So what is class? There is, to say the least, no single answer. The sociological debates (joined as well by some economists, political scientists, historians, and others) take place along several major axes. There is first of all a split between the so-called bourgeois theorists and the Marxists. Broadly speaking, bourgeois theorists in one form or another treat class as "stratification" -- as a set of differential positions on a scale of social advantage -- rather than as a set of fundamentally conflictual relations. Marxists, by contrast, work from a theoretical model in which classes are not merely sets of differentially successful people, but are derived from the specifically exploitative form of production that is capitalism, and are inherently antagonistic.

    Further debates are visible within each camp. The bourgeois theorists tend to split among themselves between those who think class should be defined by objective indicators (income, occupation, education, etc.), and those who think class should be defined in terms of how the natives themselves create social rankings, that is, in terms of something like "status"....

    The Marxists have their own splits. As Eric Olin Wright puts it, there are people who are interested in "class structure", as against the people who are interested in "class formation". Wright himself falls largely on the "class structure" side, concerned with the ways in which capitalism is or is not functioning the way Marx thought it did or should, and the implications thereof. Some of the major issues here include the implications for capitalism (as well as for Marxism and for social transformation) of the growth of the salaried middle class, of the different forms of the state, and of the different modes of relationship between state and economy. The "class formation" thinkers, for their part, are primarily concerned with the problem of how and why classes (normally, the working class) do or do not come to be self-conscious political actors....

    Let me approach this range of theoretical and methodological perspectives obliquely, by restating the problem for the present paper. It is well known that American natives almost never speak of themselves or their society in class terms. In other words, class is not a central category of cultural discourse in America, and the anthropological literature that ignores class in favor of almost any other set of social idioms -- ethnicity, race, kinship -- is in some ways merely reflecting this fact. Paul Fussell speaks of the discourse of class in America as being under a "taboo"....

    At the same time, it is clear that class is a "real" structure in American society, whether it is recogized in native discourse or not. Part of this reality is what is described in the classic Marxist account of differential relations to the means of production; some people own most of the major systems of production of wealth in America, while others produce that wealth yet garner for themselves only a small part of its value. Another part of the reality of class is one that is increasingly talked about in neo-Marxist discussions about the salaried middle class: power (administrative, regulatory, etc.) over other people's lives, whether one owns a piece of the means of production or not. And part of it, discussed most directly in studies of blacks and other poor minorities, has to do with discrimination, prejudice, stigmatization, and pain. Indeed, if one asks which aspects of the reality of class are displaced into the discourses of ethnicity, race, and gender, they are really largely the second two dimensions just noted. That is, if Americans can be said to have a discourse of class at all, it is, like that of both Marx and the bourgeois theorists, an economistic one: Americans have a discourse of money. What are not represented by the folk, and only fragmentarily represented by the class theorists, are both the power and the pain of class relations.

    It is important to note here that I take the position that class is not the only such "objective" structure of domination, that it is no more or less real than a number of others, and that it should not be constructed as more fundamental. Further, it is not distinguished from other such structures as being somehow more "material": all structures of domination are simultaneously material and cultural....

    Before going on, a few notes on the terminology I will be using and the assumptions about class that lie behind it. I begin by assuming that something we call "capitalism" has a massive reality, as a set of discourses, practices, and institutions in the world. Capitalism, in turn, is indeed central to (but not exhaustive of) the genesis of social positions of wealth and power on the one hand and poverty and social weakness on the other. When talking about class in an objectivist sense, the social positions generated by the organization and logic of capitalism may in turn be usefully clustered and glossed as "classes". Thus within such a perspective, it is possible, and it makes sense, to talk about something like "the changing structure of class" in the United States, or globally, under "late capitalism". It is highly relevant to my project, for example, that a number of social and economic theorists have argued that "the middle class" in the last two decades has been coming apart in the middle, that the rich are getting richer and the poor poorer, that there is a certain "caste-ification" of the upper middle class going on, and so forth.

    Quite apart from all this is the question of class as an identity. Here we are back in the realm of discourse, that is, of how people talk about themselves and others, and of the larger shape of the discursive field from which people draw their categories. This is, for the most part, the level at which I will operate in this article, the level of culture, ideology, discourse, habitus.... There is in fact a fair amount of slippage between the folk categories and the objectivist categories, an interesting and difficult problem in and of itself that I must set aside because of constraints of space.

