Fallibilism, coherence and realism

Abstract

It is not surprising that fallibilists invariably adopt a coherence theory of truth rather than some classical form of the correspondence theory of truth. After all, it is understandably difficult to see how what we regard as a true statement can succeed in describing the world if there is some real probability that the truth value assigned to that statement will be withdrawn as human knowledge advances in evolutionary ways. For a fallibilist, assigning truth values to a statement is the equivalent of saying that either the statement is or is not warrantedly assertible or authorizable under public and communal standards of evidence. But between a statement's being fully authorized for acceptance, or being warrantedly assertible, and its "corresponding" to the facts, or being a picture of the way the real world is, there is an important and infamous logical gap. The presence of this well-known gap implies that at any given time, statements legitimately assigned "true" may not in fact be saying the way the real world is. Hence apart from inherent difficulties present in explicating the concept of "correspondence", it is tempting for the fallibilist to regard true statements as statements that are fully warrantedly assertible, and the less said about whether true sentences could correspond to some extra-linguistic state of affairs the better. Coherence theorists need not deny that people generally regard "true" as meaning something like "corresponds with the facts", or "says the way things are". It's just that, given the conditions under which we warrant truth-value assignments to sentences or statements, there is no good reason to think that, even under the most careful conditions of assignment, true statements say the way the world is. The question is whether any statement assigned "true" need be viewed as saying the way the world is, or as corresponding to the facts. If not, then there may well be nothing in the extensional domain of "true" when defined as the correspondence theorist defines it. For this reason, coherence theorists should be viewed as making a semantic recom mendation on how we should define "true". Typically, what bothers the coherence theorist is that the world may be quite different than the way we assert it is when we are maximally justified in our assertions; and if