I recently re-read Marcel Mauss’s classic 1925 essay, The Gift: The Form and Reason for Exchange in Archaic Societies, in which Mauss (a noted sociologist, influential in anthropology, who was the nephew of Emile Durkheim) argues that gifts, operating in chains of exchange based on obligations to give, receive and reciprocate, dominated public resource management in archaic societies.
I had forgotten how frustrating I had found Mauss’s essay to be. To be sure, having recently done a lot of reading into the origins of Homo sapiens and the operation of foraging societies, I can draw on a much broader information base than he could.
As an aside, if you haven’t discovered the online lectures from CARTA (the Center for Academic Research and Training into Anthropogeny), then I highly recommend them. A series of individual, or bundled, short lectures by scholars to audiences significantly of other scholars on the latest research into the evolution of Homo sapiens is an intellectual feast. An online gift to the global community of the curious.
Back to The Gift. Mauss manages to be a bit too general at times and not general enough at other times. So, there is an insufficiency of precision in his discussions — what he means in a concrete sense is often not as clear as it should be. But he also wants to make grand claims about stages of social development that do not really fit with the evidence.
For example, he writes:
Taking a step back, the evidence for long-term exchange among Homo sapiens is about as old as the evidence for the first Homo sapiens. We are the trading ape. So, the evidence is that Adam Smith was mostly correct when he wrote in The Wealth of Nations (Book 1, Chapter 2) that:
Mauss is, however, a little too impressed with gifts as acts of exchange. I want to suggest a somewhat different take.
Pooling, connection and …
The most intimate way that Homo sapiens deal with resources is pooling, whether within the family or within the hunting band. We are a group-living species who, most likely, evolved from group-living species. All our nearest relatives are group-living species (gorillas, bonobos, chimpanzees). It is extremely unlikely that our ancestors in the Homo line, and the various common ancestors with the African great apes before that, were not also group-living species.
It is striking, however, that the four surviving African great ape species have quite different mating patterns. Gorillas are a harem species where an adult male, about twice the size of adult females, dominates a group of females. Bonobos are a group-mating species based on coalitions built around an alpha female and alpha male (often the alpha female’s son). Chimpanzees are a group-mating species based on strictly male-only, female-dominating coalitions led by an alpha male.
We are a mildly polygynous pair-bonding species whose foraging cultures tend to strongly resist attempts to establish any form of dominance by any group member. Dominance only (re)emerges among humans either when food production permits storage of food or there are variation in productive assets. Dominance behaviour can get quite extreme, however, when there is both storage of food and variation in productive assets.
Homo sapien foragers engage in sharing, in pooling, behaviour for food (particularly from hunting) far more than do our African great ape relatives.
Use of fire can also be a pooling behaviour. Indeed, the answer to ‘what is the Homo sapien diet?’ is, cooked food. We are the cooking ape.
As the evidence for systematic use of fire in our evolutionary line is older than evidence for Homo sapiens, we pooled resources and managed connections before we engaged in exchange. Indeed, it is likely that it the development of those capacities that led us to becoming the trading ape.
But Homo sapiens had connections beyond the hunting band. (This is true in most foraging societies and the few it was not true of, you really didn’t want to live in.) Human foragers also might interact with people from groups that they had no connections with. With the former, the connections need to be managed. With the latter, any non-violent interaction has to be able to operate without being sustained by connections across the groups.
By connection I simply mean: repeated, mutually acknowledged, interactions that both parties presumptively mean to continue.
Across a wide variety of human societies, managing connections has commonly entailed both gifts and shared rituals. Participating in common rituals signals wider group membership, and intensifies the emotional content of the membership. Coming together for the rituals enables people to exchange information (including, even mostly, gossip), re-invigorate existing connections and possibly make new ones. What, in the modern world, we call networking.
About exchange
Exchange emerges out of creating mutual benefit from nonviolent interaction with those we lack connections with. This may also be ritualised so as to align expectations about what frame of action applies. Rituals in general get around the talk is cheap, so a very weak signal due to lack of required commitment: the cheap talk problem. But rituals can also operate when there is a lack of a common language — the no talk problem.
Gifts can be part of an exchange ritual. Or a bid to establish connection. Or both.
If you are interacting with someone with whom you do not have a connection, and are unlikely to establish a connection, there can be no expectation of ongoing reciprocation, for there is no interaction across time of a form that can be expected to involve such reciprocation.
So, the reciprocation of giving in order to receive, receiving in order to give, has to be done on the spot as a self-contained thing, hence exchange. Each participant in the bargaining process that leads to exchange swaps something they want less for something they want more. Thus occurs gains from trade. Both bargainers benefit, both are likely to want to be able to do it again, so the expectations of exchange are adhered to.
Thus, exchange is typically something done with ‘strangers’. Those you do not have a specific connection to, so do not have expectations of reciprocation that follow from, and are part of, such connection. But that simply makes exchange a specific form and tool of interaction. One, moreover, that can be scaled up, in a way that connections generally cannot (given the time and effort that connection requires), as societies get larger and more complex.
Exchange scales up as it an information-economising form of interaction. The across-time reciprocations of connection are a much more information-dense form of interaction. Hence, as societies get larger and more complex, the information-economising form of interaction of exchange becomes much more common and tends to involve larger and more complex goods (and services). That is, exchange will scale-up both in social ambit (how often, and with how many, people you do it) and scale of resources (how much resources will be transferred across a given period of time and how large individual transfers can be).
It is also worth noting that exchange is profoundly normative. The key element, as with all systems of property rights, is not “mine!”, any chest-thumping ape can do that, the key element is “yours!”. Hence exchange is bargaining over an swap of rights. I transfer to you my ownership of/rights over this in exchange for you transferring to me your ownership of/rights over that. Homo sapiens is the only species that regularly engages in one-off acts of reciprocation with strangers of the same species due to us being a normative and communicative species. We trade because we are a pervasively normative species.
About money
Money evolves out of the information-economising nature of exchange, as money economises on information even more than exchange already does. A shared medium of exchange for transactions means you do not have to know the specific menu of goods or services the person you are interacting wants in order to exchange with them. Hence the development of transaction goods — goods you hold solely for the purpose of transacting, as people will accept them as they can also use them for transacting. Once prices are typically quoted in terms of a single transaction good, then you have full-service money.
Money, as an information-economising mechanism, greatly lowers transaction costs. It makes exchange easier, so scales exchange up even further.
