Reflective equilibrium

What Rawls says is roughly this: the role of these reflections is to provide a way of organizing our moral intuitions about political life. Our basic ideas about politics are disorganized and often inconsis­tent.

The difficulty with this view is not with the idea of reflective equi­librium; it is rather with the particular sort of theory that Rawls wants to bring into equilibrium with our intuitions. For unless there is some reason to think that the general theory supports the partic­ular claims that are derived from it, there is no reason to eliminate from our inconsistent set of intuitions just those that don't fit with the theory.

Suppose we have a theory, T₁, from which we can derive all our moral intuitions except intuition I. We can always construct a dif­ferent theory, T₂, from which we could derive all our moral intu­itions, including I, except those that are inconsistent with I. (To do this, we simply look at the class of possible worlds in which T₁ is sat­isfied—the class, W, of worlds that are morally good according to T₁—and construct T₂ as a theory that is satisfied in all the members of W in which I is true.) We cannot say I is to be rejected because it cannot be derived from a theory: it can be derived from T₂. True, T₂ may not deliver some of our other intuitions, the ones inconsistent with I, but since our intuitions are inconsistent, we have to give some of them up anyway.

We can apply this analysis to Rawls' argument. Consider Jerry, who is a utilitarian. He derives his ideas of justice from considering what will maximize human utility: call this view T₁. Rawls advocates the two principles, deriving them from his bargaining game: call this T₂. These theories both fit with our moral intuitions in many cases, as Rawls would admit. Consider now some intuition that is derivable from Rawls' theory, T₂, but not from Jerry's theory, T₁: the intuition, say, that it is right to limit the income of the richest person in order to make the poorest off slightly better off, even if the result is to make everybody in between much worse off also. Now, as I have just argued, to accept this intuition on the basis of T₂, we should need to have a reason for preferring Rawls' theory. But, as we have just seen, we have no good reason in Rawls' case to suppose that the deriva­tion of the two principles from his bargaining game offers any inde­pendent reason for supposing those principles to be right.

6.11