What Counts as Data?
IN SOCIOCULTURAL ANTHROPOLOGY, when we study culture, we often study form and not content. We study the representation of kinship, the imagery of the ordered social relationship, but not actual biological relatedness. We shy away from a discussion of the nature of madness to look at the way madness is shaped by local culture—the way it has been named, defined, treated, responded to. That is what our theory invites us to do. In David Schneider's famous first line, “This book is concerned with American kinship as a cultural system: that is, as a system of symbols” (1980 :80). In that theory, we mean by “culture” the categories a society generates around and out of its social order—concepts of witchcraft, symbols of divinity, images of the bad and the good. The definition of culture as concepts and categories which express and maintain the social order first emerged out of the early seminar room discussions in the era of British structural functionalism. It soon became instantiated in American anthropology and indeed became a professional credo of the American style of anthropology, as Talcott Parsons divided up the responsibilities of the social sciences. When Clifford Geertz, borrowing from Clyde Kluckhohn, asserted that symbols were models of and models for reality, he was enacting Parsons' division of intellectual mission from the great mélange of Harvard's Department of Social Relations: mind to the psychologists, social structure to the sociologists, culture to the anthropologists. For the British, sociocultural anthropology retained the responsibility for both social structure and culture, but still in British anthropological theory, culture remained a thing of concepts and categories, signifiers rather than the signified.
Yet what anthropologist does not have a story of his or her own stunned as-