In The Idea of a Social Science Winch, argues that, sociology is more properly conceived as a branch of philosophy than of empirical science. Winch falls victim here to the Humean assimilation of the empirical to the generalizable. He notes that much of our talk about social practice is in terms of conventions, so that explanations of social action can be given without recourse to statistical or experimental findings. But such talk depends nonetheless on the accuracy and detail with which (...) the situations in which actions occur are? recorded, and this is surely an empirical enterprise. It is the misleading conception of sociology as a discipline, characterized by common procedures, that leads Winch to espouse the assimilation of sociology to conceptual inquiry. We need to see instead that sociology embraces a group of questions and subjects so loosely connected that it would be mistaken to speak of, and idle to project, a procedure common to all of them. (shrink)