SAMFUND? Durkheim revisited i Amazonas og videre
“Society” appears a difficult notion. We use it all the time. But is it any good as an analytical concept? Sociologists seem to agree it is not. Few societies have the empirical characteristics of the bounded entity that structural-functionalist theory assumed. Constructivist notions of society as “imagined community” appear to be tied up with the existence of the State or with the spread of information technology. This leaves contemporary anthropology with “society” as a residue, the left-over from culture’s gluttonous theoretical supper. Still, social science aims to explain or understand social relations, interactions, and the processes by which structures and functions are worked into social systems as implied by the notion of society. The notion of society allows us to assume the existence of objective structures of order in the social life of people. Unlike the notion of culture, however, the notion of society has not been critically scrutinized by anthropologists. In contemporary Danish anthropology with its focus on culture and cultural representations, writers tend to simply take society for granted as the intrinsic empirical context of culture. From the perspective of Durkheimian notions of “the social”, the paper provides a brief review of interpretations that retrospectively have appeared analytical dead-ends. The author goes on to suggest that the notion of “symbolically generalized media of communication” may offer a productive opening that embraces both sides of the culture/society dichotomy in the search for structured systems of social existence whether subjectively or objectively conceived. The idea of “symbolically generalized media or communication” was originally formulated by Talcott Parsons and subsequently reworked by German sociologist Niklas Luhmann. Rather than an interrelated series of parts that make up a whole plus something else in the classic Durkheimian sense, society from this perspective appears in the form of structured sets of actions oriented by a horizon of possibilities and expectations, symbolically constituted, yet always provisional and emergent. Inspired by analyses of two different cases in Amazonian research the paper offers a brief hint at how the notion may be employed in anthropology.