Methodological individualism


I'm getting worried about methodological individualism. Yes, I know that 'society' has no life or will or organizing mind of its own, as Marx seemed to assume, and that it is just the aggregation of individuals' decisions and actions. I know that the 'price level' does not affect 'aggregate supply' or 'aggregate demand', and that these are mere statistics, summing individuals' reactions to particular prices. And I don't fall the the scientist guff that 'we can predict the behaviour of a piece of a gas, even though we don't know what any particular molecule is doing', because I know that the 'molecules' that social science deals with are individuals who are themselves so complex that their behaviour would fry the brain of the average chemist. And yet...

Former UK Prime Minister Margaret Thatcher famously (or infamously) told Women's Own magazine that "there is no such thing as society", and yes, I see her point. But she went on to say: "There are individual men and women, and there are families." Aye, there's the rub. Are we methodological individualists (the term was, I think, coined by Schumpeter, who I wrote about here recently) obliged to insist that everything comes down to the minds, thoughts, values, and actions of individuals alone? Or can we admit that relationships between individuals, like family ties, are pretty basic too? And what about culture, or history, or religion, or even class? These all shape and constrain our individual thoughts and actions. But to admit them as significant is the thin end of the methodological wedge, because these are social phenomena.

An analogy, if I may. A physicist could describe a football match in terms of kinetic energy, friction, and the forces on the ball that sent it in this direction or that. It would be a perfectly correct description, but a pretty dull one: most of us would prefer to hear the commentator talking about the skill of the players, the positioning of the teams, the tactics and strategy, the chances taken and the goals scored. The physicist's account might be the right way to talk about the workings of the Large Hadron Collider, but it's not much good for a ball game. Likewise, an individualist account of economic or social phenomena may be true in a trivial sense; but to understand what's going on, you do need to know that culture, or history, or religion do in fact shape how people act.

And again, if we do detect statistical relationships between social phenomena like a price index and a money supply figure, isn't that actually rather useful, even if only up to a point? Yes, I know that unless we refer to the individuals, we will make mistakes. A Martian observer may note that every Monday to Friday morning, Grand Central Station becomes packed with Earthlings, and predict this as a scientific law. Except that, one Monday, no Earthlings show up at all. The Martian's 'law' did not account for the fact it was a public holiday. But then, this is how science works – we make a hypothesis, then have to revise it when the unexpected happens. Sure, if we understand the motives of the actors, our predictions will be better. But just because we can't do that very easily, do we still have to throw out statistics that seem to work?