Truth pluralists argue that propositions can be true in more than one way, in different domains of discourse. A proposition about everyday average-sized objects can be true by corresponding to facts, one about legal matters can be true by being coherent with a specific set of laws, while a proposition from mathematics can be true by being provable from a set of axioms. For Crispin Wright (1992), perhaps the most important proponent of truth pluralism, propositions in a representational domain of discourse are true if they correspond to reality, while those that belong to a non-representational domain must be superassertible instead. Is the idea that reality domains and truth are correlated in this way justified? I argue that it is not and that its origins can be traced back to the works of Michael Dummett, who, without explicitly defending truth pluralism, made popular the idea that the realism/antirealism debate in metaphysics can be seen as a debate about the nature of truth. A realist view in metaphysics, which upholds the existence of mind-independent entities, aligns naturally with a correspondence theory of truth, while an antirealist view better aligns with a verificationist view on truth (one consequence of which is the falsity of the principle of bivalence). Dummett's position has been attacked by several authors, including Devitt (1984), who rejects Dummett's manifestation argument against semantic realism and makes a solid case for separating debates about the nature of truth from debates about the nature of things. My goal here is to argue that even if Dummett is right, his views do not offer sufficient support to truth pluralism. Dummett's antirealism is rooted in his views on truth, meaning and verification. According to him, the notion of truth ought to be explained “in terms of our capacity to recognize statements as true, and not in terms of a condition which transcends human capacities” (Dummett 1993: 75). Unlike truth-conditional semantics, according to which truth comes before meaning and meaning is explained in terms of truth-conditions, Dummett explains both meaning and truth in terms of verification. If different domains of discourse vary so much regarding our capacity to recognize statements as true, I argue, they would have to vary regarding the way we attach meanings to our sentences as well. However, pluralism about meaning is not a promising view. Whatever theory of truth is correct, it should be the same regardless of the domain of discourse. 

References

Devitt, M. 1984. Realism and Truth. Oxford: Blackwell.

Dummett, M. 1993. What is a Theory of Meaning (II), in The Seas of Language, Oxford University Press.

Dummett, M. 1978. Truth and Other Enigmas. Harvard University Press.

Lynch, M. 2009. Truth as One and Many. Oxford University Press. 

Wright, C. 1992. Truth and Objectivity. Harvard University Press.