To Give and to Give Not: The behavioral ecology of human food transfers

Abstract: 211 (Long), 102 (Short)

Main Text: 13,990

References: 3,179

Total Text: 17,482

Michael Gurven

Department of Anthropology

UC-Santa Barbara

Santa Barbara, CA 93106

gurven@anth.ucsb.edu, chatidye@hotmail.com

May 31, 2002

key words: behavioral ecology, costly signaling, cooperation, food sharing, foragers, reciprocal altruism

Long Abstract

The transfer of food among group members is an ubiquitous feature of small-scale forager and forager-agricultural populations. The uniqueness of pervasive sharing among humans, especially among unrelated individuals, has led researchers to evaluate numerous hypotheses about the adaptive functions and patterns of sharing in different ecologies. This paper attempts to organize available cross-cultural evidence pertaining to several contentious evolutionary models—reciprocal altruism, tolerated scrounging, and costly signaling. Debates about the relevance of these models focus primarily on the extent to which individuals exert control over the distribution of foods they acquire, and the extent to which donors receive food or other fitness-enhancing benefits in return for shares given away. Each model can explain some of the variance in sharing patterns within groups, and so generalizations that ignore or deny the importance of any one model may be misleading. Careful multivariate analyses and cross-cultural comparisons of food transfer patterns are therefore necessary tools for assessing aspects of the sexual division of labor, human life history evolution, and the evolution of the family. This paper also introduces a framework for better understanding variation in sharing behavior across small-scale traditional societies. I discuss the importance of resource ecology and the degree of coordination in acquisition activities as a key feature that influences sharing behavior.

Short Abstract

This paper attempts to organize available cross-cultural evidence pertaining to several contentious evolutionary models of food transfers in traditional populations—reciprocal altruism, tolerated scrounging, and costly signaling. Debates about the relevance of these models focus primarily on the extent to which individuals exert control over the distribution of foods they acquire, and the extent to which donors receive food or other fitness-enhancing benefits in return for shares given away. This paper introduces a framework for better understanding variation in sharing behavior across small-scale traditional societies, paying particular attention to the importance of resource ecology and the degree of coordination in acquisition activities

1. Introduction

Why do individuals give valuable resources away to others? To give or not to give is a special case of a more general dilemma--why do individuals engage in acts that incur personal costs and benefit others? Behavioral researchers are interested in discovering both the “ultimate” level evolutionary explanations for observed patterns of resource transfer across societies (Winterhalder 1997) and the “proximate” determinants that shape these and other kinds of costly pro-social behavior (Caporael et al. 1989). Anthropologists have focused on explaining the pattern of food transfer among small-scale subsistence economies. Psychologists and economists have tried to understand the motivations for altruistic, “other-regarding” behavior in western societies with market economies (e.g. Rose-Ackerman 1996; Andreoni 2001; Camerer & Thaler 1994). Behavioral biologists have studied several pro-social behaviors including food transfer (e.g. capuchin monkeys, chimpanzees, vampire bats), grooming (e.g. impala, chimpanzees, baboons), foraging (e.g. lions, African wild dogs, killer whales), and group defense. Costly pro-social behavior is viewed by all these researchers as “anomalous” (Dawes & Thaler 1988) because any behavior benefiting others at a substantial personal expense violates the “axiom of rationality” which assumes that higher levels of consumption provide higher individual utility or fitness.

One important source of information for understanding the evolution of pro-social behavior and cooperation is the rich literature on food transfers among people who meet their daily food needs from consuming wild foods and cultigens, with little access to modern markets. These are hunter-gatherers and small-scale forager-agriculturalists. The literature on food transfers among peoples practicing a subsistence economy has grown in the past twenty years. These data are useful for illustrating existing variation in cooperative sharing within and among groups, and may serve as a basis for systematic hypothesis testing.

Research among these groups is critical to resolve debates on the nature of human sociality and cooperation. First, evolutionary psychology emphasizes that the tendency for humans to cooperate, even among strangers in mock scenarios, experiments, and in real life, may be hard-wired due to a long evolutionary history of cooperative big-game hunting and food sharing (e.g. Hoffman et al. 1998; Cosmides & Tooby 1992). Common notions of fairness, equity, and punishment in many domains may have thus been shaped in the sharing context of a hunting and gathering lifestyle (Fehr & Schmidt 1999; Gintis 2000). These researchers should be concerned that assumptions made about hunter-gatherers are well-founded and that empirical results based on western, market-oriented groups are generalizable to a non-market, non-western context.

Second, if the economies of scale and the high levels of specialization found in complex societies were made possible by the development of a pro-social brain developed during a long evolutionary history of hunting and gathering, then understanding the flexibility of “pro-social” behavior may help increase our understanding of how humans have succeeded in generating cultural institutions favoring cooperative outcomes, and subsequently populating the globe. Indeed, it has been suggested that the ability to reap gains from cooperation may be responsible for the recent proliferation of Homo sapiens sapiens (Boyd and Richerson n.d.) at the expense of earlier hominid forms.

