London School of Economics and Political Science, University of London
Man, New Series, Vol. 17, No. 3 (Sep., 1982), 431-451
Greater equality of wealth, of power and of prestige has been achieved in certain hunting and gathering societies than in any other human societies. These societies, which have economies based on immediate rather than delayed return, are assertively egalitarian. Equality is achieved through direct, individual access to resources; through direct, individual access to means of coercion and means of mobility which limit the imposition of control; through procedures which prevent saving and accumulation and impose sharing; through mechanisms which allow goods to circulate without making people dependent on one another. People are systematically disengaged from property and therefore from the potentiality in property for creating dependency. A comparison is made between these societies and certain other egalitarian societies in which there is profound intergenerational inequality and in which the equality between people of senior generation is only a starting point for strenuous competition resulting in inequality. The value systems of non-competitive, egalitarian hunter-gatherers limit the development of agriculture because rules of sharing restrict the investment and savings necessary for agriculture; they may limit the care provided for the incapacitated because of the controls on dependency; they may in principle, extend equality to all mankind.
In a work published after his death, Malinowski made the splendidly forthright declaration that 'authority is the very essence of social organisation' (1960: 61). I am going to talk about a type of social organization, not understood in Malinowski's day, in which individuals have no real authority over each other. This lecture is about certain societies in which there is the closest approximation to equality known in any human societies and about the basis for that equality. I have chosen to use the term 'egalitarian' to describe these societies of near-equals because the term directly suggests that the 'equality' that is present is not neutral, the mere absence of inequality or hierarchy, but is asserted. The terms 'egality', from which 'egalitarian' is derived, was introduced into English with its present meaning in a poem by Tennyson in 1864 to suggest politically assertive equality of the French variety.1 Even today 'egalitarian' carries with it echoes of revolution, of fervor for equality in opposition to elaborate structures of inequality. But politically assertive egalitarianism is, of course, not found only in hierarchical systems under challenge and in their successor regimes. It is equally characteristic of many systems without direct experience of elaborate instituted hierarchy. Yet it may still seem surprising that equality should be asserted in certain very simply organized contemporary hunting and gathering societies which I am going to talk about, and in which, one might think, equality should be taken for granted.
In these societies equalities of power, equalities of wealth and equalities of prestige or rank are not merely sought but are, with certain limited exceptions, genuinely realized. But, the evidence suggest, they are never unchallenged. People are well of the possibility that individuals or groups within their own egalitarian societies may try to acquire more wealth, to assert more power or to claim more status than other people, and are vigilant in seeking to prevent or to limit this. The verbal rhetoric of equality may or may not be elaborated but actions speak loudly: equality is repeatedly acted out, publicly demonstrated, in opposition to possible inequality.
It is noteworthy that although very many societies are in some sense egalitarian, those which inequalities are at their minimum depend on hunting and gathering for their subsistence. For reasons which I shall seek to explain, only the hunting and gathering way of life permits so great an emphasis on equality. But there is, of course, no question of the equality being a simple product of the hunting and gathering way of life. Many hunter-gatherers have social systems in which there is very marked inequality of one sort or another, sometimes far more marked than the inequalities in certain simple agricultural or nomadic pastoral societies.
In a number of recent papers (Woodburn 1978; 1979; 1980), I have sought to classify hunting and gathering societies - that is societies in which people obtain their food from wild plant products by hunting wild animals, by fishing and gather wild roots, fruits and the honey of wild bees2 - into two major categories, those with immediate-return systems and those with delayed-return systems.
Immediate-return systems have the following basic characteristics. People obtain a direct and immediate return from their labor. They go out hunting or gathering and eat the food obtained the same day or casually over the days that follow. Food is neither elaborately processed nor stored. They use relatively simple, portable, utilitarian, easily acquired, replaceable tools and weapons made with real skill but not involving a great deal of labor.
Delayed-return systems, in contrast, have the following basic characteristics. People hold rights over valued assets of some sort, which either represent a yield, a return for labor applied over time or, if not, are held and managed in a way which resembles and has similar social implications to delayed yields on labor. In delayed-return hunting and gathering systems these assets are of four main types, which may occur separately but are more commonly found in combination with one another and are mutually reinforcing:
(1) Valuable technical facilities used in production: boats, nets, artificial weirs, stockades, pit-traps, beehives and other such artifacts which are a product of considerable labor and from which a good yield is obtained gradually over a period of months or years.
(2) Processed and stored good or materials usually in fixed dwellings.
(3) Wild products which have themselves been improved or increased by human labor: wild herds which are culled selectively, wild food-producing plants which have been tended and so on.
(4) Assets in the form of rights held by men over their female kin who are then bestowed in marriage on other men.
In principle, all farming systems, unless based on wage or slave labor, must be delayed-return for those doing the work, since the yield on the labor put into crop-growing or herding domestic animals is only obtained months or years later. Of course in all delayed-return systems there is some immediate-return activity, but it is usually rather restricted and may be treated as low-status activity. Among hunting and gathering societies, the available information suggests that both immediate-return and delayed-return systems are common. Most are surprisingly easily classified into one or the other category, but there are some which cause difficulties, as is inevitable with any simple binary system.3
Delayed-return systems in all their variety (for almost all human societies are of this type) have basic implications for social relationships and social groupings: they depend for their effective operation on a set of ordered, differentiated, jurally-defined relationships through which crucial goods and services are transmitted. They imply binding commitments and dependencies between people. For an individual to secure the yield from his labor or to manage his assets, he depends on others. The farmer, for example, will almost invariably pool his labor with others - at least with a spouse and usually during the labor peaks of the agricultural cycle with several others - but, equally important, he depends on others for the protection of his growing crops, of his use rights to the land on which they are growing and of the yield when he obtains and stores it.4 While it would, in principle, be possible to imagine situations in which individuals on their own, invested substantial amounts of labor over time on their own, protected the asset in which the labor was invested on their own, and then secured and managed the yields on their own, in practice this seems almost never to occur.
