The social life of things . , . . " Commodities in cultural perspective Edited by ARJUN APPADURAI University of Pennsylvania I ' ��� CAMBRIDGE UNIVERSITY PRESS CHAPTER 2 The cultural biography of things: commoditization as process IGOR KOPYTOFF For the economist, commodities simply are. That is, certain things and rights to things are produced, exist, and can be seen to circulate through the economic system as they are being exchanged for other· things, usually in exchange for money. This view, of course, frames the commonsensical definition of a commodity: an item with use value that also has exchange value. I shaH, for the moment, accept this definition, which should suffice for raising certain preliminary issues, and I shaH expand on it as the argument warrants. From a cultural perspective, the production of commodities is also a cultural and cognitive process: commodities must be not only produced materiaHy as things, but also culturaHy marked as being a certain kind of thing. Out of the total range of things available in a society, only some of them are considered appropriate for marking as commodities. Moreover, the same thing may be treated as a commodity at one time and not at another. And finally, the same thing may, at the same time, be seen as a commodity by one person and as something else by another. Such shifts and differences in whether and when a thing is a commodity reveal a moral economy that stands behind the objective economy of visible transactions. Of persons and things In contemporary Western thought, we take it more or less for granted that things - physical objects and rights to them - represent the natural universe of commodities. At the opposite pole we place people, who represent the natural universe of inQividuation and singularization. This conceptual polarity of individualized persons and commoditized things is recent and, culturally speaking, exceptional. People can be and have been commoditized again and again, in innumerable societies throughout history, by way of those widespread institutions known under the blanket term "slavery." Hence, it may be suggestive to approach the notion of commodity by first looking at it in the context of slavery. 64 The cultural biography of things 65 Slavery has often been defined, in the past, as the treatment of persons as property or, in some kindred definitions, as objects. More recently, there has been a shift away from this all-or-none view toward a processual perspective, in which marginality and ambiguity of status are at the core of the slave's social identity (see Meillassoux 1975, Vaughan 1977, Kopytoff and Miers 1977; Kopytoff 1982, Patterson 1982). From this perspective slavery is seen not as a fixed and unitary status, but as a process of social transformation that involves a succession of phases and changes in status, some of which merge with other statuses (for example, that of adoptee) that we in the West consider far removed from slavery. Slavery begins with capture or sale, when the individual is stripped of his previous social identity and becomes a non-person, indeed an object and an actual or potential commodity. But the process continues. The slave is acquired by a person or group and is reinserted into the host group, within which he is resocialized and rehumanized by being given a new social identity. The commodity-slave becomes in effect reindividualized by acquiring new statuses (by no means always lowly ones) and a unique configuration of personal relationships. In brief, the process has moved the slave away from the simple status of exchangeable commodity and toward that of a singular individual occupying a particular social and personal niche. But the slave usually remains a potential commodity: he or she continues to have a potential exchange value thai may be realized by resale. In many societies, this was also true of the "free," who were subject to sale under certain defined circumstances. To the extent that in such societies all persons possessed an exchange value and were commoditizable, commoditization in them was clearly not culturally confined to the world of things. What we see �n the career of a slave is a process of initial withdrawal from a given original social setting, his or her commoditization, followed by increasing singularization (that is, decommoditization) in the new setting, with the possibility of later recommoditization. As in most processes, the succer.sive phases merge one into another. Effectively, the slave was unambiguously a commodity only during the relatively short period between capture or first sale and the acquisition of the new social identity; and the slave becomes less of a commodity and more of a singular individual in the process of gradual incorporation into the host society. This biographical consideration of enslavement as a process suggests that the commoditization of other things may usefully be seen in a similar light, namely, as part of the cultural shaping of biographies. 