Lastly, reinterpretations of men’s hunting and sharing practices as mating rather than as subsistence “strategies”, have called into question traditional notions of the sexual division of labor and the origins of the family (Hawkes 1993; Bird 1999). The extent to which men’s food production and distribution strategies function as forms of family provisioning or as status display has repercussions on future depictions of the evolution of long-term pair bonds (i.e. marriage) and whether the nuclear family is best viewed as a cooperative or competitive enterprise. Whether men are an important source of calories for subsidizing women’s reproduction and child growth within the family can also impact our understanding of the evolution of fundamental human life history traits such as delayed childhood, long post-menopausal lifespans, and large brains (Hawkes et al. 1999; Kaplan et al. 2000).

Despite the growing realization that cooperation among hunter-gatherers is critical to resolving the important issues mentioned above, only a handful of ethnographic studies focus on food transfers, and few of these are systematic or quantitative, making cross-cultural comparison difficult (see below). However, references to food sharing and production in numerous ethnographies can be useful for highlighting observations that are inconsistent with particular hypotheses.

The goal of this paper is to synthesize what is known cross-culturally about within-group food transfers in light of current theory. A complete behavioral ecology of food transfers should explain the function or purpose for food transfers in the first place, as well as the social mechanisms responsible for maintaining different levels of food transfers within populations. It should also predict quantitative aspects of sharing, based on social context, local conditions, and features of resource ecology. Food sharing, for example, has been explicitly modeled as an efficient means of reducing the high daily variance in acquisition (Winterhalder 1986; Kaplan & Hill 1985; Smith 1988). Others have suggested a social purpose for food sharing, where giving acts as an honest signal of donor quality or intent (Smith & Bleige Bird 2000; Gurven et al. 2000b; Zahavi & Zahavi 1998). Since most developed models propose specific benefits to food sharing, we also require a way to specify the relative importance of each hypothetical benefit to observed patterns of food transfers.

Several theoretical models may explain trends in within-group transfers. The most prominent of these include kin selection (KS), reciprocal altruism (RA), tolerated scrounging (TS), and costly signaling (CS) (see Winterhalder 1997). Recent analyses of food sharing have led researchers to believe that several or all of these models might explain some of the variation within the same population (Hill & Kaplan 1993; Winterhalder 1997; Gurven et al. 2000a). Efforts in the past fifteen years have focused on testing alternative hypotheses that can distinguish between these models. To date, most sharing studies have focused on one or few populations. Answers to several key questions can potentially resolve important issues about the general applicability of these models to food sharing in non-market settings. These include: 1) Is (large) game a public good? Do acquirers have control over the distribution of kills? 2) Is food transferred consistently from ‘haves’ to ‘have-nots’? 3) Is giving food contingent on prior or expected future receiving? I survey available evidence on these topics, putting to rest the notion that hunter-gatherer food exchange can easily be explained by any one model. I argue that available evidence cannot rule out reciprocal altruism as an important determinant of most food transfers, nor can it entirely eliminate tolerated scrounging as an explanation of some food transfers. Nonetheless, scenarios of human life history, the sexual division of labor, and the evolution of the family that depend on a tolerated scrounging-based explanation for food sharing are on shaky ground because of the large number of food sharing observations that contradict predictions from that model. Costly signaling of genetic or phenotypic quality may also be an important influence on the production and distribution decisions of certain age and sex classes of individuals. However, many instances of food transfers seem designed to signal a willingness to cooperate, which suggests reciprocity may be a major component of food sharing behavior.

Cross-cultural analyses of sharing require a standard vocabulary for talking about sharing in different populations. Gurven et al. (2001) introduce four terms that describe different aspects of sharing. Sharing depth refers to the percentage of food production given to members of other nuclear families (e.g. 77% of all fish obtained is given to other families). Breadth is the number of other individuals or different families who receive from a given distribution, or alternatively over a given sample period (e.g. on average 4.3 families receive a portion from each deer killed). Equality reflects any disparities in amounts given to different individuals or families in the population (e.g. family B received 6.7% of the food produced by family A but family C received only 1.2% of A's total food production). Balance describes long-term differences in amounts transferred between pairs of individuals or families (e.g. family A gave 47 kg of meat but received back only 12 kg of meat from family B over a 3 month observation period). Each of these measures describes a separate domain of giving or receiving. These four measures allow detailed comparisons of sharing behavior within and across groups, and can therefore facilitate intra-cultural and cross-cultural hypothesis testing.

In this paper, I discuss transfers of all food types[1]. Early observations of extensive meat sharing among social carnivores, the absence of sharing among herbivores and frugivores (Price 1975), and the popularized role of hunting in hominid social evolution (Washburn & Lancaster 1968) have led to a biased focus on game distributions in the sharing literature. Transfers of gathered foods and other food items are either rarely mentioned in ethnographies and food sharing studies, or given only minimal treatment. Even when strong evidence suggests that transfers of game may be explained by a single model, as in the sharing of sea turtles among the Meriam according to tolerated scrounging (Bliege Bird & Bird 1997), identical patterns cannot be inferred for all other components of the diet. If the Meriam reciprocally share yams, bananas, and chicken, or if the Hadza reciprocally share roots and small game--foods which contribute significant calories to the diet--then the fact that large game may be shared according to tolerated scrounging in these societies tells only part of the story of forager food sharing.

