on 32 of 90 observation days (36 percent). Second to rain, the most important excuse given by the community for not net-hunting was that "the women have all gone into the forest" to gather mushrooms, caterpillars or to fish. From the community’s perspective women’s participation was essential for net-hunting, whereas men’s participation was not.

Figure 3 shows a significant relationship between husbands' and wives'

[INSERT FIGURE 3 HERE]

participation (r2=0.28; p=.01), but does not necessarily indicate that the husband and wife hunted together or that one spouse caused the other spouse to hunt. Of 25 days in which one or both members of a couple participated in a net-hunt, the wife participated alone 49% of the time, the husband participated alone 16% of the time, and the couple participated together 35% of the time.

Time allocation studies are another way to evaluate female versus male participation in net-hunting. Individuals’ daily activities were recorded for a total of 19,656 total person days. Activities were categorized as follows: in or near the village of Mossapoula, on a net-hunt, engaged in formal labor, engaged in day labor, in forest camp, and absent. Individuals who stayed in Mossapoula performed a variety of tasks including farming, child care, fishing, gathering, tapping raffia wine, and making raffia roof shingles. Formal labor included employment with researchers, the US-based conservation agency, the logging company based in Bayanga, a European film-maker, and a World Bank forestry survey team. Only the film-maker hired women, although several entire families established a camp near the logging operation’s field site where the men were employed. Day labor for local residents of other ethnic groups was performed by women and men. Women generally worked in manioc fields or coffee plantations, receiving payment in-kind. Men sought payment in cash for clearing fields; collecting palm nuts, palm wine, raffia leaves, rattan, or construction poles; hunting with a gun; and portering in the diamond fields. Mossapoula residents also established several temporary forest hunting camps from 10-25 km to the east of Mossapoula. Finally, individuals and families frequently moved to other Aka communities on a temporary basis.

Table 2 summarizes male and female time allocation data and indicates that the

[INSERT TABLE 2 HERE]

most important activity for women outside the community was net-hunting, to which they devoted 18 percent of their time. Men, by contrast, devoted significantly less time,11 percent, to net-hunting (t = 4.69, p < 0.01). The most active net-hunter was a woman who hunted on 43 percent of observation days while the most active male participated in net-hunts on 32 percent of observation days. The woman hunter was a 40 year-old with grown children. Her husband was the indisputable leader of the net-hunt whenever he participated and also the leader (kombeti) of more than one forest camp. She frequently hunted in his absence with his net, which at 33 m was the longest (and heaviest!) net in the village.

It is difficult to determine yields by gender as both men and women may participate in captures. Men were usually considered the owners of the net and game animals captured in the net even though women were usually the ones to seize the animals in the net. One way to estimate returns by gender is to examine net-hunt returns in relation to proportion of female versus male participation on a hunt. Do returns increase if more females participate? Figure 4 indicates no statistical

[INSERT FIGURE 4 HERE]

relationship between female participation and return rate. But the figure does suggest some patterns: Table 3 compares net-hunts in which males or females

[INSERT TABLE 3 HERE]

represented 60% or more of the participants. It demonstrates that return rates were higher when women rather than men predominated. A possible complicating factor is that the average size of the predominately female hunting group was 43.3 while the average size of the predominately male hunting group was 26.5, but no statistically significant relationship exists between number of participants and return rates. On the nine most successful net-hunts (>2.5kg/hunter), the average size of the net-hunting group was 22.6; on six of these hunts females represented over 60% of the participants and the average group size was an especially small 17.8. These hunts also took place from the forest camp 10 km from the village where the camp size was smaller and game more abundant. But Figure 4 also demonstrates that net-hunts are not very successful when few males participate--some of the least successful hunts were when more than 85% of the participants were women. Overall, net-hunts are most successful when women and men complement each other.

In addition to the four primary game species exploited by net-hunters (blue duikers, brush-tailed porcupines, bay duikers, Peters duikers), women captured one python (Python sebae)weighing 15 kg and measuring over three meters in length, one long-nosed mongoose (Herpestes naso), and three tortoises (Kinixys erosa).