As money is typically used across time periods, it is an asset. A transaction good that was “used up” in the process of transacting would not be much use. Hence the standard obligation that issuers of banknotes accept to replace damaged notes or coins so as to protect their value as transaction goods. (Not having any liability or obligation other than such replacement is the great advantage of fiat currency.) But money is an asset because, as a transaction good, it is used across time periods. Being an asset is not money’s essential feature, however, it is a consequence of its essential features. We can tell this as, during periods of hyperinflation, when money is, by orders of magnitude, the worst asset to hold, people still use money, because of money’s information-economising properties. They simply unload it as quickly as possible.
I don’t much like the unit of account, medium of exchange, store of value description of money as it mystifies money unnecessarily. Defining money as a good held for the purpose of transacting that prices are quoted in (either directly or through accepted rates of exchange), and so is used across time periods, fits into ordinary commercial patterns and economic discourse rather better. The unit-of-account and store-of-value characteristics flow from money’s use as a transaction good. But money is not just a medium of exchange, as it is also used to quote prices and obligations in.
People sometimes talk of odd cases where something is only used as a unit of account. That, however, only works if there are agreed rates of exchange with things that are exchanged. Something used purely as a unit of account is not really money because it is not a transaction good, more a transaction aid. Economising on information in exchange is the crucial, transaction good, element of money, with the addition of the unit-of-account function turning it into full-service money.
The information-economising transaction-good approach to money also fits better with use of commodities as so-called ‘primitive money’. Any convergence of empirical expectations around a particular transferable commodity for what people will accept in exchange can turn that commodity into a transaction good. Hence the vast array of commodities that have been used as transaction goods. I distilled the following list from A. Hingston Quiggin’s A Survey of Primitive Money. They have all been used as ‘primitive money’ (i.e. transaction goods):
Notes and coins: branded moneyNotes and coins are branded money. If the branding is what gives them their transaction utility, the range and readiness of people willing to accept them as transaction goods, then they are just tokens. (The operation of supply and demand in their use as transaction goods gives them their exchange value.) If coins are branded bullion, then that opens up the possibility of a gap between their nominal, branded, value and their commodity value.
Notes can be claims on bullion, as in a silver standard or a gold standard, but that is a complexity we need not deal with here.
If the branding authority has sufficient incentives to be concerned with the branded value of coins outside their jurisdiction, they are likely to keep the commodity content of the coins stable. Indeed, if the coins establish a reputation for reliability as a bullion brand due to the stability of their bullion content, then the readiness to accept can give them a premium above their bullion value because of their ease of acceptance. That is, the brand can increase the information-economising nature of the money, and so its transaction utility, and thus its exchange value.
The less the exchange value of the branding of the coin, the more the potential incentive there is to melt the bullion content down for other uses. But any branding is likely to provide some extra transaction utility, and so increase the value of the coin above its bullion value.
The less the branding authority has reason to be concerned with the branded value of coins outside their jurisdiction, the greater the temptation to attempt to profit from creating (more of) a wedge between nominal value and commodity value. Debasement reduces the bullion content without changing the nominal branding. To the extent that the branding authority can force acceptance at the nominal value, and that there is a lag in perceiving the change in commodity value and/or circulating supply, it can profit from debasement.
Private operators can do something similar, typically by ‘clipping’ the coin, shaving off some of the bullion. Hence the ridges or milled edges on modern coins, introduced into English coins by Sir Isaac Newton when he was Warden of the Mint.
The more effort is required to assess the bullion content, the higher the threshold at which people will go with the nominal value rather then going to the effort of assessing the bullion content. So, the lower the preceding value of the metal content of the coins, the more the direct incentive to debase but the less the likely profit per coin from doing so. Turning copper coins into tokens is relatively easy, while official debasing of silver coins is more likely than gold coins.
The classic response to debasement, or at least to variable branding reliability, is Gresham’s law. This is the tendency for bad money (less bullion content) to drive out good money (more bullion content) as people economise on bullion in response to debasement by the branding authority, or debasement by others. Both official debasement and private debasements are also cases of economising on bullion.
Debasement and Gresham’s law, as forms of economising on bullion, both rely on the condition that:
Economising on information is a pervasive element in the use and dynamics of money. Attention and cognitive capacity are both scarce items, so economising on them has value.
The point about the cognitive effort and attention threshold for worrying about nominal value versus commodity value can also apply to token money across time. There is considerable economising-on-cognitive-effort value in being able to simply use the nominal value of money in monetary calculations without any adjusting calculations. If, however, the money is changing in exchange value over time, then there is an incentive to adjust the nominal value for setting and judging prices. At some point (depending on rate of change in exchange value and the scale of exchange and the ease of doing the calculations), it is worth doing those adjustment calculations. Below that point, it is not.
Ignoring changes that are not big enough to trip the cognitive attention-and-effort threshold is called by economists ‘money illusion’. As if attention and cognitive capacity are unconstrained, so there is no cost in their use.
The notion that a rational being does not economise on attention and use of cognitive capacity, as if these are not scarce items, is a pretty remarkable concept of rationality.
Gifts
Going back to the evolution of human societies, there is no reason to presume a gifts-first, exchange-later pattern of social development. As soon as people seek to have nonviolent but beneficial interactions with people they are not connected to, exchange can be expected to develop. Exchange evolves where connections do not reach.
Gifts will, however, be very important in connection-dense societies, for the more connection-dense a society, the less need for exchange. Moreover, gifts are rather more information-dense than ordinary exchange. Indeed, a key element of a gift is acknowledgement — I see you, this is why I am giving you this specific thing appropriate to you. The more well-targeted the gift, the stronger the effect.
Gifts both rely on, and reinforce, the personalised information flows of connection. They are very much an intended statement of connection. There is more than echo of this in that, in common law, the intent to give is a necessary element for a legally recognised transfer.
The differences between gift and exchange is why I am not entirely comfortable with Mauss treating gifts as a form of exchange. I prefer to see gifts as connection-management acts of personalised acknowledgement. Hence gifts are used to create flows of resources in information-dense social environments.
Exchange does not really seem to be what is going on in such transfers in such societies. So-called gift exchange seems to be better described as networks of reciprocation. Especially as, as is set out below, gifts operate in essentially a permanent state of imbalance, avoiding the very state of completion that is central to exchange. Moreover, it is much more fruitful to think of exchange as an information-economising mechanism while gifts are embedded in information flows. With money pushing the information-economising a step further, thereby extending the operation and density of exchange. If money itself is given as a gift, there is, unlike exchange, no specific, often immediate, matching reciprocation.
But what about Mauss’s analysis of the obligations to give, to receive and to reciprocate? What about how gifts tended to dominate resource transfers in archaic societies?