66 Igor Kopytoff The biographical approach Biographies have been approached in various ways in anthropology (for a survey, see Langness 1965). One may present an actual biography, or one may construct a typical biographical model from randomly assembled biographical data, as one does in the standard Life Cycle chapter in a general ethnography. A more theoretically aware biographical model is rather more demanding. It is based on a reasonable number of actual life histories. It presents the range of biographical possibilities that the society in question offers and examines the manner in which these possibilities are realized in the life stories of various categories of people. And it examines idealized biographies that are considered to be desirable models in the society and the way real-life departures from the models are perceived. As Margaret Mead remarked, one way to understand a culture is to see what sort of biography it regards as embodying a successful social career. Clearly, what is seen as a well-lived life in an African society is different in outline from what would be pronounced as a well-lived life along the Ganges, or in Brittany, or among the Eskimos. It seems to me that we can profitably ask the same range and kinds of cultural questions to arrive at biographies of things. Early in this century, in an article entitled "The genealogical method of anthropological inquiry" (1910), W. H. R. Rivers offered what has since become a standard tool in ethnographic fieldwork. The thrust of the article - the aspect for which it is now mainly remembered - is to show how kinship terminology and relationships may be superimposed on a genealogical diagram and traced through the social-structure-intime that the �iagram mirrors. But Rivers also suggested somethingl else: that, for example, when the anthropologist is in search of inheritance rules, he may compare the ideal statement of the rules with the actual movement of a particular object, such as a plot of land, through the genealogical diagram, noting concretely how it passes from hand to hand. What Rivers proposed was a kind of biography of things in terms of ownership. But a biography may concentrate on innumerable other matters and events. In doing the biography of a thing, one would ask questions similar to those one asks about people: What, sociologically, are the biographical possibilities inherent in its "status" and in the period and culture, and how are these possibilities realized? Where does the thing come from and who made it? What has been its career so far, and what do people consider to be an ideal career for such things? What are the recognized "ages" or periods in the thing's "life," and what are the The cultural biography of things 67 cultural markers for them? How does the thing's use change with its age, and what happens to it when it reaches the end of its usefulness? For example, 'among the Suku of Zaire, among whom I worked, the life expectancy of a hut is about ten years. The typical biography of a hut begins with itshousing a couple or, in apolygynoushousehold, a wife with her children. As the hut ages, it is successively turned into a guest house or a house for a widow, a teenagers' hangout, kitchen, and, finally, goat or chicken house - until at last the termites win and the structure collapses. The physical state of the hut at each given age corresponds to a particular use. For a hut to be out of phase in its use makes a Suku uncomfortable, and it conveys a message. Thus, to house a visitor in a hut that should be a kitchen says something about the visitor's status; and if there is no visitors' hut available in a compound, it says something about the compound-head's character - he must be lazy, inhospitable, or poor. We have similar biographical expectations of things. To us, a biography of a painting by Renoir that ends up in an incinerator is as lragic, in its way, as the biography of a perspn who ends up murdered. That is obvious. But there are other events in the biography of objects that convey more subtle meanings. What of a Renoir ending up in a private and inaccessible collection? Of one lying neglected in a museum basement? How should we feel about yet another Renoir leaving France for the United States? Or for Nigeria? The cultural responses to such biographical details reveal a tangled mass of aesthetic, historical, :fmd even political judgments, and of convictions and values that shape our attitudes to objects labeled "art." Biographies of things can make salient what might otherwise remain ,obscure. For example, in situations of culture contact, they can show what anthropologists have so often stressed: that what is significant about the adoption of alien objects - as of alien ideas - is not the fact that they are adopted, but the way they are culturally redefined and put to use. The biography of a car in Africa would reveal a wealth of cultural data: the way it was acquired, how and from whom the money was assembled to pay for it, the relationship of the seller to the buyer, the uses to which the car is regularly put, the identity of its most frequent passenger,s and of those who borrow it, the frequency of borrowing, the garages to which it is taken and the owner's relation to the mechanics, the movement of the car from hand to hand over the years, and in the end, when the car coIJapses, the final disposition of its remains, All of these details would reveal an entirely different biography from that of a middle-classAmerican, or Navajo, or French peasant car. One brings to every biography some prior conception of what is to 68 Igor Kopytoff be its focus. We accept that every person has many biographies psychological, professional, political, familial, economic and so forth - each of which selects some aspects of the life history and discards others. Biographies of things cannot but be similarly partial. Obviously, the