The calculations of net-hunt returns as grams of meat per hunter ignore the wide range of other forest products for consumption or sale that were collected during the hunt in addition to game, returns that were not quantified. Food items other than honey were gathered principally by women and likely tended to exceed the caloric and protein value of game meat obtained: "payo" (Irvingia excelsa) and other nuts, caterpillars, mushrooms, "koko" leaves (Gnetum buchholzianum), fruit, and wild yams (Dioscorea spp.). Women as well as men gathered additional forest products for sale or domestic use: "kosa" (Manniophyton fulvum) bark strips for making cord and nets, "ngongo" (Megaphrynium macrostachyum) leaves for wrapping food or roofing huts, and rattan (Eremospatha sp. and Ancistrophyllum secundiflorum) (Noss 1995, 1997).

Economic Alternatives

If women are interested in providing on a daily basis as evolutionary theory suggests, why did women not dedicate themselves to farming or helping other village women with their fields? These activities are generally more reliable and efficient than hunting (Hames 1989), and are often the preferences Aka women demonstrate in more "traditional"/undeveloped areas. The Mossapoula study was not undertaken to evaluate tradeoffs of alternative subsistence activities, so data to calculate return rates for each alternative activity were not collected, but it is possible to qualitatively evaluate other options. First, productivity of farming in the area was lowered by frequent losses to crop-raiding elephants. Second, in a community the size of Mossapoula, Aka farmers faced significant losses to other Aka community members who took with or without permission (Sarno 1998). When it was time to harvest a product all the relatives showed up and expected shares. Finally, farming sometimes meant competing with horticulturalists for relatively limited lands available for farming, since the National Park and Reserve prohibit farming more than 500m from the main road (Renner 1998). Access to farmland was unequal because villagers perceived the Aka as "primitive" and sometimes took over fields after Aka had cleared them.

Why didn't Aka females provide more day labor for villagers? As demonstrated in Table 2, this was a relatively important activity for Aka women, but since local horticulturalists viewed Aka as less than human they often tried to exploit their labor. As Bayanga turned increasingly to a market economy, agricultural work as day laborers drew the Aka into relations that were frequently exploitative and sometimes abusive.

Why did females not take up other small game hunting, such as hunting with tobas (small traps for mongoose and porcupine)? Cultural precedents influenced decision making as "traditional" Aka females seldom hunted with tobas or other small game techniques by comparison to net-hunting. Also, individuals or small groups conducted other small game hunting. Women may have preferred to hunt in larger groups to deter the possibility of harassment or possible rape by men from other areas. Agta women also prefer to hunt in groups whereas men hunt individually (Estioko-Griffin 1986), and Matses and Ache women accompany their spouses (Hurtado et al., 1985, Romanoff 1983).

It is also important to consider cultural precedents in evaluating economic tradeoffs. Cultural precedents are more than knowledge or experience. All Aka knew how to farm and have had some experience farming either on their own or for villagers. However, very few Aka had fields. Females net-hunted, in part, because African forest forager women seem to have had a long history of hunting and cultural precedents have social and emotional dimensions. Culture tends to make one feel that their behaviors are universal and natural. For instance, we eat particular foods as children and over time we feel like these are the best foods; we feel uncomfortable (even disgusted) if we are asked to eat a new food (i.e., caterpillars). One might call these food precedents. If one goes to a food market in a foreign country, one is likely to purchase foods based upon these food precedents. Precedents regarding task allocation are cultural in that we tend to feel that members of one gender are better at a task than members of the other gender. Task allocation precedents influence subsistence decisions, in particular when individuals are undergoing change.

The data demonstrate the significance and efficiency of Mossapoula Aka women net-hunters. Mothers, grandmothers, single and married women all participated in net-hunting. As previously stated, it is not surprising that women can hunt, but they do not do so in all contexts. Most of the contexts for female hunting predicted by the evolutionary and cultural theoretical orientations occurred among the Mossapoula Aka: women received relatively high caloric returns from hunting (but we do not know precisely how this compares to other subsistence options); game animals were acquired synchronously; hunting took place with other adults; Aka women had access to the means/technology of efficient hunting; Aka male ideological/political control of women was minimal; and, cultural precedents existed that enabled women to obtain knowledge and experience in hunting. We were unable to adequately evaluate the evolutionary prediction that women with infants net-hunt less frequently. We can only state that younger women (20-35 years-old) net-hunted just as frequently as older adult women (36-65 years old), and it is more likely that the younger women had infants. This suggests net-hunting may be more consistent with infant care by comparison to other hunting techniques (e.g., stalking). But Mossapoula Aka had smaller nets and did not hunt as long as net-hunters in other regions, which suggests women may modify the technologies and methods so they can carry an infant, basket and net.