Taking the resource-transfer point first, for gifts to dominate resource transfers (outside the areas of life dominated by pooling of resources) just tells us that such societies are information dense and dominated by connection. In a sufficiently information-dense social environment, gifts will “work” for the reasons Mauss notes. Indeed, a problem with his analysis is that he is so impressed with the social scale of the gifting networks he identifies, he does not think quite hard enough about the boundaries of those gifting networks and how far out those boundaries can reach and why.
This failure is striking given that Mauss does draw attention to a boundary problem that persistently turns up in myths, legends and folk tales: the boundary of the failure to invite. Which is to say, the failure to properly manage connection. This is the classic case when the cursed gift occurs. Think of the uninvited evil godmother in the Sleeping Beauty story. If you fail to include someone you should have, if you fail to acknowledge somebody you should have acknowledged, bad things will happen.
Of course they will, because your society as a whole, and your position in particular, relies on effective management of connection.
As for the obligations to give, receive and reciprocate, connections were managed by gifting networks. You give to show the connection is important to you. You give to the correct people in the correct way to show that you know your place, and your obligations in from being in that place. You are obliged to receive so that other people can manage their connections with you correctly. You are obliged to reciprocate to show that you also value the connection, that you are actively committed to it.
A description of the Japanese art of giving, ochugen, expresses this nicely:
One reason we can manage so many connections is that we are good at reading cues. Hence people who are bad at reading cues are disadvantaged in social interactions. But those cues are a better, less information burdensome guide, the stronger the framework of expectations is. That is, if there are agreed schemas of belief that generate agreed scripts of action.
The advantage of using structures and discourses to align expectations (both expectations of what people will do, empirical expectations, and expectations of what people expect you should do, normative expectations) generates the discourses around gifts and giving that Mauss is so impressed by. An evolving normative framework, set out and reinforced in myths, legends and gossip, helps people align their expectations. Hence the development of obligations to give, receive and reciprocate. You have to play the connections game in such societies and the structure of obligations are the “currency” within which you play it.
Connections operate across time, this is a fundamental feature of them. So, as Mauss indicates, inherently involve the notion of credit (and its converse, debt).
Yes, resources go in both directions, as having both parties benefit from the connection is required for it to continue and to be stable enough to continue. But it is better to think of gifts as a flow of resources rather than an exchange of resources. A flow of resources that both manages connections and provides much of the reason to have connections. A flow that, moreover alternates between patterns of credit and debt, so is almost always in a state of imbalance, not balance.
Mauss notes that reciprocation, being a matter of credit, can involve a notion of interest, and at quite high rates of interest. That is, as time passes, the level of expected reciprocation comes to exceed what was originally received. Such putative rates of interest can be, Mauss tells us, in the order of 30–100 per cent per annum.
If giving is an act of connection management, it entails a notion of credit. Having received, you have also show that the connection matters to you and that you are a person worthy of your status, who understands what is expected and can meet those expectations. The longer it takes you to reciprocate, the more the giver’s credit, and your debt as receiver, mounts. The effective interest rate becomes a measure of the period of imbalance in obligation and of commitment to the connection. The rolling pattern of imbalance, of shifting credit and debit, both manifests the connection and is a an expression of commitment to the connection across time.
You cannot reciprocate immediately. That stops it being a gift and turns into an exchange, an act without the information burden of connection and that robs the giver of the ability to express their commitment to the connection. But you have to reciprocate eventually, otherwise you are denying the importance of the connection. So, you acknowledge the period of imbalance and continue the binding cycle of imbalance by giving back with interest, by giving back greater than you received, thereby keeping the cycle of imbalance, and commitment to the continuation of the connection, going. Thus does each party move back and forth from credit to debt and signal their commitment to the connection. I put you in debt so you can put me in debt in return. I achieve a credit, so you can achieve a credit in turn. If we reached a state of balance, then neither of us has an outstanding resource commitment to the connection, which is thus put at risk.
Again, this does not really look like exchange, it looks like connection management via fluctuating, and deliberately alternatively unbalanced, flow of resources. A lack of balance alternating between the parties via resource flow, where the rhythm of imbalance is the key.
In acts of exchange, by contrast, the interaction is self-contained; that is the point. We both walk away from the act of exchange feeling like winners because there is no continuing connection that requires an ongoing resource commitment from the exchange. That is the, information-economising, point.
In very hierarchical societies, giving, and giving-up, can be an act of prestige, even dominance. Hence the deliberate destruction of goods in the potlatch. This is, in some ways, a manifestation of a society too driven by connection, too interconnected, and not enough open to exchange. Everyone knows too much about others, is connected too thoroughly with others, is too concerned by status embedded in connections, for exchange to operate sufficiently as a path for gaining value from resources. (Or, possibly, the lack of accepted opportunities for exchange is forcing everything into connections.)
Connection can be so dominant, that giving, and giving-up, can be an act not merely of prestige, but of dominance. Not merely of status granted by others (prestige) but an act of dominance over others. The structuring rules of connections require reciprocation, but if I am so giving-dominant that I give in a way you cannot match, then your loss of status can be so great, I can have a credit with you (that implies a debt to me) so great, that you are forced into bondage. Again, exchange is not the most useful way to think of what is happening here. Overwhelming someone with the flow of resources, so as to bankrupt their capacity to be independent, seems a better way to think of such dominance plays.
Particularly as it is very much not a gains-from-trade situation. It is much more of a winning-the-game situation.
Created within the dynamics of connection, the notion of credit and debt can be adapted to people with whom one does not have a connection, or not a sufficient connection. Thus collateral for a loan is what you do if there is no connection between creditor and debtor, or the connection is not strong enough to carry the risk involved in the loan.
Some exchanges create connections. Contracts create connections because they operate across time as the specified exchange cannot happen all at once. Contracts thus have a managing-the-connection element, especially as no contract can be fully specified to cover all circumstances. The trick is to structure the contract enough, and well enough, that connection management within the contract can handle the rest due to expectations and incentives aligning enough to do so. Hence the law sometimes treats contracts as simply exchanges that can be “paid off” with appropriate value and sometimes as a promise that has to be fulfilled. With a major factor to be considered being whether it is reasonable to force a continuing connection between parties whose interaction has broken down.
Employment contracts create particularly complex connections. A major reason why wages are “sticky” is that cutting wages obviously risks poisoning an employer’s connections with their employees. It is usually going to be much more in an employer’s interest to end a series of employment connections (i.e. sack people) rather than poison continuing connections with ongoing employees, creating a much more fraught series of connection-management problems. Especially as sackings can warn remaining employees that things are serious.
