Chapter 1 — Three Ways of Talking about Value

If one reads a lot of anthropology, it is hard to escape the impression that theories of value are all the rage of late. One certainly sees references to “value” and “theories of value” all the time—usually thrown out in such

a way as to suggest there is a vast and probably very complicated literature lying behind them.[1] If one tries to track this literature down, however, one quickly runs into problems. In fact it is extremely difficult to find a systematic “theory of value” anywhere in the recent literature; and it usually turns out to be very difficult to figure out what body of theory, if any, that any particular author who uses the term “value” is drawing on. Sometimes, one suspects it is this very ambiguity that makes the term so attractive.

What I’d like to do in this chapter is offer some suggestions as to how this situation came about. I think it has something to do with the fact that anthropology has been caught in a kind of theoretical limbo. The great theoretical dilemmas of twenty years ago or so have never really been resolved; it’s more like they were shrugged off. There is a general feeling that a theory of value would have been just the thing to resolve most of those dilemmas, but such a theory never really materialized; hence, perhaps, the habit of so many scholars acting as if one actually did exist.

It will become easier to see why a theory of value should have seemed to hold such promise if one looks at the way the word “value” has been used in social theory in the past. There are, one might say, three large streams of thought that converge in the present term. These are:

1. “values” in the sociological sense: conceptions of what is ultimately good, proper, or desirable in human life

2. “value” in the economic sense: the degree to which objects are desired, particularly, as measured by how much others are willing to give up to get them

3. “value” in the linguistic sense, which goes back to the structural linguistics of Ferdinand de Saussure (1966), and might be most simply glossed as “meaningful difference”

When anthropologists nowadays speak of “value”—particularly, when they refer to “value” in the singular when one writing twenty years ago would have spoken of “values” in the plural—they are at the very least implying that the fact that all these things should be called by the same word is no coincidence. That ultimately, these are all refractions of the same thing. But if one reflects on it at all, this is a very challenging notion. It would mean, for instance, that when we talk about the “meaning” of a word, and when we talk about the “meaning of life,” we are not talking about utterly different things. And that both have something in common with the sale-price of a refrigerator. Now, putting things this way raises obvious objections. A skeptic might reply: it may well be that all these concepts do have something in common, but if so, that “something” would have to be so utterly abstract and vague that pointing it out is simply meaningless. In which case the ambiguity really is the point. But I don’t think this is so. In fact, if one looks back over the history of anthropological thought on each of the three sorts of value mentioned above one finds that in almost every case, scholars trying to come up with a coherent theory of any one of them have ended up falling into terrible problems for lack of sufficient consideration of the other ones.

Let me give a brief sketch of these histories, one at a time:

I: Clyde Kluckhohn’s value project

The theoretical analysis of “values” or “systems of values” is largely confined to philosophy (where it is called “axiology”) and sociology (where it is what one is free from when one is “value-free.”) It is not as if anthropologists haven’t always used the term. One can pick up a work of anthropology from almost any period and, if one flips through long enough, be almost certain to find at least one or two casual references to “values.” But anthropologists rarely made much of an effort to define it, let alone to make the analysis of values a part of anthropological theory. The one great exception was during the late 1940s and early ‘50s, when Clyde Kluckhohn and a team of allied scholars at Harvard embarked on a major effort to place the issue of values at the center of anthropology. Kluckhohn’s project, in fact, was to redefine anthropology itself as the comparative study of values.

Nowadays, the project is mainly remembered because it managed to find its way into Talcott Parson’s General Theory of Action (Parsons and Shils 1951), meant as a kind of entente cordiale between sociology, anthropology, and psychology, which divided up the study of human behavior between them. Psychologists were to investigate the structure of the individual personality, sociologists studied social relations, and anthropologists were to deal with the way both were mediated by culture, which comes down largely to how values become esconced in symbols and meanings. Kluckhohn’s main anthropological work had been among the Navaho, but he conceived the notion of doing a comparative study of values that focused on the county of Rimrock, New Mexico (1951b, 1956; Vogt and Albert 1966), which was divided between five different communities: Navaho, Zuñi, Mormon, Texan, and Mexican-American. Its existence, Kluckhohn thought, provided as close as one could get in anthropology to a controlled experiment: a chance to see how five groups of people with profoundly different systems of value adapted to the same environment. He sent off five students, one to study each (and in fact quite a number of the next generation of American anthropologists were involved in the Rimrock study at one time or another), while he remained behind at Harvard, leading a seminar on values and working out a succession of working papers that aimed to define the terms of analysis.

So what, precisely, are values? Kluckhohn kept refining his definitions. The central assumption though was that values are “conceptions of the desirable”—conceptions which play some sort of role in influencing the choices people make between different possible courses of action (1951a:395). The key term here is “desirable.” The desirable refers not simply to what people actually want—in practice, people want all sorts of things. Values are ideas about what they ought to want. They are the criteria by which people judge which desires they consider legitimate and worthwhile and which they do not. Values, then, are ideas if not necessarily about the meaning of life, then at least about what one could justifiably want from it. The problem though comes with the second half of the definition: Kluckhohn also insisted that these were not just abstract philosophies of life but ideas that had direct effects on people’s actual behavior. The problem was to determine how.

Of course when one speaks of values in the traditional sense, this is not so difficult. By this I mean, in the sense in which one might say that the Navaho community in Rimrock places a high value on something it calls “harmony,” or the Texan, on something it calls “success.” Normally “value analysis,” such as it is, consists of identifying such terms and interpreting them, figuring out precisely what “harmony” or “success” means to the people in question, and placing these definitions in a larger cultural context. The problem though is that such terms tend to be highly idiosyncratic. Kluckhohn was interested in the systematic comparison of values.