While the contexts are consistent with theoretical predictions, it is not clear if the contexts are specific to Mossapoula Aka or if they apply to all Aka or all tropical forest net-hunters. Table 4 summarizes the generalizability of the various contexts that

[INSERT TABLE 4 HERE]

contribute to frequent female hunting. First (A in Table 4), it is important to point out that relatively greater gender equality and flexibility exist among mobile foragers by comparison to farmers, pastoralists and sedentary foragers where resources are stored and defended (Lee 1998). For instance, African forest foragers acquired net-hunting technology from Bantu-speaking farmers (Vansina 1986), but women farmers do not participate in farmers' net-hunts while they do among foragers. This may result from a general pattern of greater gender flexibility, but it is also probable that cultural precedents existed for women hunters before they adopted nets (e.g., the mota cooperative bow-and-arrow hunt among the Efe foragers [Bailey 1991]). In central Africa, farmers are much more likely than foragers to place female taboos on hunting technologies (e.g., nets, spears, cross-bows, guns). Extensive sharing and cooperation in subsistence and child care are also central to understanding female hunting, and the frequency and scope of sharing are greater among foragers than they are among farmers (Kelly 1995, Lee 1998).

Second (B in Table 4), most contexts that predict female hunting are also found among other African forager net-hunters (by comparison to African foragers that use other hunting technologies such as bow-and-arrow or spears). Regular female participation in net-hunting, weak male control ideologies, low risk of net-hunt failure by the group and the other items listed in the table are common to most African net-hunting forest foragers (Mbuti, Aka, Bakola/Bagyeli, Bongo). A few contexts are specific to the Aka (C in Table 4)--level of sharing and female control of division and distribution of game. Female control of the division and distribution of their hunting efforts appears to be an important factor in understanding female hunting. If women do not control the distribution of game they may be less interested in hunting. Data on the frequency and scope of sharing are limited, but the single systematic study of sharing among African forest foragers (Kitanishi 1996) together with descriptive accounts suggest that the Aka share the animals captured on the net hunt more widely than other forager net hunters.

Finally (D in Table 4), several cultural and environmental contexts specific to the Mossapoula Aka contribute to the persistence of regular female hunting. Nets are smaller, net hunts do not last very long and success rates are relatively high by comparison to other net-hunting groups. Human population densities are low (<1 person/km2) (Carroll 1988) and hunting by non-Aka is restricted by park guards patrolling the National Park and Special Reserve (Renner 1998).

The presence of a conservation agency, several scientific researchers, a logging company, and a film-maker in the region also dramatically effect Mossapoula Aka decisions regarding hunting. The employers of these groups have cultural precedents regarding the sexual division of labor just as the Aka have cultural precedents. The employers are Euro-Americans or Western-trained Africans from farming families and they usually hire Aka men as wage laborers because they view men as more appropriate for these positions. We do not know exactly why they prefer men (i.e., their gender ideologies), but cultural precedents for hiring men are apparent as only men are hired by the logging company, scientists and the conservation agency. It is particularly interesting that only men are hired for tourist net-hunts. Only the film-maker hired women because he wanted to film entire families.

In summary, most contexts identified in Table 4 are consistent with evolutionary and cultural predictions, but a few contexts are seldom discussed in the literature: 1) modifications of net-hunting technology and hunt organization; 2) female control over the distribution of game; and, 3) cultural precedents of those who hire wage laborers.

Both evolutionary and cultural theories regarding forager sexual division of labor are useful in that they correctly predicted several contexts of Mossapoula Aka frequent female hunting. While the data are generally consistent with the theoretical predictions, the details of the Aka case question some of the assumptions of the theories and identify factors that contribute to a more precise understanding of both cultural and evolutionary predictions.

Evolutionary theory indicates that women are likely to hunt for synchronously acquired small or medium sized game because they are more likely to be shared within the nuclear family than are large game. Aka women are more likely to hunt small and medium-sized synchronously acquired game as predicted, but they share the game extensively with others, in a similar way that the "show off" hypothesis predicts for big game-hunting males. Hawkes et al.'s evolutionary theory did not consider the method of acquiring game animals and assumed most animals were captured by individual or small group methods that she was familiar with among the Aché and Hadza. The theory also does not explain the extensive sharing of game found among the Aka. An evolutionist might predict that women have less to gain reproductively than men by "showing off" (i.e., extra matings), but may have a lot to gain in cooperation and friend-making. Cooperative subsistence techniques also generally lead to more extensive sharing (Hames 1989). But it is also essential to consider cultural precedents for the distribution of game as a long culture history exists among the Aka for dividing and distributing the game from net-hunts in this way. There is no indication that Mossapoula Aka division of net-hunting returns is dramatically different from that of Aka in other areas where both men and women participate in net hunting.