Until quite recently most anthropological research has been conducted in relatively small-scale, delayed-return, pastoral, agricultural and hunting and gathering societies and here we find the familiar kinship commitments and dependencies; lineages, clans and other kinship groups; marriages in which women are bestowed in marriage by men on other men; marriage alliances between groups. Immediate-return systems have only recently begun to be properly investigated and hence their social arrangements are still relatively unfamiliar. Societies which fall into this category include the Mbuti Pygmies of Zaire (Turnbull 1965; 1966); the !Kung Bushmen (San) of Botswana and Namibia (Lee 1979; Marshall 1976; Lee & DeVore 1976; Wiessner 1977); the Pandaram and Paliyan of south India (Morris 1975; Gardner 1980); the Batek Negritos of Malaysia (K. M. Endicott 1974; 1979; K. L. Endicott 1979) and the Hadza of Tanzania (Woodburn 1968a; 1968b; 1970; 1972) among whom my own fieldwork was conducted and about whom I can talk with most confidence.5 Most of my illustrations will be drawn from material on the Hadza and the !Kung.
The characteristics of these immediate-return systems I have spelt out in some detail elsewhere. Here all I intend is an outline sufficient to provide a background for my discussion of how these societies promote equality. The social organization of these societies has the following basic characteristics:
(1) Social groupings are flexible and constantly changing in composition.
(2) Individuals have a choice of whom they associate with in residence, in the food quest, in trade and exchange, in ritual contexts.
(3) People are not dependent on specific other people for access to basic requirements.
(4) Relationships between people, whether relationships of kinship or other relationships, stress sharing and mutuality but do not involve long-term binding commitments and dependencies of the sort that are so familiar delayed-return systems.
I should stress, as I have before (Woodburn 1980: 111), that I am not seeking to reduce social organization in hunter-gatherer or other societies to no more than a mere epiphenomenon of technology, the work process and the rules governing the control of assets. All I am saying is this: in a delayed-return system there must be organization having the very general characteristics I have outlined. The particular form the organization will take cannot be predicted, nor can one say that the organization exists in order to control and apportion these assets because, once in existence, the organization will be used in a variety of ways, which will include the control and apportionment of assets, but which are not otherwise determined. In societies without delayed yields and assets, we do not find delayed-return social organization.
All the six immediate-return systems listed are egalitarian, profoundly egalitarian, though not all quite in the same way or to the same extent; delayed-return systems are far more variable, but to the best of my knowledge not one of them is egalitarian to the same extent as any one of the immediate-return systems. There is no doubt that, whatever else I may be defining by these categories, I certainly am marking off one set of societies that resemble one another in their realization of a remarkable degree of equality.
What is perhaps surprising is that these societies systematically eliminate distinctions - other than those between the sexes - of wealth, of power and of status. There is here no disconnection between wealth, power and status, no tolerance for inequalities in one of these dimensions any more than in the others. I have exempted relations between men and women from this sweeping assertion. In fact, formal relationships between men and women are quite variable in these societies, although in all of them women have far more independence than is usual in delayed-return systems. But since I have talked specifically about male-female relations (1978), I have decided to leave them out of the discussion today. In the present article, all the general statements I make about relationships should be taken unless otherwise stated as referring only to adult males.
Let us now see how these systems operate in practice.
Mobility and flexibility. In all of these six societies nomadism is fundamental. There are no fixed dwellings, fixed base camps, fixed stores, fixed hunting or fishing apparatus - such as stockades or weirs - or fixed ritual sites to constrain movements. People live in small camp units containing usually a dozen or two people and moving frequently.
These small nomadic camp units are associated with particular areas, usually described in the literature as territories, large enough to provide for subsistence requirements during the annual cycle. Each area at any one time will usually contain one or more camps: camp size and the number of camps vary seasonally. In some cases rights are asserted over its natural resources by the people most closely associated with the area. There is variation between these societies in the extent to which such rights are asserted, but what seems clear is that in every case individuals have full rights of access to camps in several of these areas and there is no question of tightly defined groups monopolizing the resources of their areas and excluding outsiders. People can and do move from one camp to another and from one area to another, either temporarily or permanently and without economic penalty. Lee describes how the composition of !Kung camps which usually contain between ten and thirty individuals changes from day to day. Intercamp visiting is, he says, the main source of fluctuation, but each year about 13 percent of the population makes a permanent residential shift from one camp to another. Another 35 percent divides its period of residence equally among two or three different camps which may or may not be within the same area (1979: 54).
For the Hadza the situation is relatively simple. Like the !Kung, individual Hadza identify strongly with particular areas but, unlike the !Kung, Hadza do not assert rights to the areas with which they are associated. Anyone may live, hunt and gather wherever he or she likes without restriction - both within the area with which he or she is mainly associated and anywhere else in Hadza country. The camp units in which people live are not fixed entities: there is constant movement in and out while a camp remains at one site: when the site is changed people may move together to one or more new sites or all or some may choose to move to an existing camp elsewhere. There are continuities in the composition of these local groupings but none which seriously limit individual freedom of movement (Woodburn 1968b; 1972).
In all these societies nomadic movement of all types, both within and outside the local area, is apparently not seen as a burdensome necessity but positively as something healthy and desirable in itself. I have discussed elsewhere (Woodburn 1972) how neither the frequency nor the spatial patterning of Hadza moves can be interpreted in terms of ecological factors alone, although probably such flexible movement does, among other things, rapidly accomplish a rational distribution of people in relation to resources available at any particular time. What it also does is to allow people to segregate themselves easily from those with whom they are in conflict, without economic penalty and without sacrificing any other vital interests. Most important of all for the present discussion is the way that such arrangements are subversive for the development of authority. Individuals are not bound to fixed areas, to fixed assets or to fixed resources. They are able to move away without difficulty and at a moments notice from constraint which others may seek to impose on them and such possibility of movement is a powerful leveling mechanism, positively valued like other leveling mechanism, positively valued like other leveling mechanisms in these societies.