2. Models of food sharing

Imagine a male forager with a fresh kill, or a female forager with a basket of fruits or roots. Each must decide (or have decided for them): 1) how much to give to others (depth); 2) How many families should receive a share (breadth) and 3) how much should be given to each of n other locally available individuals (equality)? Each model discussed below gives ceteris paribus conditions which predict when sharing should occur. These differ in the kinds of benefits returned to donors, and the manner in which these benefits are paid.

2.1. Kin selection-based nepotism (KS) Because biological kin share a percentage of ego’s genes through descent, kin-selected food sharing should favor biased transfers toward kin. The conditions which favor kin-selected sharing can be defined by a simple version of Hamilton’s rule (1964), as rB>C. An individual should give to kin when the benefits, B, to a recipient, weighted by Wright’s coefficient of relatedness, r, outweigh costs, C, to the donor.[2] B and C should be measured as impacts on survival and fertility, although these parameters have not been measured in any food sharing study among humans, and in only several cases among other organisms (Wilkinson 1984). It is important to remember that merely showing that kin receive food does not demonstrate nepotism, especially when the majority of one’s neighbors and peers may be related to some degree to any acquirer. A weak test of nepotism predicts that kin should at least receive more than non-kin, and close kin (r=0.5, offspring, parents) should receive more than distant kin (r=0.25, grandparents, grandchildren, r=0.125, first cousins). A stronger test must show that a kin bias is not just due to reciprocal altruism or tolerated scrounging.[3]

2.2. Reciprocal altruism (RA) One may also give portions of food to individuals with whom one has shared with in the past, and is likely to receive shares from in the future. The critical aspect of RA is that giving in the present is an incentive for receipt in the future. This is the concept of contingency (de Waal 1996; Gurven et al. 2000a; Hames 2000). Although tit-for-tat, as modeled via an iterated Prisoner’s Dilemma (Axelrod & Hamilton 1981), is often equated with RA in the game theory literature, tit-for-tat is only one manifestation of RA. A donor who gives a share to an unrelated individual may not know when he may receive a share in return, nor how much he is likely to receive, but may nonetheless give the morsel away, as long as time-discounted expected returns outweigh the costs of the initial sharing. RA, as well as KS and TS (below), are likely when B is significantly greater than C. Thus, the reciprocal transfer of unequal amounts of food is consistent with RA and expected from bargaining theory under a variety of conditions (see also Boyd 1992; Frean 1996). RA found in traditional societies may reflect a type of health insurance, where long-term benefits only sometimes outweigh the costs of giving (Gurven et al. 2000b).[4] Trade is a form of RA where the products given and received are in different currencies. Thus, meat for sex, fish for carbohydrates, honey for social deference, and fruit for assistance in clearing a field are examples of trade. While both trade and in-kind reciprocity yield net benefits to the donor, only in-kind reciprocity has the effect of risk- or variance-reduction in daily intake of specific food types (Hawkes 1993).

2.3. Tolerated scrounging (TS) If individuals get smaller increments of value from consuming additional portions of food, then remaining food portions will eventually be worth more to hungry individuals than to the sated acquirer. When one is unable to maintain control of a resource without paying a substantial cost to defend ‘surplus’ food, an acquirer should cede portions to other individuals if this price of defense is greater than the additional value that could be gained from consuming those extra pieces (Blurton Jones 1987). The acquirer should cede portions until all potential contenders have equal marginal consumption value or utility (Winterhalder 1996). Thus, TS describes food flows from haves to have-nots, when food given away is not contingent on shares received. If a producer can control who receives and how much, or if marginal value is linear or increasing (due to trade for example), then TS is unlikely to explain food transfers. As in RA, medium to large-sized items which are acquired intermittently are most susceptible to sharing by TS.

2.4. Costly signaling (CS) The food quest often involves tasks that require great risk, skill, stamina and vigor. If success in these tasks is due to certain valued characteristics of the acquirer, then engaging in those tasks may represent an honest signal of phenotypic quality. They are honest because they are not easily faked, and they can therefore provide reliable information about some quality of the acquirer. Although less explored, sharing can also be an honest signal of intent, either to initiate or maintain cooperative relations with other individuals. Thus, this different kind of signaling is consistent with RA models (Alexander 1987; Frank 1985; Gurven et al. 2000b).

