28. Choices About Choice

One pillar of Cold War philosophy is the use of rational choice theory as a fundamental account of human rationality—the equivalent, in weird ways, of Kantian critique. Do I choose to repeal rational choice theory? I do. But no repeal without replacement. So with what might I choose to replace rational choice?

According to rational choice theory, the chooser first establishes a set of feasible alternatives, sequences of events that can be triggered by her action. She ranks them according to how they contribute to her overall utility, and then opts for the highest.

She is thus dissociated from the alternatives themselves. She makes them into alternatives in the first place (by evaluating their “feasibility”), and can then choose or not choose any of them. She can even choose none of them and walk away from the game altogether, because a game is —only a game. A choice is “rational,” then, in the context of a game, i.e. when you are able not to choose any or all of the alternatives.

Some choices are indeed game-ish. I leave the drugstore with the toothpaste I chose to buy, and the voting booth having chosen for whom to vote. But viewing choice in terms of games also has limits. Some impressive philosophers thought that it presupposed an abrogation of causality that cannot be explained without heavy metaphysics (Kant) or theology (Augustine, Descartes).  Moreover, our important decisions are sometimes not made that way.

This leads some people to say that they are not made at all. I saw a movie about Le Chambon-sur-Lignane, the French village that over the course of the Nazi occupation managed to hide more Jews than there were people in the village. An elderly couple was asked how they decided to do this, and the wife said, shockingly for me: “on n’a jamais vraiment décidé:” “we never really ‘decided.’” She went on to say that, as French Protestants, they knew what it was to be persecuted.

My Hollywood friend Bernie Gordon had more movie credits restored to his name after the blacklist ended than even Dalton Trumbo. He tried to explain why he decided not to name names, which would have saved him, and he too said there was no such decision: It just wasn’t something I was going to do (paraphrased).

These people did not make rational choice decisions. They did not first formulate various alternatives and then opt for one. They simply did what they had to do, once it was clear what that was. Doing that was somehow necessary for them: the old couple had to be French Protestants; Bernie Gordon had to be Bernie Gordon. They were not obliged to do so; they simply could not do otherwise. It’s not that they identified with one of the alternatives before them; they were identified as one of the alternatives.

Here then is another view of choice: that it is the recognition of necessity. The “moment of decision” is when you realize that this is what you have to do, whether there are feasible alternatives to it or not. The “decision” was made before you were aware of it.

This view of choice has an impressive philosophical pedigree: start with Hegel, Spinoza, and Aristotle. It has also attracted notice from neuroscience, especially via the Libet experiments, which are said to show that decisions are made before we are aware of them. There have been numerous attempts to rescue freedom, in the rational choice sense, from Libet’s work. The rescues are sometimes successful: I bought Crest instead of Colgate just the other day. But I do not see how Bernie Gordon could have chosen to name names and still be Bernie Gordon. Neither did he. And those decisions, the ones that bear on who we really are, are the important ones.

So I want to repeal the view that the kind of choice involved in rational choice theory is the only kind, and replace it with a pluralistic theory of different types of choices.  I choose to have a choice among theories of choice

But I would like, someday, to get rid of the rational choice model altogether. I’m worried about the metaphysical lifting it requires—and the political downfalls it brings.