In both these cases, connections are operating across time and within, in effect, the “gaps” in exchange. Time is, of course, a key element in connection: whatever else connections are, they are things that operate across time periods. Contracts are for exchanges that require connection as they operate across time periods. Loans also create connections across time, though they can otherwise be completely self-contained.
Later in his essay, Mauss tries to discern elements of gift exchange lingering on in legal systems. The effort comes across as a bit strained, in part because his discussion is somewhat under-specified. Moreover, his attempt to connect gift exchange to solidarity in contemporary societies through state action is just sad. Modern states do not, and cannot, remotely operate according to the patterns he discerns in archaic societies. States are not personal connection devices, no matter how some attempt to romanticise them in that way. (This is not to deny that connections can be very important in understanding the operation of the state apparatus, but that is connections between individuals within state, and with outside actors, not the state itself.)
If, however, we see gifts as flows of resources expressing and manifesting the management of connections, then the later echoes in legal systems are clearer. But those later echoes rest in connections, and their management, far more than they do in the shifting balances of credit and debit, via the resource flows of gifts, in information- and connection-dense societies that Mauss memorably grapples with but does not really get to the heart of.
Marriage
Marriage provides a fascinating nexus between the resource modes of pooling, connection and exchange. The classic human pooling is men and women bringing different skills and abilities to the task of raising children via marriage. Yet marriage can often be understood as an exchange. In patrilineal kin group systems, kin groups trade out the fertility of their daughters. In matrilineal kin group systems, kin groups trade out the protect-and-provide capacities of their sons.
The protect-and-provide for sex-and-fertility exchange (plus domestic preparation services such as cooking or weaving) has generally been at the base of marriage systems, though with considerable variability in the expected level of the wife’s contribution to subsistence. The expectation of wifely contribution to subsistence was generally high in foraging and hoe-farming societies. It was usually lower, possibly much lower, in plough-farming and pastoralist societies. This typically weakened women’s social leverage in such societies. Hence plough-farming and pastoralist societies have tended to be distinctly more patriarchal than foraging and hoe-farming societies.
The exchange aspect in marriages could be very explicit in bride price (paid to bride’s family), dowry (paid to the groom), dower (paid to the bride). Bride price and dower pay for/acknowledge the fertility of the bride, dowry pays for/acknowledges the status of the groom. They all register relative scarcity.
Bride price and dowers are typical features of polygynous societies. These perennially have a shortage of prospective brides compared to grooms. For a women getting married took her, but not her husband, out of the marriage market, as he could take another wife but she could not take another husband. If the marriage is viewed as a contract, then the sense of an exchange of services could extend to notions of payment by the husband for the wife’s sexual services.
Bride wealth signals that the groom has the assets and connections to sustain the role of husband and father. For the other thing marriage does is create connections. The only feature common to all marriages across all human societies that anthropologists have been able to find is that marriage creates in-laws, it creates a kin connection. Bride price, dower and dowry all provide a way of creating a balance so as to enable the connection to be worth investing in by both families or kin groups. The value of the (continuing) benefit of the bride’s fertility (or the husband’s status) is acknowledged by the other party so that the marriage, and the connections it creates, and the pooling it is supposed to entail, can take place from a place of balance. A good match does all that.
Even though they are about allowing a connection to be established, bride price, dowry and dowers are exchanges for something rather than gifts. They are exchanges (resources for fertility or status) to create sufficient balance for the connection to be established and sustained. If the social gap between the parties is too large, then either the connection-via-marriage will not happen or the gap will be a point of continuing strain.
Although dowries are illegal in India, the continuing status of higher caste, particularly Brahmin, husbands, and the expansion, due to spreading prosperity, of the range of families who aspire to a higher caste husband for their daughters has not only meant dowries have persisted, it has led to considerable and continuing rise in the cost of dowries. Rather grimly, this seems to have led to a particular pattern of wife-beating so that the husband can extort a dowry “top up” from the bride’s family to match the rise in the price of dowries since the marriage.
I had forgotten how frustrating I had found Mauss’s essay to be. To be sure, having recently done a lot of reading into the origins of Homo sapiens and the operation of foraging societies, I can draw on a much broader information base than he could.
As an aside, if you haven’t discovered the online lectures from CARTA (the Center for Academic Research and Training into Anthropogeny), then I highly recommend them. A series of individual, or bundled, short lectures by scholars to audiences significantly of other scholars on the latest research into the evolution of Homo sapiens is an intellectual feast. An online gift to the global community of the curious.
Back to The Gift. Mauss manages to be a bit too general at times and not general enough at other times. So, there is an insufficiency of precision in his discussions — what he means in a concrete sense is often not as clear as it should be. But he also wants to make grand claims about stages of social development that do not really fit with the evidence.
For example, he writes:
The victory of rationalism and mercantilism was needed before the notions of profit and the individual, raised to the level of principles, were introduced. P.76. (Norton paperback edition)Well, thousands of cuneiform tablets from Mesopotamia say that’s not the case.
Taking a step back, the evidence for long-term exchange among Homo sapiens is about as old as the evidence for the first Homo sapiens. We are the trading ape. So, the evidence is that Adam Smith was mostly correct when he wrote in The Wealth of Nations (Book 1, Chapter 2) that:
This division of labour, from which so many advantages are derived, is not originally the effect of any human wisdom, which foresees and intends that general opulence to which it gives occasion. It is the necessary, though very slow and gradual consequence of a certain propensity in human nature which has in view no such extensive utility; the propensity to truck, barter, and exchange one thing for another.Smith is clearly correct on the importance of the division of labour, though not on its origins, which predate exchange. The ways in which we Homo sapiens are a dimorphic species makes sense on the basis of a division of labour between males and females that we see in all foraging societies. (Men do the riskier foraging; the hunting, and such gathering, as is not compatible with minding the kids. Women do the safer foraging; the gathering, and such hunting, as is compatible with minding the kids.) Smith is also correct in identifying among humans a “propensity to truck, barter and exchange”.
Mauss is, however, a little too impressed with gifts as acts of exchange. I want to suggest a somewhat different take.
Pooling, connection and …
The most intimate way that Homo sapiens deal with resources is pooling, whether within the family or within the hunting band. We are a group-living species who, most likely, evolved from group-living species. All our nearest relatives are group-living species (gorillas, bonobos, chimpanzees). It is extremely unlikely that our ancestors in the Homo line, and the various common ancestors with the African great apes before that, were not also group-living species.
