Is not positivism a genuine alternative to the root-metaphor hypotheses?
Let me state this question at greater length in the words in which it was given to me by one critic: “Although positivism is not a hypothesis respecting the actual nature of the world, since it maintains that this nature is unknowable, does it not remain as a genuine alternative to all the specialized world hypotheses? Can it be refuted or put out of court in any other way than through a consideration of the nature of probability and the application of probability to theories? And does not your view of the inadequacy of all the world hypotheses give it at least a provisional standing?” [695] My answer is that if positivism is undogmatic it cannot, with the data we have, be a hypothesis respecting the nature of the world either affirmatively or negatively, but can only be an interest in, or a preference for, the results of multiplicative corroboration. It therefore does not assert that the actual nature of the world is unknowable. As soon as positivism makes such an assertion it becomes
