Abstract
Most contemporary studies of science operate with some notion of scientific specialty communities as the basic units within which science is socially and technically organized. This paper presents a critique of scientific communities as sociological constructs which appear to be largely irrelevant to scientific work. Furthermore, the paper criticizes the prevailing quasi-economic models of such collectives for what appears to be a naive internalism and functionalism compared with the realities of scientific everyday life as they concern scientists themselves. It is argued that the arenas of action within which scientific (laboratory) inquiry proceeds are transepistemic — that is, they in principle include scientists and non-scientists, and encompass arguments and concerns of a `technical' as well as a `non-technical' nature. The paper also argues that the transepistemic connection of research is built into scientific inquiry (and thereby into the products of research) through the decision criteria invoked in laboratory work. The paper draws upon one year of observation in a scientific laboratory in Berkeley, California, which provides the grounds and the illustrations for the theoretical arguments presented.