It is striking, however, that the four surviving African great ape species have quite different mating patterns. Gorillas are a harem species where an adult male, about twice the size of adult females, dominates a group of females. Bonobos are a group-mating species based on coalitions built around an alpha female and alpha male (often the alpha female’s son). Chimpanzees are a group-mating species based on strictly male-only, female-dominating coalitions led by an alpha male.
We are a mildly polygynous pair-bonding species whose foraging cultures tend to strongly resist attempts to establish any form of dominance by any group member. Dominance only (re)emerges among humans either when food production permits storage of food or there are variation in productive assets. Dominance behaviour can get quite extreme, however, when there is both storage of food and variation in productive assets.
Homo sapien foragers engage in sharing, in pooling, behaviour for food (particularly from hunting) far more than do our African great ape relatives.
Use of fire can also be a pooling behaviour. Indeed, the answer to ‘what is the Homo sapien diet?’ is, cooked food. We are the cooking ape.
As the evidence for systematic use of fire in our evolutionary line is older than evidence for Homo sapiens, we pooled resources and managed connections before we engaged in exchange. Indeed, it is likely that it the development of those capacities that led us to becoming the trading ape.
But Homo sapiens had connections beyond the hunting band. (This is true in most foraging societies and the few it was not true of, you really didn’t want to live in.) Human foragers also might interact with people from groups that they had no connections with. With the former, the connections need to be managed. With the latter, any non-violent interaction has to be able to operate without being sustained by connections across the groups.
By connection I simply mean: repeated, mutually acknowledged, interactions that both parties presumptively mean to continue.
Across a wide variety of human societies, managing connections has commonly entailed both gifts and shared rituals. Participating in common rituals signals wider group membership, and intensifies the emotional content of the membership. Coming together for the rituals enables people to exchange information (including, even mostly, gossip), re-invigorate existing connections and possibly make new ones. What, in the modern world, we call networking.
About exchange
Exchange emerges out of creating mutual benefit from nonviolent interaction with those we lack connections with. This may also be ritualised so as to align expectations about what frame of action applies. Rituals in general get around the talk is cheap, so a very weak signal due to lack of required commitment: the cheap talk problem. But rituals can also operate when there is a lack of a common language — the no talk problem.
Gifts can be part of an exchange ritual. Or a bid to establish connection. Or both.
If you are interacting with someone with whom you do not have a connection, and are unlikely to establish a connection, there can be no expectation of ongoing reciprocation, for there is no interaction across time of a form that can be expected to involve such reciprocation.
So, the reciprocation of giving in order to receive, receiving in order to give, has to be done on the spot as a self-contained thing, hence exchange. Each participant in the bargaining process that leads to exchange swaps something they want less for something they want more. Thus occurs gains from trade. Both bargainers benefit, both are likely to want to be able to do it again, so the expectations of exchange are adhered to.
Thus, exchange is typically something done with ‘strangers’. Those you do not have a specific connection to, so do not have expectations of reciprocation that follow from, and are part of, such connection. But that simply makes exchange a specific form and tool of interaction. One, moreover, that can be scaled up, in a way that connections generally cannot (given the time and effort that connection requires), as societies get larger and more complex.
Exchange scales up as it an information-economising form of interaction. The across-time reciprocations of connection are a much more information-dense form of interaction. Hence, as societies get larger and more complex, the information-economising form of interaction of exchange becomes much more common and tends to involve larger and more complex goods (and services). That is, exchange will scale-up both in social ambit (how often, and with how many, people you do it) and scale of resources (how much resources will be transferred across a given period of time and how large individual transfers can be).
It is also worth noting that exchange is profoundly normative. The key element, as with all systems of property rights, is not “mine!”, any chest-thumping ape can do that, the key element is “yours!”. Hence exchange is bargaining over an swap of rights. I transfer to you my ownership of/rights over this in exchange for you transferring to me your ownership of/rights over that. Homo sapiens is the only species that regularly engages in one-off acts of reciprocation with strangers of the same species due to us being a normative and communicative species. We trade because we are a pervasively normative species.
About money
Money evolves out of the information-economising nature of exchange, as money economises on information even more than exchange already does. A shared medium of exchange for transactions means you do not have to know the specific menu of goods or services the person you are interacting wants in order to exchange with them. Hence the development of transaction goods — goods you hold solely for the purpose of transacting, as people will accept them as they can also use them for transacting. Once prices are typically quoted in terms of a single transaction good, then you have full-service money.
Money, as an information-economising mechanism, greatly lowers transaction costs. It makes exchange easier, so scales exchange up even further.
As money is typically used across time periods, it is an asset. A transaction good that was “used up” in the process of transacting would not be much use. Hence the standard obligation that issuers of banknotes accept to replace damaged notes or coins so as to protect their value as transaction goods. (Not having any liability or obligation other than such replacement is the great advantage of fiat currency.) But money is an asset because, as a transaction good, it is used across time periods. Being an asset is not money’s essential feature, however, it is a consequence of its essential features. We can tell this as, during periods of hyperinflation, when money is, by orders of magnitude, the worst asset to hold, people still use money, because of money’s information-economising properties. They simply unload it as quickly as possible.
I don’t much like the unit of account, medium of exchange, store of value description of money as it mystifies money unnecessarily. Defining money as a good held for the purpose of transacting that prices are quoted in (either directly or through accepted rates of exchange), and so is used across time periods, fits into ordinary commercial patterns and economic discourse rather better. The unit-of-account and store-of-value characteristics flow from money’s use as a transaction good. But money is not just a medium of exchange, as it is also used to quote prices and obligations in.
People sometimes talk of odd cases where something is only used as a unit of account. That, however, only works if there are agreed rates of exchange with things that are exchanged. Something used purely as a unit of account is not really money because it is not a transaction good, more a transaction aid. Economising on information in exchange is the crucial, transaction good, element of money, with the addition of the unit-of-account function turning it into full-service money.
The information-economising transaction-good approach to money also fits better with use of commodities as so-called ‘primitive money’. Any convergence of empirical expectations around a particular transferable commodity for what people will accept in exchange can turn that commodity into a transaction good. Hence the vast array of commodities that have been used as transaction goods. I distilled the following list from A. Hingston Quiggin’s A Survey of Primitive Money. They have all been used as ‘primitive money’ (i.e. transaction goods):
Beads (whether made of stone, shell, glass or whatever), beeswax, buffaloes, camphor, cattle, cloth (from silk to wadmal, a coarse wool fabric), coca leaves, cocoa beans, coconuts, strings of coconut discs, dye cakes, feather coils, gold dust, weighted gold, grain (notably barley and rice), human heads, logwood (mahogany), tool metals (iron, copper, tin, bronze) in various shapes, plaited palm-fibre rings, pigs, porcelain jars, balls of rubber, salt (including stamped salt cakes), seeds, shells (especially cowries), silver in lumps or shaped, animal skins, slaves, carved stone (a tool material), tea (including in bricks), teeth, tobacco.When it comes to economising on information in exchange, it seems to be a case of, whatever works.