Access to means of coercion. Another important factor in this context is the access which all males have to weapons among the !Kung, Hadza, Mbuti and Batek. Hunting weapons are lethal not just for game animals but also for people. There are serious dangers in antagonizing someone: he might choose simply to move away but if he feels a strong sense of grievance that his rights have been encroached upon he could respond with violence. Lee gives a number of important case histories of !Kung murders showing clearly that there are contexts in which individuals are prepared to use their poisoned arrows (1979: 370-400). Hadza recognize not just the danger of open public violence, where at least retaliation may be possible, but also the hazard of being shot when asleep in camp at night or being ambushed when out hunting alone in the bush (Woodburn 1979: 252).6 Effective protection against ambush is impossible. Those of you who have seen the film about the Hadza which I was involved in making (Woodburn & Hudson 1966) may remember Salida, the successful hunter of an impala in the film and of very many other animals in ordinary life. He is now dead and is believed by the Hadza to have died in such an ambush. Only his bones were found. The Hadza had theories about who the murderer might be but there was much uncertainty; the cause of the conflict is said to have been a dispute over a woman.7 No action was taken. The important in all this is that, with such lethal weapons available to all men, with the possibility of using them for murder undetected, with the likelihood that even if detected no action will be taken,8 with the knowledge that such weapons have indeed been used for murder in the past, the dangers of conflict between men over claims not only to women but more generally to wealth, to power or to prestige are well understood.
Yet there have been instances over the years of Hadza men who have in spite of the apparent risks demonstrated that they are not averse to attempting to dominate other Hadza, to order them about, to take their wives or to plunder their possessions. What is striking is that these instances are typically backed by coercive powers derived from outside Hadza society and have only been effective to the extent that such men have been able to override the crucial limiting mechanisms - the mobility of the victims and their individual capacity to retaliate - which would in normal circumstances be sufficient to prevent such predation (Woodburn 1979: 262-4).
In normal circumstances the possession by all men, however physically weak, cowardly, unskilled or socially inept, of the means to kill secretly anyone perceived as a threat to their own well-being not only limits predation and exploitation; it also acts directly as a powerful leveling mechanism. Inequalities of wealth, power and prestige are potential source of envy and resentment and can be dangerous for holders where means of effective protection are lacking.
What we have here is direct and immediate access to social control, access which is not mediated through formal institutions or through relationships with other people. it is directly analogous to, and matched by, the direct and immediate access, again not normally mediated through formal institutions or through relationships with other people, which people have to food and other resources.
Access to food and other resources. I have already discussed how, within the general pattern of nomadic movement, individuals are able to avoid constraint by their freedom to detach themselves from others at a moment's notice without economic or other penalty. But let us now look more closely at the rights which individuals enjoy without which such action would not be practicable. What are an individual's entitlements to food and other resources and how are these entitlements taken up?
In all these societies individuals have direct access, limited by the division of labor between the sexes, to the ungarnered resources of their country. Whatever the system of territorial rights, in practice in their own areas and in other areas with which they have ties, people have free and equal access to wild foods and water; to all the various raw materials they need for making shelters, tools, weapons and ornaments; to whatever wild resources they use, processed or unprocessed, for trade.
Among the !Kung each area and its resources are used both by a core of men and women with long-standing associations with the area, who identify with it rather than with other areas, and by a wide range of other people who have come from other areas, some temporarily and some more permanently, and who are in most cases linked to one or more of the core members or other residents by a kinship of affinal tie (Marshall 1976; Lee 1979). Anyone with such a link who comes to live with the people of the area cannot, in practice, be refused full access to its resources provided that he or she observes certain minimal rules of politeness. As Marshall explains, newcomers share equally while they live there. No core member or anyone else has the right to withhold resources from the newcomer or to take a larger share (1976: 189).
Among the !Kung, this relative freedom of access operates in spite of the fact that people long associated with an area claim to be 'owners' (k"ausi) of it and in particular of its plant and water resources. The !Kung notion of 'ownership' is clearly a broad one and seems here to mean association with, involvement in, identification with the area rather than narrow possession of it. Lee tells us that the usual term for hunter in !Kung is !gaik'au, for which he gives the literal translation 'hunt owner' (1979: 206). This suggests that the term k"au ('owner') can indicate association and not just possession. In general in these societies, and even among the !Kung where land rights might at first sight appear to be important in constraining movement and access, what association with a particular locality seems usually to provide is a means of identifying oneself and others, a way of mapping out social relations spatially, rather than a set of exclusive rights. Among the Hadza, people identify strongly with their own areas but place far more emphasis than the !Kung do on an individual's rights of access to resources everywhere, both in his own area and elsewhere.
The boundaries of Hadza area are, as one would expect from what has already been said, undefined: in effect there are no boundaries (Woodburn 1968b: 104). According to Marshall, !Kung boundaries between areas are rather closely defined in localities where important wild plants can be gathered (1976: 187-8). Lee subsequently suggests that, at least in the region where he worked, territories are not clearly defined or bounded even in relation to plant foods: he tells us that he believes 'the !Kung consciously strive to maintain a boundaryless universe because this is the best way to operate as hunter-gatherers in a world where group size and resources vary from year to year' (1979: 335). If there were a rather rigid principle of recruitment to local groups combined with rather rigid boundaries between the areas used by local groups, material inequalities between local groups would inevitably develop as populations and resources fluctuated through time. Double flexibility - over group boundaries and territorial boundaries - obviously limits, in a bad year, the dangers of local food shortages or of destructive over-exploitation of sources of wild food. But given low population density and the limited pressure on resources even in bad years I think it would be appropriate to give less attention to the nutritional benefits of all this flexibility in very rare times of crisis and to stress its day-to-day significance as a leveling mechanism.
Not just among the !Kung but generally in these societies, this double flexibility directly limits the development of local variation in wealth or in standard of living. If in one area people are eating better than in another, or obtaining food more easily than in another, then, other things being equal, movement of people couple perhaps with adjustment of boundaries, is likely over time to tend to even out such unacceptable discrepancies.
