Title: On social currencies and human economies
Subtitle: Some notes on the violence of equivalence
Author: David Graeber
Date: 2012
Source: Retrieved on 28th November 2021 from davidgraeber.org
Notes: Published in Social Anthropology (2012) 20, 4 411–428. doi:10.1111/j.1469–8676.2012.00228.x


      The matter of debt

      Lele: blood debts

      Tiv: flesh-debts

      Way of death


Abstract: In this essay I propose a category of ‘human economies’ to refer to those where the primary focus of economic life is on reconfiguring relations between people, rather than the allocation of commodities. Currencies that used to be labelled ‘primitive money’, but which are primarily used to effect this, would better be called ‘social currencies’. These social currencies are often seen as inadequate substitutes for human beings, not so much ways of discharging debts as of recognising the existence of a debt that cannot be paid. By reconsidering some classic anthropological cases (the Lele, the Tiv) in the light of the slave trade, we might catch a glimpse of the violence required to transform such social currencies into commercial currencies by which debts can be entirely cancelled out.

Key words: money, debt, reciprocity, violence, Africa


Once, the study of ‘primitive money’ – shell currencies, feather currencies, wampum, Fijian whale teeth, Tiv iron bars, and so forth – was the stock and trade of economic anthropology. In recent decades there’s been almost nothing written on the subject. James Carrier’s otherwise comprehensive Handbook of economic anthropology (2010) not only considers the matter not worth a chapter, it contains not a single mention of wampum, or trade beads, anywhere in the book. We don’t even know what to call such items any more. ‘Primitive currencies’ or ‘primitive valuables’ will obviously no longer do; French economists who propose to re-label them ‘paleo-moneys’ (Servet 1981, 1982) or even ‘savage money’ (Rospabe´ 1995) are hardly better. Akin and Robbins’ (1999) decision, when dealing with traditional money-forms in Melanesia, to use the term ‘local currencies’ (as opposed to state ones), while at least not obviously offensive, seems an obvious a place-holder, something to use until a better term comes up.

One reason it’s hard to come up with better terminology is that there’s no consensus on what, if anything, actually makes a string of Indian Ocean trade beads, or California woodpecker scalps, different from a shekel or a pound. Is it something in the nature of the object? Or is it a matter of the kind of transactions in which it is used? Or is it the conceptual apparatus, the cosmological assumptions, the notions of value, surrounding it? Or, alternately, are we just looking at a meaningless default category, where the term ‘primitive currencies’ is applied to any widely circulating valuable that is not used primarily for commercial purposes?

I would like to propose a solution. The key distinction here is between currencies that are used primarily to further the exchange of material goods, and those primarily used to transform social relationships. The first can be referred to as commercial currencies, because even though they may be used for social purposes, their primary purpose is seen to lie in buying, selling, renting or otherwise disposing of alienable property. The second should best be referred to as ‘social currencies’. They may also be used to buy and sell material goods (often they are, but not always), but their primary purpose is seen to lie in arranging marriages, resolving conflicts, consoling mourners, making treaties, assembling followers for military expeditions or competitive feasts, making gifts or rewarding services. For this reason, I also propose to call those economic systems in which social currencies predominate ‘human economies’. I must emphasise (much though I really shouldn’t have to) that this is not because they are necessarily more humane, in the sense of less brutal, more caring, than impersonal market economies. Some certainly are; others are extraordinarily brutal and destructive. What distinguishes human economies is merely that they recognise that the chief business of any social system – or, indeed, of any system of the production and distribution of material goods – is the creation and mutual fashioning of human beings. Indeed, one could well argue that it’s only the emergence of commercial currencies that made it possible to imagine an ‘economy’ in the sense we are used to using it today at all – that is, an autonomous domain of human activity primarily concerned with creating and allocating material possessions, and not primarily about the creation of people and social relations – let alone, see how people behave within that domain as a model for human aims and aspirations more generally. Historically, the endless repetition of the ‘myth of barter’ has played a key role in making it possible to imagine that such an autonomous sphere of activity exists, and, of course, of creating the institutional arrangements that could make it possible.

The matter of debt

I became interested in the problem of social currencies when trying to solve a logical conundrum having to do with the origins of money – that is, money in the current, predominant, commercial sense of the term. Anthropologists have been pointing out for at least a century that the famous story about the origins of money from the ‘inconveniencies of barter’ proposed by Adam Smith in The wealth of nations (1776), and that still appears regularly in economic textbooks, is a myth. Certainly, no economy based on barter of the ‘I’ll give you twenty chickens for that cow’ variety has ever been found. Yet at the same time, it’s not precisely clear how a social currency could transform into a commercial one, or if, indeed, that ever happened.

The matter becomes even more complex when one takes into account that such historical evidence we do have about early commercial currencies – for instance, from Bronze Age Mesopotamia and Egypt – indicates that money was used much more as an abstract unit of account to support credit systems than as an actual ‘medium of exchange’. In other words, the real question seems to be, not how some kind of currency emerged from the difficulties of barter, but rather, how social responsibilities first came to be quantifiable, that is to be enumerated as debts, and how that same logic could then be extended to material goods. Since anthropologists have not been asking these kind of grand historical questions, the closest we have to a suggested answer derives from a numismatist, Philip Grierson (1977), who points to the importance of legal codes. Early Medieval Welsh law codes, for example, listed the money value of every object likely to be found in a typical house, from the pots to the rafters, despite the apparent absence of markets where any of these items could be bought or sold. The reason was the need to pay fines and penalties that were denominated in cattle, but (since not everyone had cattle) payable in anything of value acceptable to the other side. This situation seems to have been widespread.

It makes a lot of sense that the need to establish precise proportion values – and at minimum definition this is all that money is, the means of establishing such proportional values conceptually, whether it exists as a physical currency or not – would be unlikely to arise from, say, gift exchange, with its emphasis on at least the pretence of open- handed generosity, but would be needed for legal negotiations, intended to head off feuds, where each party would be minimally inclined to compromise, and problems of commutation would frequently come up. The ethnographic record provides a good deal of confirmation. One reason this ‘wergeld theory’ of the origins of money (as it has come to be known) has not become better established outside certain heterodox economic circles is that in the hands of most of those who have taken it up, it has become entangled in another body of theory, usually referred to as ‘primordial debt theory’, which hearkens back not to ethnographic evidence but to a certain tradition of reading Vedic and Homeric literature. This is one that assumes money originates in sacrificial ritual, and that ultimately taxes are a secularisation of the existential debt humans are felt to owe to the gods (Laum 1924; Malamoud 1988; Aglietta and Orle´an 1995; The´ret 1995; Ingham 2004). Systems of fines and penalties, in this conception, become simply the means by which this fundamental prior ‘debt to society’ first becomes refracted into specific debts to individuals. The problem of course is that such an approach is quite impossible to square with most ethnographic cases, where sacrificial ritual – let alone sacrifice framed in these particular ideological terms – often does not appear at all. In fact the whole seems better approached as an attempt to create a counter-myth to the myth of barter, based on a Durkheimian notion of society as God, than as a genuine theory of the origins of money.