One of the cultural theories points out the importance of gender politics and taboos on women's use of efficient weapons, but the theory does not discuss the importance of female control in the distribution of game animals. The Aka data extend our understanding of the importance of female access to and control of resources. Friedl (1975) indicated that female access to meat resources explained variability in male dominance (i.e., , male dominance increased as men's meat contribution to the diet increased) in foraging societies, but utilized reproductive reasons (i.e., babies as constraining hypothesis) to explain why females did not hunt.

The Aka data question some of the assumptions of the evolutionary perspective and extend our understanding of one "cultural" perspective, but what is missing from the literature is any attempt to integrate these perspectives. Attempts to integrate disparate theoretical orientations are not popular because they seldom make anyone happy--each group tends to defend its turf--but we attempt to do so here as a synthesis seems to be an obvious step in theoretical development.

Brightman (1996) does an excellent job of evaluating the weaknesses of earlier "physiological" explanations for the sexual division of labor among foragers and demonstrates the importance of gender power relations for explaining why women do not hunt. His analysis, however, is limited because he does not attempt to explain why forager men in all parts of the world try to limit women's access to the best hunting technologies. His explanations could be enhanced by incorporating recent theoretical contributions on the evolution of human patriarchy (Hrdy 1997, Smuts 1994). The recent work points out that in contrast to males in most nonhuman primate societies, human males tend to control both resources and political power. Smuts (1994: 24) identifies several factors that led to pronounced patriarchy in humans and that are rare in primates: patrilocaity (most primates are female philopatric), male-male alliances for raiding of other males, male control over resources, increased hierarchy formation among men and the evolution of language and its power to create male-biased ideologies. She also points out that female choice--females selecting men with more resources--reinforces male control over females. It is necessary to understand the evolutionary factors that have contributed to the development of human patriarchy if we are going to understand gender inequality and gender politics. Brightman's descriptions of gender inequality among foragers are consistent with the history of patriarchy outlined by the evolutionists; the latter extends his theory because it explains why most forager men try to control politics and hunting technology. The show-off hypothesis also suggests that hunting is important to a forager man's ability to attract mates so it makes evolutionary sense that controlling hunting technology would be especially important to men.

While feminist evolutionists have examined the developmental history of patriarchy, human behavioral ecologists have not incorporated this work or Brightman's research on gender power relations into their explanatory models of the sexual division of labor. HBE points out that men and women have different reproductive trade-offs in gender allocation of labor, but they do not address gender differences in political power and access to technology. Bird (1999) notes that existing HBE studies explain female variability in time allocation of food production according to reproductive status, but HBE does not explain the differences in the resource choices of men and women. Incorporating gender power relations into HBE would enhance our understanding of the contexts of female hunting (i.e., when male control is low). Male control of technology is another consequence of males' greater concern with and investment in mating effort.

Evolutionists also tend to ignore the concept of culture. From a HBE perspective, culture is simply another way humans adapt to their environments and it does not have particular properties of its own. Thus Bird's (1999) summary and critique of the HBE literature on the forager sexual division of labor never mentions "culture". A cultural approach assumes otherwise--cultural precedents/schema are not assumed to be linked to reproductive interests and are more likely to result from relatively unique and arbitrary culture histories. For instance, in the New World, forager men are more likely than forager women to contribute the majority of the calories to the diet whereas in the Old World men and women are more likely to contribute equally or women contribute more than men (Hewlett unpublished data). This suggests gender task allocation is linked to particular culture histories rather than different natural environments: peoples that migrated into the Americas emphasized and had cultural precedents for male hunting. If we are to understand the contexts of the sexual division of labor in a particular area it is necessary to understand the culture history of task allocation (i.e., cultural precedents) in the group and the region.

In many instances the absence or presence of cultural precedents is consistent with evolutionary predictions. For instance, cultural precedents do not exist for female big game hunting and this is consistent with the show-off hypothesis. The cultural precedent for female net-hunting is also consistent with evolutionary predictions as yields are high. This is what one might expect as culture is often characterized as being "adaptive". But cultural precedents influence subsistence decisions in ways that are not predicted by evolutionary hypotheses. For instance, the cultural precedent for Aka female net-hunting makes it a viable subsistence option in a rapidly changing environment whereas Baka forest forager women 500m away on the other side of the Sangha River do not hunt for small or medium-sized game because they lack the cultural precedents. As social-economic conditions change, Baka women increase the amount of day labor they provide to village women. Also, game is abundant in the Bayanga area and net-hunting is not particularly difficult to learn, but women from the other resident ethnic groups do not net-hunt on their own or with Aka women because cultural precedents guide their decisions. Finally, cultural precedents pattern how one feels about performing particular subsistence activities and it is clear that one reason Aka women hunt is that they feel comfortable doing so.