In order to compare such concepts, Kluckhohn and his disciples ended up having to create a second, less abstract level of what he called “value orientations.” These were “assumptions about the ends and purposes of human existence,” the nature of knowledge, “what human beings have a right to expect from each other and the gods, about what constitutes fulfillment and frustration” (Kluckhohn 1949:358–59). In other words, value orientations mixed ideas of the desirable with assumptions about the nature of the world in which one had to act. The next step was to establish a basic list of existential questions, that presumably every culture had to answer in some way: are human beings good or evil? Should their relations with nature be based on harmony, mastery, or subjugation? Should one’s ultimate loyalties be to oneself, to a larger group, or to other individuals? Kluckhohn did come up with such a list; but he and his students found it very difficult to move from this super-refined level to the more mundane details of why people prefer to grow potatoes rather than rice or prefer to marry their cross-cousins—the sort of everyday matters with which anthropologists normally concern themselves.

At this point the story takes on something of the color of tragedy. Almost everyone involved felt that the Rimrock study was a failure; in writing up their conclusions, the fieldworkers found it almost impossible to develop common terms. Even while Kluckhohn’s disciples—notably philosopher Edith Albert—were continuing to pour out essays gushing with scientific confidence in the late ‘50s and early ‘60s, Kluckhohn himself seems to have spent the last years of his life plagued by a sense of frustration, an inability to find the breakthrough that would make a real, systematic comparative study of values possible—or anyway, to relate it properly to action (Albert 1956, 1968, Kluckhohn 1951a, 1961, F. Kluckhohn and Strodtbeck 1961). It was all the more frustrating because Kluckhohn saw his project in many ways a last-ditch effort to rescue American anthropology from what almost everyone perceived as the theoretical doldrums. Where British anthropologists had always conceived their discipline as a branch of sociology, the North American school founded by Franz Boas had drawn on German culture theory to compare societies not primarily as ways of organizing relations between people but equally, as structures of thought and feeling. The assumption was always that there was, at the core of a culture, certain key patterns or symbols or themes that held everything together and that couldn’t be reduced to pure individual psychology; the problem, to define precisely what this was and how one could get at it. One is left with a strange, rather contradictory picture, since this was also the time when Boasian anthropology was at the height of its popular influence and academic authority, flush with Cold War money, at a time when their books were often read by ordinary Americans, but at the same time, was burdened with a growing feeling of intellectual bankruptcy. Kluckhohn’s effort to reframe anthropology as the study of values could be seen as a last-ditch effort to salvage the Boasian project; it is nowadays seen as yet another dead end. The consensus of those who even bother to talk about the episode (Edmonson 1973, Dumont 1982), though, is that there was nothing inherently wrong with the project itself: rather, it failed for the lack of an adequate theory of structure. Kluckhohn wanted to compare systems of ideas, but he had no theoretical model of how ideas fitted together as systems. In his last few years he became increasingly interested in the idea of borrowing models from linguistics, but the tools then available were simply not up to the challenge. His critics seem to imply that if he or his project had lasted a few more years, until structuralist models burst on the scene in the late ‘60s, everything might have been different.

Be this as it may, the project had no intellectual successors. This is not to say, of course, that anthropologists no longer talk about “values.” Some regional subdisciplines are veritably obsessed with particular values (especially those dealing with regions without too much elaborate social structure, such as clans or lineage systems): most notoriously the anthropology of the Mediterranean, which has been for most of its history focused on “honor.” But there has been next to nothing on “values” in general. This is true even of scholars working in Kluckhohn’s own intellectual tradition. Some of the most influential American cultural theorists of the ‘60s and ‘70s—I am thinking here especially of Clifford Geertz and David Schneider—were in many ways continuing in it, but they moved in very different directions.

In a way this is rather too bad. For all its sterility in practice, there is something appealing about Kluckhohn’s key idea: that what makes cultures different is not simply what they believe the world to like, but what they feel one can justifiably demand from it. That anthropology, in other words, should be the comparative study of practical philosophies of life. Actually the closest parallel to it in the social sciences was probably Max Weber’s comparative study of world religions, which also was concerned with delineating a limited number of possible ways for thinking about the meaning of human existence and then trying to understand the implications for social action of each. It’s possible that his work may even be due for something of a revival: there have been some recent efforts, for example by Charles Nuckolls (1999), to integrate such value analysis with psychological approaches in anthropology. But for present purposes, the important thing is that the first great effort to come up with an anthropological theory of values ran most definitively aground; and that anthropological concerns with such issues started developing, in the ‘60s, in two opposite directions: one that looked to economics, the other, to linguistics.

II: the maximizing individual

Practically from the beginnings of modern anthropology, there have been efforts to apply the tools of microeconomics to the study of non-Western societies. There are several reasons it seemed an obvious thing to do. First, because (apart from linguistics) economics has always been the social science that could make the most plausible claim that what it was doing was anything like a natural science; it has long had the additional advantage of being seen as the very model of “hard” science by the sort of people who distribute grants (people who themselves usually have some economic training). It also has the advantage of joining an extremely simple model of human nature with extremely complicated mathematical formulas that non-specialists can rarely understand, much less criticize. Its premises are straightforward enough. Society is made up of individuals. Any individual is assumed to have a fairly clear idea what he or she wants out of life, and to be trying to get as much of it as possible for the least amount of sacrifice and effort. (This is called the “mini/max” approach. People want to minimize their output and maximize their yields.) What we call “society”—at least, if one controls for a little cultural “interference”—is simply the outcome of all this self-interested activity.