However dubious as an attempt at historical reconstruction, this literature does raise at least one important, and challenging, point. Sacrifice was rarely, if ever, conceived as a matter of settling an existential debt to the gods; even those who did see it this way saw sacrifice, rather, as an acknowledgement of debts that could not be paid. And as Philippe Rospabe´ (1995), a French economist-turned-anthropologist working broadly within this tradition, has argued quite convincingly, the same could be said of many of the classical uses of social currencies as well. Payments for homicide are the most obvious examples. In almost every case, great emphasis is placed on the fact that the fine, whether paid in Nuer cattle, Haudensaunee wampum, or some other social currency, is in no sense to be considered an equivalent for the victim’s life; only a life can be the equivalent for a life and even that is not really adequate for the loss of a relative. Rather the payment is itself a kind of sacrifice, in the sense of token of sincerity, giving something up to demonstrate a desire to make amends. Suggesting the payment was in any way equivalent to the life would be considered an insult, and likely to provoke further hostilities. Primordial debt theorists tend to sidestep the whole question of marriage payments, but as Rospabe´ notes, a case could well be made these are not too different: since the 1920s, anthropologists have emphasised that paying bridewealth is not brideprice, not a matter of purchasing a wife, and as Rospabe´ himself observed, those societies that do practice sister-exchange, such as the Tiv, are equally inclined to insist that only a woman’s life can be considered equivalent to another woman’s life, so that payments in copper bars are really just a place-holder, a recognition of such a debt, until some future generation where a sister or ward can be provided (1995: 13).

If so, the conceptual question does indeed become: how does a token of recognition that one cannot pay a debt turn into a form of payment by which a debt can be extinguished? Without such a conceptual break, it’s hard to see how a human economy could ever have produced a commercial economy in the first place. Although the historical origins of commercial economies in West Asia, India or China took place so long before the appearance of written records that we can never really move beyond speculation, I do think the ethnographic record provides us some suggestive analogies – essentially, by showing us what happens when human economies are abruptly incorporated into larger commercial ones. Probably the most dramatic example here is the West African slave trade in the 17th–18th centuries, which happened recently enough that some of its traces can still be observed in colonial ethnographies.

In what follows I will reconsider two classic cases of 1950s ethnography: the Lele of the Republic of Congo, made famous by the work of Mary Douglas, and the Tiv, known by the work of Paul and Laura Bohannan. Each is known for both their use of social currencies (Lele raffia and camwood, Tiv brass rods) and elaborate ideologies of debt. Yet the differences are even more significant, and seem to have much to do with their very different historical situation: the Lele appear to have been largely unaffected by the slave trade, and thus provide something of a glimpse of how earlier human economies on the Atlantic side of central Africa might have been organised; the Tiv were clearly making strenuous efforts to avoid being drawn into the trade that engulfed so many of their neighbours. What emerges most of all from these comparisons (Lele, Tiv, the Cross River societies neighbouring the Tiv) is that the conceptual break in question can only be achieved through extraordinary violence.

Lele: blood debts

The Lele were, when Mary Douglas carried out her initial ethnography in the 1950s, a relatively small group of perhaps 10,000 souls, living on a stretch of rolling country near the Kasai River. Lele women grew maize and manioc; the men thought of themselves above all else as intrepid hunters, although they spent most of their time weaving and sewing together cloth from the fibres of the local raffia palms. This raffia cloth was what the area was really known for. It was not only used for every sort of clothing, but also widely exported. Internally, it functioned as a sort of currency. Still, raffia was not used in markets (there were no markets), and, as Douglas discovered to her great inconvenience, within a village, it couldn’t be used to acquire food, tools, tableware or almost any other material good. Rather, it was the quintessential social currency.

Informal gifts of raffia cloth smooth all social relations: husband to wife, son to mother, son to father. They resolve occasions of tension, as peace-offerings; they make parting gifts, or convey congratulations. There are also formal gifts of raffia which are neglected only at risk of rupture of the social ties involved. A man, on reaching adulthood, should give 20 cloths to his father. Otherwise he would be ashamed to ask his father’s help for raising his marriage dues. A man should give 20 cloths to his wife on each delivery of a child (Douglas 1958: 112; also 1982: 43)

Cloth was also used for various fines, fees and to reward curers. For instance, if a man’s wife reported a would-be seducer, it was customary to reward her with 20 cloths for her fidelity (it was not required, but not doing so was considered decidedly unwise); if an adulterer was caught, he was expected to pay 50 or 100 to the woman’s husband, if the husband and lover disturbed the peace of the village by fighting before the matter was settled, each would have to pay two in compensation, and so forth.

Raffia tended to flow upwards. Young people were always giving little gifts of cloth as a mark of respect to fathers, mothers, uncles and the like. These gifts were hierarchical: it never occurred to those receiving them that they should have to reciprocate in any way. As a result, male elders usually had a bit extra lying around, and young men, who could never weave quite enough to meet their needs, would have to turn to elders when they had to make a major payment (fines, doctors to assist in childbirth, fees for a cult society). They were thus always slightly beholden to their elders. Still, most also had a range of friends or kin they could turn to for help.

Marriage was particularly expensive, since the negotiations usually required several bars of camwood. If raffia cloth was the small change of social life, camwood – a rare imported wood used for the manufacture of cosmetics – was the high denomination currency. A hundred raffia cloths was equivalent to three to five bars. Few individuals owned much camwood, usually just little bits to grind up for their own use. Most was kept in each village’s collective treasury. Still, this was not bridewealth. The Lele were matrilineal; men could not use either camwood or raffia to acquire rights in women or their rights over children. These could only be acquired through the system of what Douglas calls ‘blood-debts’.

It’s important to understand how this system worked, since everything was organised around it.