CS of phenotypic quality is similar to the show-off hypothesis (Hawkes 1991; 1992), but differs in two important ways. First, it does not require TS-based sharing. It therefore does not assume that sharing is determined only by resource package size and asynchronicity in acquisition. Second, CS avoids the second-order collective action problem of who should reward generous sharers, because those that choose sharers as mates, allies, or other social partners, do so as a response to the advertised qualities of those individuals, and NOT as a form of payback for transferred food or as an encouragement for the good provider to stay with the social group (Smith & Bliege Bird 2000). Thus, donors should not resent a lack of giving on behalf of past recipients, nor should recipients feel obliged to return benefits to a donor. Applications of the show-off hypothesis have only been invoked to explain men’s foraging and sharing decisions, and with respect to large game, due to the proposed mating benefits accorded high status, even though signaling benefits may also include alliance building, social support, and mating opportunities for offspring. It is not invoked to explain food transfers by men of other resources (e.g. fruits, roots, honey, firewood) nor of food transfers by women.

3. Predictions of sharing models

The relevance of these models with respect to any particular society is difficult to assess because many predictions are consistent with several of the models. An analysis of the specific costs and benefits of sharing necessary to compare the impact of each model would require a level of estimation unseen in existing quantitative analyses. For this paper, I focus on several key predictions that are most useful for distinguishing among the four models:

3.1. Producer Control An assumption of TS is that producers have little to no control over who receives shares of items they acquire because these items are relinquished to those with greater need. TS asserts that only relative need and power should have any influence on the direction of food transfer. Without producer control, any agent-centered model that tries to understand directed transfers as a function of individual payoffs is suspect, unless the ‘goals’ of the appropriate decision-maker(s) correspond with those of the acquirer. Thus, lack of producer control over redistribution is inconsistent with KS and RA, but is consistent with CS.

3.2. Need The principal determinant of food flows in TS is the need of potential recipients relative to that of the acquirer. Assuming equal ability to defend resources (resource holding potential), food portions should flow to recipients until all possess the same marginal value of consumption (Winterhalder 1996). TS therefore directs food flows from haves to the have-nots, and in the simplest scenario (i.e. no differential information or travel costs, equal marginal values for additional portions), egalitarian distributions among all recipients (including the acquirer) are expected. Any strong bias in food sharing, towards kin (KS), neighbors, specific individuals (RA), etc. is therefore inconsistent with this assumption, unless these preferred recipients show greater relative need than other potential recipients or can obtain benefits at a smaller cost (smaller traveling or monitoring costs, for example). According to CS, we should also not find biased transfers toward privileged others based on need, because the payoffs to signaling derive only from the honest display of production to a wide audience, and not from giving to specific individual.

3.3. Contingency Only RA requires that food be given on condition of expected future receipt. Producers giving more to specific people should receive more back from those people, and similarly, those who do not give should not receive. This requires some form of punishment or ostracism of “defectors”. If shares are returned in the future, the net present value of expected future shares should at least compensate for the present costs of giving. As mentioned above, a contingency effect is generally inconsistent with TS[5]. Although CS does not require contingency among specific pairs of individuals, someone, perhaps other than recipients, is required to provide a benefit to offset the costs of giving up food to signal quality. Thus, according to CS, donors should not be angry or upset if recipients do not return favors, nor should recipients feel obligated to return those favors. It is important to emphasize that CS requires a generalized payback from others, whereas only RA requires a payback from past recipients. KS provides automatic benefits through increased inclusive fitness, while TS avoids a cost and thus provides no benefit.

Much theoretical work and ethnographic discussion on sharing has focused on function--reducing the risk of daily food shortfalls or reducing intake variance due to variance in acquisition (Winterhalder 1986; Smith 1988). It is important to realize that RA, TS, KS, and CS can all produce these effects, thus demonstrating group-level benefits from food sharing practices is not revealing.

The importance of surveying what is known about foragers in relation to these individual-oriented models has become evident in light of the issues raised in the beginning of this paper, particularly the recent arguments over men’s foraging goals (Hawkes 1993; Hill & Kaplan 1993), the sexual division of labor (Bird 1999), and the evolution of a human life history (Kaplan et al. 2000; Hawkes et al. 1998). If foragers lack producer control and if nothing is given in return for that which is received, then the production of large, asynchronously acquired resources, (i.e. wild game or any moderately large, valuable resource) is a partial public good, because others cannot be excluded from receiving shares. Food production, or allocation to the public good, is thus viewed as a collective action problem because non-producers consume portions without paying any production costs. Without producer control and contingency, the traditional notion of hunting as a family provisioning strategy is therefore suspect. It is then argued that men hunt and share game widely as a form of mating effort, vis à vis the show-off hypothesis and CS of phenotypic quality.

