There’s an important way in which sociological theory is unavoidable. I mean this in the sense in which Alexander describes the problems of action and order as non-optional: “every theory takes some position on both” (Alexander 1987: 12). This is an empirical statement about sociological practice as much as anything else. Against those who dismiss ‘general theory’ as unnecessary, I’d argue that the sorts of questions encompassed by it inevitably emerge in any kind of sociological reasoning – to talk of a general problem of order is simply to affirm the (potential) value of our treating those questions we can’t avoid in a reflective and systematic way.
If we assume there are patterns to social life susceptible to explanation then we are necessarily committed, however inchoately, to a view of these patterns as having some ontological status. The possibility of a pattern presumes a distinction between identifiable regularities and some broader sum of activity in relation to which the regularity is (potentially) identifiable. This distinction in turn invites questions of the relationship between the former and the latter: how does (identifiable) order emerge out of (observable) activity? Once we ask this question, we’re effectively talking about structure and agency. This doesn’t commit us to any one understanding of ‘structure’, ‘agency’ or the relationship between them. But it does leave us within the space of questions which the discourse of structure and agency attempts to treat systematically. The notion of a ‘space of questions’ I’m invoking doesn’t imply deterministic constraints. It creates openings to escape the space of questions by seeking to transcend the dichotomies encountered e.g. both Bourdieu and Giddens seek to do this in different ways. We can find alternative ways to characterize the space of questions. What I think of as structure and agency seems to be seen by Jeffrey Alexander in the 1980s as two related questions: the problem of order and the problem of action.
On my understanding, sociological theory should be concerned with the systematic elaboration of this space of questions with a view to the amelioration of problems that impede empirical research and the construction of conceptual tools which contribute to it. This entails ‘translation’ work in order to bring divergent perspectives into dialogue with each other. It invites empirical work looking at how theoretical ideas are applied in practice and their contributions (or lack thereof) to empirical research. It invites conceptual work to establish meta-evaluative criteria upon which to establish what constitutes a contribution to sociological theory and what does not.
In essence I’m arguing for the unavoidability of sociological theory on praxeological grounds. I’m suggesting that certain presuppositional categories are intrinsic to sociological practice as a purposive activity orientated towards particular classes of objects. As Alexander puts it, “the real world puts terribly strict limits on our theorizing” (Alexander 1987: 5). We encounter these ‘limits’ through purposive activity of a very particular sort and I’m interested in how the activities which I’m subsuming under the category of ‘sociological practice’ mediate our encounter with these limits on theorizing. I share Alexander’s view that theoretical reasoning has “relative autonomy” in relation to the real world but I want to understand in a much more substantive way how relative it is. My underlying claim is that the theoretical constructs I’m invoking (‘structure and agency’, ‘the problem of order’, ‘ the problem of action’ etc) are systematic articulations of issues raised by a practical engagement with the social world motivated by specific purposes. We are drawn to take positions and/or make assumptions on these issues as a necessary condition of sociological practice. So why not try to do this systematically?