First of all, Lele assumed that all deaths, except perhaps of the extremely old, were caused by some sort of human action. A culprit had to be identified, if necessary by divination, and that culprit was then held by the village to owe a life to the victim’s closest male kin. This life-debt could only be paid by transferring a young woman from his family, his sister or her daughter, to be the victim’s ward, or ‘pawn’.

The status of pawnship was inherited by a woman’s children, which meant most males were also someone’s pawn; but no one would accept a male pawn in payment of blood-debts: the whole point of the game was to acquire as many pawns as possible, by acquiring a young woman, who would then go on to produce additional pawn children. Douglas’ Lele informants emphasised that any man would naturally want to have as many of these as possible:

Every man is always aware that at any time he is liable for a blood-debt. If any woman he has seduced confesses his name in the throes of child-birth, and subsequently dies, or if her child dies, or if anyone he has quarreled with dies of illness or accident, he may be held responsible Even if a woman runs away from her husband, and fighting breaks out on her account, the deaths will be laid at her door, and her brother or mother’s brother will have to pay up. Since only women are accepted as blood-compensation, and since compensation is demanded for all deaths, of men as well as of women, it is obvious that there can never be enough to go around. Men fall into arrears in their pawnship obligations, and girls used to be pledged before their birth, even before their mothers were of marriageable age. (Douglas 1973: 144–5)

In other words, the whole arrangement turned into an endlessly complicated chess game – one reason, Douglas remarks, why the term ‘pawn’ seems singularly apropos. As a result just about all adult Lele males were both someone else’s pawn, and engaged in a constant effort to secure, swap, marry off or redeem other pawns. Every major drama or tragedy of village life would ordinarily lead to a transfer of rights in women. Almost all of those women would eventually get swapped again.

Several points need to be emphasised here. First of all, what were being traded were, quite specifically, human lives. Douglas calls them ‘blood-debts’ but ‘life-debts’ would be more appropriate. Even a man saved from drowning, who thus ‘owed his life’ to his rescuer, would be normally expected to turn over his sister or ward as a pawn.

Secondly, nothing could substitute for a human life: no amount of raffia cloth, camwood bars, goats, transistor radios or anything else could ever be considered equivalent to a human life or used to acquire a pawn.

Finally, in practice, ‘human life’ meant ‘woman’s life’ – or even more specifically, ‘young woman’s life’. Ostensibly this was because only young women could become pregnant and bear other pawns, but even Mary Douglas, who was in no sense a feminist, admitted that the whole arrangement did seem to operate as if it were one gigantic apparatus for asserting male control over women. This was true above all because women themselves could not own pawns. They could only be pawns. In other words: when it came to life-debts, only men could be either creditors or debtors. Young women were thus the credits and the debits – the pieces being moved around the chessboard – while the hands that moved them were invariably male.

For a man, being a pawn could be advantageous, since it meant one effectively had two different families to turn to for help – that of one’s mother’s brother, and that of one’s ‘lord’. For this reason, Douglas’ informants insisted pawnship had nothing in common with slavery. Slaves were war captives, typically non-Lele, and as a result had neither family nor patron to protect them.

Many young women were born pawns, already assigned to some man for eventual marriage – but Douglas insists they had plenty of space to negotiate (Douglas 1966: 150). They regularly took lovers, or threatened to, and in addition, a young Lele woman had one unique and powerful card to play: everyone knew that if she felt her preferences were completely ignored, for instance, if she was assigned to be some old man’s third or fourth wife, she always had the option of fleeing to a rival community to become a ‘village-wife’ (Douglas 1951; 1963: 128–40).

Douglas explains this institution – which seems peculiar to the Lele – as a way of dealing with the fact that older men tended to monopolise the available pool of wives:

Everyone recognized that the young unmarried men coveted the wives of their seniors. Indeed, one of their pastimes was to plan seductions and the man who boasted of none was derided. Since the old men wished to remain polygynists, with two or three wives, and since adulteries were thought to disrupt the peace of the village, Lele had to make some arrangement to appease their unmarried men. Therefore, when a sufficient number of them reached the age of eighteen or so, they were allowed to buy the right to a common wife. (Douglas 1963: 76; cf. 1951: 11)

After paying an appropriate fee in raffia cloth to the village treasury, they were permitted to build a collective house, and then they were either allotted a wife to put in it, or allowed to form a party that would try to steal one from a rival village. Alternately, if one showed up as a refugee, they would ask the rest of the village for the right to accept her: this was invariably granted. This common wife is what’s referred to as a ‘village wife’. The position of village wife was perfectly respectable. In fact, a newly married village wife was treated very much like a princess.[1] She was not expected to plant or weed, fetch wood or water, or even to cook; all the household chores were done for her by her eager young husbands, who spent much of their time hunting in the forest, vying to bring her the choicest delicacies, or plying her with palm wine. A new village wife could help herself to others’ possessions, appropriate their raffia; she was expected to make all sorts of mischief, to the bemused indulgence of all concerned. She was also expected to make herself sexually available to all members of the age-set – perhaps ten or twelve different men – at first, pretty much whenever they wanted her.

Over time, a village wife would normally settle down with just three or four of her husbands, and finally, just one. Still, technically, she was married to the village as a whole. If she had children, the village was, collectively, their father, and as such expected to bring them up, provide them with resources, and eventually, to get them properly married off.

This is important because it had the effect of ensuring that villages, at least in that one limited capacity, became corporate entities, fictive persons in much the same way as contemporary corporations that could own property and enter into contracts as if they were individual human beings. That’s why villages had collective treasuries full of raffia and camwood bars in the first place. But as a result, the responsibilities of a village, as fictive person, had a tendency to expand. At any time a given village was likely to have several village-wives. This usually meant that over time it would accumulate a fair number of children and grandchildren. As a result, villages could – and generally did – also end up holding pawns, demanding or paying blood-debts, and otherwise, act like extremely powerful patrons, or ‘lords’.

Still, villages were not really individuals. They were made up of large numbers of individuals, some of whom could band together and make war. This added another element, which in a way transformed everything. Villages could make claims to pawns, or debts, much like any individual; but unlike individuals, they were also able to back up those claims with the implicit threat of force.