Bronislaw Malinowski was already complaining about this sort of thing in 1922, in what is arguably the first book-length work of economic anthropology: Argonauts of the Western Pacific. Such a theory would do nothing, he said, to explain economic behavior in the Trobriand Islands:

Another notion which must be exploded, once and for ever, is that of the Primitive Economic Man of some current economic textbooks... prompted in all his actions by a rationalistic conception of self-interest, and achieving his aims directly and with the minimum of effort. Even one well established instance should show how preposterous is this assumption. The primitive Trobriander furnishes us with such an instance, contradicting this fallacious theory. In the first place, as we have seen, work is not carried out on the principle of the least effort. On the contrary, much time and energy is spent on wholly unnecessary effort, that is, from a utilitarian point of view. (Malinowski 1922:60)

Malinowski takes up the example of the attitude Trobriand men have toward their yam gardens: the endless energies they pour into vying to make their garden the most tidy and attractive (effort that is in strictly “economic” terms entirely useless). The whole point of gardening was to show off how much effort a man could sink into it; as a result, half the yields ended up rotting for lack of anyone to eat them. What’s more, those that were eaten were not eaten by the gardener himself:

The most important point about this is, however, that all, or almost all the fruits of his work, and certainly any surplus which he can achieve by extra effort, goes not to the man himself, but to his relatives-in-law. Without entering into details... it may be said that about three quarters of a man’s crops go partly as tribute to the chief, partly as his due to his sister’s (or mother’s) husband and family (Malinowski 1922:60–61).

In other words, rather than “economizing” their efforts, Trobriand men are actively trying to perform unnecessary labor; then they give the products away to their sister’s families. There’s not even any direct reciprocity involved, since the man’s own family is fed not by his sister’s family but the brothers of his own wife.

Such examples could be multiplied endlessly, and, in the early days of anthropology, they were. It didn’t make much difference. Every decade or so has seen at least one new attempt to put the maximizing individual back into anthropological theory, even if economic theory itself usually ends up having to bend itself into ribbons in order to do so.

In fact, the effort to reconcile the two disciplines is in many ways inherently contradictory. This is because economics and anthropology were created with almost entirely opposite purposes in mind. Economics is all about prediction. It came into existence and continues to be maintained with all sorts of lavish funding, because people with money want to know what other people with money are likely to do. As a result, it is also a discipline that, more than any other, tends to participate in the world it describes. That is to say, economic science is mainly concerned with the behavior of people who have some familiarity with economics—either ones who have studied it or at the very least are acting within institutions that have been entirely shaped by it. Economics, as a discipline, has almost always played a role in defining the situations it describes.[2] Nor do economists have a problem with this; they seem to feel it is quite as it should be. Anthropology was from the beginning entirely different. It has always been most interested in the action of those people who are least influenced by the practical or theoretical world in which the analyst moves and operates. This was especially true in the days when anthropologists saw themselves as studying savages; but to this day anthropologists have remained most interested in the people whose understanding of the world, and whose interests and ambitions, are most different than their own. As a result, it is generally carried out completely without a thought to furthering those interests and ambitions. When Malinowski was trying to figure out what Trobriand gardeners were trying to accomplish in acting as they did, it almost certainly never even occurred to him that whatever that was, reading his book might make them better able to accomplish it. In fact, when an anthropologist discovers that anyone is using anthropological texts in this way—say, as a guide for how to perform their own rituals—they are usually quite disturbed.

Economics, then, is about predicting individual behavior; anthropology, about understanding collective differences.
As a result, efforts to bring maximizing models into anthropology always end up stumbling into the same sort of incredibly complicated dead ends. The classic case studies of economic anthropology, for instance—Franz Boas’ reports on the Kwakiutl potlatch (1897, etc.) or Malinowski’s on Trobriand kula exchange (1922)—concerned systems of exchange that seemed to work on principles utterly different from the observers’ own: ones in which the most important figures seemed to be not so much trying to accumulate wealth as vying to see who could give the most away. In 1925, Marcel Mauss coined the phrase “gift economies” to describe them.
Actually, the existence of gifts—even in Western societies—has always been something of a problem for economists. Trying to account for them always leads to some variation of the same, rather silly, circular arguments.

Q: If people only act to maximize their gains in some way or another, then how do you explain people who give things away for nothing?

A: They are trying to maximize their social standing, or the honor, or prestige that accrues to them by doing so.
Q: Then what about people who give anonymous gifts?
A: Well, they’re trying to maximize the sense of self-worth, or the good feeling they get from doing it.

And so on. If you are sufficiently determined, you can always identify something that people are trying to maximize. But if all maximizing models are really arguing is that “people will always seek to maximize something,” then they obviously can’t predict anything, which means employing them can hardly be said to make anthropology more scientific. All they really add to analysis is a set of assumptions about human nature. The assumption, most of all, that no one ever does anything primarily out of concern for others; that whatever one does, one is only trying to get something out of it for oneself. In common English, there is a word for this attitude. It’s called “cynicism.” Most of us try to avoid people who take it too much to heart. In economics, apparently, they call it “science.”[3]

Still, all these dead ends did produce one interesting side effect. In order to carry out such an economic analysis, one almost always ends up having to map out a series of “values” of something like the traditional sociological sense—power, prestige, moral purity, etc.,—and to define them as being on some level fundamentally similar to economic ones. This means that economic anthropologists do have to talk about values. But it also means they have to talk about them in a rather peculiar way. When one says that a person is choosing between having more money, more possessions, or more prestige, what one is really doing is taking an abstraction (“prestige”) and reifying it, treating it as an object not fundamentally different in kind from jars of spaghetti sauce or ingots of pig iron. This is a peculiar operation, because in fact prestige is not an object that one can dispose of as one will, or even, really, consume; it is rather an attitude that exists in the minds of other people.[4] It can exist only within a web of social relations. Of course, one might argue that property is a social relation as well, reified in exactly the same way: when one buys a car one is not really purchasing the right to use it so much as the right to prevent others from using it—or, to be even more precise, one is purchasing their recognition that one has a right to do so. But since it is so diffuse a social relation—a contract, in effect, between the owner and everyone else in the entire world—it is easy to think of it as a thing. In other words, the way economists talk about “goods and services” already involves reducing what are really social relations to objects; an economistic[5] approach to values extends the same process even further, to just about everything.

