What is a Social Structure? Please, anyone!!

Over the years I've become convinced that this is one of the central questions in all social sciences. If I didn't know it would be a counterproductive strategy, I would simply stop everything else to work on it. I think this is the heart of the distinction between "Continental" and "Anglo-Saxon" social theory and of many other crucial questions. Problem is, I don't think I'm getting any closer to a satisfactory solution. This is really the first of a bunch of posts that are little else than a cry of despair.

Human life (and, for that matter, non-human life as well) flows through structures. In my University, we just changed President, recently in our school we changed Dean and we will soon have a new Department Chair. No one believes that these new individuals will remake the University, the School or the Department in ways that will make the current ones unrecognizable. We experience our surroundings, the institutions around us, the organizations to which we belong, even our familiar life as structures where individuals are replaceable at least to some degree. 

This is nothing new. Phenomenologists have made a living out of this (at least since Berger and Luckmann if not since Husserl himself), this is Sociology 101. 

The problem is that, to my knowledge, we don't have:

a. A good definition of what IS a structure.

b. A good account of how structures emerge, change and collapse.

I'm enough of a rational-choice theorist to try a crack at b. In that capacity, I could try to show that structures are necessary for social life because they, for instance, reduce our cognitive demands. In a world of bounded rationality, it is easier to navigate our social milieu if we are surrounded by structures so that we don't have to start from scratch everything each time the teller at a bank calls in sick. But this may explain why we need structures, not how they emerge. Similarly, Searle has argued that institutional facts (a topic akin to what I have in mind) operate on the basis of our Background abilities. (Here "Background" is capitalized because it is a technical term for Searle.) But, again, this doesn't tell us why we have this structure or that one, and not the other one out there. 

Continental social theorists may say that the very question is unintelligible. After all, they point out that these things exist, in the elegant phrase, "always already." In other words, it may seem futile to try to DERIVE a structure from something else because everything exists within a structure. There is nothing prior to a structure. For people who believe this, and, to be sure, it is a good point, the exercises we do in economics of imagining exchanges between two agents, economies with one or two goods, two or three traders and so on, are pointless because they will never be able to generate a structure; those traders, agents, citizens, voters, what have you, always already exist against a structure that makes them traders, agents, citizens, voters and what have you.

Granted. But if we take this to its logical conclusion, it means that we can never offer an account for the emergence of structures. I can offer a "genetic account" that is, I can tell how one specific structure in history developed out of another one. But notice what we are giving up if this is all there is to it: we are giving up the possibility of knowing the roads not taken. A full understanding of a structure should tell us what other possible forms it might take and why it doesn't take them.

Let me illustrate with a nagging question I've been pondering about. Bourdieu convinced me that power is the central social phenomenon. And, sure enough, power always exist within a structure. There is no power outside of a structure, except perhaps for the power of brute force. So, as far as I can follow it, the literature influenced by Bourdieu has taken up the task of describing how power is transformed from one type of capital to the next, from one field to the next. I like the way Bourdieu makes power fungible so that you can transform physical capital into symbolic capital and into political capital and so on. He does it much better than what I just said, of course. 

But the question that, as far as I can tell he cannot answer is: what determines the total amount of power/capital? Is power like matter-energy that it can only be transformed? That seems unlikely. More likely is that different structures generate different amounts of power. I surmise that modern bureaucracies have enormous levels of power that were undreamed of in previous eras. But if we want to know how this happens, we need to know how is power generated, how is capital generated. But this is not a historical question. It is an analytical one. We need an analytical account of how structures come into being to answer this.

It's getting late and I'm not coming across as lucid as I thought I would. Anyway, I'm sure I'll keep wrestling with this for a long time so there's no harm in stopping now.