This latter deserves emphasis because when ordinary ‘lords’ made claims on their clients, they most certainly could not. Men as patrons or husbands could sometimes put a great deal of physical pressure on women, threatening them with beatings if they denied sex to their husbands, for example, but they really could not use physical compulsion on other men. There was an almost complete lack of any systematic means of coercion (Douglas emphasises this frequently: 1960: 4; 1963: 145–6, 168–73; 1964: 303). This was the main reason, she adds, that pawnship was so innocuous. Without an apparatus of coercion, no one was in a position to completely ignore another’s feelings. What’s more, in everyday affairs, Lele put great stock on gentle and agreeable behaviour (1963: 70–1). Men might have been regularly seized with urges to throw themselves at each other in fits of jealous rage (often they had good reason to), but they very rarely did. And if a fight did break out, it was considered the responsibility of everyone nearby to immediately break it up, and submit the affair to public mediation.

Between villages it was completely different. Villages were fortified, and age-sets could be mobilised to act as military units. Here, and only here, did organised violence enter the picture. True, when villages fought, too, it was always over women. (All Lele Douglas talked to expressed incredulity at the very idea that grown men, anywhere, could ever come to blows over anything else.) But in the case of villages, it could come to an actual war. If another village’s elders ignored one’s claims to a pawn, one’s young men might organise a raiding party and kidnap her, or carry off some other likely young women to be their collective wives. This might lead to deaths, and further claims for compensation. ‘Since it had the backing of force,’ Douglas observes drily, ‘the village could afford to be less conciliatory towards the wishes of its pawns and to act more arbitrarily’ (1963: 170).

It’s at exactly this point, too, where the potential for violence enters in the picture, that the great wall constructed between the value of lives and money can suddenly come toppling down.

Sometimes when two clans were disputing a claim to blood compensation, the claimant might see no hope of getting satisfaction from his opponents. The political system offered no direct means for one man (or clan) to use physical coercion or to resort to superior authority to enforce claims against another. In such a case, rather than abandon his claim to a pawn-woman, he would be ready to take the equivalent in wealth, if he could get it. The usual procedure was to sell his case against the defendants to the only group capable of extorting a pawn by force, that is, to a village.

The man who meant to sell his case to a village asked them for 100 raffia cloths or five bars of camwood. The village raised the amount, either from its treasury, or by a loan from one of its members, and thereby adopted as its own his claim to a pawn. (1963: 171)

Once he held the money, his claim was over, and the village, which had now bought it, would proceed to organise a raid to seize the woman in dispute.

In other words, it was only when violence was brought into the equation that there was any question of buying and selling human beings. The ability to deploy force, hence to cut through the endless maze of preferences, obligations, expectations and responsibilities that mark ordinary human relationships and all have to be taken so painstakingly into consideration when negotiating an ordinary marriage or exchange of pawns, also made it possible to overcome what is otherwise the first rule of all economic relationships: that human lives can only be exchanged for other human lives. Significantly, the amount paid – a hundred cloths, or equivalent amount of camwood – was also the price of a slave (1963: 36, 1982: 46–7). Slaves were, as I mentioned, war captives. There never seem to have been very many of them; most did not live long,since most were de stined for sacrifice at some important man’s funeral (1963: 36). Still, the mere fact of their existence established that life could have a price. But if objects of wealth could be transformed into people, it was only through the mediation of violence.

Tiv: flesh-debts

Lele currencies were quintessential social currencies, used to mark every visit, every promise, every important moment in a man or woman’s life. It is surely significant, too, that raffia cloth was, in Douglas’ day, the primary form of human clothing, for both men and women; and that the red paste made from camwood bars was the main substance used as make-up, by both men and women, to beautify themselves each day. Money-stuffs were also the materials used to shape people’s physical appearance, to make them appear mature, decent, attractive, dignified to their fellows; to turn a mere naked body into a proper social being.

This is by no means unusual. In human economies, money typically consists of objects otherwise used primarily as adornment to the person. There are exceptions, of course, like cattle. But ordinarily social currencies take the form of beads, shells, feathers, dog or whale teeth, gold and silver rather than barley, tobacco or salt (see Graeber 2001). Even the famous ‘brass rods’ used by the Tiv were used mainly as raw material for the manufacture of jewellery, or just twisted into hoops and worn at dances. But is this not use of objects that create the exterior, social aspect of the person entirely appropriate for a currency seen as in some sense the equivalent of a human being, but at the same time, in the most important ways, as always falling short?

All this was quite explicit in the case of the Tiv.

The Tiv are an acephalous society from the Benue River region of central Nigeria, most famous in the anthropological literature for maintaining what Paul Bohannan famously described as three separate ‘spheres of exchange’ (1955, 1959, cf. Dorward 1976; Guyer 2004: 27–31. When they did their research in the 1950s, ordinary, everyday economic activity was mostly the affair of women, who dominated local marketplaces where okra, shovels, sandals and the like could be bought and sold using colonial currency. This represented the first sphere. Men concerned themselves with what they considered higher transactions of the sort that could be conducted using the traditional Tiv currency, which were considered the second sphere: here too, currency consisted of what were in effect two denominations: a kind of locally made cloth called tugudu, widely exported, and, for major transactions, bundles of imported brass rods. Currency could be used to acquire certain flashy and luxurious items (cows, purchased foreign wives), but mainly for the give and take of political affairs, hiring curers, acquiring magic, gaining initiation into cult societies. In such political matters, Tiv were even more resolutely egalitarian than the Lele: successful old men with their numerous wives might have lorded it over their sons and other dependants within their own house compounds, but beyond that, there was no formal political organisation of any sort. The third sphere, finally, consisted exclusively in rights in women: in this case not ‘pawns’ but what are called in the literature ‘wards’. Where Lele pawns were women turned over in compensation for the taking of life, Tiv wards were women turned over in compensation for the giving of life. The Tiv were patrilineal, and the only entirely proper form of marriage was held to be sister exchange; any other form left a debt, to be paid by transfer of control over a woman who would thus become a ‘ward’; as among the Lele, the system soon became an infinitely complicated chess game played by men over rights in women.

In principle, these three spheres – of ordinary consumption goods, masculine prestige goods and rights in women – were completely separate. No amount of okra could get you a brass rod, just as, in principle, no number of brass rods could give you full rights to dispose of a woman in marriage. In practice, there were always ways to convert upwards or downwards. But they were never considered true equivalents. A man with no sister, or ward, could attain a wife by constantly plying her parents or guardians with gifts of brass rods, but such a wife was never considered truly his.