But on what basis? In reality, the only thing spaghetti sauce and prestige have in common is the fact that some people want them. What economic theory ultimately tries to do is to explain all human behavior—all human behavior it considers worth explaining, anyway—on the basis of a certain notion of desire, which then in turn is premised on a certain notion of pleasure. People try to obtain things because those things will make them happy or gratify them in some way (or at least because they think they will). Chocolate cheesecake promises pleasure, but so does the knowledge that others do not consider you obese; rational actors regularly weigh one against the other. It is this promise of pleasure economists call “value.”

In the end, most economic theory relies on trying to make anything that smacks of “society” disappear. But even if one does manage to reduce every social relation to thing, so that one is left with the empiricist’s dream, a word consisting of nothing but individuals and objects, one is still left to puzzle over why individuals feel some objects will afford them more pleasure than others. There is only so far one can go by appealing to physiological needs.[6] In the end, faced with explaining why in some parts of the world most people are indifferent to the pleasures of chocolate cheesecake but excited by those of salted prune drinks, or why in others obesity is considered attractive, economists, however begrudgingly, usually admit they do have to bring some notion like society or culture back in.

It was just these kind of issues that lay behind the Formalist-Substantivist debate that preoccupied economic anthropology in the 1960s (Polanyi 1957, 1959, 1968; Dalton 1961, Burling 1962, Cook 1966, etc.). Nowadays, most consider this debate to have been rather pointless—and indeed, the theoretical basis of both positions has been largely discredited—but the basic issues have never, really, been resolved. Let me provide a rapid summary.

The terms “formalism” and “substantivism” were actually both invented by Hungarian economist Karl Polanyi. Polanyi’s most famous work, The Great Transformation, was an account of the historical origins, in eighteenth and nineteenth century England, of what we now refer to as “the market.” In this century, the market has come to be seen as practically a natural phenomenon—a direct emanation of what Adam Smith once called “man’s natural propensity to truck, barter and exchange one thing for another.” Actually, this attitude follows logically from the same (cynical) theory of human nature that lies behind economic theory. The basic reasoning— rarely explicitly stated—runs something like this. Human beings are driven by desires; these desires are unlimited. Human beings are also rational, insofar as they will always tend to calculate the most efficient way of getting what they want. Hence, if they are left to their own devices, something like a “free market” will inevitably develop. Of course, for 99% of human history, none ever did, but that’s just because of the interference of one or another state or feudal elite. Feudal relations, which are based on force, are basically inimical to market relations, which are based on freedom; therefore, once feudalism began to dissolve, the market inevitably emerged to take its place.[7]

The beauty of Polanyi’s book is that it demonstrates just how completely wrong that common wisdom is. In fact, the state and its coercive powers had everything to do with the creation of what we now know as “the market”—based as it is on institutions such as private property, national currencies, legal contracts, credit markets. All had to be created and maintained by government policy. The market was a creation of government and has always remained so. If one really reflects on the assumptions economists make about human behavior, it only makes sense that it should be so: the principle of maximization after all assumes that people will normally try to extract as much as possible from whoever they are dealing with, taking no consideration whatever of that other person’s interests—but at the same time that they will never under any circumstances resort to any of the most obvious ways of extracting wealth from those towards whose fate one is indifferent, such as taking it by force. “Market behavior” would be impossible without police.

Polanyi goes on to describe how, almost as soon as these institutions were created, men like Smith, Malthus, and Ricardo appeared, all drawing analogies from nature to argue that these new forms of behavior followed inevitable, universal laws. It is the study of these laws that Polanyi refers to as economic “formalism.” Polanyi is perfectly willing to admit that formal methods are appropriate for understanding how people will behave within such a market. But in most societies, such institutions did not exist; one simply cannot talk about an “economy” at all, in the sense of an autonomous sphere of behavior that operates according to its own internal logic. Rather, one has to take what he calls a “substantive” approach and examine the actual process through which the society provides itself with food, shelter, and other material goods, bearing in mind that this process is entirely embedded in society and not a sphere of activity that can be distinguished from, say, politics, kinship, or religion.

The Substantivist school of economic anthropology (its chief exponent was Polanyi’s student, George Dalton) was thus basically empirical. One takes a given society, looks at how things are distributed, and tries to understand the principles. The main result was a list of new forms of exchange and distribution, all of which did not seem to operate on principles of economic maximization, to add to the gift economies with which anthropologists were already familiar. These included the notion of redistributive economies, the phenomena of ports of trade (neutral enclaves in which merchants of different countries could do business according to pre-established exchange rates: [Polanyi, Arensberg and Pearson 1957]), the notion of spheres of exchange (Firth 1959, Bohannon 1955, 1959; Bohannon and Bohannon 1968), and Marshall Sahlins’ spheres of sociability (Sahlins 1972).

All this was a definite contribution to human knowledge. The problem was the overall theoretical armature. It is one thing to say “societies” have different ways of distributing goods. It is another to explain what particular members of the society in question think they’re doing when they give gifts, or demand bridewealth, or exchange saffron for ivory in a port of trade. This is precisely what their opponents were quick to point out. Because almost immediately, the Substantivist challenge was met by a counter-offensive by selfproclaimed Formalists (for example, Burling 1962, Cook 1966). Formalists claimed that Polanyi had misunderstood what economics was actually about. It did not depend on the presence or absence of something called “the economy.” Economics was concerned with a certain type of human behavior called “economizing.” People economize when they make choices between different uses for scarce resources in an attempt to minimize their outputs and maximize rewards. (Yes, they said, this did involve some a priori assumptions about human nature, but everyone has to work from some assumptions: the ultimate test is whether the resulting theories produce results.) The point of social science is not comparing different forms of social system but understanding what motivates human beings to act the way they do.

Here they did have a point. For the most part, Substantivists didn’t try to explain anything; they just created taxonomies. Insofar as they did invoke a larger theory, it was generally some variant on Durkheimian functionalism. Where economists saw the shape of society largely as the outcome of individual decisions, Functionalists represented society as an active force in its own right—even, as something close to a conscious, purposive agent, though its only purpose seems to be a sort of animalistic self-preservation. For a Durkheimian, economic institutions can be seen as a means of social integration—one of the ways society creates a network of moral ties between what would otherwise be a chaotic mass of individuals—or, if not that, then at least the means by which “society” allocates resources. The obvious question is how “society” motivates people to do this. Without some theory of motivation, one is left with a picture of automatons mindlessly following whatever rules society lays down for them, which at the very least makes it difficult to understand how society could ever change.