Notes and coins: branded moneyNotes and coins are branded money. If the branding is what gives them their transaction utility, the range and readiness of people willing to accept them as transaction goods, then they are just tokens. (The operation of supply and demand in their use as transaction goods gives them their exchange value.) If coins are branded bullion, then that opens up the possibility of a gap between their nominal, branded, value and their commodity value.
Notes can be claims on bullion, as in a silver standard or a gold standard, but that is a complexity we need not deal with here.
If the branding authority has sufficient incentives to be concerned with the branded value of coins outside their jurisdiction, they are likely to keep the commodity content of the coins stable. Indeed, if the coins establish a reputation for reliability as a bullion brand due to the stability of their bullion content, then the readiness to accept can give them a premium above their bullion value because of their ease of acceptance. That is, the brand can increase the information-economising nature of the money, and so its transaction utility, and thus its exchange value.
The less the exchange value of the branding of the coin, the more the potential incentive there is to melt the bullion content down for other uses. But any branding is likely to provide some extra transaction utility, and so increase the value of the coin above its bullion value.
The less the branding authority has reason to be concerned with the branded value of coins outside their jurisdiction, the greater the temptation to attempt to profit from creating (more of) a wedge between nominal value and commodity value. Debasement reduces the bullion content without changing the nominal branding. To the extent that the branding authority can force acceptance at the nominal value, and that there is a lag in perceiving the change in commodity value and/or circulating supply, it can profit from debasement.
Private operators can do something similar, typically by ‘clipping’ the coin, shaving off some of the bullion. Hence the ridges or milled edges on modern coins, introduced into English coins by Sir Isaac Newton when he was Warden of the Mint.
The more effort is required to assess the bullion content, the higher the threshold at which people will go with the nominal value rather then going to the effort of assessing the bullion content. So, the lower the preceding value of the metal content of the coins, the more the direct incentive to debase but the less the likely profit per coin from doing so. Turning copper coins into tokens is relatively easy, while official debasing of silver coins is more likely than gold coins.
The classic response to debasement, or at least to variable branding reliability, is Gresham’s law. This is the tendency for bad money (less bullion content) to drive out good money (more bullion content) as people economise on bullion in response to debasement by the branding authority, or debasement by others. Both official debasement and private debasements are also cases of economising on bullion.
Debasement and Gresham’s law, as forms of economising on bullion, both rely on the condition that:
there are two (or more) types of money which are of equivalent value for some purposes and of different value for others.For ordinary users of coins, facing variable bullion content in coins of the same nominal value, there was a trade-off between economising on bullion and economising on attention and cognitive effort.
Economising on information is a pervasive element in the use and dynamics of money. Attention and cognitive capacity are both scarce items, so economising on them has value.
The point about the cognitive effort and attention threshold for worrying about nominal value versus commodity value can also apply to token money across time. There is considerable economising-on-cognitive-effort value in being able to simply use the nominal value of money in monetary calculations without any adjusting calculations. If, however, the money is changing in exchange value over time, then there is an incentive to adjust the nominal value for setting and judging prices. At some point (depending on rate of change in exchange value and the scale of exchange and the ease of doing the calculations), it is worth doing those adjustment calculations. Below that point, it is not.
Ignoring changes that are not big enough to trip the cognitive attention-and-effort threshold is called by economists ‘money illusion’. As if attention and cognitive capacity are unconstrained, so there is no cost in their use.
The notion that a rational being does not economise on attention and use of cognitive capacity, as if these are not scarce items, is a pretty remarkable concept of rationality.
Gifts
Going back to the evolution of human societies, there is no reason to presume a gifts-first, exchange-later pattern of social development. As soon as people seek to have nonviolent but beneficial interactions with people they are not connected to, exchange can be expected to develop. Exchange evolves where connections do not reach.
Gifts will, however, be very important in connection-dense societies, for the more connection-dense a society, the less need for exchange. Moreover, gifts are rather more information-dense than ordinary exchange. Indeed, a key element of a gift is acknowledgement — I see you, this is why I am giving you this specific thing appropriate to you. The more well-targeted the gift, the stronger the effect.
Gifts both rely on, and reinforce, the personalised information flows of connection. They are very much an intended statement of connection. There is more than echo of this in that, in common law, the intent to give is a necessary element for a legally recognised transfer.
The differences between gift and exchange is why I am not entirely comfortable with Mauss treating gifts as a form of exchange. I prefer to see gifts as connection-management acts of personalised acknowledgement. Hence gifts are used to create flows of resources in information-dense social environments.
Exchange does not really seem to be what is going on in such transfers in such societies. So-called gift exchange seems to be better described as networks of reciprocation. Especially as, as is set out below, gifts operate in essentially a permanent state of imbalance, avoiding the very state of completion that is central to exchange. Moreover, it is much more fruitful to think of exchange as an information-economising mechanism while gifts are embedded in information flows. With money pushing the information-economising a step further, thereby extending the operation and density of exchange. If money itself is given as a gift, there is, unlike exchange, no specific, often immediate, matching reciprocation.
But what about Mauss’s analysis of the obligations to give, to receive and to reciprocate? What about how gifts tended to dominate resource transfers in archaic societies?
Taking the resource-transfer point first, for gifts to dominate resource transfers (outside the areas of life dominated by pooling of resources) just tells us that such societies are information dense and dominated by connection. In a sufficiently information-dense social environment, gifts will “work” for the reasons Mauss notes. Indeed, a problem with his analysis is that he is so impressed with the social scale of the gifting networks he identifies, he does not think quite hard enough about the boundaries of those gifting networks and how far out those boundaries can reach and why.
This failure is striking given that Mauss does draw attention to a boundary problem that persistently turns up in myths, legends and folk tales: the boundary of the failure to invite. Which is to say, the failure to properly manage connection. This is the classic case when the cursed gift occurs. Think of the uninvited evil godmother in the Sleeping Beauty story. If you fail to include someone you should have, if you fail to acknowledge somebody you should have acknowledged, bad things will happen.
Of course they will, because your society as a whole, and your position in particular, relies on effective management of connection.