Just as among the Lele, though, there was only one real exception to this principle – the only way to establish an actual equivalence between valuables and human beings was through the mediation of violence. It was possible to buy a slave, that is, a woman kidnapped in a raid from some distant country (Akiga Sai 1939: 121, 158–60), and marry her. Slaves after all had no parents, or could be treated as if they didn’t; they had been forcibly removed from all those networks of mutual obligation and debt in which ordinary people became who they were. This was why they could be bought and sold. But it also meant such a woman was completely one’s wife, unlike one merely acquired from her parents through brass rods.

The reverse side of this was that anyone who did, in practice, manage to game the system, and convert upwards, was seen as if they were, effectively, practitioners of violence. Converting upwards was normally accomplished by manipulating debts. Say someone was sponsoring a marriage feast and was short of supplies; a neighbour might come to their aid, then later, discretely, ask for a bundle or two of rods in repayment. To be able to ‘turn chickens into cows’, as the saying went, and eventually, use one’s wealth and prestige to acquire wives, required a ‘strong heart’ – that is, an enterprising and charismatic personality. But ‘strong heart’ had another meaning too. There was believed to be a certain actual biological substance called tsav that grew on the human heart. This was what gave certain people their charm, their energy and powers of persuasion. Tsav therefore was both a physical substance, and that invisible power that allows certain people to bend others to their will (Abraham 1933: 26; Akiga Sai 1939: 246; Bohannan 1958: 3; Downes 1977: 27).

The problem was held to be that it was also possible to augment one’s tsav through artificial means, by consuming human flesh. At least at the time our ethnographic sources refer to, most Tiv appear to have been veritably obsessed by the suspicion that some of their neighbours – and particularly prominent men who became de facto political leaders – were, in fact, secret cannibals. Men who built up their tsav by such means, the stories went, gained extraordinary powers: the ability to fly, to become impervious to weapons. They became able to send out their souls at night to kill their victims in such a way that their victims did not even know that they were dead, but would wander about, confused and feckless, until they could be harvested for their cannibal feasts. They became, in short, terrifying witches.

The mbatsav or society of witches was always looking for new members and the way to accomplish this was to trick people into eating human flesh. A witch would take a piece of the body of one of his own close relatives, who he had murdered, and place it in the victim’s food. If the man was foolish enough to eat it, he would contract a ‘flesh debt’, and the society of witches ensured that flesh debts are always paid. According to Akiga Sai, an early Tiv account:

Perhaps your friend, or some older man, has noticed that you have a large number of children, or brothers and sisters, and so tricks you into contracting the debt with him. He invites you to eat food in his house alone with him, and when you begin the meal he sets before you two dishes of sauce, one of which contains cooked human flesh (Akiga Sai 1939: 257)

If you eat from the wrong dish, but you do not have a ‘strong heart’ – the potential to become a witch – you will become sick and flee from the house in terror. But if you have that hidden potential, the flesh will begin to work in you. That evening, you will find your house surrounded by screeching cats and owls. Strange noises will fill the air. Your new creditor will appear before you, backed by his confederates, and demand your life in payment. The only way out is to pledge a member of your own family as substitute. This is possible, because you will find you have terrible new powers, but they must be used as the other witches demand. One by one, you must kill off your brothers, sisters, children; their bodies will be stolen from their graves by the college of witches, brought back to life just long enough to be properly fattened, tortured, killed again, then carved and roasted for yet another feast.

The flesh debt goes on and on. The creditor keeps on coming. Unless the debtor has men behind him who are very strong in tsav, he cannot free himself from the flesh debt until he has given up all his people, and his family is finished. Then he goes himself and lies down on the ground to be slaughtered, and so the debt is finally discharged. (Akiga Sai 1939: 260)

On a superficial level, it’s easy to see what’s happening here. The Tiv lived in a landscape dotted with compounds, each organised around a single older man with his numerous wives, children and assorted hangers-on. Outside, there was no formal political organisation whatsoever, or institutional hierarchies of any kind. They were a people where men aspired to absolute authority within their households, but were suspicious of any sort of authority beyond it. To be a successful man meant to be able to manipulate debt so as to be able to convert food (the first sphere) into objects of prestige (the second), prestige into control over women (the third) so as to become the head of a household with numerous descendants. Is it surprising, then, they also felt that the very ‘strong heart’ that made this possible also threatened to make that ability to manipulate debt tumble back on itself; that the ‘standardized nightmare’ (Wilson 1951) of their society was one where the successful head of household ended up locked in a flesh debt where he was forced to convert those very descendants back into food.

Way of death

Still, the emphasis on debt is puzzling, since nothing in the ethnography suggests that in the 1950s debt was a pervasive concern of everyday Tiv life. Here I think we have to turn to a larger historical context.

The early history of the Tiv is difficult to reconstruct, but they appear to have arrived in the Benue River valley and adjacent lands sometime around 1750 – that is, during a time when all of what’s now Nigeria was being torn apart by the Atlantic slave trade. Early stories told how the Tiv, during their migrations, used to paint their wives and children with simulated smallpox scars, so that potential raiders would be afraid to carry them off. They established themselves in a notoriously inaccessible stretch of country, and offered up ferocious defence against periodic raids from neighbours to their north and west (Abraham 1933: 17–26; Akiga Sai 1954; Bohannan 1954). Some of these raids were not entirely unsuccessful. It’s probably not insignificant that the nearby Jukun kingdom, which made a series of ultimately unsuccessful attempts to conquer the Tiv in the 18th century, disposing many Tiv captives to slave-dealers on the coast in the process, was also seen, in later times, as the real origin of the ‘organisation’ of the mbatsav (Abraham 1933: 19, 31–5; Curtin 1969: 255, 298; Latham 1973: 29; Tambo 1976: 201–3).

One might also consider the actual origin of the famous copper bars used as social currency.

Copper bars had been used for money in this part of Africa for centuries; often, it seemed, they were used not just for social purposes but broken up into small change for use in ordinary commercial transactions (Jones 1958; Latham 1971; Northrup 1978: 157–64; Herbert 2003: 196). Ibn Battuta saw people using copper bars to buy everyday wares in marketplaces in the nearby Niger region as far back as the 1340s. Most of the bars current in 18th- and 19th-century Tivland, on the other hand, were not local products. They were mass-produced in factories in Birmingham, and imported through the port of Old Calabar at the mouth of the Cross River, by slave-traders based in Liverpool and Bristol. The Tiv were unusual in restricting these bars to social purposes. In all the country adjoining the Cross River – that is, in the region directly to the south of the Tiv territory – they were still used as everyday currency.