Of course, the Formalists, as I have already noted, could not do much better. They were working with tools originally designed to predict individual behavior in a market setting; by twisting them around, they could sometimes predict the behavior of individuals in other cultures but not the values that motivated them, or, for that matter, the shape of society as a whole. At their most ambitious, a Formalist might try to demonstrate how, if one starts with a collection of people living in, say, Baluchistan, and a scattershot collection of “values” (food, sex, prestige, not being tortured in hell for all eternity, etc.,) one could then show how the existing shape of Baluchi society emerged as the result of the strategies people adopted to secure them. This is pretty much what Frederick Barth proposed anthropologists should do, calling his approach Transactionalism (1966; cf. Kapferer 1976). Transactionalism was probably the most ambitious attempt to apply the principles of formal economics to anthropology, and it caused something of a stir in the late 1960s. The obvious question though, was—even if it was possible to create a model that would thus generate the entire Baluchi lineage system or the structure of a West African kingdom from the right collection of values—what would be the point? What would one then know that one had not known before one started? The result would not even be an historical reconstruction, but a purely logical model that need have nothing to do with the actual historical origins of the societies in question.

Most anthropologists nowadays would wonder what the point is in even going into all this; the Substantivist-Formalist debate is considered definitively passé. But there is a point. It seems to me that these basic issues have never been resolved. Those who start by looking at society as a whole are left, like the Substantivists, trying to explain how people are motivated to reproduce society; those who start by looking at individual desires end up, like the Formalists, unable to explain why people chose to maximize some things and not others (or otherwise to account for qestions of meaning). In fact, though scholars have drifted off to other concerns, the same problems keep reemerging. As we shall see, a lot of what passes for the newest and most refreshing poststructural theory nowadays is largely warmed-over Transactionalism, minus the fancy economic formulas, with some even fancier linguistic formulas pasted on instead.

III: Structuralism and linguistic value

Linguists have long been in the habit of speaking of the meaning of a word as its “value.” From quite early in the history of anthropology, there have been efforts connect this usage to other sorts of value. One of the most interesting can be found in Evans-Pritchard’s The Nuer (1940:135–38): a discussion of the “value” of the word cieng, or “home.” For a Nuer, EvansPritchard notes, the “value” of this word varies with context; a speaker can use it to refer to one’s house, one’s village, one’s territory, even (when speaking to a foreigner) Nuerland as a whole. But it is more than a word; the notion of “home,” on any of these levels, also carries a certain emotional load. It implies a sense of loyalty, and that can translate into political action. Home is the place one defends against outsiders. So we are talking about value in the sociological, “values” sense as well. “Values,” Evans-Pritchard says, “are embodied in words through which they influence behavior” (135). Or, alternatively, the notion of “home,” when it serves to determine who one considers a friend, and who an enemy, in the case of potential blood-feuds, “becomes a political value” as well. Note here how “value” slips back and from “meaning” to something more like “importance”: one’s home is essential to one’s sense of oneself, one’s allegiances, what one cares about most in life.

This was a fascinating start, but it never really went anywhere. When a contemporary anthropologist speaks of the value of words, instead, they are almost invariably referring back to the ideas of Ferdinand de Saussure, founder of modern, structural linguistics.

In his Course of General Linguistics (1916 [1966]), Saussure argued that one could indeed speak of any word as having a value, but that this value was essentially “negative.” By this he meant that words take on meaning only by contrast with other words in the same language. Take for example the word “red.” One cannot define its meaning, or “value,” in any given language, without knowing all the other color terms in the same language; that is, without knowing all the colors that it is not. We might translate a word in some African language as “red,” but its meaning (or value) would not be the same as the English “red” if, say, that other language does not have a word for “brown.” People in that language might then be in the habit of referring to trees as red. The most precise definition of the English “red,” then, would be: the color that is not blue, not yellow, not brown, etc. It follows then that in order to understand the “value” of any one color term one must also know those of all the others in that language: the meaning of a term is its place in the total system.[8] Saussure’s arguments of course had an enormous impact on anthropology, and were the most important influence on the rise of Structuralism—which took off from Saussure’s suggestion that all systems of meaning are organized on the same principles as a language, so that technically, linguistics should be considered just one sub-field of an (as yet non-existent) master discipline that he dubbed semiology, the science of meaning.[9]

As these examples suggest, Saussure’s approach was more about vocabulary than grammar, more about nouns and adjectives than verbs. It was concerned with the objects of human action more than with the actions themselves. Not surprising, then, that those who tried to follow Saussure’s lead and actually create this non-existent science tended to be most successful when exploring the meaning of physical objects (Barthes 1967; Baudrillard 1968; Sahlins 1976). Objects are defined by the meaningful distinctions one can make between them. To understand the meaning (value) of an object, then, one must understand its place in a larger system. Just as the value of “red” is determined negatively, by all the other colors it is not, if one were to analyze the meaning of, say, a turtleneck worn underneath one’s jacket, one must examine the full set of other things one person might be wearing: that is, wearing a turtleneck means that one is not wearing a shirt and tie beneath one’s jacket, but that neither is one wearing a Tshirt, or nothing at all. Again, the meaning of one element makes sense only in terms of its contrast with other possible elements within the same system. This is a crucial consideration, because it means nothing can be analyzed in isolation. In order to understand any one object, one must first identify some kind of total system. This became the trademark of Structuralism: the point of analysis was always to discover the hidden code, or symbolic system, which (language-like) tied everything together.

