Debt, Violence, and Impersonal Markets David Graeber

If The Great Transformation will be remembered for anything a century from now, it willbe as the definitive rejoinder to the great liberal myth. This is, of course, the assumption that there is something natural about what Polanyi called "self-regulating markets", that they arise of their own accord as long as state interference doesn't prevent them. Polanyi examined the very period when this ideology first emerged, and managed to demonstrate just how crucial government interference was in creating "the self-regulating market" to begin with—just as it has continued to be necessary to maintain it.

One need hardly point out that in the current, neoliberal age, Polanyi's insights are more relevant than ever. The ideology that Polanyi felt was gone forever in the '40s has returned with a vengeance—returned to reap a terrible vengeance, in fact, on the most vulnerable of the people of the earth. Yet at the same time the intellectual landscape has shifted dramatically. Among what passes as the intellectual opposition, grand sweeping theory in the Polanyian tradition has fallen largely out of favor. At the same time, the high theorists of neoliberalism—at least, the most sophisticated of them—often appear more than happy to incorporate many of Polanyi's insights. Most will, if pressed, be happy to admit that "the market" isn't really an empirical object at all, that when they refer to "markets" they are really talking about abstract models, constructed by selecting only certain features of reality and intentionally ignore all others; and that of course one needs constant political work to maintain conditions where those models will take on any semblance of empirical form. Of course, when giving policy advice, these same economists will then turn around and declare that "the market"—now transformed from an abstract model to a quasi-deity—will punish those who disregard its sovereign dictates.

When arguments don't even have to make logical sense, critique might well seem to lose its point. Nonetheless, it strikes me that new theoretical tools would be helpful here—if only because how we conceptualize the moment has everything to do with how we imagine alternatives. Polanyi wrote at an historical moment when it seemed that the very governments that had created self-regulating markets seemed to be in the process of transcending them. Today we've seen those same social democratic regimes often leading the way in stripping away social protections, and anti-capitalist movements increasingly moving away from any notion that the state—which is, after all, basically a means of organizing violence—can help solve anything.

What I would like to do in this essay then is to make a few suggestions about how we might begin to reconceptualize Polanyi's approach to economic history from this, rather different, historical perspective. This means coming up with new terms to supplement, and to some degree supplant, Polanyi's distinctions between reciprocity, redistribution, and market, special and general purpose monies, and introduce distinctions between what I will call "human economies" and different sorts of market, some dominated by credit institutions, others by anonymous exchange of metal bullion. In both cases, I want at least to consider the importance of war and violence as critical elements in allowing the transformations of one form into theother.1 The easiest way to begin to reconstruct this history, I think, is by looking at the history of money. 1: Value versus Debt 
The approach to economic history I will propose here has larger theoretical implications. While this is not the place to develop them in any detail, it seems to me that we have come to the point where we have largely moved past the hoary opposition between individual and society, and might better begin instead from an opposition between value and debt: that is, between webs of dyadic relations based on various forms of (usually mutual) obligation, and the creation of virtual arenas for the realization of human creativity.

This point may seem obscure. Perhaps the best way to explain it is to explain how I came to it. 
In 2001, I wrote a book which among other things tried to develop a new approach to some of the intricate problems of Marxian value theory. My key point was that our distinction between "value" (in the economic sense) and "values" (in the social sense) really turns on the commoditization of labor. Where human energies are directed at profit, or wages, we are in the domain of "the economy" or "the market", which operates according to the law of value. When we enter into other pursuits, such as domestic life (housework being probably the most important form of unremunerated labor in industrialized societies), or religion, politics, and so on, we are suddenly said to jump into the domain of "values": this is precisely where people begin to talk about "family values", religious faith, political ideals, the pursuit of beauty, patriotism and so on. All these are seen as commitments that ought to be uncorrupted by the market. At the same time, they are also seen as utterly unique, effectively, incommensurable. It would be absurd to search for a mathematic formula that could allow one to calculate just how much personal integrity it is right to sacrifice in the pursuit of art, or how to balance responsibilities to God and to your family.

The entire argument here turns on money being an impersonal abstraction. "Value" is that which money measures. Money is a generic substance whose only quality is that is can be precisely counted; aside from its denomination, one banknote is precisely the same as any other. Therefore no particular dollar bill can develop a unique history. It is pure potentiality. Without such a generic medium, one is left with a series of unique historical crystallizations. 
1 This might be considered an extension of an argument about the similarities between impersonal market relations and violence itself as forms of radical simplification (Graeber 1996).