sheer physical biography of a car is quite different from its technical biography, known in the trade as its repair record. The car can also furnish an economic biography - its initial worth,· its sale and resale price, the rate of decline in its value, its response to the recession, the patterning over several years of its maintenance costs. The car also offers several possible social biographies: one biography may concentrate on its place in the owner-family's economy, another may relate the history of its ownership to the society's class structure, and a third may focus on its role in the sociology of the family's kin relations, such as loosening family ties in America or strengthening them in Africa. But all such biographies - economic, technical, social - may or may not be culturally informed. What would make a biography cultural is not what it deals with, but how and from what perspective. A culturally informed economic biography of an object would look at it as a culturally constructed entity, endowed with culturally specific meanings, and classified and reclassified into culturally constituted categories. It is from this point of view that I should like to propose a framework for looking at commodities - or rather, speaking processually, at commoditization. But first, what is a commodity? . The singular and the common 1 assume commodities to be a universal cultural phenomenon. Their existence is a concomitant of the existence of transactions that involve the exchange of things (objects and services), exchange being a universal feature of human social life and, according to some theorists, at the very core of it (see, for example, Homans 1961; Ekeh 1974; and Kapferer 1976). Where societies differ is in the ways commoditization as a special expression of exchange is structured and related to the social system, in the factors that encourage or contain it, in the long-term tendencies for it to expand or stabilize, and in the cultural and ideological premises that suffuse its workings. What, then, makes a thing a commodity? A commodity is a thing that has use value and that can be exchanged in a discrete transaction for a counterpart, the very fact of exchange indicating that the counterpart has, in the immediate context, an equivalent value. The coun- The cultural biography of things 69 terpart is by the same token also a commodity at the time of exchange. The exchange can be direct or it can be achieved indirectly by way of money, one of whose functions is as a means of exchange. Hence, anything that can be bought for money is at that point a commodity, whatever the fate that is reserved for it after the transaction has been made (it may, thereafter, be decommoditized). Hence, in the West, as a matter of cultural shorthand, we usually take saleability to be the unmistakable indicator of commodity status, while non-saleability imparts to a thing a special aura of apartness from the mundane and the common. In fact, of course, saleability for money is not a necessary feature of commodity status, given the existence of commodity exchange in non-monetary economies. I refer to the transaction involving commodities as discrete in order to stress that the primary and immediate purpose of the transaction is to obtain the counterpart value (and that, for the economist, is also its economic function). The purpose of the transaction is not, for example, to open the way for some other kind of transactio'n, as in the case of gifts given to initiate marriage negotiations or to secure patronage; each of these cases is a partial transaction that should be considered in the context of the entire transaction. While exchanges of things usually involve commodities, a notable exception is the exchanges that mark relations of reciprocity, as these have been classically defined in anthropology. Here, gifts are given in order to evoke an obligation to give back a gift, which in turn will evoke a similar obligation - a never-ending chain of gifts and obligations. The gifts themselves may be thirigs that are normally used as commodities (food, feasts, luxury goods, services), but each transaction is not discrete and none, in principle, is terminal. To be saleable for money or to be exchangeable for a wide array of other things is to have something in common with a large number of exchangeable things that, taken together, partake of a single universe of comparable values. To use an appropriately loaded even if archaic term, to be saleable or widely exchangeable is to be "common" - the opposite of being uncommon, incomparable, unique, singular, and therefore not exchangeable for anything else. The perfect commodity would be one that is exchangeable with anything and everything else, as the perfectly commoditized world would be one in which everything is exchangeable or for sale. By the same token, the perfectly decommoditized world would be one in which everything is singular, unique, and unexchangeable. The two situations are ideal polar types, and no real economic 70 Igor Kopytoff system could conform to either. In no system is everything so singular as to preclude even the hint of exchange. And in no system, except in some extravagant Marxian image of an utterly commoditized capitalism, is everything a commodity and exchangeable for everything else within a unitary sphere of exchange. Such a construction of the world - in the first case as totally heterogeneous in terms