The direct and immediate access to food and to other resources which people enjoy is important in other ways. Without seeking permission, obtaining instruction, or being recognized as qualified (except by sex), individuals in these societies can set about obtaining their own requirements as they think fit. They need considerable knowledge and skill but this is freely available to all who are of the appropriate sex and is not, in general, transmitted by formal (or even informal) instruction: rather it is learned by participation and emulation. In most, but not all, of these societies neither kinship status nor age is used as a qualification to obtain access to particular hunting and gathering skills or equipment.9 A Hadza boy who wishes to try to hunt large game with bow and poisoned arrow can do so without restriction as soon as he wishes to do so and is able to make or obtain the necessary arrows. More important still, any person - man, woman or child - who seeks to obtain his or her requirements either individually or in association with others can do so without entering into commitments to and dependencies on kin, affines or contractual partners. Adults of either sex can readily, if they choose, obtain food to feed themselves adequately and are, in spite of the rules of the division of labor, potentially autonomous. It is not rare, at least among the Hadza to find individuals, usually males, living entirely on their own as hermits for long periods. Of course in practice people living within the community do not live simply on self-acquired food: there are pooling and sharing practices, some of them very important, which I shall describe separately. What matters here is the lack of dependence on sharing or pooling of resources: a Hadza woman out gathering with other women will consume much of the proceeds on the spot and, out of what she brings back to camp, little, if any, may go to her husband. Similarly a Hadza man out hunting will expect to feed himself by picking and eating berries and by consuming any small animal he may kill. Only food surpluses cross the sexual boundary. The Hadza are perhaps an extreme instance: in some other immediate-return societies much more food is used socially - is brought back to camp and consumed jointly by people of both sexes. Only among the net-hunting Mbuti (as distinct from the archer Mbuti) does co-ordinated activity provide a substantial amount of food and even in this instance many sources of food are available to and obtained by individuals.
The net-hunting Mbuti apparently gather wild fruits and roots in a manner not very different from the Hadza (Turnbull 1966: 166-8). But hunting is a co-operative venture: women and children drive game animals, large and small, into a semi-circle of nets set up by the men. The yield is shared out among the various participants as soon as they return to camp (1966: 157-8). I must stress that this co-operation in the hunt is of a very specific sort. Within the rather broad limits set by the optimal numbers for an efficient hunt (1966: 154), anyone who is present may participate according to his or her sex role and is entitled to a proportion of the yield. There is no commitment to participate and no basis for exclusion from participation. Each hunt is complete in itself and participation today apparently carries no obligation to participate tomorrow. This is fundamentally different from co-operation in any agricultural system where the members of the productive group are not an ad hoc aggregation but are a set of people bound by more enduring ties of kinship or of contract.
What I want to stress here in this lack of dependence on specific others, is the implication for authority and, most obliviously, for domestic authority. The process of production is not in general controlled and directed by the household head or, if it is, the control is not authoritarian and is better described as limited co-ordination by consent. Indeed among the Hadza I would say there are no household heads (Woodburn 1968b: 109). Older children and young unmarried adults in these societies are not dependent on the senior generation for access to property, to food or to resources though they may receive some property, food and resources from them. Among both Hadza and !Kung, children do relatively little work and what they do is done at their choice rather than under parental direction. Neither the parent-child nexus, nor the relationship between generations more generally, provide either a model or a training ground for relationships of authority and dependency; indeed they provide an alternative model for and training in personal decision-making and in the possibilities of self-reliance, in sharing but not dependency on sharing.
The point can perhaps be made most clear if a simple contrast is drawn with some agricultural and pastoral societies living in south eastern Africa. Using detailed ethnographic evidence on the Southern Bantu, Richards long ago drew attention to the importance of access to food and other resources for an understanding of inequality: she stressed how, in the absence of other valuables, possession or control of food is singularly important as a means of differentiating on member of the community from another and as a source of power (1932: 89). The authority of a man as household head over his unmarried sons, both immature and adult, his right to direct their labor and to demand obedience and respect from them is linked with
the father's possession and control over the food supply - the cattle-herd and their produce, and in general, the grain supply too. The head of the family is … bound to support his sons … The receipt of food marks the dependence of the child on the father … Unless he can earn money [in employment], … he simply cannot acquire food except from his parents' hands … one woman alone - his mother, or her substitute - must cook [it for him]. His dependence is thus displayed concretely by the receipt of actual food … (1932: 77)
In immediate-return hunting and gathering societies the household head has no comparable role as real or symbolic provider, as the source of most good things. It is, I am sure, not accidental that neither !Kung nor Hadza usually place much emphasis on formal meal times. A great deal of food is eaten informally throughout the day (Lee 1979: 199). Marshall records that 'Meat is not habitually cooked and eaten as a family meal among the !Kung … The men, women, and children may cook their pieces when and as they wish, often roasting bits in the coals and hot ashes and eating them along at odd times' (1976: 302).
Sharing. The genuine equality of opportunity that individuals enjoy in their access to resources, limited only by the division of labor between the sexes, does not, of course, ensure equality of yield. The quantities of all the various items which individuals obtain, either on their own or jointly with other people, vary greatly depending on skill, on luck, on persistence, on capacity to work and on other factors. It is at this point that the most crucial controls on the development of inequality come into action.