TABLE 1. Worldwide Ethnographic Sample

HUNTER-GATHERERS (31)

Africa North America

Hadza (Hawkes et al. 1991; 2001; Dogrib (Helm 1972)

Marlowe n.d.) Central Eskimo (Damas 1972; Balikci 1970)

Dobe !Kung (Lee 1972) Mistassini Cree (Rogers 1972)

G/wi Bushmen (Silberbauer 1981; Washo (Price 1975)

Tanaka 1980) Tolowa (Gould 1981)

Nyae Nyae !Kung (Marshall 1976) Tututni (Gould 1981)

Kutse Basarwa (Kent 1993) Coast Yurok (Gould 1981)

Mbuti Pygmies (Ichikawa 1983; Shoshoni (Fowler 1986; Steward 1938)

Harako 1976)

Efe Pygmies (Bailey 1991)

Aka Pygmies (Bahuchet 1990;

Kitanishi 1996; 1998)

South America Australia

Pilaga (Henry 1951) Gunwinggu (Altman 1987)

Yora/Yaminahua (Hill and Kaplan 1989) W. Desert Aborig. (Gould 1981; Myers 1988)

Ache (Kaplan and Hill 1984; 1985) Yolngu (Peterson 1993)

Siriono (Holmberg 1969) Pintupi (Myers 1988)

Hiwi (Gurven et al. 2000a)

Kaingang (Henry 1941) Southeast Asia

Ayoreo (Bugos and McCarthy 1984) Agta (Peterson 1978; Griffin 1984)

Lengua (Grubb 1911) Onge (Bose 1964)

Batek (Semang) (Endicott 1988)

FORAGER-AGRICULTURALISTS (11)

South America Africa

Maimande (Aspelin 1979) Basarwa Kung (Cashdan 1985)

Yanomamo (Hames 2000; 1990) Tswana/Kalanga (Cashdan 1985)

Yuqui (Stearman 1989)

Ache (Gurven et al. 2000b; 2001; 2002)

Chácobo (Prost 1980) Islands

Ifaluk (Betzig and Turke 1986; Betzig 1988;

Sosis 1997)

Meriam (Bliege Bird and Bird 1997;

Bliege Bird et al. n.d.)

Batak (Cadelina 1982)

Lamalera (Alvard and Nolin 2002)

4. The cross-cultural record

Table 1 lists all the hunter-gatherer and forager-agriculturalist groups for which I was able to find explicit quantitative or qualitative descriptions of food transfer patterns. Quantitative studies are in italics. Of the 41 groups listed, 29% are from South America, 24% from Africa, and the remaining from Australia, North America, and Southeast Asia. While these percentages may not accurately reflect the worldwide representation of foragers and small-scale non-market economies, this list includes all available studies that I could find in the literature. Information on each topic discussed below was not available for all groups listed in Table 1, and so omission of a group for a specific topic does not necessarily imply an absence of that behavior in the group.

4.1. Do producers have control over distributions?

Descriptions of widespread sharing where everyone present in camp sometimes receives portions of a kill (e.g. Western Desert Aborigines, Ache, G/wi), where kills are handed over and butchered by individuals other than the hunter (e.g. Ache, Efe Pygmies, Gunwinggu, Ona), where specific cultural rules delineate which classes of individuals receive specific portions of game animals (e.g. Copper Eskimo, Aka Pygmies, Gunwinggu and Western Desert Aborigines), or where hunters receive no more than other band members (Ache, Batak), have led some investigators to conclude that hunters exert little influence over the distribution of game (Dowling 1968; Hawkes 1993; Bird 1999). Without producer control, the question “Why bother sharing if the spoils go to other people?” is a legitimate concern because food may then be viewed as a public good. As argued above, if exclusions are possible due to a moderate level of producer control over the character of distributions, then game is not a public good. Observing the extent of producer control is confounded by a lack of understanding how distribution decisions are made in the context of the conflicting push and pull of interested parties. It is also confounded by the implicit assumptions that a lack of control is signified by a hunter’s receiving 1/n and that complete control is viewed as an ability to hoard 100% of a resource. However, keeping 1/n does not signify a lack of control if the acquirer decides that 1/n is the optimal portion to keep, given the expected payoffs to sharing. Even when hunters relinquish complete control of game, as among the Ache, such abandonment may be voluntary, as Ache do not relinquish control when at the reservation (Gurven et al. 2002).

Producer control of distribution is indicated by several common ethnographic distribution patterns. Many studies report biased distributions, preferential shares to acquirers and their families, or more frequent sharing to close kin outside the nuclear family at the expense of more distant kin and unrelated individuals [Gunwinggu (Altman 1987), Copper and Netsilik Eskimo (Damas 1972), Pilaga (Henry 1951), Hiwi (Gurven et al. 2000a), Kaingang (Henry 1941), Batek (Endicott 1988), Pintupi (Myers 1988), Washo (Price 1975), Yanomamo (Hames 1990), Basarwa (Cashdan 1982), Ifaluk (Sosis 1997), Agta (Griffin 1984), Ache (at reservation) (Gurven et al. 2001), and Machiguenga (Kaplan pers comm)]. While it is possible that close kin may be more likely to live in closer proximity than other individuals (and hence more likely to demand shares), the few studies which examine both kinship and distance reveal that close kin receive more than other individuals, even when controlling for residential distance [Hiwi (Gurven et al. 2000a, Ache (at settlement) (Gurven et al. 2001)]. An additional bias common in many forager societies is the bride service tradition, where young men must provide meat for their new wife and in-laws [!Kung (Leacock & Lee 1980); Yanomamo (Ritchie 1991); Hadza (Woodburn 1998)].