As for the obligations to give, receive and reciprocate, connections were managed by gifting networks. You give to show the connection is important to you. You give to the correct people in the correct way to show that you know your place, and your obligations in from being in that place. You are obliged to receive so that other people can manage their connections with you correctly. You are obliged to reciprocate to show that you also value the connection, that you are actively committed to it.
A description of the Japanese art of giving, ochugen, expresses this nicely:
Gift-giving is more than just buying gifts for people on birthdays and special holidays, it is a way to show gratitude to those you are indebted to and to show appreciation for those you care about and/or respect. Gifts are brought to houses that are being visited whether its just for a short visit, or if its an overseas visit. Parents will bring snacks or treats over to houses where their children would be visiting for playdates, families would bring omiyage (gifts) over to family member’s houses that they have not seen in awhile, and over to people’s houses in which they would be residing. Gift-giving serves an important function in Japanese culture as not only a way to show appreciation, but as a way to strengthen and maintain relationships, and in some cases, show closeness or fondness for another.The more structured connections, the more empirical and normative expectations are aligned, the more stable connections will be. Moreover, the more structured connections are, the more expectations are aligned, the less of an information burden connections are. If one can rely on accepted patterns of behaviour, there is a lot less search and other information burdens involved.
One reason we can manage so many connections is that we are good at reading cues. Hence people who are bad at reading cues are disadvantaged in social interactions. But those cues are a better, less information burdensome guide, the stronger the framework of expectations is. That is, if there are agreed schemas of belief that generate agreed scripts of action.
The advantage of using structures and discourses to align expectations (both expectations of what people will do, empirical expectations, and expectations of what people expect you should do, normative expectations) generates the discourses around gifts and giving that Mauss is so impressed by. An evolving normative framework, set out and reinforced in myths, legends and gossip, helps people align their expectations. Hence the development of obligations to give, receive and reciprocate. You have to play the connections game in such societies and the structure of obligations are the “currency” within which you play it.
Connections operate across time, this is a fundamental feature of them. So, as Mauss indicates, inherently involve the notion of credit (and its converse, debt).
Yes, resources go in both directions, as having both parties benefit from the connection is required for it to continue and to be stable enough to continue. But it is better to think of gifts as a flow of resources rather than an exchange of resources. A flow of resources that both manages connections and provides much of the reason to have connections. A flow that, moreover alternates between patterns of credit and debt, so is almost always in a state of imbalance, not balance.
Mauss notes that reciprocation, being a matter of credit, can involve a notion of interest, and at quite high rates of interest. That is, as time passes, the level of expected reciprocation comes to exceed what was originally received. Such putative rates of interest can be, Mauss tells us, in the order of 30–100 per cent per annum.
If giving is an act of connection management, it entails a notion of credit. Having received, you have also show that the connection matters to you and that you are a person worthy of your status, who understands what is expected and can meet those expectations. The longer it takes you to reciprocate, the more the giver’s credit, and your debt as receiver, mounts. The effective interest rate becomes a measure of the period of imbalance in obligation and of commitment to the connection. The rolling pattern of imbalance, of shifting credit and debit, both manifests the connection and is a an expression of commitment to the connection across time.
You cannot reciprocate immediately. That stops it being a gift and turns into an exchange, an act without the information burden of connection and that robs the giver of the ability to express their commitment to the connection. But you have to reciprocate eventually, otherwise you are denying the importance of the connection. So, you acknowledge the period of imbalance and continue the binding cycle of imbalance by giving back with interest, by giving back greater than you received, thereby keeping the cycle of imbalance, and commitment to the continuation of the connection, going. Thus does each party move back and forth from credit to debt and signal their commitment to the connection. I put you in debt so you can put me in debt in return. I achieve a credit, so you can achieve a credit in turn. If we reached a state of balance, then neither of us has an outstanding resource commitment to the connection, which is thus put at risk.
Again, this does not really look like exchange, it looks like connection management via fluctuating, and deliberately alternatively unbalanced, flow of resources. A lack of balance alternating between the parties via resource flow, where the rhythm of imbalance is the key.
In acts of exchange, by contrast, the interaction is self-contained; that is the point. We both walk away from the act of exchange feeling like winners because there is no continuing connection that requires an ongoing resource commitment from the exchange. That is the, information-economising, point.
In very hierarchical societies, giving, and giving-up, can be an act of prestige, even dominance. Hence the deliberate destruction of goods in the potlatch. This is, in some ways, a manifestation of a society too driven by connection, too interconnected, and not enough open to exchange. Everyone knows too much about others, is connected too thoroughly with others, is too concerned by status embedded in connections, for exchange to operate sufficiently as a path for gaining value from resources. (Or, possibly, the lack of accepted opportunities for exchange is forcing everything into connections.)
Connection can be so dominant, that giving, and giving-up, can be an act not merely of prestige, but of dominance. Not merely of status granted by others (prestige) but an act of dominance over others. The structuring rules of connections require reciprocation, but if I am so giving-dominant that I give in a way you cannot match, then your loss of status can be so great, I can have a credit with you (that implies a debt to me) so great, that you are forced into bondage. Again, exchange is not the most useful way to think of what is happening here. Overwhelming someone with the flow of resources, so as to bankrupt their capacity to be independent, seems a better way to think of such dominance plays.
Particularly as it is very much not a gains-from-trade situation. It is much more of a winning-the-game situation.
Created within the dynamics of connection, the notion of credit and debt can be adapted to people with whom one does not have a connection, or not a sufficient connection. Thus collateral for a loan is what you do if there is no connection between creditor and debtor, or the connection is not strong enough to carry the risk involved in the loan.
Some exchanges create connections. Contracts create connections because they operate across time as the specified exchange cannot happen all at once. Contracts thus have a managing-the-connection element, especially as no contract can be fully specified to cover all circumstances. The trick is to structure the contract enough, and well enough, that connection management within the contract can handle the rest due to expectations and incentives aligning enough to do so. Hence the law sometimes treats contracts as simply exchanges that can be “paid off” with appropriate value and sometimes as a promise that has to be fulfilled. With a major factor to be considered being whether it is reasonable to force a continuing connection between parties whose interaction has broken down.
Employment contracts create particularly complex connections. A major reason why wages are “sticky” is that cutting wages obviously risks poisoning an employer’s connections with their employees. It is usually going to be much more in an employer’s interest to end a series of employment connections (i.e. sack people) rather than poison continuing connections with ongoing employees, creating a much more fraught series of connection-management problems. Especially as sackings can warn remaining employees that things are serious.