The principal occasions in which individuals in these societies are brought into association with valued assets which could be accumulated or distributed to build status are when large game animals are killed. And it is then that the most elaborate formal rules dissociating the hunter from his kill and denying him the privileges of ownership are brought to bear. Leveling mechanisms come into operation precisely at the point where the potential for the development of inequalities of wealth, power and prestige is greatest. Among the Hadza and the !Kung hunting success among adult men seems to be very variable. A high proportion of animals are killed by a small proportion of men (Lee 1979: 242-4). Techniques for drying meat and converting it into relatively lightweight stores of biltong are known. Yet successful individual hunters are specifically denied the opportunity to make effective use of their kills to build wealth and prestige or to attract dependents. Lee has reported how !Kung are expected to be self-deprecating about their hunting successes; boasting is met with scorn (1979: 243-6). Turnbull (1966: 183) tells us that 'some Mbuti men, because of exceptional hunting skill, may come to resent it when their views are disregarded, but if they try to force those views they are very promptly subjected to ridicule'. A Hadza returning to camp having shot a large animal is expected to exercise restraint. He sits down quietly with the other men and allows the blood on his arrow shaft to speak for him. If the animal has only been wounded, the name of the species may not be mentioned directly: the effect of uttering its name would, Hadza believe, allow it to recover from the arrow poison and escape. Special respectful terms must be used and there should be a minimum of comment. Similarly, to talk of a dead animal before it has been dismembered is believed to put at risk the animal's fat which the Hadza value more highly than lean meat. One effect of these rules is, certainly, to deny the hunter praise which he might otherwise expect to receive.10
For the !Kung the owner of the first arrow to hit the animal effectively is the owner of the kill with the right to distribute it. Arrow-lending is common and the owner of the arrow is often someone other than the hunter himself. Lee explains that meat distribution brings prestige but it also brings the risk of accusations of stinginess or improper behavior if the distribution is not to everybody's liking (1979: 247-8). So hunters are not reluctant to hunt sometimes with someone else's arrow and to pass over the responsibility for distributing the kill to him. When discussing this practice Marshall suggests that 'the society seems to want to extinguish in every way possible the concept of the meat belonging to the hunter' (1976: 297).
The meat of the kill is widely shared within the camp unit. Among the Hadza the best portions (which differ depending on which species has been killed) belong to the initiated men and may not under any circumstances be eaten by the hunter on his own. For the Hadza this would be a particularly heinous offense which would be likely to result in violence towards the hunter; it would also, Hadza believe, cause him to become seriously ill and perhaps even die. The hunter's rights to the initiated men's meat are identical to those of each of other initiated men: the meat must be eaten in secret by the initiated men as a group until finished. The rest of the meat is described as people's meat and is distributed first at the kill site among those who have come to carry in the meat, then back at camp it is distributed among thse who remained behind and then, finally, when it is cooked, it is consumed by those who happen to be present and not simply by the person or the family grouping to whom the meat was allocated. The hunter himself will often not be involved in dismembering the carcass and in distributing the meat but he and his wife's mother and father, if they are present, will receive substantial shares. Among the !Kung the meat is distributed and redistributed in waves of sharing through the camp: everyone receives a share.
It has often been suggested that meat-sharing is simply a labor-saving form of storage. The hunter surrenders his rights to much of his kill in order to secure rights over parts of the kills of other hunters in the future. There are problems with this formulation: as I have already mentioned, hunting success is unequal. Donors often remain on balance donors and may not receive anything like an equivalent return. Entitlement does not depend in any way on donation. Some men who are regular recipients never themselves contribute. Instead of seeing the arrangement as being in the interest of the donor, I think we should be clear that it is imposed on the donor by the community. Instead of seeing the transaction as a form of reciprocal exchange, I would suggest we treat it as analogous to taxation on incomes of the successful in our own society. The successful pay more than the less successful and are obliged to do so. They are not able to establish greater claims in future through having paid more tax and do not derive much prestige from having contributed more to the tax pool than they have withdrawn from it in benefits. The analogy may sound rather crude: certainly the hunter derives more prestige from contributing an animal than any taxpayer does from paying his taxes, but it does bring out the important fact that we are dealing here with a socially imposed leveling mechanism and not a mere practical convenience for the hunter.
Vegetable foods, which are less highly valued and which are obtained more easily and more predictably and in amounts which correspond more closely to the needs of the gatherers, are less widely shared but are not narrowly reserved for each gatherer and her immediate family. Among the Hadza it would be out of the question for a woman to hoard food while others are hungry. Sharing rights for pregnant women are particularly emphasized by the Hadza: they have the right to ask anyone for food at any time and are believed to be at risk if they are refused.
Sanctions on the accumulation of personal possessions. Clothing, tools, weapons, smoking pipes, bead ornaments and other similar objects are personally held and owned. At least in the case of the three African societies, they are in general relatively simple objects, made with skill but not elaborately styled or decorated and not vested with any special significance. They can be made or obtained without great difficulty. Rules of inheritance are flexible and no-one depends on receiving such objects either by inheritance or by formal transmission from close kin of the previous generation during their lifetime.
Everywhere we find that there are sanctions against accumulation. This cannot be explained, as so many writers have mistakenly suggested, simply in practical terms: nomadic peoples who have to carry everything they possess are concerned that their possessions should be readily portable so that they can be carried with ease when the time comes to move camp, but sanctions against accumulation go far beyond meeting this requirement and apply even to the lightest objects such as beads, arrowheads of supplies of arrow poison.
The transmission of possessions between people. Hadza use a distinctive method for transmitting such personally owned objects between people which has profound consequences for their relationships. In any large camp men spend most of their time gambling with one another, far more time than is spent obtaining food. They gamble mainly for metal-headed hunting arrows, both poisoned and non-poisoned, but are also able to stake knives, axes, beads, smoking pipes, cloth and even occasionally a container of honey which can be used in trade. A few personally-owned objects cannot be staked, because, Hadza say, they are no sufficiently valuable. These are a man's hunting bow, his non-poisoned arrows without metal heads used for hunting birds and small animals, and his leather bad used for carrying his pipes and tobacco, arrowheads and other odds and ends. These objects excluded from gambling share two characteristics: first, they maintain a man's capacity to feed and protect himself and secondly, they are made from materials available in every part of the country. In constract many of the objects used in gambling are made, at least in part, of materials not available in every part of the country. For example, poisoned arrows incorporate a head made from scrap metal obtained in trade from non-Hadza, poison which is made from the sap of one tree or the seeds of another which are local in their distribution and entirely absent from large areas of the country, and, if possible, a lightweight shaft of a particular species of shrub even more restricted in its range. Stone smoking pipes are made from a type of soapstone which is again only found in certain localities.