Almost inevitably, though, the question became how to connect this sort of value to value in either of the other two senses. In the early days, when Structuralism was a new idea that seemed to offer resolutions for almost any outstanding problem in social theory, it seemed self-evident that it should be possible to do so. Hence Marshall Sahlins (former Substantivist, newfound Structuralist) concluded his famous analysis of the Western clothing system in Culture and Practical Reason (1976) by suggesting that one could only understand economic value, too, as the product of meaningful distinctions. To understand why people want to buy things, he said, we have to understand the place that thing has in a larger code of meaning.

Production for gain is the production of a symbolically significant difference; in the case of the consumer market, it is the production of an appropriate social distinction by way of concrete contrast in the object. The point is implicit in the apparent ambiguity of the term “value,” which may refer to the price of something or the meaning of something (as the differential concept of a word), or in general to that which people hold “dear,” either morally or monetarily. Anthropologists, incidentally, are quite familiar with this ambiguity, if not always entirely conscious of it, since many adopt it to illustrate the universality of rational economic behavior—even where market exchange is specifically absent. The people are nevertheless economizing their resources: it’s just that they are interested in “values” other than the material—brotherhood for example (1976:213–14).

So, value in each sense is ultimately the same, just as the Formalists were often forced to admit. Things are meaningful because they are important. Things are important because they are meaningful.

Sahlins goes on to observe that Saussure himself made a similar analogy, and suggests the passage in which he does so should be the basis for any future economic anthropology. To understand the value of a five-franc piece, Saussure had written, one must be able to understand (a) something different with which it can be “exchanged,” i.e., a loaf of bread, and (b) something similar to which it can be “compared,” i.e., a one franc piece, or other denominations of money.

In the same way a word can be exchanged for something dissimilar, an idea; besides, it can be compared with something of the same nature, another word. Its value is therefore not fixed so long as one simply states that it can be “exchanged” for a given concept, i.e., that it has this or that signification: one must also compare it to similar values, with other words that stand in opposition to it (Saussure in Sahlins 1976: 214–215).

Perhaps the best one can say about this passage is that in the flush of enthusiasm that often follows the discovery of powerful new techniques for understanding reality—such as Structuralism was in the ‘60s and ‘70s—there is good reason to put common sense on hold and see how far these techniques can take you. Still, this seems to be the point at which they reached their limits. I mean, really: what does it mean to say that when you use a word, you are “exchanging” it for a concept? In what way does this really resemble paying a shopkeeper for a loaf of bread? Most of all, what sort of “comparison” are we really talking about here? After all, when one observes that a loaf of bread costs five francs, and a steak-frites costs twenty, one is not simply observing that the bread and steak-frites are different. One is more likely to be emphasizing the fact that one is worth more. This is why one can say an element of evaluation is involved. This is also precisely what makes money unique—that it can indicate exactly how much more one is worth than the other[10]—and precisely what Saussurean models cannot account for. The latter provides a way to understand how the world is divided up, how objects are grouped into categories based on their differences with other sorts of object—and Sahlins is of course right when he says that in a consumer society, marketing is often a matter of creating symbolic distinctions between products that are otherwise virtually identical, like two different brands of corn flake or detergent—but this in itself does not explain why people are willing to spend money on them. People do not buy things simply because they recognize them as being different than other things in some way. Even if they did, this would do nothing to explain why they are willing to spend more on certain things than others.

On Structuralism, the results are by now more or less in. The general consensus is that its greatest weak point is evaluation. Many have pointed out for example that Structuralist literary critics have often provided brilliant analyzes of the formal principles underlying a novel or a poem, discovering all sorts of hidden patterns of meaning, but that they provided no insight at all into whether the novel or poem in question was any good. Similarly, Structuralist approaches in anthropology—as exemplified in the works of Claude Levi-Strauss (1949, 1958, 1962, 1966)—tend to focus on how members of different cultures understand the nature of the universe, and for this they can be remarkably revealing; but the moment one tries to understand how, say, one thing is seen as better—preferable, more desirable, more valuable—than another, problems immediately emerge. As a result, the great dilemma of Structuralism has been how to move on from understanding people’s passive contemplation of the world (Geertz’s “cerebral savage”), to their active participation in it.[11]

Actually, no one has done more than Marshall Sahlins toward thinking a way out of this box, often with spectacular results (Sahlins 1981, 1985, 1988, 1991). So perhaps I am not being especially fair to him in singling out this one, very early, text. But it’s also true that since, he has tended to abandon talk of value entirely. The only author who has made a consistent effort to develop a theory of value along Structuralist lines is Louis Dumont (1966, 1971, 1982, 1986). His work thus deserves more detailed consideration.

Dumont is of course best known for having been almost single-handedly responsible for popularizing the concept of “hierarchy” in the social sciences. His notion of value, in fact, emerges directly out of his concept of hierarchy.

Classical Structuralism, according to Dumont, was developed as a technique meant to analyze the formal organization of ideas, not values. Carrying out a structural analysis means, first, identifying certain key conceptual oppositions—raw/cooked, pure/impure, masculine/feminine, consanguinity/affinity, etc.—and then mapping out how these relate to one another, say, within in a series of myths or rituals, or perhaps an entire social system. What most Structuralists fail to realize, Dumont adds, is that these ideas are also “values.” This is because with any such pair of terms, one will be considered superior. This superior term always “encompasses” the inferior one. The notion of encompassment is in turn the key to Dumont’s notion of hierarchy. One of his favorite illustrations is the opposition of right and left. Anthropologists having long noted a tendency, which apparently occurs in the vast majority of the world’s cultures, for the right hand to be treated as somehow morally superior to the left (Hertz 1907, Needham 1973). In offering a handshake, Dumont notes, one must normally extend one hand or the other. The right hand put forward thus, in effect, represents one’s person as a whole—including the left hand that is not extended (Dumont 1983, see Tcherkezoff 1983.) Hence, at least in that context, the right hand “encompasses” or “includes” the left, which is also its opposite. (This is what he calls “encompassing the contrary.”) This principle of hierarchy, he argues, applies to all significant binary oppositions—in fact Dumont rejects the idea that two such terms could ever be considered equal, or that there might be any other principle of ranking, which as one might suspect has created a certain amount of controversy, since it pretty obviously isn’t true.[12]