It is worth noting that it will be difficult to distinguish evolutionary and cultural explanations. For instance, why do Aka men rather than women occupy most of the wage labor positions? Evolutionists might hypothesize that Aka men are more interested in these positions because, like hunting of large game animals, they attract status and show-off value. Men can bring in the "big bucks" rather than the big buck (Lupo, personal communication). Formal positions are rare for anyone in the country, but especially for Aka since farmers tend to view them as primitive. The cash income allows men to purchase a variety of city goods, such as cloth, shoes, radios and knives as well as Western medicines, canned foods and alcohol. Paychecks are similar to large game animal captures in that they are utilized by many people within a day or so.

The "cultural" ethnographic decision-makers on the other hand would argue that Aka men obtain these positions because the people hiring the Aka have their own cultural precedents regarding the sexual division of labor, which usually favors males. Detailed and systematic research are needed to distinguish evolutionary and cultural predictions.

An understanding of the contexts of female hunting is clearly linked to understanding both evolutionary and cultural factors related to the sexual division of labor among foragers. The Aka data presented in this paper and the limited review of the evolutionary and cultural perspectives suggest that the following four factors are especially important for understanding female hunting and the sexual division of labor among foragers:

1. differential male and female reproductive strategies

2. male control ideologies and level of patriarchy

3. culture history and precedents for female hunting

4. social and natural environments

The first factor is "biological" (i.e., genetic) and part of our mammalian heritage. Males and females have different reproductive interests and strategies: males concentrate time and energy in mating effort rather than parenting effort, while females concentrate more time and energy in parenting rather than mating effort. The second and third are "cultural" (i.e., symbolic and transmitted non-biologically generation to generation) while the fourth emphasizes the site-specific characteristics of the natural (e.g., availability, size and quality of game, seasonality) and social (e.g., number and types of neighbors, demographic features of the group and their neighbors, ideologies of neighbors) environment. The four factors are not free-standing as they interact and influence each other. For instance, differential male and female reproductive strategies contribute to male control ideologies and those ideologies in turn influence the establishment of cultural precedents in the sexual division of labor. This does not mean that everything "cultural" flows from reproductive strategies. Cultural precedents may be arbitrary and have a history of their own as suggested in the cultural histories of net versus bow hunting. The social and natural environment is a very large and complex group of factors that can influence the level of male control ideologies or the persistence or change of cultural precedents. The identification of these four factors underlines the importance of understanding biological, cultural and environmental factors if one seeks to explain female hunting or forager gender task allocation.

This study has contributed to ethnography and theory regarding the sexual division of labor among foragers. The study: 1) systematically documented a forager group where women net-hunted more frequently than men; 2) identified environmental and cultural contexts that contributed to frequent female hunting; 3) evaluated evolutionary and cultural predictions regarding the contexts of female hunting; 4) suggested modifications to both evolutionary and cultural theories that deal with female hunting and sexual division of labor among foragers; and, 5) offered a synthesis of evolutionary and cultural approaches to the sexual division of labor.

1. The Aka are also known as BaAka, BiAka, Bayaka, BaBenzele or BaBinga.

Field research was assisted by a grant from the Joint Committee on African Studies of the Social Sciences Research Council and the American Council of Learned Societies with funds provided by the Ford, Mellon, and Rockefeller Foundations. Additional funds were provided by two Grants-in-Aid of Research from Sigma Xi, The Scientific Research Society; and by the World Wildlife Fund under the United States Department of Agriculture Agreement No. 93-G-155. We would also like to thank Robert Moise, Daou Joiris and Kiyoshi Takeuchi for providing unpublished data, as well as Karen Lupo, Clare Wilkinson-Weber and several anonymous reviewers for comments on earlier drafts.

Bahuchet, S., 1975, Rapport sur une Mission Effectueé en Saison Sèche en Lobaye (République Centrafricaine): Observations sur la Vie d’une Famille de Pygmées Bayaka. Journal d’Agriculture Tropicale et de Botanique Appliquée 22:177-97.