The game involves tossing a handful of bark discs against a tree and reading the result which depends on which way up the various discs fall. Basically it is a game of chance and players have relatively little opportunity of influencing the result. Skill in throwing does play some part but the effect is reduced by banning the winner of the previous throw from acting as thrower next time.
In the course of a day in a large dry-season camp there will usually be hundreds of contests each one involving the transfer of a piece of property from the loser or losers to the winner. During the day many objects pass through competitors' hands, but as the game is a game of chance the randomizing effect means that they will usually end the day without a substantial net gain or loss. If the player ends the day with some winnings, usually they will include some objects he wants to keep and use. He restakes objects he does not want and keeps back, as far as he can, objects he wants. From time to time people do gain substantial winnings and are then subject to great pressure to continue to compete so that other competitors can win back their lost possessions. Often winners, in an attempt to keep their winnings intact for long than would otherwise be possible, move to another camp, but they are then usually followed by some of the other competitors. I can recall only one instance of a man who succeeded in keeping most of his winnings for a period of as long as a few weeks and he was quite exceptional person who withstood the pressures on him and rationed his subsequent play whenever he won.
I have discussed Hadza gambling at greater length than might at first seem justified because its effects are so important. It is the major means by which scarce and local objects are circulated throughout the country: much intercamp visiting is stimulated by gambling and winnings are constantly on the move. Objects such as stone pipes which are made in one part of the country circulate out to other areas where they are withdrawn from the game and put to use.
This circulation is accomplished, then, not through some form of exchange which would bind participants to one another in potentially unequal relationships of kinship or contract. The transactions are neutralized and depersonalized by being passed through the game. Even close kin and affines gamble with each other and the game acts against any development of one-way flows and dependency in relationships between them.
Individual effort, craft skill and, particularly, the skill of trading with outsiders are quite variable. The attraction of gambling mobilizes effort and skill but distributes its proceeds at random in a way which subverts the accumulation of individual wealth by the hard-working or by the skilled. It further subverts any tendency to regional differentiation within Hadza country based on valuable local resources which are in demand in other areas. It is paradoxical that a game based on the desire to win and, in a sense, to accumulate should operate so directly against the possibility of systematic accumulation. Its leveling effect is very powerful.
!Kung transmit personal possessions in a quite different way which has far less leveling potential. Each !Kung enters into formal exchange partnerships, known as hxaro, with a number of other people with whom systematic exchanges of personal possessions and of hospitality take place. Far from subverting relationships as Hadza gambling does, these exchanges create ties involving some mutual commitment. However, they seem to be organized in ways which stress the equal relationship between partners and provide little opportunity for property accumulation or the development of patron-client type relations between partners (Wiessner 1977). Marshall shows how the gifts are simple tokens of generosity and friendly intent. The trivial debts incurred are quite unlike the binding jural obligations at the center of delayed-return systems. And as Marshall says 'no one was dependent on obtaining objects by gift giving' (1976: 308).
Both among the !Kung and among the Hadza individuals with any objects for which they appear to have no immediate need are under the greatest pressure to give them up and many possessions are given away almost as soon as they are obtained and usually, I think, without any expectation of return.11 Among the Hadza the attitude of property often seems very casual. I well remember early on in my fieldwork being pressed very strongly to to give people hoop iron to make arrowheads: I obtained the iron with some difficulty, gave out a piece to each man in the camp and was rather affronted when I discovered that, because I had given out more than could immediately be used, some of it had simply been thrown away. Limitations on the uses to which property can be put, especially if no gambling is going on at the time, in context in which saving and accumulation are so actively discouraged and pressure for de facto equality of wealth is so great, mean that people often seem to place surprisingly little value both on their own and on other people's possession.
Leadership and decision-making. In these societies there are either no leaders at all or leaders who are very elaborately constrained to prevent them from exercising authority or using their influence to acquire wealth or prestige.12 A Hadza camp at any particular time is often known by the name of well-known man then living in it (for example, /ets'a ma Durugida - Durugida's camp). But this indicates only that the man is well enough known for his name to be a useful label, and not that he acts as either a leader or a representative of the camp (Woodburn 1968b: 105). Hadza decisions are essentially individual ones: even when matters such as the timing of a camp move or the choice of a new site are to be decided, there are no leaders whose responsibility it is to take the decisions or guide people towards some general agreement. Sporadic discussion about moving does occur but usually it takes the form of announcements by some individual men that they are going to move and where they are going to move to. Other men will often defer a decision about whether to stay, whether to accompany those who are moving, or whether to move elsewhere, until the move actually begins. Certainly some, particularly older men with large families who are often those after whom camps are named, are likely to have more influence on the outcome than others and to precipitate by their individual decisions the movement of an entire camp or segment of a camp. But I would not describe them as leaders (Woodburn 1979: 253).
The !Kung also name camps after individuals, whom Lee describes as leaders but leaders with a very limited role. 'In group discussions these people may speak out more than others, maybe be deferred to by others, and one gets the feeling that their opinions hold a bit more weight than the opinions of other discussants' (1979: 343). What is particularly striking is that personal qualities suggesting that a !Kung individual is ambitious for power or wealth exclude such a person from the possibility of leadership.
None is arrogant, overhearing, boastful, or aloof. In !Kung terms these traits absolutely disqualify a person as a leader and may engender even stronger forms of ostracism … Another trait emphatically not found among traditional camp leaders is a desire for wealth or acquisitiveness … Whatever their personal influence over group decisions, they never translate this into more wealth or more leisure time than other group members have (1979: 345) … . Their accumulation of material goods is never more, and is often much less, than the average accumulation of the other households in their camp (1979: 457).
Leaders should ideally be 'modest in demeanor, generous to a fault, and egalitarian … ' (1979: 350). Leaders such as these pose no threat to egalitarian values and indeed could be said to display and reinforce egalitarianism.