of valuation and, in the second, as totally homogeneous - would be humanly and culturally impossible. But they are two extremes between which every real economy occupies its own peculiar place. We can accept, with most philosophers, linguists, and psychologists, that the human mind has an inherent tendency to impose order upon the chaos of its environment by classifying its contents, and without this classification knowledge of the world and adjustment to it would not be possible. Culture serves the mind by imposing a collectively shared cognitive order upon the world which, objectively, is totally heterogeneous and presents an endless array of singular things. Culture achieves order by carving out, through discrimination and classification, distinct areas of homogeneity within the overalI heterogeneity. Yet, if the homogenizing process is carried too far and the perceived world begins to approach-too closely the other pole in the case of goods, that of utter commoditization -.:. culture's function of cognitive discrimination is undermined. Both individuals and cultural collectivities must navigate somewhere between the polar extremes by classifying things into categories that are simultaneously neither too many nor too embracing. In brief, what we usually refer to as "structure" lies between the heterogeneity of too much splitting and the homogeneity of too much lumping. In the realm of exchange values, this means that the natural world of singular things must be arranged into several manageable value classes - that is, different things must be selected and made cognitively similar when put together within each category and dissimilar when put into different categories. This is the basis for a well-known economic phenomenon - that of several spheres of exchange values, which operate more or less independently of one another. The phenomenon is found in every society, though Westerners are most apt to perceive it in uncommercialized and unmonetized economies. The nature and structure of these spheres of exchange varies among societies because, as we can expect with Durkheim and Mauss (1963; original publication 1903), the cultural systems of classification reflect the structure and the cultural resources of the societies in question. And beyond that, as we may expect with Dumont (1972), there's also some tendency to impose a hierarchy upon the categories. The cultural biography of things 71 Spheres of exchange A concrete example of an economy with clearly distinct spheres of exchange will help the discussion. In what is a classic analysis of a "multi-centric economy," Bohannan (1959) describes three such spheres of exchange as they operated before the colonial period among the Tiv of central Nigeria: (a) the sphere of subsistence items - yams, cereals, condiments, chickens, goats, utensils, tools, and the like; (b) the sphere of prestige items - mainly cattle, slaves, ritual offices, special cloth, medicines, and brass rods; and (c) the sphere of rights-inpeople, which included rights in wives, wards, and offspring. The three spheres represent three separate universes of exchange values, that is, three commodity spheres. Items within each were exchangeable, and each was ruled by its own kind of morality. Moreover, there was a moral hierarchy among the spheres: the subsistence sphere, with its untrammeled market morality, was the lowest, and the rightsin-people sphere, related to the world of kin and kin-group relations, was the highest. In the Tiv case (in contrast to that of many other similar systems), it was possible to move - even if in a rather cumbersome manner - between the spheres. Brass rods provided the link. In exceptional circumstances, people relinquished, unwillingly, rods for subsistence items; and, at the other end, one could also initiate with rods some transactions in the rights-in-people sphere. The Tiv considered it satisfying and morally appropriate to convert "upward," from subsistence to prestige and from prestige to rights-in-people, whereas converting "downward" was shameful and done only under extreme duress. The problem of value and value equivalence has always been a philosophical conundrum in economics. It involves the mysterious process by which things that are patently unlike are somehow made to be alike with respect to value, making yams, for example, somehow comparable to and exchangeable with a mortar or a pot. In the terms we have been using here, this involves taking the patently singular and inserting it into a uniform category of value with other patently singular things. For all the difficulties that the labor theory of value presents, it at least suggests that while yams and pots can conceivably be compared by the labor required to produce them (even while allowing for the different investment in training that the labor represents in each case), no such common standard is available in comparing yams to ritual offices or pots to wives and offspring. Hence, the immense difficulty, indeed impossibility, of lumping all such disparate items into a single commodity sphere. This difficulty provides 72 Igor Kopytoff the natural basis for the cultural construction of separate spheres of exchange. The culture takes on the less sweeping task of making valueequivalence by creating several discrete commodity spheres - in the Tiv case, palpable items of subsistence created by physical labor, as opposed to the prestige items of social