Bahuchet, S., 1979, Pygmées de Centrafrique: Etudes Ethnologuqies, Historiques, et Linguistiques sur les "Ba-Mbenga" (Aka/Baka) du Nord/Ouest du Bassin Congolais. Paris: SELAF.

Bahuchet, S., 1991, Les Pygmées Aka et la Forêt Centrafricaine: Ethnologie Ecologique. Paris: Société d’Etudes Linguistiques et Anthropologiques de France.

Bahuchet, S. and H. Guillaume, 1982, Aka-Farmer Relations in the Northwest Congo Basin. Pp. 189-211 in Politics and History in Band Societies (ed. by E. Leacock and R. Lee). Cambridge: Cambridge University Press.

Bailey, R.C., 1991, The Behavioral Ecology of Efe Pygmy Men in the Ituri Forest, Zaire. Ann Arbor, Michigan: Museum of Anthropology.

Bailey, R.C. and R. Aunger, Jr., 1989, Net Hunters vs. Archers: Variation in Women’s Subsistence Strategies in the Ituri Forest. Human Ecology 17:273-97.

Biesele, M. and S. Barclay, 2001, Ju'/hoan Women's Tracking Knowledge and its Contribution to Their Husbands' Hunting Success. African Study Monographs 26: 67-84.

Bird, R., 1999, Cooperation and Conflict: The Behavioral Ecology of the Sexual Division of Labor. Evolutionary Anthropology 8: 65-75.

Brightman, R., 1996, The Sexual Division of Foraging Labor: Biology, Taboo, and Gender Politics. Comparative Studies in Society and History 38: 687-729.

Carroll, R.W., 1988, Relative Density, Range Extension, and Conservation Potential of the Lowland Gorilla in the Dzanga-Sangha Region of Southwestern Central African Republic. Mammalia 52:309-23.

Cashdan, E.A., 1985, Coping with Uncertainty: Reciprocity among the Basarwa of Northern Botswana. Man 20:454-74.

D'Andrade, R. and C. Strauss (eds.), 1992, Human Motives and Cultural Models. NY: Cambridge UP.

Delobeau, J.M., 1989, Yamonzombo et Yandenga: les Relations entre les Villages Monzombo et les Campaments Pygmées Aka dans la Sous-Préfecture de Mongoumba (Centrafrique). Paris: SELAF.

Demesse, L., 1978, Changements Techno-Economiques et Sociaux chez les Pygmées BaBinga (nord Congo et sud Centrafrique). Paris: SELAF.

Demesse, L., 1980, Techniques et Economie des Pygmées Babinga. Paris: Institut d’Ethnologie.

Estioko-Griffin, A., 1985, Women as Hunters: The Case of an Eastern Cagayan Agta Group. Pp. 18-32 in The Agta of Northeastern Luzon: Recent Studies (ed. by P.B. Griffin and A. Estioko-Griffin). Cebu City, Philippines: University of San Carlos.

Estioko-Griffin, A., 1986, Daughters of the Forest. Natural History 95:36-43.

Estioko-Griffin, A.A. and P.B. Griffin, 1981, Woman the Hunter: The Agta. Pp. 121-51 in Woman the gatherer (ed. by F. Dahlberg). New Haven: Yale University Press.

Friedl, E., 1975, Women and Men: An Anthropologist's View. NY: Holt, Rinehart, and Winston.

Gladwin, C.H., 1989, Ethnographic Decision Tree Modeling. NY: Sage

Goodman, M.J., P.B. Griffin, A.A. Estioko-Griffin and J.S. Grove, 1985, The Compatibility of Hunting and Mothering among the Agta Hunter-Gatherers of the Philippines, Sex Roles 12: 1199-1209.

Hames, R.B., 1989, Time, Efficiency, and Fitness in the Amazon Protein Quest. Research in Economic Anthropology 11: 43-85.

Harako, R., 1976, The Mbuti as Hunters: A Study of Ecological Anthropology of the Mbuti Pygmies. Kyoto University African Studies 10:37-99.

Harako, R., 1981, The Cultural Ecology of Hunting Behavior among the Mbuti Pygmies in the Ituri Forest, Zaïre. In Omnivorous Primates: Gathering and Hunting in Human Evolution, R.S.O. Harding and G. Teleki (Eds.). New York: Columbia University Press. Pp. 499-555.