It is hardly surprising that Tiv were suspicious of such items. Almost everywhere else, they were also the currency of the slave. During the 1760s alone, perhaps 100,000 Africans were shipped down the Cross River to Calabar and nearby ports, where they were put in chains, placed on British, French or other European ships, and shipped across the Atlantic – part of perhaps 1.5 million exported from the Bight of Biafra during the whole period of the trade (Eltis et al. 2000; Lovejoy and Richardson 1999: 337). Some had been captured in wars, raids or simply kidnapped. The majority, though, were carried off because of debts.

In fact the Atlantic Slave Trade as a whole was a gigantic network of credit arrangements. Ship-owners based in Liverpool or Bristol would acquire goods on easy credit terms from local wholesalers, expecting to make good by selling slaves (also on credit) to planters in the Antilles and America, with commission agents in the city of London ultimately financing the affair through the profits of the sugar and tobacco trade (Sheridan 1958; Price 1980, 1989, 1991). Ship-owners would then ship their wares to African ports like Old Calabar. Calabar was the quintessential mercantile city-state, dominated by an African elite who dressed in European clothes, built themselves European-style houses, and in some cases even sent their children to England to be educated.

On arrival, European merchants would negotiate the value of their cargoes in the copper rods that served as the currency of the port. The cargoes themselves consisted of cloth, iron and copper ware, incidental goods like beads, and substantial numbers of firearms. The goods were then advanced to Calabar’s merchant elite, again on credit, who assigned them to their own agents to move upstream.

The obvious problem was how to secure the debt. The trade was an extraordinarily duplicitous and brutal business, and merchants who often doubled in the interior as no more than raiders and kidnappers were also notoriously bad credit risks. As a result, a system quickly developed where European captains would demand security in the form of pawns.

It would seem that, with the development of commercial towns on the West African coast, institutions that must have originally resembled Lele pawnship, or Tiv wards, had gradually transformed into what was effectively a form of debt peonage. We don’t know precisely how it happened, but the process was clearly well under way even before Europeans appeared on the scene in the 16th century. Debtors would pledge a family member as surety for a loan; the pawn would then become a dependent in the creditor’s household, working his fields or tending to his household chores – their persons acting as security and their labour, effectively, substituting for interest. Still, there are clear signs of a historic connection: for instance, if a girl was pledged, the creditor generally had the option of marrying her when she reached maturity, thus cancelling the debt, exactly as among the Lele. And critically, pawns were clearly distinguished from slaves. The difference only became blurred once it became the custom for the masters of slaving ships, on advancing goods to their African counterparts, to demand pawns – for instance, two of the merchants’ own dependents for every three slaves to be delivered, preferably, including at least one or two members of the merchants’ families (Lovejoy and Richardson 1999: 349–51; 2001). This was in practice not much different than demanding the surrender of hostages, and at times created major political crises if captains, tired of waiting for delayed shipments, decided to take off with a cargo of pawns instead.

Upriver, debt pawns also played a major part in the trade. In the Cross River region, this trade seems to have had two phases. The first was one of absolute terror and utter chaos, in which raids were frequent, and anyone travelling alone risked being kidnapped by roving gangs of thugs and sold to Calabar. Villages lay abandoned; many fled into the forest; men would have to form armed parties to work the fields (Equiano 1789: 6–13). This period was relatively brief. The second began when representatives of local merchant societies began establishing themselves in communities up and down the region, offering to restore order. The most famous of these was the Aro Confederacy, who, calling themselves ‘Children of God’, and backed by heavily armed mercenaries and the prestige of their famous Oracle at Arochukwu, created their own justice system, with the Oracle acting as a kind of regional court of high appeal (see Jones 1939; Ottenberg 1958; Afigbo 1971; Ekejiuba 1972; Isichei 1976; Northrup 1978; Dike and Ekejiuba 1990; Nwauwa 1991). This system was notoriously harsh, and itself seems to have functioned above all to either reduce as many villagers as possible into slavery by judicial means, or to assign penalties (always denominated in brass rods) so hefty that culprits would be forced to sell themselves or members of their families into slavery.

These same merchant societies also assisted in the dissemination of a secret society called Ekpe, most famous for sponsoring magnificent masquerades and for initiating its members into arcane mysteries, but that also acted as a covert mechanism for the enforcement of debts. In Calabar itself, the Ekpe society operated primarily as a means of enforcing contracts and collecting debts (Latham 1973: 38). But it was open to anyone willing to pay the hefty initiation fees – which were also exacted in the brass rods the merchants themselves supplied. In the town the fee schedule for each grade looked like this (from Walker 1875: 120):

1. Nyampi
2. Oku Akana
3. Brass
4. Makanda
300 boxes brass rods, each £2 = £735, for the first four grades
+ 5. Makara
6. Mboko Mboko
7. Bunko Abonko
8. Mboko Nya Ekpo
9. Ekpe
50 boxes brass rods for each of the lower grades.

In town, membership became the chief mark of honour and distinction. Entry fees were no doubt less exorbitant in small, distant communities, but the effect was the same: thousands ended up in debt to the merchants, whether for the fees required for joining, or for the trade goods they supplied (mostly cloth and metal put to use creating the equipment and costumes for the Ekpe performances), debts that they thus themselves became responsible for enforcing on themselves. These debts, too, were regularly paid in people, ostensibly, yielded up as pawns. But in these cases the line between pawns and slaves soon became effectively non-existent.

In the countryside, practices varied. In many areas, copper rods became general purpose money. In the Afikpo district (Ottenberg and Ottenberg 1962: 124), on a remote part of upper Cross River, we learn that copper bars, supplied by the merchant societies, were not used to buy food but restricted to social purposes, ‘for gifts and for payments in funerals, titles, and other ceremonies’. Most of those payments, titles and ceremonies however were tied to the secret societies that the merchants themselves had brought to the area:

In the old days, if anybody got into trouble or debt in the upper parts of the Cross River, and wanted ready money, he used generally to ‘pledge’ one or more of his children, or some other members of his family or household, to one of the Akunakuna traders who paid periodical visits to his village. Or he would make a raid on some neighboring village, seize a child, and sell him or her to the same willing purchaser. (Partridge 1905: 72)

The passage only makes sense if one recognises that debtors were also, owing to their membership in the secret societies, also the debt collectors. The seizing of a child can only be a reference to the local practice of ‘panyarring’, current throughout West Africa, by which creditors despairing of repayment would simply sweep into the debtor’s community with a group of armed men and seize anything – people, goods, domestic animals – that could be easily carried off, then hold it hostage as security. It was actually a quite sensible expedient in an environment with no central authority, where people tended to feel an enormous sense of responsibility towards other members of their community, and very little responsibility towards anyone else. In the case cited above, the debtor would, presumably, be calling in his own debts – real or imagined – to those outside the organisation, in order not to have to send members of his own family.