maneuvering, as opposed to the' more intimate domain of the rights and obligations of kinship. The drive to commoditization From this perspective, a multi-centric economy such as that of the Tiv is not an exotically complicated rendering of a straightforward exchange system. It is rather the opposite - a feat of simplification of what is naturally an unmanageable mass of singular items. But why only three spheres and not, say, a dozen? The commoditization seems to be pushed to the limits permitted by the Tiv exchange technology, which lacked a common denominator of value more convenient than brass rods. One perceives in this a drive inherent in every exchange system toward optimum commoditization - the drive to extend the fundamentally seductive idea of exchange to as many items as the existing exchange technology will comfortably allow. Hence the universal acceptance of money whenever it has been introduced into nonmonetized societies and its inexorable conquest of the internal economy of these societies, regardless,of initial rejection and of individual unhappiness about it - an unhappiness well illustrated by the modern Tiv. Hence also the uniform results of the introduction of money in a wide range of otherwise different societies: more extensive commoditization and the merger of the separate spheres of exchange. It is as if the internal logic of exchange itself pre-adapts all economies to seize upon the new opportunities that wide commoditization so obviously brings with it. One may interpret Braudel's recent work (1983) in this light - as showing how the development in early modern Europe of a range of new institutions shaped what might be called a new exchange technology and how this, in turn, led to the explosion of commoditization that was at the root of capitalism. The extensive commoditization we associate with capitalism is thus not a feature of capitalism per se, but of the exchange technology that, historically, was associated with it and that set dramatically wider limits to maximum feasible commoditization� Modern state-ordered, noncapitalist economies certainly show no signs of being systematically exempt from this tendency, even though they may try to control it by political means. Indeed, given their endemic shortages and ubiquitous black markets, commoditi- The cultural biography of things 73 zation in them expands into novel areas, in which the consumer, in order to purchase goods and services, must first purchase access to the transaction. Commoditization, then, is best looked upon as a process of becoming rather than as an all-or-none state of being. Its expansion takes place in two ways: (a) with respect to each thing, by making it exchangeable for more and more other things, and (b) with respect to the system as a whole, by making more and more different things more widely exchangeable. Singularization: cultural and individual The counterdrive to this potentialiilln�sh of commoditization is cultu-ie. In the sense that commoditization homogenizes value, while the essence'of culture is discrimination, excessive commoditization is anticultural - as indeed so many have perceived it or sensed it to be. And if, as Durkheim (1915; original publication 1912) saw it, societies need to set apart a certain portion of their environment, marking it as "sacred," singularization is one means to this end. Culture ensures that some things remain unambigiously singular, it resists the commoditization of others; and it sometimes resingularizes what has been commoditized. In every society, there are things that are publicly precluded from being commoditized. Some of the prohibitions are cultural and upheld collectively. In state societies, many of these prohibitions are the handwork of the sta)k, with the usual. intertwining between what serves the society at large, what serves the state, and what serves the specific groups in control. This applies to much of what one thinks of as the symbolic inventory of a society: public lands, monuments, state art collections, the paraphernalia of political power, royal residences, chiefly insignia, ritual objects, and so on. Power often asserts itself symbolically precisely by insisting on its right to singularize an object, or a set or class ot�ects. African chiefs and kings reserve to themselves the right to certain animals and animal products, such as the skins and teeth of spotted wild cats. The kings of Siam monopolized albino elephants. And British monarchs have kept their right to dead whales washed ashore. There may be some practical side to these royal pretensions, which ecological and cultural materialists will no doubt diligently discover. What these monopolies clearly do, however, is to expand the visible reach of sacred power by projecting it onto additional sacralized objects. Such singularization is sometimes extended to things that are nor- 74 Igor Kopytoff mally commodities - in effect, commodities are singularized by being pulled out of their usual commodity sphere. Thus, in the ritual paraphernalia of the British monarchy, we find a Star of India that, contrary to what would normally have happened, was prevented from becoming a commodity and eventually singularized into a "crown jew.el." Similarly, the ritual paraphernalia of the kings of the Suku of Zaire included standard trade items from the past, such as eighteenthcentury European ceramic drinking mugs brought by the Portuguese, carried by the