Hart, J.A., 1978, From Subsistence to Market: A Case Study of the Mbuti Net Hunters. Human Ecology 6:325-53.

Hawkes, K., 1991, Showing Off: Tests of an Hypothesis about Men’s Foraging Goals. Ethology and Sociobiology 12:29-54.

Hawkes, K., J.F. O'Connell and N.G. Blurton Jones, 1997, Hadza Women's Time Allocation, Offspring Provisioning, and the Evolution of Long Postmenopausal Life Spans, Current Anthropology 38: 551-577.

Hewlett, B., 1977, Notes on the Mbuti and Aka Pygmies of Central Africa. M.A. thesis, California State University, Chico.

Hewlett, B., 1991, Intimate Fathers: The Nature and Context of Aka Pygmy Paternal Infant Care. Ann Arbor: University of Michigan Press.

Hewlett, B., 1996, Cultural Diversity among African Pygmies, In Cultural Diversity Among Twentieth-Century Foragers: An African Perspective, S. Kent (Ed.). Cambridge: Cambridge University Press. Pp. 215-244.

Hewlett, B. and L.L. Cavalli Sforza, 1988, Cultural Transmission Among Aka Pygmies. American Anthropologist 88: 922-934.

Hrdy, S.B., 1997. Raising Darwin's Consciousness: Female Sexuality and the Prehominid Origins of Patriarchy. Human Nature 8:1-49.

Hudson, J.L., 1990, Advancing Methods in Zooarchaeology: An Ethnoarchaeological Study among the Aka. Ph.D. diss., University of California, Santa Barbara.

Hurtado, A.M., K. Hawkes, K. Hill and H. Kaplan, 1985, Female Subsistence Strategies among Ache Hunter-Gatherers of Eastern Paraguay. Human Ecology 13:1-28.

Ichikawa, M., 1983, An Examination of the Hunting-Dependent Life of the Mbuti Pygmies, Eastern Zaïre. African Study Monographs 4:55-76.

Joiris, D.V., 1994, Elements of Techo-Economic Changes among the Sedentarized Bagyeli Pygmies (South-West Cameroon). African Study Monographs 15: 83-95.

Kelly, R.L., 1995, The Foraging Spectrum. Washington, D.C.: Smithsonian Institution Press.

Kitanishi, K., 1995, Seasonal Changes in the Subsistence Activities and Food Intake of the Aka Hunter-Gatherers in Northeastern Congo. African Study Monographs 16:73-118.

Kitanishi, K., 1996, Variability in the Subsistence Activities and Distribution of Food among Different Aged Males of the Aka Hunter-Gatherers in Northeastern Congo. African Study Monographs 17:35-57.

Kitanishi, K., 1998, Food Sharing among the Aka Hunter-Gatherers in Northeastern Congo. African Study Monographs Supplement 25: 3-32.

Lee, R.B., 1998, Hunter-Gatherer Studies and the Millennium: A Look Forward (and Back). Paper presented at the International Conference on Hunting and Gathering Societies, Osaka, Japan.

McCreedy, M., 1994, The Arms of the Dibouka, In Key Issues in Hunter-Gatherer Research, E.S. Burch and L.J. Ellanna (Eds.). Oxford: Berg. Pp. 15-34.

Motte, E., 1982, Les Plantes chez les Pygmées Aka et les Monzombo de la Lobaye (Centrafrique): Contribution à une Etude Ethnobotanique Comparative chez des Chasseurs-Cueilleurs et des Pêcheurs-Cultivateurs dans un même Milieu Végétal. Paris: SELAF.

Mukhopadhyay. C.C., 1980, The Sexual Division of Labor in the Family: A Decision-Making Approach. Ann Arbor: Univ. Mich. Microfilms.

Mukhopadhyay, C.C., 1984, Testing a Decision Process of the Sexual Division of Labor in the Family. Human Organization 43:227-242.

Mukhopadhyay, C.C. and P. Higgins, 1988, Anthropological Studies of Women's Status Revisited: 1977-1987. Annual Reviews in Anthropology 17: 461-495.

Noss, A.J., 1995, Duikers, Cables, and Nets: The Cultural Ecology of Hunting in a Central African Forest. Ph.D. dissertation., University of Florida, Gainesville.

Noss, A.J., 1997, The Economic Importance of Communal Net Hunting among the BaAka of the Central African Republic. Human Ecology 25:71-89.

Noss, A.J., 1998, The Impacts of BaAka Net Hunting on Rainforest Wildlife. Biological Conservation 86:161-7.