Expectations of sharing are usually greatest in camp, which leaves the option for some hunters to consume small portions of their catch at or near the kill site prior to transporting it back to a communal camp. Indeed, hunters are frequently allowed to consume internal organs and marrow from animals they kill at the kill site [e.g. the !Kung (Speth 1990) the G/wi (Silberbauer 1981), the Nyae Nyae !Kung (Marshall 1976), the Hadza (Woodburn 1998) and the Batek (Endicott 1988)], where “no one begrudges them this right” (ibid:117). Several Lengua men gorged themselves full of ostrich eggs, returning to camp with only a few, so that they wouldn’t have to share with those who were not producing enough (Grubb 1911:190). Ache hunters, for example, could potentially bring family members to cook and consume meat at the kill site, but this never happens. In all of these groups, much food is transported to camp, an observation that is consistent with a desire to share food[6].

A higher percentage of big game is distributed to more families than small game in all groups where the effect of package size has been examined [Hiwi (Gurven et al. 2000a, Ache (Kaplan and Hill 1985; Gurven et al. n.d.), Dobe !Kung (Lee 1979), Kutse (Kent 1993), Yanomamo (Hames 1990), G/wi (Silberbauer 1981), Nyae Nyae Kung (Marshall 1976), Ifaluk (Sosis 1997), Aka (Kitanishi 1998)], which suggests either greater opportunities for hunters to gain benefits through increased exchange (due in part to diminishing returns to hoarding for the acquirer) or that producers have increasingly less control over distributions. Even if greater sharing depth and breadth were indicative of declining producer control, producers often receive significantly more than 1/n, thereby making the production of large resource packages worthwhile. During one season in 1987, a Gunwinggu family composed only 20% of the band, provided 41% of the band’s total calories, and kept twice as much as the other household cluster (Altman 1987). Similarly, Hiwi and Ache families represented 3% and 5% of their village settlement populations in 1990 and 1998, and kept 20% or more of what they acquired, including meat, giving the rest to fewer than 6 other nuclear families (out of 23 and 36, respectively) (Gurven et al. 2000ab). While Yora families divide game equally on forest trips, they keep about 40% of acquired game at the village settlement, giving the rest to 3 (out of 10) other families (Hill & Kaplan 1989). About 69% of acquired meat was kept within the family of Yuqui hunters, with the rest given to about 5 other hunters out of 15 (Stearman 1989). Yanomamo hunters kept twice as much food for their families than was given to each other family (Hames 2000). Similarly, Hadza hunters’ share of large game items are almost twice as large as those given to others (Hawkes et al. 2001).

If hunger gives others claim to shares, thereby reducing producer control, then it is unclear why smaller resource items are frequently kept within the nuclear family of the acquirer even though others may be hungry. Small game, such as steenbok, duikers, and tortoises, are frequently consumed within an acquirer’s family among the Dobe !Kung (Lee 1972) or those “people close to the hunter” among the G/wi (Silberbauer 1981), even though the size of some of these small animals is comparable to those which observe wide sharing among other groups, such as the Ache. Thus, as reported among Western Desert Aborigines, even small game meat is distributed as tiny portions so that “everyone in camp gets a share” (Gould 1981:432).

Others’ hunger levels should also increase during periods of food scarcity. According to TS, any increased demand for food should increase the breadth and/or depth of sharing. Case reports of the Ik (Turnbull 1972), the Ojibwa Indians (Bishop 1978) and the Northern Shoshone (Moulton & Dunlay 1983) however, demonstrate less sharing during stressful times. The Batak share with significantly fewer households during the pre-harvest season when food is scarce. The average geographical distance between sharing households during this time is about one-half the distance during more plentiful seasons (Cadelina 1982).

Another common pattern among the subset of groups where men hunt cooperatively is for game to be distributed initially among all participants in the hunt [Netsilik Eskimo (Damas 1972), Nyae Nyae !Kung (Marshall 1976), NW Coast Indians (Gould 1981), Ifaluk (Sosis 1997), Pintupi (Myers 1988), Washo (Price 1975), Mbuti (Ichikawa 1983), Aka (Kitanishi 1996; 1998), Efe (Bailey 1991), Shoshone and Paiute (Fowler 1986), Hiwi (Hill pers. comm.)]. Several ethnographies are explicit about subsequent exclusive ownership of meat shares upon initial receipt in a primary distribution, regardless of whether or not others have received their own shares [Mbuti (Ichikawa 1983), Nyae Nyae !Kung (Marshall 1976), Kaingang (Henry 1941), Efe (Bailey 1991)]. This is exemplified by Marshall’s statement about the Nyae Nyae !Kung that “when an individual receives a portion of meat, he owns it outright for himself. He may give and share it further as he wishes, but it never becomes family or group property” (1976:363). Similarly, Bailey writes that while cooperatively acquired game is shared among Efe hunters, meat acquired by solitary hunters is “entirely his to allocate as he pleases” (Bailey 1991:100).