Suku to their present area, and sacralized in the process. Another way to singularize objects is through restricted commoditization, in which some things are confined to a very narrow sphere of exchange. The Tiv system illustrates the principle. The few items in the prestige sphere (slaves, cattle, ritual offices, a special cloth, and brass rods), though commodities by virtue of being exchangeable one for the other, were less commoditized than the far more numerous· items of the subsistence sphere, ranging widely from yams to pots. A sphere consisting of but two kinds of items - as in the classic model of the Trobriand kula exchange sphere of arm bands and bracelets represents an even greater degree of singularization. The Tiv exchange sphere of rights-in-person achieved a singular integrity by a different though related principle, that of the homogeneity of its components. The two upper Tiv spheres, it m�y be noted, were more singular, more special, and hence more sacred than the lowest sphere, containing tlIe many objects of mundance subsistence. Thus the moral hierarchy of the Tiv exchange spheres corresponded to a gradient of singularity. If sacralization can be achieved by singularity, singularity does not guarantee sacralization. Being a non-commodity does not by itself assure high regard, and many singular things (that is, non-exchangeable things) may be worth very little. Among the Aghem of western Cameroon, with exchange spheres not unlike those of the Tiv, one could detect yet another and lower sphere, one below that of marketable subsistence items. Once, when trying to find out the precolonial exchange value of various items, I asked about the barter value of manioc. The response was indignant scoffing at the very idea that such a lowly thing as manioc should have been exchangeable for anything: "One eats it, that's all. Or one gives it away if one wants to. Women may help out one another with it and other such food. But one doesn't trade it." Lest the outburst be misunderstood and sentimentalized, let me stress that the indignation was not about a suggested commercial corruption of a symbolically supercharged staple, on the order, say, of bread among Eastern European peasants. The The cultural biography of things 75 Aghem are and were a commercially minded people, with no disdain for trade. The scoffing was rather like what an Aghem would get from a Westerner whom he asked about the exchange value of a match he proffers to light a stranger's cigarette. Manioc was part of a class of singular things of so little worth as to have no publicly recognized exchange value. To be a non-commodity is to be "priceless" in the full possible sense of the term, ranging from the uniquely valuable to the uniquely worthless. In addition to things being classified as more or less singular, the�e is also what might be called terminal commoditization, in which further exchange is precludeo by fiat. In many societies, medicines are so treated: the medicine man makes and sells a medicine that is utterly singular since it is efficacious only for the intended patient. Terminal commoditization also marked the sale of indulgences in the Roman Catholic Church of half a millennium ago: the sinner could buy them but not resell them. In modern Western medicine, such terminal commoditization is achievt;Jk'legally; it rests on the prohibitiol1 against reselling a prescribed drug and against selling any medicine without proper licensing. There are otherexamples of legal attempts to restrict recommoditization: paperbound books published in Great Britain often carry a bewildering note forbidding the buyer to resell it in any but the original covers; and in America, an equally mystifying label is attached to mattressd and cushions, forbidding their resale. Other factors besides legal or cultural fiat may create terminal commodities. Most consumer goods are, after all, destined to be terminalor so, at least, it is hoped by the manufacturer. The expectation is easily enough fulfilled with such things as canned peas, though even here external circumstances can intrude; in times of war shortages, all sorts of normally consumable goods begin to serve as a store of wealth and, instead of being consumed, circulate endlessly in the market. With durable goods, a second-hand market normally develops, and the idea that it does may be fostered by the sellers. There is an area of our economy in which the selling strategy rests on stressing that the commoditization of goods bought for consumption need not be terminal: thus, the promise that oriental carpets, though bought for use, are a "good investment," or that certain expensive cars have a "high resale value." The existence of terminal commoditization raises a point that is central to the analysis of slavery, where the fact that a person has been bought does not in itself tell us anything about the uses to which the-person may then be put (Kopytoff 1982:223ff). Some purchased people ended up in the mines, on plantations, or on galleys; others 76 Igor Kopytoff became Grand Viziers or Imperial Roman Admirals. In the same way, the fact that an object is bought or exchanged says nothing about its subsequent status and whether it will remain a commodity or not. But unless formally decommoditized, commoditized things remain potential commodities - they contiQ!JS! to have an exchange value, even if they have been effectively withdrawn from their exchange