Truth and Objectivity

by Brian Ellis

Published 26 April 1990
This book traces an argument from an ontology to a theory of truth, and back again. The ontology is that of scientific realism; the theory of truth is an evaluative theory. In the course of the argument it is shown that the correspondence and redundancy theories of truth most commonly accepted by scientific realists are either incompatible with scientific realism or unsatisfactory as theories of truth. Scientific realists cannot believe in the sorts of entities required as truth bearers in most versions of the correspondence theory, or in the relationships between possible worlds to which true model and conditional propositions are supposed to correspond. Redundancy theories are unsatisfactory because they fail to explain the value of truth. The only kind of truth theory compatible with scientific realism which gives a satisfactory account of the value of truth is an evaluative theory - one which identifies truth with what it is right epistemically to believe. To develop such a theory, it is necessary to spell out the preferred system of epistemic preferences, or values, with reference to which truth is to be defined.
From the perspective of a scientific realist, the most fundamental of these are likely to be for the kinds of beliefs or belief systems which have contributed most to our survival as a species. When truth is defined with reference to such a system of values, it appears that most of our scientific inductive strategies can be justified as more or less effective strategies for coming to know the truth. The truth theory thus justifies the ontology.