So: meaning arises from making conceptual distinctions. Conceptual distinctions always contain an element of value, since they are ranked.[13] Even more important, the social contexts in which these distinctions are put into practice are also ranked. Societies are divided into a series of domains or levels, and higher ones encompass lower ones—they are more universal and thus have more value. In any society, for instance, domestic affairs, which relate to the interests of a small group of people, will be considered subordinate to political affairs, which represent the concerns of a larger, more inclusive community; and likely as not that political sphere will itself be considered subordinate to the religious or cosmological one, where priests or their equivalents represent the concerns of humanity as a whole before the powers that control the universe.[14] Perhaps the most innovative aspect of Dumont’s theory is the way that the relations between different conceptual terms can be inverted on different levels. Since Dumont developed his model in an analysis of the Indian caste system, this might make a good illustration. On the religious level, where Brahmans represent humanity as a whole before the gods, the operative principle is purity. All castes are ranked according to their purity, and by this standard Brahmans outrank even kings. In the subordinate, political sphere in which humans relate only to other humans, power is the dominant value, and in that context, kings are superior to Brahmans, who must do as they say. Nonetheless Brahmans are ultimately superior, because the sphere in which they are superior is the most encompassing.[15]

None of this, of course, applies to contemporary Western society, but according to Dumont, the last three hundred years or so of European history have been something of an aberration. Other societies (“one is almost tempted to say, ‘normal ones’”) are “holistic,” holistic societies are always hierarchical, ranked in a series of more and more inclusive domains. Our society is the great exception because for us, the supreme value is the individual: each person being assumed to have a unique individuality, which goes back to the notion of an immortal soul, which are by definition incomparable. Each individual is a value unto themselves, and none can be treated as intrinsically superior to any other. In most of his more recent work in fact (Dumont 1971, 1977, 1986) Dumont has been effectively expanding on Polanyi’s arguments in The Great Transformation, arguing that it was precisely this principle of individualism that made possible the emergence of “the economy.”

One could go further. In France, there is by now a Dumontian school of anthropology, largely made up of his devoted students, and its approach to traditional, non-Western societies (normal, hierarchical ones that is) is in many aspects a new form of Substantivism. If anything, it is more radical than the original in its uncompromising rejection of anything that smacks of methodological individualism.[16] The main difference is that it has tossed out the functionalist assumption that economic institutions act to integrate society, and put in its place the Saussurean notion that you have understand a total system of meaning in order for any particular part of it to make sense. Either way, it means that the first step in analysis is to identify some totality. The Dumontians call their project one of “comparing wholes,” by which they mean not so much symbolic systems as societies taken as totalities structured around certain key values. (Or, as Dumont puts it, “ideas-values.”)

Note how even in Dumont’s original analysis of India, the use of the term value covers quite a range. Purity, for example, is clearly a value of the “cultural values” variety with which Kluckhohn concerned himself, a conception of what people should want to be like; power, on the other hand, seems more like one of the values Formalists came up with when trying to account for what people actually seem to want, even if they don’t necessarily admit to it.[17] The claim is that both are ultimately “ideas-values” that can be analyzed in Saussurean terms, as part of an overall system of meaning.

The best illustration of how all this works itself out in practice can be found in a book called Of Relations and the Dead (1994), co-written by four of Dumont’s students: Daniel de Coppet, who worked among the ‘Aru’Aru in the Solomon Islands (1969, 1970, 1982, 1985, 1992), Cecile Barraud in the Moluccan village of Tanebar-Evav (1979), Andre Iteanu among the Orokaiva of Papua New Guinea (1983a, 1983b, 1990), and Robert Jamous among the Berbers of the Moroccan Rif. The idea of the book is to compare each society as a total system.

In every case, the societies turn out to be structured around two or three key values. The highest is the one that defines its members’ place in the cosmos as a whole. So among Jamous’ Berbers, while important men spend much of their time defending and increasing their honor through various forms of aggressive exchange, ranging from dramatic gift-giving to the exchange of violence in blood-feuds, honor is not the highest value. The highest is baraka, which can roughly be translated divine grace and is immanent in the holy men who resolve feuds and generally mediate human relations with God. In Barraud’s Moluccan village, life is taken up in a series of marital exchanges, but this takes place on what is ultimately a lower level of value called haratut—roughly, island society in relation to its own divine ancestors—while the most important level is that of lor, or “law,” in which the society of both living and dead is bound to other societies. The two Melanesian cases are even more complicated, since the values are not named—the authors make the interesting (and profoundly structuralist) suggestion that key values receive names only when a society is aware that other societies with different values exist. When it does not, members of that society will not distinguish the nature of their social order from the nature of the cosmos as a whole, and the values are seen as inhering in the very fabric of reality. Among the ‘Aru’Aru, for example, the three key values are embodied in three basic constituents of every human being: body, breath, and ancestral “image” (the last is the only element to survive a person’s death). These in turn correspond to the three most important objects of exchange: taro, pork, and shell money. According to De Coppet, ‘Aru’Aru ritual life is largely made up of an intricate web of exchanges, in which taro, pork, and shell money change hands as a way of building up and breaking down human personae, creating new ones with marriages or dissolving them in funerals, and, on the highest level, reproducing the relations between human beings and their ancestors.

In societies such as these, the authors argue, it is utterly absurd to talk about individuals maximizing goods. There are no individuals. Any person is himself made up of the very stuff he exchanges, which are in turn the basic constituents of the universe.

