Abstract

Both in theory development and in everyday reasoning, scientists often find themselves subscribing to inconsistent beliefs. This dissertation explicates the role such inconsistent belief sets or proposals play in scientific reasoning toward consistent replacement theories. I reject the idea that a paraconsistent logic, that is a deviant deductive logic in which the consequence class of inconsistent premisses is not the universal set of all premisses, is the implicit basis for the use and revision of inconsistent proposals. Various paraconsistent logics are discussed. Some are shown to be as intolerant of inconsistency as classical logic, and others are shown to be simply uninterpretable. An adequate explication of the role of inconsistent proposals in science must take into account the nature of the confirmation of each constituent statement in the proposal. Such proposals serve scientific inquiry as a part of what I term the "projected constrained replacement strategy." This strategy involves using the proposal itself, the confirming evidence for each of its constituent statements, and the kind of confirmation relation that exists between that evidence and each statement to project a schematic replacement theory. The projected replacement theory is consistent but incomplete. Nevertheless, it is constructed so as to retain as much as possible of the content of the original, inconsistent proposal as we have some rational reason for believing to be true. The projection constitutes an explicit representation of the kind of replacement theory that is sought. Reasoning with the projected replacement provides information about the constraints that any proposed interpretation for its uninterpreted parts must satisfy. In that way, it serves as a heuristic guide to the final, complete replacement theory for the original proposal. Several historical cases, including Bohr's theory of spectral emission, Einstein's derivation of Planck's radiation law, and wave-particle duality are reconstructed as instances of implementation of the projected replacement strategy