These are, of course, not the only contexts in which equality is expressed and leveling mechanisms operate: to do justice to the subject it would be necessary to go much further and in particular to explore the expression of egalitarianism in religious belief and practice. But I think I have said enough to show that we have here the application of a rigorously systematic principle: in these societies the ability of individuals to attach and to detach themselves at will from groupings and from relationships, to resist the imposition of authority by force, to use resources freely without reference to other people, to share as equals in game meat brought into camp, to obtain personal possessions without entering into dependent relationships - all these bring about one central aspect of this specific form of egalitarianism. What it above all does is to disengage people from property, from the potentiality in property rights for creating dependency. I think it is probable that this specialized development can only be realized without impoverishment in societies with a simple hunting and gathering economy because elsewhere this degree of disengagement from property would damage the operation of the economy. Indeed the indications are that this development is intrinsic, a necessary component of immediate-return economies which occurs only in such economies.
It is time that I discussed very briefly how these societies compare with others that have been described as egalitarian. Of course we are all familiar with the application of the term to those modern complex societies, including our own, in which egalitarian sentiments are commonly expressed in political discourse. But let us leave these aside. Anthropologists, at least those conducting research outside Europe, have more commonly used the term as a simple synonym for acephalous societies, societies without rulers, societies without formal political offices. Delayed-return systems within this category, in contrast to immediate-return systems which I have been discussing in some detail, are egalitarian in a very different sense. To generalize about so wid and variable category is not really satisfactory but a few obvious points can be made: first, the community of political equals is usually a community of property-holding household heads whose relations with their wives and female kin and with junior male kinsmen within the household are far from equal. Intergenerational inequality and especially the inequality of holder and heir is usually stressed. The household head has the right and the duty to maintain and control the assets held by the household and to direct the labor of its members.
Equality between household heads is in many of these systems only a starting-point, a qualification to compete in a strenuous competition for wealth, power and prestige. Writing about Papua New Guinea, Forge (1972: 533) tell us that 'to be equal and stay equal is an extremely onerous task requiring continual vigilance and effort. Keeping up with the Joneses may be hard work, but keeping up with all other adult males of a community is incomparably harder. The principal mechanism by which equality is maintained is equal exchange of things of the same class or of identical things. Basically all presentations of this type are challenges to prove equality.' The outcome of such challenges is striking inequality. Writing about the same area Burridge tells us what happens to the casualties. 'Rubbish-men', as impoverished people are called,
were men who had either opted out of the status competition, or who habitually failed to meet their obligations. Often, they lived on the fringes of a settlement, between village proper and the forest or bush outside: a placement which reflected their position in the moral order. Because they had failed to meet their obligations, or were unable or had chosen not to participate in the process whereby proper men were supposed to demonstrate their moral nature, they hardly rated as men, were not fully moral beings in the locally recognized sense. Treated with a charitable contempt which they accepted with resigned equanimity, rubbish-men were generally left to their own devices, working by and for themselves at a markedly lower level of subsistence than others (1975: 95).
This type of competitive equality contrasts dramatically with the non-competitive equality of the systems I have been describing. There equality does not have to be earned or displayed, in fact should not be displayed, but is intrinsically present as an entitlement of all men. There are no casualties of the principle of equality among the Hadza13 or the !Kung, none of whose moral worth is destroyed by poor economic performance or lack of personal competitiveness. Egalitarianism is asserted as an automatic entitlement which does not have to be validated.
New Guinea is something of an extreme instance but even the less aggressive egalitarianism of east African pastoral societies, whose members do not engage in the competitive displays of wealth so characteristic of New Guinea, are still highly competitive in comparison with the !Kung and the Hadza and have far more in common socially with New Guinea than they do with these hunter-gatherers. There are marked differences in wealth in cattle and in standard of living. They too have casualties. Household heads who through bad management or bad luck lose all or most of their herd, lose at the same time the essential qualification for equality with herd owners. They have then to work as low-status clients or else move out of the pastoral economy.
Burnham has recently discussed what he describes as the structural conservatism of nomadic pastoral societies (1979: 349-60). In a valuable paper he argues that fluidity of local grouping and spatial mobility are remarkably resistant to change. They promote egalitarianism and greatly inhibit the development of both political centralization and class stratification. Within sub-Saharan Africa nomadic pastoral societies such centralization and stratification have emerged only in situations of conquest and/or substantial sedentarization.
I would argue that the nomadic hunting and gathering societies I have discussed are even more profoundly conservative. Fluidity of local grouping and spatial mobility, here combined with and reinforced by a set of distinctive egalitarian practices which disengage people from property, inhibit not only political change but any for of intensification of the economy. There is no easy transition from non-competitive egalitarianism to competitive egalitarianism.
Lee has described the difficulties some !Kung have encountered in their efforts to live by agriculture and by keeping domestic animals (1979: 409-14). The difficulties are not technical ones: some individual !Kung have for years worked for neighboring farmers and have become knowledgeable about farming techniques. The overwhelming difficulties lie in the egalitarian leveling mechanisms. There is no way that farming can be carried on without some accumulation, without stores of grain and of agricultural tools, and the major difficulty for the few individual !Kung who wish to make a real effort to farm is not that they themselves are unable to exercise self-restraint and to build up their stocks but that they are unable to restrain their kin and affines from coming to eat the harvested grain. Exactly the same has occurred again and again in the Hadza government-run settlements. Almost everyone understands basic agricultural techniques and almost everybody is prepared to carry out some cultivation. But those few who apply their labor systematically and skillfully and obtain a good crop have found that their fields are raided by other Hadza even before grain is harvested, and once it has been harvested those with grain in store are under relentless pressure to share it with other Hadza rather than to ration its use so that it will last until the next harvest is obtained. In the face of such obstacles, even the successful farmer is likely to give up. If it is so difficult for these egalitarian hunter-gatherers to take up agriculture nowadays with so many pressures on them to settle, it is even less likely that they would have been able to convert to agriculture in the past.14 I have suggested elsewhere that if we are to understand the development of agriculture from hunting and gathering, we ought to look at delayed-return hunter-gatherers who have the values and the organization to facilitate the transition.