They also admit that all four societies have their “great men”—Melanesian “big men,” Bedouin “men of honor,” important lineage heads in the Moluccas—and that these are always, those who have achieved mastery of that society’s most important form of exchange. But the values they are trying to maximize are never the ultimate values of that society. Always, there are two levels, so that while on the lower one, the “values” involved may resemble the sort a Transactionalist might come up with—“honor,” “power,” wealth, etc.—on the most important level they are more values in Kluckhohn’s sense, ideas about what is ultimately important in life. So, they note, from the point of view of society, great men only exist in order to sponsor certain forms of cosmological ritual—ritual that in turn serves to reproduce society as a whole, along with its key values. While this somewhat contradicts Dumont’s own statements that the value he is dealing with has nothing whatever in common with the economic sort (economists look at preferences; hierarchical values are about intrinsic superiority), it means that this model does embrace all three of the main ways of talking about value; though strictly subordinating one to a synthesis between the two others. As for why great men perform these rituals: well, this is not for the purpose of self-aggrandizement, but simply because they believe it is the right thing to do.

More than any other approach, the Dumontians provide the promise of a grand synthesis of theories of value—in their case, through a sort of supercharged Substantivism. The question: at what cost? In order to do so, they have had to make a strict division between “modern” societies, in which people are individuals and seek economic values, and “holistic” ones, in which they do not. Hence there is a fundamental break between the sort of society in which most anthropologists live and the sort they study. Second, they reintroduce all the notorious problems of functionalism. For example, to speak of societies as “wholes” does seem to imply there are clear borders between them, and they exist in relative isolation.[18] Examining history shows this is very rarely the case. Even more, it becomes almost impossible to see how these societies can ever change. In fact, one of Dumont’s most notorious arguments is that the Indian caste system cannot, by definition, change. Its structure is fixed; therefore, it can either continue, or it can collapse and be replaced by an entirely different system: like a chair eaten away by termites, it will maintain the same form until it falls apart (1970:219). These are the main reasons why anthropologists rejected functionalism to begin with.

conclusions

At this point, the reader should at least have some idea of the history the term “value” evokes. It is a term that suggests the possibility of resolving ongoing theoretical dilemmas; particularly of overcoming the difference between what one might call top-down and bottom-up perspectives: between theories that start from a certain notion of social structure, or social order, or some other totalizing notion, and theories that start from individual motivation. Reconciling the two has been a perennial problem for social theory.

Of course, there are those who would question whether there’s much of a point in grand theory to begin with. Some scoff at the very notion, arguing that all anthropology really has to offer to the world is ethnography, the description of other societies and other ways of life. There is no doubt that this is a very important part of what we do: keeping a record, one might say, of cultural and social differences, a compendium of what being human has meant, in different times and places (and hence, perhaps of human possibilities). It is hard to deny that if anyone is reading our books, say, two hundred years from now, this is what they’re most likely to be looking for. The conventional reply is that every ethnography always implies a theory. Since even the most apparently bland and factual descriptions turn out to be based on all sorts of a priori assumptions about what is important and relevant, and these, on what human beings, or human society, are fundamentally about, the real choice then is between thinking about such questions explicitly, or leaving them implicit—in which case, one will inevitably end up drawing on one’s own culture’s unstated folk beliefs. The usual result is one or another sort of economism. And the more one deals with human motivations, the more of a problem this becomes. A more recent variant of this attitude, that it’s the “grand” part of “grand theory” that’s objectionable, has—as we’ll see in the next chapter—resulted in much the same problems.

Economics, of course, has a very clear notion of what it is trying to do, and of what constitutes a successful analysis (does it or does it not predict what happens?). One way to look at the history of anthropological theory is to ask the same question. What is it anthropologists in any given period were trying to figure out? The discipline has clearly gone through stages in this regard. At all stages one gathered data. But for a nineteenth century evolutionist, for instance, the point of gathering the data about a particular society was to determine where it stands in a grand historical series, and to discover how its existence might reveal something about the universal history of mankind. For a functionalist, it was a matter of showing how a given practice or institution contributed to social stability (which did carry with it the tacit but rarely stated assumption that without such institutions society would collapse into some kind of Hobbesian chaos). For a structuralist, the point of analysis was to show how social forms were made up of symbolic elements that hang together as a total system of meaning. For all, however, the ultimate point was the same: to delineate some kind of logically coherent system, which meant moving away from individual action—and, in doing so, left the empty space into which economistic theories were always trying to crawl.

By the early ‘80s, there was a general consensus that this was the great problem of the day: how to come up with a “dynamic” theory of structuralism, one that could account for the vagaries of human action, creativity, and change. The way it was usually phrased was as a matter of moving from langue to parole, from language (“the code” of meaning, however conceived) to speech. It was at this point that value really came to the forefront of intellectual debate. For reasons that should be obvious by now, a theory of value seemed to be just what was needed to bridge the gap: to bring together society and human purposes, to move from meaning to desire.

It is interesting that in these arguments, virtually no one mentioned the legacy of Kluckhohn. His work was considered definitively outdated. If anything, this shows the extent to which structuralism really has come to set the terms of debate. However primitive the models Kluckhohn actually produced, he did at least open up the possibility of looking at cultures as not just different ways of perceiving the world, but as different ways of imagining what life ought to be like—as moral projects, one might say. This was so far from the approaches most theorists were starting from that it seemed utterly irrelevant.

Anyway, all this perhaps provides an explanation for both the continued popularity of the term “value,” and the lack of a concrete theory behind it. Anthropology didn’t really resolve the dilemmas of the early ‘80s. For the most part, it just skipped over them. The discipline moved on to other issues: concerning the politics of ethnographic fieldwork, memory, the body, transnationalism, and so forth. Structuralism faded out of prominence, then gradually came to seem ridiculous; theories that concentrated on power (Foucault) or practice (Bourdieu) largely replaced it; there was (and is) a general feeling that the debate was over. Hence, the tendency to act as if such a theory does, in fact, exist.

As we will see in the next chapter, though, most of the new theories that seem to have made the old arguments irrelevant are, at least in many of their aspects, little more than retooled versions of the same old thing. Nor do I think that ignoring the problem is necessarily the best way to make it go away.