Middle-range Theories and the unification problem in social science

Résumé En

This chapter aims at evaluating the relevance, within the current philosophical and sociological contexts, of the search both for "middle-range theories" (as Robert Merton famously called them) and for a unification of these theories. In his time Merton's view was, on the one hand, a crucial complement to both quantitative and qualitative studies, remarkably illustrated by his colleague at Columbia, Paul Lazarsfeld, and, on the other hand, an alternative to "Grand Theories" such as Talcott Parsons', whose inconvenience was to be too abstract at the risk of being merely verbal. Many contemporary theories in sociology have similar flaws. They are either purely statistical surveys or mere ethnographic case studies, on the one hand, or all-encompassing theories, on the other hand. But not only have middlerange theories been recently rediscovered in the social sciences, some leading figures in philosophy of science have even argued that biologists and physicists regularly use similar methodologies very fruitfully even though they are not fully aware of doing so. Although these philosophers have sometimes gone so far as to support the value of "disunity" in science, which may result from the coexistence of local or partial "middle-range theories", I want to make a plea for unifying middle-range theories in sociology which do not fall back in the false abstraction of Grand Theories.