Abstract

A coherent theory of relations was a critical part of Russell’s metaphysics. In Appearance and Reality Bradley posed a problem that sits squarely in the way of any doctrine of “external” relations. Russell, determined to advance such a doctrine, tried several times to find a way around the paradox and apparently believed he had succeeded by making use of one of his inventions, the theory of logical types.Gilbert Ryle and Alan Donagan have advanced an argument that I read, over the objections of its authors, as a special case of Bradley’s. In this paper I argue that the ad hoc solution suggested by Donagan to the special problem is one that Russell had already indicated a willingness to accept but that the general problem of the paradox remains.What finally prevents Russell from solving the paradox is a combination of his refusal to abandon the claim that relations are constituents of facts and the necessity of distinguishing a relational fact from its converse. Following some hints that Russell left, I do some reconstruction, showing how the theory of types would (and should) have been applied had Russell followed through on his own insights. The result, I suggest, is a truly Russellian theory that escapes Bradley’s paradox.