Abstract
In this paper, I address the concept of action, one of the most contested categories in sociological theory. The historical analysis takes Weberian and Parsonsian theories as its point of departure.
Their failure to develop a consistent action frame of reference have stimulated theorists who attempted to elaborate synthetic general theories of action; amongst them the structuration theory of Anthony Giddens, the multidimensional theoretical synthesis of Jeffrey C. Alexander, and the theory of communicative action of Jürgen Habermas, were perhaps the most prominent.
The intention of all these authors is to interconnect both the perspective of the systems analysis and the perspective of the action paradigm. The fact that the systems paradigm and the action paradigm have followed its disarray in social inquiry.
Although they are aware of the deficiencies of earlier approaches, they also fail in transcending the subject/object, action/structure, observer/participant dualisms in their generalised categorical schemes. The basic premise of action that the actor "could have always done otherwise" cause that all attempts at synthesis have to take account of the voluntaristic aspects of action.
Even if they were successful in transcending those dualisms they tried to avoid, there would be one apparently residual dualism of possible/actual actions, which they were not able to address properly. Their presuppositional framework of categories based on action can, in the logic of their argument (in most part proceeding through mere conceptual innovation), only explain, or describe, possible actions, not the actual ones.
However, although action is an open, inconsistent and contradictory concept, it does not follow, I try to argue, that it is in principle untheorizable. And that the way out of the dilemma of action does not lead us to its dismissal, but to more attempts to reconstruct it.