Such expedients were not always effective. Often debtors would be forced to pawn more and more of their own children or dependents, until finally, there was no recourse but to pawn themselves (Harris 1972: 128). And of course, at the height of the slave trade, ‘pawning’ had become little more than a euphemism. The distinction between pawns and slaves had largely disappeared. Debtors, like their families before them, ended up turned over to the Aro, then to the British, and finally, shackled and chained, crowded into tiny slaving vessels, and sent off to be sold in plantations across the sea.

If the Tiv, then, were haunted by the vision of an insidious secret organisation that lured unsuspecting victims into debt traps, whereby they themselves became the enforcers of debts to be paid with the bodies of their children, and ultimately, themselves – one reason was because this was, literally, happening to people who lived no more than a few hundred miles away. Nor is the use of the phrase ‘flesh debt’ especially inappropriate. Slave-traders might not have been reducing their victims to meat, but they were certainly reducing them to nothing more than bodies.

What was remarkable that all this was done, the bodies extracted, through the very mechanisms of the human economy, premised on the principle that human lives are the ultimate values, to which nothing could possibly compare. Instead, all the same institutions – fees for initiations, means of calculating guilt and compensation, social currencies, debt pawnship – were turned into their opposite; the machinery was, as it were, thrown into reverse; and, as the Tiv also perceived, the very gears and mechanisms designed for the creation of human beings collapsed on itself, and became the means for their destruction.

As the above examples reveal, the change could only be effected by violence – in the case of the Atlantic slave trade, what is almost certainly the greatest and most catastrophic outbreaks of commercial violence in the history of the world. Yet at the same time, I think the very intensity of the catastrophe can help lay bare some of the mechanisms by which human economies could have, in many other times and places in human history, overcome the conceptual barriers between social currencies, as tokens of a debt that cannot be paid, and commercial currencies, as means of cancelling debts in their entirety. One thing is clear: the change was effected by violence. Above all, it was only violence that could rip a human being entirely from the web of unique human relations that thereby made her a unique individual, a daughter, sister, wife, lover, friend, so as to make her the exact equivalent of anyone else. But of course, this violence was already present even when lives could only be equivalent to other lives. Among the Lele, men could not be compelled to do anything they did not agree to do, but women could still be beaten if they completely refused to comply with the system that rendered them exchangeable. Among the Tiv, Akiga Sai is even more explicit:

Under the old system an elder who had a ward could always marry a young girl, however senile he might be, even if he were a leper with no hands or feet; no girl would dare to refuse him. If another man were attracted by his ward he would take his own and give her to the old man by force, in order to make an exchange. The girl had to go with the old man, sorrowfully carrying his goat-skin bag. If she ran back to her home her owner caught her and beat her, then bound her and brought her back to the elder. The old man was pleased, and grinned till he showed his blackened molars. ‘Wherever you go,’ he told her, ‘you will be brought back here to me; so stop worrying, and settle down as my wife.’ The girl fretted, till she wished the earth might swallow her. Some women even stabbed themselves to death when they were given to an old man against their will; but in spite of all, the Tiv did not care. (1939: 161)

Anthropologists have spent much of the 20th century studying kinship systems, often creating elaborate and elegant diagrams to understand what Le´vi-Strauss so famously called ‘the exchange of women’ (1949). It was only after feminist authors like Gayle Rubin (1975) began to point out just how coercive such systems ultimately are, how much violence lay beneath them, that anthropologists suddenly seem to have concluded that the entire subject was no longer particularly interesting. Yet it would appear that it is precisely through elaborating on this underlying violence, through the transformation of pawns into peons, for example, that systems of debt could begin to take what we would now consider commercial form: that is, as a series of quantifiable, fully exchangeable equivalents, and that social currencies could become money in the familiar sense of the term.


Abraham, R. C. 1933. The Tiv people. Lagos: Government Printer.

Afigbo, A. E. 1971. ‘The Aro of southeastern Nigeria: a socio-historical analysis of legends of their origins’, African Notes 6: 31–46.

Aglietta, M. and A. Orle´an (eds.) 1995. Souverainete´, le´gitimite´ de la monnaie. Paris: Association d’E´ conomie Financie‘re (Cahiers finance, e´thique, confiance).

Akiga Sai, B. 1939. Akiga’s story; the Tiv tribe as seen by one of its members. Translated and annotated by Rupert East. London: Published for the International African Institute by the Oxford University Press.

Akiga Sai, B. 1954. ‘The “descent” of the Tiv from Ibenda Hill’ (translated by Paul Bohannan), Africa: Journal of the International African Institute 24: 295–310.

Akin, D. and J. Robbins 1998. An introduction to Melanesian currencies: agencies, identity, and social reproduction, in D. Akin and J. Robbins (eds.), Money and modernity: state and local currencies in Melanesia, 1–40. Pittsburgh: University of Pittsburgh Press.

Bohannan, P. 1954. ‘The migration and expansion of the Tiv’, Africa: Journal of the International African Institute 24: 2–16.

Bohannan, P. 1955. ‘Some principles of exchange and investment among the Tiv’, American Anthropologist 57: 60–7.

Bohannan, P. 1958. ‘Extra-processual events in Tiv political institutions’, American Anthropologist 60: 1–12.

Bohannan, P. 1959. ‘The impact of money on an African subsistence economy’, Journal of Economic History 19: 491–503.

Carrier, J. 2010. Handbook of economic anthropology. Cheltenham: Edward Elgar.

Curtin, P. D. 1969. The Atlantic slave trade: a census. Madison, WI: University of Wisconsin Press.

Dike, K. O. and F. Ekejiuba 1990. The Aro of south-eastern Nigeria, 1650–1980: a study of socio-economic formation and transformation in Nigeria. Ibadan: University Press.

Dorward, D. C. 1976. ‘Precolonial Tiv trade and cloth currency’, International Journal of African Historical Studies 9: 576–91.

Douglas, M. 1951. ‘A form of polyandry among the Lele of the Kasai’, Africa: Journal of the International African Institute 21: 1–12. (As Mary Tew.)

Douglas, M. 1958. ‘Raffia cloth distribution in the Lele economy’, Africa: Journal of the International African Institute 28: 109–22.