While frequent protestations often make distributions the subject of strife, the occurrence of demand sharing (Peterson 1993; Woodburn 1998) does not imply a lack of producer control due to high costs of defending resources. Henry (1951) reports that Pilaga families are able to bias food towards specific households despite the objections of other individuals. Among the Siriono, “one may be accused of hoarding food, but the other members of the extended family can do little about it except to go out and look for their own” (Holmberg 1969:88). People do not have automatic claim to others’ acquisition among the Pintupi, where “sharing often takes place only on request” (Myers 1988). Aka Pygmies often do not share food, and “…distribution within the camp is actually voluntary…the family chooses whether or not it shares its meals and with whom it shares…temporary disappointment is evident when a household is left out of a distribution” (Bahuchet 1990:38). While the Agta are reported to share most foods equally among available families, they often set aside separate portions of meat to be used in trading for carbohydrates with non-Agta neighbors (Griffin 1984).

4.2. Does food flow according to need?

Much has been written about the emphasis placed on generosity, and the “moral obligation” to help others in need among traditional societies (Barnard and Woodburn 1988), exemplified by the Chácobo proverb, “If you are a human being, then you will share what you have with those who are in need” (Prost 1980:64). Marshall writes that among the Nyae Nyae !Kung “…if there is hunger, it is commonly shared. There are no distinct haves and have-nots” (1976:357). The complementary side of praising generosity is condemning stinginess. “The most serious accusations one !Kung can level against another are the charge of stinginess and the charge of arrogance.” (Lee 1979:458). Similarly, one of the most serious Ache insults is to call somebody mella (a non-giver). The Yanomamo are “so preoccupied with the possessions (including food) of others…anyone who has more than a day’s supply of anything is a potential target of an accusation of stinginess if he does not share” (Hames 1990:103). Lengua who insist on keeping food for themselves are similarly “hated and terrorized by others” (Grubb 1911:190). These descriptions support the view that social dynamics in small-scale societies are organized by an ethic of ‘assertive’ egalitarianism (Woodburn 1982) and that ‘demand sharing’ equalizes differences due to production ability. Because strong pooling norms reduce variance in benefits as well as costs, certain leveling mechanisms have been proposed as cultural means of limiting the arrogance and wealth accumulation of hunters (or anyone for that matter) (Dowling 1968; Woodburn 1982; Wiessner 1996). These include ridicule of a hunter’s prowess [!Kung (Lee 1979)], taboos against hunters consuming portions of their own kills [e.g., Ache (Clastres 1972), Hadza (Woodburn 1982), Ona (Bridges 1948)], and explicit sharing rules [e.g., Central Eskimo (Damas 1972), Gunwinggu (Altman 1987)]. Additionally, it has often been stated that refusing to give shares to others upon request is “the ultimate sin” (Prost 1980:52), and that even when food is not obligatorily indebted to others, requests for shares are rarely denied [e.g. Batek (Endicott 1988), Pintupi (Myers 1988), Kaingang (Henry 1941), Kutse (Kent 1993)].

These cultural notions manifest themselves in ways that encourage egalitarianism. Anecdotes of horticulture failing among the Hadza (Woodburn 1982), Batek (Endicott 1988), Hiwi (Hill pers. comm.), and Agta (Headland 1986) due to incessant pressures on the hardest-working to give away the bulk of their production, are consistent with assertive egalitarianism. The fact that men still hunt even though some selfish benefits may be denied via various leveling mechanisms suggests that hunters either retain additional portions (as argued above), gain other benefits through reciprocity or trade, or obtain mating or other benefits through costly signaling (see below)[7].

While norms regarding ideal distributions are prevalent cross-culturally, they do not necessarily eliminate producer control or producer advantage, nor do they indicate that givers do not gain any advantage by helping needy individuals. Cultural rules or expectations need not mesh with daily transactions (Pennington & Harpending 1993). Indeed, Altman and Peterson (1988) report that explicit sharing rules for dividing large macropods among the Gunwinggu account for only 50% of game items. Among the Aka, estimates of the percentage of different game items shared to other individuals differed substantially from the amounts predicted by sharing rules (Kitanishi 1998). Extensive descriptions of quarrels over food distributions among the !Kung, the Siriono, and the Yanomamo are also testament to the fact that rules do not always cleanly predict behavioral outcomes.