To end I should like to return briefly again to the nature of equality in these systems and its association with certain other values. Are people in these societies, who are relatively free from want and free from many of the forms of competition which are so evident in other societies, humane, altruistic and caring towards those who are not able to care for themselves? The answer is not simple but it is possible to say a little. In delayed-return systems people are bound to their close kin and to certain other associates in relationships which commonly involve the constant exchange of goods and services in fulfillment of obligations of one sort or another. People are committed to one another both in the sense that they have material obligations to one another and in the wider sense that they accept a measure of responsibility for one another. The potentiality for autonomy and the limitation of obligations to specific other people - as opposed to the generalized obligation to share - certainly seem to reduce the sense of commitment that people feel to others, at least in the Hadza instance where the movement of goods between people is so often depersonalized by the gambling game. There are instances in which the Hadza have abandoned the seriously ill when they moved camp, leaving them with their possessions and with food and water but knowing that they were unlikely to be able to provide for themselves. I was very surprised by the neglect of a previously popular grandmother in one of the settlements when she became senile and I think it less likely that this would have occurred within a household of one of the neighboring agricultural societies. I have to allow, though, for the fact that the total number of instances of neglect or abandonment that have come to my attention is small and that such instances are more public among the Hadza than they are in societies where they can occur behind closed doors.
In another respect I can report with more confidence. Hadza society is open. People who are able to associate themselves at will with whatever area and whatever camp they choose do not impose social boundaries between themselves and others. These is no basis for exclusion. I think this is best brought out by their treatment of lepers. Among neighboring sedentary societies lepers are not able to participate effectively in social life and are commonly ostracized or even expelled (these days usually to a leprosarium). There are a very small number of Hadza lepers who appear to be treated exactly like everyone else although the Hadza are very well aware of the dire consequences of the disease and the fact that it can be transmitted from one person to another. The principle is entirely clear. It is not that the Hadza are particularly sympathetic or humane towards lepers, who may well be mercilessly teased about their clumsiness: the principle is that Hadza society is open and there is simply no basis for exclusion. Equality is, in a sense, generalized by them to all mankind but, sadly, few of the rest of mankind, so enmeshed in property relations, would be willing to extend parity of esteem to hunter-gatherers who treat property with such a lack of seriousness.
1 The term was used earlier with a different meaning (Murray et al. 1970)
2 In some, people obtain part of their food from wild products and part directly in exchange for wild products.
3 In using this simple, indeed over-simplified, dichotomous categorization into immediate-return and delayed-return systems, I should stress that it is designed as an initial, rough-and-ready basis for looking at what seem to be crucial variables for understanding social organization. The job of specifying them more precisely and of disentangling their implications is still to come if it is accepted that the timing of yields on labor is significant for social organization. I recognize, of course, that the four types of assets listed for delayed-return systems are qualitatively different and certainly do not have identical implications for social organization though they do, it seems to me, have enough in common for it to be useful to group them together initially. I recognize too that some yields are part-immediate and part-delayed, that some immediate yields are much more immediate than others and that there is always some immediate-return activity in delayed-return systems (most strikingly in the case of Australian Aboriginal societies) but also that there is some delayed-return activity in immediate-return systems. In taking the matter further, the valuable comments of the following colleagues will be taking into account: Andrade 1979; Dodd 1980; Firth pers. comm. 1981; Ingold 1980; Ndagala pers. comm. 1981; Wiessner 1980.
4 If some undemanding agricultural crop is grown in an area where land is plentiful, if it does not require systematic application of labor over time and if it is unsuitable for storage, then this dependency may be small and the organizational requirements more limited.
5 I have listed here only a few of the more relevant references. Additional references to these societies are given in Woodburn 1980.
6 Out of the twenty-two !Kung homicides which Lee records, seventeen occurred during spontaneous open public violence while five were the result of premeditated sneak attacks (1979: 382). Ambushes are possibly less likely to occur among the !Kung where 'a person's footprints are as well known as his face' (Marshall 1976: 188.) Hadza are expert trackers but are usually not able to recognize one another's footprints.
7 The murdered man was a rather aggressive character and had in the past been involved in disputes over women. Years earlier he himself made threats in my presence that he would kill a man - not his suspected assailant - whom he accused of committing adultery with his wife.
8 Among the !Kung some killings lead to retaliatory killings, others do not. According to Lee's figures, four killings led to retaliatory killings (some of them multiple) while seven did not (1979: 389).
9 Mbuti are an exception here: for them age distinctions are of some importance in the work process (Turnbull 1966).
10 I am not, of course, suggesting that this is the main significance of such rules.
11 The unremitting demands Hadza make on one another are highly conspicuous and often go beyond asking for things for which the owner has no immediate need. A man who obtains a ball of tobacco, a shirt or a cloth by trading with or begging from non-Hadza is unlikely to keep it for long unless he is very determined and willing to make himself unpopular. He will be asked for it endlessly. The pressures on outsiders (including anthropologists) to give away their possessions is equally great. Lee writes of how he was repelled by the !Kung's 'nagging demands for gifts, demands that grow more insistent the more we give' (1979: 458).
12 I leave aside leaders imposed on these societies by outsiders or who derive their power from links with outsiders (Lee 1979: 348-50; Woodburn 1979: 261-4).
13 As I mention at the end of this article, there are casualties of one sort: the seriously ill, the senile and others unable to care for themselves and to claim equality, may not be treated well.
14 In an important recent article Cashdan discusses the contrast between equality among the !Kung and inequality (especially in wealth) among the //Gana, a Central Kalahari population derived from mixed marriages between immigrant Bakgalagadi men and San (Bushmen) women. The //Gana supplement their basic hunting and gathering subsistence with some farming. They are nomadic but from a home base. She attributes //Gana inequalilty not to the development of any formal organization of inequality but to 'the inevitable result of the lifting of the constraints that produce strict egalitarianism among other Kalahari hunter-gatherers' (1980: 119-20). I differ from her in believing that egalitarian values, so deeply built into traditional hunting and gathering systems in the area, would be unlikely to be so easily shed were it not for the fact that the community is of mixed origin and drew part of its membership from people with experience of faming and delayed-return.