sphere and deactivated, so to speak, as commodities. This deactivation leaves them open not only to the various kinds of singularization I have mentioned so far, but also to individual, as opposed to collective, redefinitions. lfitJie Bamenda area of western Cameroon, people prized large decorated calabashes �hat came over the border from Nigeria. The conduit for them was the Aku, a pastoral group \whose women used them extensively and normally were willing to sell them. I had acquired several in this way. Yet one day I failed completely to convince an Aku woman to sell me a standard calabash to which she had added some minor decorations of her own. Her friends told her that she was being silly, arguing that for the money she could get a far better and prettier calabash. But she would not budge, no more than does that ever-newsworthy man in our society - part hero, part fool - who refuses to sell his house for a million dollars and forces the skyscraper to be built around it. And there is also the opposite phenomenon: the ideological commoditizer, advocating, say, the sale of public lands as a way of balancing the budget, or, as I have seen in Africa, calling for the sale of some piece of chi�fly paraphernalia in order to provide a tin roof for the schoolhouse. What these mundane examples show is that, in any society, the individual is often caught between the cultural structure of commoditization and his own personal attempts to bring a value order to the universe of things. Some of this clash between culture and individual is inevitable, at least at the cognitive level. The world of things lends itself to an endless number of classifications, rooted in natural features and cultural and idiosyncratic perceptions. The individual mind can play with them all, constructing innumerable classes, different universes ofcommon value, and changing spheres of exchange. Culture, by contrast, cannot be so exuberant, least so in the economy, where its classifications must provide unambiguous guidance to pragmatic and coordinated action. BuLif the clash is inevitable, the social structures within which it takes place vary, giving it different intensities. .. In a society like the precolonial Tiv or Aghem, the culture and the economy were in relative harmony; the economy followed the cultural classifications, and these catered successfully to the individual cog� nitive need for discrimination. By contrast, in a commercialized, mo- The cultural biography of things 77 netized, and highly comm�ditized society, the value-homogenizingdrive of the exchange system has an enormous momentum, producing results that both culture and individual cognition often oppose, but in inconsistent and even contradictory ways. Complex societies I said above that the exchange spheres are, to us, more visible in noncommercial, non-monetized societies like the Tiv than in commercial, monetized ones like our own. Partly this is a matter of noticing the exotic and taking the familiar for granted. But it is more thap that. Certainly, in our society, some discrete spheres of exchange exist and are nearly unanimously accepted and approved. Thus, we are adamant about keeping separate the spheres of material objects and persons (a matter I shall elaborate on later). We also exchange dinners and keep that sphere discrete. We blandly accept the existence of an exchange sphere of political or academic favors, but would be as shocked at the idea tf monetizing this sphere as the Tiv were at first at the idea of monetizing their marriage transactions. Like the Tiv, who carefully moved from the sphere of mundane pots to that of prestigeful titles' by using the mediation of brass rods, so do our financiers cautiously navigate between exchange spheres in such matters as gift-giving to qniversities. A straight money donation in general funds, if it is of any size, is suspect because it looks too much like purchasing influence, and such donations, when made, are normally anonymous or posthumous. A money donation in installments would be particularly suspect, implying the donor's power to withhold the next check. But converting a large donation into a building moves the money into a nearly decommoditized sphere, freezes the gift into visible irrevocability, a.nd shields the donor from suspicion of continuous undue influence on the university. Putting the donor's name on the building thus honors not simply the donor but also the university, which declares in doing so that it is free of any lingering obligations to the specific donor. The values underlying such transactions are, on the whole, societywide, or at least are held by the groups who wield cultural hegemony in our society and define much of what we are apt to call our public culture. "Everyone" is against commoditizing what has been publicly marked as singular and made sacred: public parks, national landmarks, the Lincoln Memorial, George Washington's false teeth at Mount Vernon. Other singularizing values are held by more restricted groups. We have explicit exchange spheres recognized only by segments of society, 78 Igor Kopytoff such as professional and occupational groups, which subscribe to a common cultural code and a specially focused morality. Such groups constitute the networks of mechanical solidarity that tie together the parts of the organic structure of the wider society, the latter being ruled in most of its activities by commodity principles. Let