Marx as we all know saw the value of money as ultimately rooted in human capacities for creative action, or "labor power". He also argued that it was only through the institution of wage labor that such creative potential itself becomes a commodity. One interesting concomitant is that as a result, wage laborers—who are after all working in order to get money—are effectively working in order to obtain symbolic tokens that represent the importance of their own work. Money, then, is a symbol that effectively brings into being the very thing it represents. As such it comes to seem the source of the value of the labor, rather than something having been produced by it. The premise of the book was that any system of value tends to operate this way. Value is simply the way that we represent the meaning or importance of our own actions to ourselves. Our actions become meaningful and important by becoming part of some larger social totality, real or imagined; this must also necessarily happen through some material medium: if not money, then treasures, tokens, performances, privileges, and so on. The medium can be almost anything, but its nature has very definite implications as to how this realization of value takes place. With a quantifiable abstraction like money, one can develop systems of abstract value; when the most important tokens of value are unique but permanent heirlooms, betokening "fame", one might end up with something more like a kula system (Munn 1986); when they areelaborate, but ephemeral, ritual performances that express "beauty", one can end up withsomething more like the Kayapo rituals described by Turner (1984, 1985, 1987). Nonetheless,there are always certain constants. One is that since value can only be realized in the eyes of others, what we think of as "society" largely emerges as the audience for different projects forthe realization of value. From the perspective of the actor, at least, "society" is simply all thosewhose opinions he actually cares about. It is always to a certain degree an imaginary totality. Another is that the tokens through which they are realized tend to become fetishized, in the sensethat from the point of view of the actors, they are seen as the source of that which they motivate.The desire to acquire tokens of honor inspires honorable behavior; the desire to attain tokens offaith, or certificates of educational attainment, comes to inspire piety or learning, even to organize the form such actions take. The result, as in the case of money, is that it often seems as if these tokens, rather than the human actions aimed at acquiring them, are what brings piety orlearning into being in the world to begin with—since, from the point of view of the actors, this often might as well be true.

Value theory then is about how desire becomes social. It is about how our actions becomemeaningful by being reflected back at us in the form of representations—ultimately, of those very actions—that come to seem their aim and origin. And this is about how different conceptions of "society" are constantly being thrown up, like shadows on a wall, as a necessary part of that process.

The main weakness in this approach, I soon discovered, was its treatment of money. Like Marx, I emphasized the anonymous, impersonal qualities of money. These do exist. There'sabsolutely no way to know where a dollar bill in one's pocket has been; the result is that thehistory of objects bought and sold by dollar bills tend to be effaced as well. This is of course, thekey to Marx's conception of fetishism, where objects come to seem to embody the intentions oftheir designers and producers, since one has no way of knowing who those people actually were.

The problem is that, while this may be true of cash, most transactions in contemporarysocieties do not employ cash; and the largest, most significant transactions almost never do— unless, that is, they are criminal in nature. There is a reason why bank robbers and drug kingpinsare the only people who prefer to operate with suitcases full of hundred dollar bills. Ordinary monetary transactions do indeed leave a history, since they usually operate through credit and, as 
law enforcement agents are well aware, it is quite possible to keep exact and detailed tabs on the movements of any citizen simply by monitoring their bank and credit card transactions. Whilethis does not change Marx's main point about commodity fetishism—I still don't have the slightest idea who was involved in creating and assembling my cell phone or my toaster—it means that money is a far more complex object than we might otherwise assume. Where somesee money as wiping away the possibility of memory, Keith Hart, for example, insists instead that money "is mainly… an act of remembering, a way of keeping track of the exchanges which we enter into with the rest of humanity" (1999:234).2

It seems to me Hart is a good place to start on a reconsideration of this problem becausehe's one of the few authors who looks at money neither as a means of recording history nor as a means of effacing history, but rather sees the peculiar quality of money as lying in the fact that it is an unstable suspension of both: 
Look at a coin from your pocket. On one side is 'heads'—the symbol of the political authority which minted the coin; on the other side is 'tails'—the precisespecification of the amount the coin is worth as payment in exchange. One sidereminds us that states underwrite currencies and the money is originally a relation between persons in society, a token perhaps. The other reveals the coin as a thing,capable of entering into definite relations with other things, as a quantitative ratio independent of the persons engaged in any particular transaction. In this latterrespect money is like a commodity and its logic is that of anonymous markets (Hart 1986:638).

Marx, of course, made the famous argument that in fetishism, what are actually relationsbetween persons are displaced and made to appear as if they were relations between things. Mauss' distinction between gifts and commodities actually works by an analogous logic: atransaction is a gift if it is largely concerned with the relations between persons, a commodityexchange, if what is being established is instead equivalence between things. What Hart ispointing out is that this distinction is inscribed into the very nature of money itself, so much so that economists have produced completely contradictory theories as to what money even is. Onthe one hand we find the familiar "Metallist" or "commodity" theory of money (what Hart wouldcall the "tails" approach), that sees money as having first emerged from the inconveniences ofbarter. We've all heard this story.3 At first human beings bartered useful objects directly one for another; after a while, they came to realize that it would be much easier simply to denominate asingle commodity as a means to pay for every other one. For various reasons, precious metals seemed the most convenient choice. According to this view (e.g., Samuelson 1947), modern economies are still really just elaborate systems of barter, a way for economic actors to tradeuseful commodities for others, with money merely serving as a convenient technology ofexchange. This view is, effectively, economic orthodoxy: the overwhelming majority of

2 In fact, the very word is derived from memory: the English "money" ultimately derives fromthe temple of Juno Moneta in ancient Rome, where coins were struck during the Punic Wars— Moneta being the goddess of Memory and mother of the muses (1999:15, 256).3 This theory of the origin of money already appears in Adam Smith, though in itscanonical version it was most famously laid out by Jevons (1875) and Menger (1892). 
professional economists accept it, despite there being virtually no evidence that anything like thisever happened.