In both these cases, connections are operating across time and within, in effect, the “gaps” in exchange. Time is, of course, a key element in connection: whatever else connections are, they are things that operate across time periods. Contracts are for exchanges that require connection as they operate across time periods. Loans also create connections across time, though they can otherwise be completely self-contained.
Later in his essay, Mauss tries to discern elements of gift exchange lingering on in legal systems. The effort comes across as a bit strained, in part because his discussion is somewhat under-specified. Moreover, his attempt to connect gift exchange to solidarity in contemporary societies through state action is just sad. Modern states do not, and cannot, remotely operate according to the patterns he discerns in archaic societies. States are not personal connection devices, no matter how some attempt to romanticise them in that way. (This is not to deny that connections can be very important in understanding the operation of the state apparatus, but that is connections between individuals within state, and with outside actors, not the state itself.)
If, however, we see gifts as flows of resources expressing and manifesting the management of connections, then the later echoes in legal systems are clearer. But those later echoes rest in connections, and their management, far more than they do in the shifting balances of credit and debit, via the resource flows of gifts, in information- and connection-dense societies that Mauss memorably grapples with but does not really get to the heart of.
Marriage
Marriage provides a fascinating nexus between the resource modes of pooling, connection and exchange. The classic human pooling is men and women bringing different skills and abilities to the task of raising children via marriage. Yet marriage can often be understood as an exchange. In patrilineal kin group systems, kin groups trade out the fertility of their daughters. In matrilineal kin group systems, kin groups trade out the protect-and-provide capacities of their sons.
The protect-and-provide for sex-and-fertility exchange (plus domestic preparation services such as cooking or weaving) has generally been at the base of marriage systems, though with considerable variability in the expected level of the wife’s contribution to subsistence. The expectation of wifely contribution to subsistence was generally high in foraging and hoe-farming societies. It was usually lower, possibly much lower, in plough-farming and pastoralist societies. This typically weakened women’s social leverage in such societies. Hence plough-farming and pastoralist societies have tended to be distinctly more patriarchal than foraging and hoe-farming societies.
The exchange aspect in marriages could be very explicit in bride price (paid to bride’s family), dowry (paid to the groom), dower (paid to the bride). Bride price and dower pay for/acknowledge the fertility of the bride, dowry pays for/acknowledges the status of the groom. They all register relative scarcity.
Bride price and dowers are typical features of polygynous societies. These perennially have a shortage of prospective brides compared to grooms. For a women getting married took her, but not her husband, out of the marriage market, as he could take another wife but she could not take another husband. If the marriage is viewed as a contract, then the sense of an exchange of services could extend to notions of payment by the husband for the wife’s sexual services.
Bride wealth signals that the groom has the assets and connections to sustain the role of husband and father. For the other thing marriage does is create connections. The only feature common to all marriages across all human societies that anthropologists have been able to find is that marriage creates in-laws, it creates a kin connection. Bride price, dower and dowry all provide a way of creating a balance so as to enable the connection to be worth investing in by both families or kin groups. The value of the (continuing) benefit of the bride’s fertility (or the husband’s status) is acknowledged by the other party so that the marriage, and the connections it creates, and the pooling it is supposed to entail, can take place from a place of balance. A good match does all that.
Even though they are about allowing a connection to be established, bride price, dowry and dowers are exchanges for something rather than gifts. They are exchanges (resources for fertility or status) to create sufficient balance for the connection to be established and sustained. If the social gap between the parties is too large, then either the connection-via-marriage will not happen or the gap will be a point of continuing strain.
Although dowries are illegal in India, the continuing status of higher caste, particularly Brahmin, husbands, and the expansion, due to spreading prosperity, of the range of families who aspire to a higher caste husband for their daughters has not only meant dowries have persisted, it has led to considerable and continuing rise in the cost of dowries. Rather grimly, this seems to have led to a particular pattern of wife-beating so that the husband can extort a dowry “top up” from the bride’s family to match the rise in the price of dowries since the marriage.
In summary
Gifts take place within, or to establish, connections through a flow of imbalance. This is very different from exchange, the point of which is completion, and thus exchange typically creates a balance of satisfaction.
Indeed, the Chinese state traditionally re-characterised trade as tribute precisely to avoid an implication of balance in status. Thus the Son of Heaven would receive tribute in horses and, in return, graciously provide gifts of silk and treasure. It was absolutely not, everyone was to understand, an exchange of silk for horses, nor treasure payments for peace on the border, buying off potential raiders. The Son of Heaven was therefore absolutely not a merchant nor the equal of any steppe khan or khagan. Hence, the Son of Heaven entered into tributary arrangements, not trade ones. The status claims that such tributary-trade exchanges were wrapped in did not, however, make them not exchange, nor the alleged gifts other than purchases and pay-offs.
Gifts are typically embedded in resource flows so as to create and sustain connections. They are marked by a flow of shifting resource imbalance in an information-dense environment to make connections work. They are not information-economising acts of completion that require no ongoing connection beyond the participants, so they are not acts of exchange.
So, no, gifts are not an archaic form of exchange, nor are they exchange within archaic societies.
This essay is part of the intellectual scaffolding for a book to be published by Connor Court looking at the social dynamics of marriage. (There might also be some suggestion that economists and anthropologists should talk more.) As this essay is something of a work in progress, it is subject to ongoing fiddling.
Gifts take place within, or to establish, connections through a flow of imbalance. This is very different from exchange, the point of which is completion, and thus exchange typically creates a balance of satisfaction.
Indeed, the Chinese state traditionally re-characterised trade as tribute precisely to avoid an implication of balance in status. Thus the Son of Heaven would receive tribute in horses and, in return, graciously provide gifts of silk and treasure. It was absolutely not, everyone was to understand, an exchange of silk for horses, nor treasure payments for peace on the border, buying off potential raiders. The Son of Heaven was therefore absolutely not a merchant nor the equal of any steppe khan or khagan. Hence, the Son of Heaven entered into tributary arrangements, not trade ones. The status claims that such tributary-trade exchanges were wrapped in did not, however, make them not exchange, nor the alleged gifts other than purchases and pay-offs.
Gifts are typically embedded in resource flows so as to create and sustain connections. They are marked by a flow of shifting resource imbalance in an information-dense environment to make connections work. They are not information-economising acts of completion that require no ongoing connection beyond the participants, so they are not acts of exchange.
So, no, gifts are not an archaic form of exchange, nor are they exchange within archaic societies.
This essay is part of the intellectual scaffolding for a book to be published by Connor Court looking at the social dynamics of marriage. (There might also be some suggestion that economists and anthropologists should talk more.) As this essay is something of a work in progress, it is subject to ongoing fiddling.