Douglas, M. 1960. ‘Blood-debts and clientship among the Lele’, Journal of the Royal Anthropological Institute of Great Britain and Ireland 90: 1–28.

Douglas, M. 1963. The Lele of the Kasai. London: Oxford University Press.

Douglas, M. 1964. ‘Matriliny and pawnship in Central Africa’, Africa: Journal of the International African Institute 34: 301–13.

Douglas, M. 1966. Purity and danger: an analysis of concepts of pollution and taboo. London: Routledge and Kegan Paul.

Douglas, M. 1982. In the active voice. London: Routledge and Kegan Paul. Downes, R. M. 1977. Tiv religion. Ibadan: Ibadan University Press.

Ekejiuba, F. I. 1972. ‘The Aro trade system in the nineteenth century’, Ikenga 1(1): 11–26, 1(2): 10–21. Eltis, D., S. D. Behrent, D. Richardson and H. S. Klein 2000. The transatlantic slave trade: a database. Cambridge: Cambridge University Press.

Equiano, O. 1789. The interesting narrative of the Life of Olaudah Equiano: or, Gustavus Vassa, the African. New York: Modern Library Edition, 2004.

Graeber, D. 2001. Toward an anthropological theory of value: the false coin of our own dreams. New York: Palgrave.

Grierson, P. 1977. The origins of money. London: Athlone Press.

Guyer, J. I. 2004. Marginal gains: monetary transactions in Atlantic Africa. Chicago, IL: University of Chicago Press.

Harris, R. 1972. ‘The history of trade at Ikom, Eastern Nigeria’, Africa: Journal of the International African Institute 42: 122–39.

Herbert, E. W. 2003. Red gold of Africa: copper in precolonial history and culture. Madison, WI: University of Wisconsin Press.

Ingham, G. 2004. The nature of money. Cambridge: Polity Press. Isichei, E. 1976. A history of the Igbo people. London: Basingstoke. Jones, G. I. 1939. ‘Who are the Aro?’, Nigerian Field 8: 100–3.

Jones, G. I. 1958. ‘Native and trade currencies in southern Nigeria during the eighteenth and nineteenth centuries’, Africa: Journal of the International African Institute 28: 43–56.

Latham, A. J. H. 1971. ‘Currency, credit and capitalism on the Cross River in the pre-colonial era’, Journal of African History 12: 599–605.

Latham, A. J. H. 1973. Old Calabar 1600–1891: the impact of the international economy upon a traditional society. Oxford: Clarendon Press.

Laum, B. 1924. Heiliges Geld: Eine historische Untersuchung ueber den sakralen Ursprung des Geldes. Tu¨ bingen: J. C. B. Mohr.

Le´vi-Strauss, C. 1949. Les structures e´le´mentaires de la parente´. Paris: EHESS.

Lovejoy, P. F. and D. Richardson 1999. ‘Trust, pawnship, and Atlantic history: the institutional foundations of the Old Calabar slave trade’, American Historical Review 104: 333–55.

Lovejoy, P. F. and D. Richardson 2001. ‘The business of slaving: pawnship in Western Africa, c. 1600–1810’, Journal of African History 42: 67–84.

Malamoud, C. (ed.) 1988. Lien de vie, noeud mortel. Les repre´sentations de la dette en Chine, au Japon et dans le monde indien. Paris: EHESS.

Northrup, D. 1978. Trade without rulers: pre-colonial economic development in south-eastern Nigeria. Oxford: Clarendon Press.

Nwauwa, A. O. 1991. ‘Integrating Arochukwu into the regional chronological structure’, History in Africa 18: 297–310.

Ottenberg, S. 1958. ‘Ibo oracles and intergroup relations’, Southwestern Journal of Anthropology 14: 295–317.

Ottenberg, S. and P. Ottenberg 1962. Afikpo markets: 1900–1960, in P. Bohannan and G. Dalton (eds.), Markets in Africa, 118–69. Chicago, IL: Northwestern University Press.

Partridge, C. 1905. Cross River natives: being some notes on the primitive pagans of Obubura Hill District, Southern Nigeria. London: Hutchinson & Co.

Price, J. M. 1980. Capital and credit in British overseas trade: the view from the Chesapeake, 1700–1776. Cambridge, MA: Harvard University Press.

Price, J. M. 1989. ‘What did merchants do? Reflections on British overseas trade, 1660–1790’, The Journal of Economic History 49: 267–84.

Price, J. M. 1991. Credit in the slave trade and plantation economies, in B. L. Solow (ed.), Slavery and the rise of the Atlantic system, 313–17. Cambridge: Cambridge University Press.

Rospabe´, P. 1995. La Dette de Vie: aux origines de la monnaie sauvage. Paris: Editions la De´couverte/MAUSS.

Rubin, G. 1975. The traffic in women: notes on the ‘political economy’ of sex, in R. Reiter (ed.), Toward an anthropology of women. New York: Monthly Review Press.

Servet, J.-M. 1981. ‘Primitive order and archaic trade. Part I’, Economy and Society 10: 423–50. Servet, J.-M. 1982. ‘Primitive order and archaic trade. Part II’, Economy and Society 11: 22–59. Sheridan, R. B. 1958. ‘The commercial and financial organization of the British slave trade, 1750–1807’, The Economic History Review, New Series 11: 249–63.

Smith, A. 1776. An inquiry into the nature and causes of the wealth of nations. Oxford: Clarendon Press (1976 edition).

Tambo, D. C. 1976. ‘The Sokoto Caliphate slave trade in the nineteenth century’, International Journal of African Historical Studies 9: 187–217.

The´ret, B. 1995. L’E´ tat, la finance et le social. Souverainete´ nationale et construction europe´enne (direction de publication). Paris: E´ ditions la De´couverte.

Walker, J. B. 1875. ‘Notes on the politics, religion, and commerce of Old Calabar’, The Journal of the Anthropological Institute of Great Britain and Ireland 6: 119–24.

Wilson, M. H. 1951. ‘Witch beliefs and social structure’, The American Journal of Sociology 56: 307–31.

[1] Some village wives were literally princesses, since chiefs’ daughters invariably chose to marry age sets in this way. The daughters of chiefs were allowed to have sex with anyone they wanted, regardless of age-set, and also had the right to refuse sex, which ordinary village wives did not. Princesses of this sort were rare: there were only three chiefs in all Lele territory. Douglas estimates that the number of Lele women who became village wives on the other hand was about 10% (1951).