There is quantitative evidence that giving does indeed reflect the relative need of recipients. Among Ache (Kaplan & Hill 1985; Gurven et al. 2001), Maimande (Aspelin 1979) and Hiwi (Gurven et al. 2000), shares are given in proportion to the number of consumers within the recipient family. These observations are consistent with both TS and RA. Families with high dependency tend to be net consumers while those with low dependency are net producers among the Batak (Cadelina 1982). Among the G/wi, the largest shares of game are first given to families with kids, then to those without kids, and the smallest shares are given to single individuals (Silberbauer 1981). There is also some description of younger Ache, Gunwinggu, Efe, Kutse and Agta hunters ceding portions of game to older men who may bias distributions in their favor, with the end result that older hunters with more children (and hence greater caloric demand) benefit more from sharing than do the younger hunters with small or no families. Furthermore, prolific hunters often subsidize other band members, and often give away more than they receive back [Yuqui (Stearman 1989), Ache (Gurven et al. 2000b), Hiwi (Gurven et al. 2000a), Kutse (Kent 1993), Efe (Bailey 1991)]. Even at a permanent Ache settlement where cultivated foods constitute the majority of the daily diet, higher producers give an increasingly higher proportion of their production away to members outside their nuclear family (unpublished data), consistent with the notion of a progressive tax on income (Woodburn 1982).

There is, however, little question that limitations on the kinds and amounts of benefits that accrue to good hunters exist, and that self-interest models which ignore constraints of group living will not completely explain variation in food sharing patterns. Group living implies a series of trade-offs where high producers may compromise their production in exchange for some other group-derived benefit, such as defense, trade, and increased mating access. If individuals are free to move among bands or villages (except for transaction costs), then these group-derived benefits (and not risk-reduction) must influence the perceived costs and benefits of sharing decisions when donors give more than they receive. Empirical studies need to explore the possibility that consistently generous individuals may receive prestige, support, or social insurance (discussed below), and that these social benefits have fitness consequences, before concluding that generous donors give according to TS. Alternatively, recent cultural group selection models may also offer insight into the evolution of costly giving, as opposed to other costly displays of phenotypic quality (Wilson 1998; Boyd & Richerson 2001).

Need is a salient component of sharing, but it does not dictate the entire character of daily distributions. Necessity for food can be due to differential abilities, knowledge, luck, or high dependency ratio, and there is no reason to expect the same patterns of distribution for all four causes of need (see Section 5). Furthermore, biases in distributions mentioned in the previous section, as well as the influence of proximate factors, such as population size and privacy, can all influence the salience of need in food transfer decisions.

Need may also not correlate with sharing outcomes when individuals differ in what is referred to by biologists as ‘resource holding potential’ (RHP). RHP includes physical prowess, authority, social influence, or any ability that can allow an individual to defend resources more easily, or to extort resources from other less powerful individuals. According to TS, only powerful individuals can avoid relinquishing shares to hungry individuals. RHP has never been measured in any society, especially since any single factor, such as muscular strength, fighting ability, or age, may not accurately predict RHP. Many observations, however, are inconsistent with RHP-based predictions. People often save plates of food for absent individuals, even though other group members may not receive any portions. Hungry children often receive food from adults other than their parents. Village chiefs and influential individuals often give away more food than they receive.

4.3. Do donors get back more utility than they give away?

The notion that giving is conditional upon expectations of future receiving (based perhaps on past receiving) is difficult to test. Sahlins’s (1972) “generalized reciprocity” implies that in-flows and out-flows should balance over the course of people’s lives, but that daily giving is done without reference to any accounting procedure. As pointed out by Hawkes (1992), this general anthropological description of reciprocity differs from the way RA is commonly used among biologists and evolutionary anthropologists. The maintenance of RA requires that beneficiaries give a return benefit back to the original donor. Several factors are crucial in determining how much is returned to pay back a donor: the cost to the donor of giving, the benefit to the recipient, the time delay into the future when a benefit is returned, and the benefit to the original donor of receiving in the future. A suitable condition for RA occurs when benefits to recipients greatly outweigh the costs to donors—precisely the need-based condition compatible with TS and KS. One problem with identifying and measuring contingency lies in the choice of an appropriate time frame over which reciprocation should occur (Hawkes 1992; Gurven et al. 2000a). At which point is a lack of reciprocation considered a defection? Does giving back half of what one was given constitute an act of reciprocation or defection?

Economic bargaining theory offers an appropriate way for understanding contingency and RA (Ståhl 1972; Hill & Kaplan 1993; Sosis 1998; Gurven et al. 2000a). Donors should give as long as the expected future benefit outweighs the current costs of giving relative to other options; thus, the exchange of unequal quantities is often consistent with RA. Figure 1 shows an Edgeworth box representing the exchange of A’s present production for B’s future production. Concave curves radiating from the lower left and upper right corners represent the utility A and B derive from consuming some combination of A’s (or B’s) present and B’s (or A’s) future production. The oval region in the interior represents the “bargaining zone”. A and B can both expect to gain if the final bargain is struck anywhere in this region, although they may not benefit equally. Where the final bargain is struck should be influenced by the relative bargaining power of the interactants, which reflects the expected cost from giving and benefit from receiving a specific quantity of food. These costs and benefits could vary with the amount of existing wealth, influence, production ability, status, or number of dependent offspring. Thus, exchange does not have to be perfectly balanced in order to be perceived as beneficial to involved parties and maintained by RA.

FIGURE 1. An Edgeworth Box of Food Exchange