Ranged against it is a variety of heretical, "Chartalist" approaches that rely on the other side of Hart's coin. These assume that money did not arise from individual actors trying tomaximize their material advantage, but rather, from public institutions aiming to calculate and manage social obligations: that money arises, in effect, from debt. The paradigm is Knapp's "State Theory of Money" (1928), where he argued that money arose not as a medium of exchange but as a unit of account (and secondarily, means of payment), specifically, as a means of assessing and levying tax payments. Money, here, is a way of managing debt, starting with the debt that subjects or citizens were assumed to have to their sovereign. In order to do so, the statemust establish the nominal units of account, and fixes the conversion rates between commodities.

Moreover, as colonial regimes were to rediscover in the 18th and 19th centuries,demanding cash payments from one's subjects is the most effective way to encourage a marketin goods and services, and this might often have been at least half the point. It is in fact much easier, from the point of view of a government, to create a market for goods and services, and then buy what it needs, than to requisition everything directly, either in kind or in labor. The key point though is, as Michael Innes (1913, 1914) originally put it, that "money is debt": the stateissues tokens of its own obligations that become validated and go into general use by citizensseeking to cancel their debts with one another, because the state is willing to accept them tocancel debts which (it has declared) citizens owe to it.4

The Chartalist view has always been in a minority among professional economists—even though almost all the historical evidence seems to support it. Still, it has its exponents, especially amongst the followers of John Maynard Keynes. However, the two camps have always, as Hartnoted, tended to state their positions in absolute terms, arguing money is purely one thing or theother. Hence Keynesians end up arguing for state-managed manipulation of the money supply as a tool of policy, while "monetarists" insist the government's role is simply to back up a stablecurrency but otherwise let the market do its work, and policy tends to swing back and forth wildly between them.

As Hart observes, for the most part anthropologists have simply ignored these debates. They have had especially little to say about the phenomenon of debt. This is in a way surprising,since anthropologists have over the years had a great deal to say about social obligation.Structural functionalist anthropology was, more than anything else, an elaborate system ofmapping out "rights and duties" (two concepts which are, like credit and debt, themselves two sides of the same coin.) In fact, it seems to me that such oppositions between theories of value and theories of debt open up a much more interesting set of theoretical problems than more familiar (and increasingly sterile) divisions between "individual" and "society". The Metallistview, for example, doesn't begin with one individual who confronts society: it begins with a series of dyadic relations (mainly buyer-seller) and then tries to see how an endless network ofsuch relations can gradually produce an imaginary totality it calls "the market". The Chartalist 
4 Innes also noted that banks, which specializing in the canceling credits against debts,developed as intermediaries with the state: in every case we know about, it wasgovernments (even, in the case of Medieval Europe, no longer existing governments: seeEinaudi 1953) that were seen as establishing the abstract units of exchange, just as they were seen as establishing systems of weights and measures. 
view starts from the state—an entity that I have argued always begins primarily as a utopian project (Graeber 2003)—and works its way down to the regulation of networks of obligation.The state in this view creates money in much the same way as it regulates justice: as a means of balancing moral accounts.

This in turn raises two particularly sticky conceptual questions. The first is about theorigin of the idea of debt. How do social obligations, rights and duties that people have with oneanother, end up becoming attached themselves to objects of material wealth, so that the meretransfer of such objects can often render one person entirely at another's command? The second is even larger: how does one relate a theory of value to a theory of debt? It is possible toconceive what we call "societies" as an endless web of inter-personal relations; it is possible to conceive them as imaginary totalities that serve as arenas for the realization of value. It is very difficult to understand them as both at the same time.

I cannot solve all these problems here. But I want to attempt an outline of what a theory of debt might look like, because I think it should be critical to the larger task of conceptualizing the current historical moment in a way that allows for alternatives. Certainly the problem is profoundly under-theorized. The modern state, after all, is often said to have emerged withdeficit financing; the economies of wealthy countries are now driven largely by consumer debt; international relations are increasingly dominated by the debt bondage of the poor to the IMF and World Bank and by the debt of the United States to East Asia. Yet there is remarkably little written about the nature of debt itself. It's a question of particular political interest, it seems to me, since debt has long been one of the chief ways in which relations based on exploitation and even violence have come to seem moral in the eyes of those living inside them. Throughouthistory, there have been classes of people who essentially live off the labors of others; in aremarkably large number of cases, they appear to have managed to convince the latter that it is they who are somehow in their debt. Yet they do not do this, normally, as a class. They do so through an endless multiplication of individual—or, more accurately, dyadic—ties.