    We may start at the lower end of the social scale. The choice of terms is immediately loaded. The phrase "working class" is rarely used and, indeed, usually rejected by people we might think of "objectively" as members of the working class. The chemical workers in Elizabeth, New Jersey, so brilliantly studied by David Halle, for example, think/speak of themselves as "working men" but reject the phrase "working class". In class language in fact, they think of themselves as "middle class", in contrast to people below them on the social scale, whom they designate as "lower class". The structure of the discourse, moreover, is heavily racially loaded: the "working class" (in social science/Marxist discourse) who call themselves "middle class" (in American cultural terms) are coded white; the "lower class" ( in American cultural language, more or less coterminous with certain objectivist usages) are coded as racial and ethnic minorities.

    In the terminology I will be using, the largely-white-working-class-that-thinks-of-itself-as-middle-class will normally be labeled "lower middle class", for reasons I have discussed elsewhere. I will use "lower class", in an extended version of the folk sense, to refer to the segment of the working class that is usually poorer and usually non-white, compared to the "lower middle class"....

    Turning to the middle class, it iscommonplace that the vast majority of Americans think of themselves as "middle class". There is also a folk lexicon of subdivisions for this category -- "upper middle", "lower middle", and just plain "middle" -- and American natives will occasionally use these terms, though in general they do not like to subdivide the middle-class category. Further, the term "lower-middle class" is very much disliked by those who might be so categorized, apparently because of the presence of the word "lower"; "lower-middle class" is to "(upper) middle class" as "lower class" is to "working class".

    The plain "middle class" is the most slippery category. It is either used as the modest self-label for the upper middle class, which quickly gives way to an acknowledgement that one is probably "upper" middle class after all; or it is the covering label for the lower-middle class as just discussed. Either way there is almost no "there" there; to be plain middle class is almost to be "really" something else, or on the way to somewhere else. At the same time, the "middle class" is the most inclusive social category; indeed, it is almost a national category. In many usages it means simply all those Americans who have signed up for the Amercian dream, who believe in a kind of decent life of work and family, in the worth of the "individual;" and the importance of "freedom", and who strive for a moderate amount of material success. It is everybody except the very rich and the very poor....

    Crosscutting these racial and ethnic categories in turn was the virtually unmarked, though never completely invisible, divider of class.... Here I will simply say that the project focuses on various fractions and transformations of "the middle class", and I will employ the standard American folk categories to label them. Thus Wequahic HS in the fifties covered the segment in the class spectrum that includes the "(middle-) middle class" and the "lower-middle" (or "upper-working") class. The better-off families of the "upper-middle class" tended to move away to the suburbs, while the lower end of the working class, and anything below that, was largely located in other neighborhoods.

    Class differences can be tracked retrospectively in a variety of ways: by parents' occupations, by whether the parents owned or rented their houses, and by the elementary school attended (which in turn indexes the microneighborhood the person grew up in within the larger Weequahic neighborhood). The elementary school attended, and its corresponding neighborhood, tended to be the marker most consciously noticed among the kids. While in a general way there were probably more Jews in the upper half of this basically two-tier system, and more non-Jews (both white and African American) in the lower half, in fact there were divisions within all groups, including -- although in some ways more hidden -- among the Jews. People in the lower tier, I came to realize, were often acutely aware of where they were, and the largely casual exclusions and snobberies exercised by those in the upper tier were still -- years later -- intensely felt.

    Class in turn underlies, though never fully exhausts, the two major social structural formations found in most American high schools at least since the fifties. The first is the categorical division between the kids who were considered respectable citizens of the school and the kids who were considered relatively alienated troublemakers.... In Weequahic there was no general name for the respectables, who I think tended to be equated with the Jews. Troublemakers, or people who adopted a certain oppositional style, were called "hoods"; in practice, these were drawn largely from the white, non-Jewish ethnic groups.... The ethnicizing of these categories in Weequahic did not negate their strong, although unmarked, class alignments either.

    Class also underlies, but does not fully determine, those other ubiquitous groupings in American high schools, the "clicks" (cliques, but always pronounced "click"). Clicks were subsets of groups of friends predicated on other markers of success in America -- physical attractiveness, money, "popularity", athletic ability, leadership/political savvy, and so on.... Clicks are specifically sets of the self-styled coolest (I cannot really think of a better word) kids, those with the largest amounts of these various forms of cultural and material capital....

    High schools are peculiar in forcing all of these socially differentiated people to be together in one place and to more or less deal with one another on a daily basis. But the various divisions of ethnicity, race, class, and class subsets like clicks and hoods lay down lines that will define the dispersal patterns of this community, once people are liberated from coresidence....