me lead into my discussion by looking at an activity in one such group: the collection of African art among American Africanists. In the simpler days of thirty or more years ago, African art picked up randomly in the course of fieldwork was placed entirely in a closed sphere with a sacred cast. The objects collected were greatly singularized; they were held to have for their collector a personal sentimental value, or a purely aesthetic one, or a scientific one, the last supported by the collector's supposed knowledge of the object's cultural context. It was not considered entirely proper to acquire an art object from African market traders or, worse, from European traders in Africa, or worse still, from dealers in Europe or America. Such an object, acquired at second hand, had little scientific value, and it was vaguely contaminated by having circulated in a monetized commoditysphere - a contamination that was not entirely removed by keeping it thereafter in the same category as the objects "legitimately" acquired in the field. The exchange sphere to which African art objects belonged was extremely homogeneous in content. It was permissible to exchange them for other African (or other "primitive art") objects. One could also give them as gifts. Students returning from the field usually brought one or two as gifts to their supervisors, thus inserting them into another circumscribed sphere, that of academic patronclient relationships. The morality governing the sphere did not allow for them to be sold, except at cost to a museum. Nevertheless, as among the Tiv, for whom it was permissible though shameful to sell a brass rod for food, so here extreme need justified "liquidation" on the commercial art market, but it had to be done with appropriate discretion and it was certainly seen as converting "downward." As Douglas and Isherwood (1980) show, the public culture in complex societies does provide broadly discriminating / value markings of goods and services. That is, the public culture offers discriminating classifications here no less than it does in small-scale societies. But these must constantly compete with classifications by individuals and by small networks, whose members also belong to other networks expounding yet other value systems. The discriminating criteria that each individual or network can bring to the task of classification are extremely varied. Not only is every individual's or network's version of exchange spheres idiosyncratic and different from those of others, The cultural biography of things 79 but it also shifts contextually and biographically as the originators' perspectives, affiliations and interests shift. The result is a debate not only between people and groups, but within each person as well. To be sure, the seeds for such debates also exist in societies like the precolonial Tiv, but there the culture and the economy joined hands to provide an approved model of classification. In a commercialized, heterogenous, and liberal society, the public culture defers most of the time to pluralism and relativism and provides no firm guidance, while the only lesson the economy can teach is that of the freedom and dynamism that ever-wider commoditization clearly brings with it. The results can be partly glimpsed in what/has happened to African art collecting over the past quarter century. The rules have been loosened in some of the same ways that monetization, according to Bohannan, loosened the rules among the Tiv - namely, by merging the previously distinct exchange spheres. There are, for example, no strictll7es now on buying an African art object at an auction in America, let alone from an African trader in Africa. Monetization in itself has become less contaminating as it has become more seduct ' ive, for no one can remain unaware that these objects are what every newspaper and magazine calls "collectibles." But the most noticeable change has been, quite simply, to make the rules less clear and more open to individual interpretations and to idiosyncratic systems of values. Where before the professiOI}al culture decreed that the value of these objects was sentimental when it was not scientific, now sentimental value is conferred as a matter of individual choice, perhaps more sincerely but also less widely. At the same time, puritans have arisen, thundering about the immorality of any kind of circulation of these objects and calling for their complete singularization and sacralization within the closed boundaries of the society that produced them. In brief, the rules of the professional culture have become less tight and the rules of propriety have become more idiosyncratic. The widespread rejection, since the 1960's, of the very idea of cultural restraints has, here as elsewhere, opened the door to a great variety of definitions by individuals and small groups. What I am arguing here is that the crucial difference between complex and small-scale societies does not lie simply in the extensive commoditization in the former. There have been, we must not forget, small-scale societies in which commoditization (helped by indigenous money) was very extensive, such as the Yurok of northern California (Kroeber 1925) or the Kapauku of western New Guinea (Pospisil 1963). The peculiarity of complex societies is that their publicly recognized commoditization operates side by side with innumerable 80 Igor Kopytoff , schemes of valuation and sil1.gl!!