Noss, A.J., 2000, Cable Snares and Nets in the Central African Republic. In Hunting for Sustainability in Tropical Forests, J.G. Robinson and E. Bennett (Eds.). New York: Columbia University Press. Pp. 282-304

Ravenstein, E.G., 1901, The strange adventures of Andrew Battell of Leigh, in Angola and the adjoining regions. London: Hakluyt Society.

Renner, D., 1998, Situation Socio-Foncière, Stratégies d’Utilisation des Ressources Naturelles et Gestin Participative du Terroir: Cas de Bayanga, WWF-US, Bangui, Central African Republic.

Romanoff, S., 1983, Women as Hunters among the Matses of the Peruvian Amazon. Human Ecology 11:339-43.

Sarno, L., 1998, Social Organization, Access to Land and Natural Resources: The BaAka of Yandoumbé, WWF-US, Bangui, Central African Republic.

Smith, E.A., 2000, Three Styles in the Evolutionary Study of Human Behavior, In Adaptation and Human Behavior, L. Cronk, N. Chagnon and W. Irons (eds.) Hawthorne, NY: Aldine de Gruyter

Smuts, B., 1994, The Evolutionary Origins of Patriarchy. Human Nature 6: 1-32.

Soffer, O. J.M. Adovasio and D.C. Hyland, 1997, Textiles and Cordange: A Preliminary Assessment, In Pavlov I 1957 Excavations, J. Svoboda (Ed.). The Czech Republic: Archeologicky Ustav AV Brno, pp. 403-424.

Tanno, T., 1976, The Mbuti Net Hunters in the Ituri Forest, Eastern Zaïre--Their Hunting Activities and Band Composition. Kyoto University African Studies 10:101-35.

Trivers, R. 1972, Parental Investment and Sexual Selection, In Sexual Selection and the Descent of Man 1871-1971, B. Campbell (ed.). Chicago: Aldine de Gruyer, pp. 136-179.

Turnbull, C.M., 1961, The Forest People. London: Chatto and Windus.

Vansina, J. 1986. Do Pygmies Have a History? Sprache und Geschiche in Afrika 7.1: 431-445.

Wilkie, D.S. and B. Curran., 1991, Why Do Mbuti hunters Use Nets? Ungulate Hunting Efficiency of Archers and Net-Hunters in the Ituri Rain Forest. American Anthropologist 93:680-9.

Table 1. A comparative study of net-hunting among Congo Basin foragers.

Table 2. Mossapoula Aka time allocation by sex.

Note: N=90 observation days. Columns do not total 100 percent because not all individuals were on the records list for the full 90-day observation period.

Table 3. Differences in returns when there are more males or females on a net hunt.

Table 4. Cultural and Environmental Contexts of Female Hunting

1. Gender flexibility and versatility relatively common.

2. Extensive sharing in childcare and subsistence activities are relatively common.

B. Contexts common to most African forager net-hunters

1. Risk of net-hunt failure by the group is low.

2. Cooperation with other adults is common.

3. Most game animals are small or medium-sized.

4. A cultural precedent exists for regular female participation in net hunting which provides women extensive knowledge and experience with hunting.

5. Net-hunting is compatible with infant care (by comparison to spear or bow-and-arrow hunting of other African foragers).

6. Women have access to efficient hunting technology.

7. Male control ideologies are not as strong as in other forest foragers, in part because men and women cooperate in a broad range of subsistence and child care tasks.

1. Women control the division and distribution of game captured on net-hunts.

2. Sharing with other families appears to be more extenseive than among other net- hunting groups.

1. Nets are smaller and lighter than nets in other areas so they can be carried with an infant and/or basket.

2. Net-hunts do not last as long as net-hunts in other areas, but they are more intensive (i.e., more casts of nets).

3. Risk of failure by the individual or group is low. Game is relatively plentiful and return rates are higher than in most other Aka and African forest forager groups.

4. The presence of a conservation agency, logging company and a film-maker provide alternative resources and different cultural precedents regarding the division of labor.

5. Settlement size is large (300 individuals) so it is relatively easy to obtain enough individuals to net hunt.

Legends to Figures

Figure 1. Dzanga-Sangha National Park and Special Reserve (from Doungoubé 1990).

Figure 2. Proportion of days men and women in five age groups net-hunted.

Figure 3. Relationship between percentage of net-hunts participated in by husband and wife.

Figure 4. Relationship between percentage of women on net-hunt and total yield in grams per hunter.