Realism and Truth: A Comment on Crispin Wright's Truth and Objectivity

Abstract

You can be a realist about any area of talk or thought, so everyone agrees, only if you treat the discourse as knowledge-seeking or cognitive in character. You must see participation in the discourse, not as a matter of giving orders or expressing wishes, for example, but rather as an exercise in offering answers to various implicit and explicit questions. But what is it, then, actually to be a realist about the discourse? What does realism involve over and beyond such cognitivism? Crispin Wright gives this familiar sort of question an interesting cast. He takes the concept of truth as that which is fixed by the disquotational schema 'P is true if and only if PF and argues, by a distinctive route, that this concept directs us to a property distinct from that of warranted assertibility; this is the property of minimal truth which he characterises as 'substantial', if metaphysically 'lightweight'. As I asked what it is for a discourse to satisfy realism as well as being cognitivist or knowledge-seeking, he asks what it is for the discourse to satisfy realism as well as being apt for minimal truth: as well as being capable of being true or false in his minimal sense. He takes aptness for minimal truth to be a matter of satisfying the 'constraints of syntax and discipline' associated with assertoric form (29, 35-36) and so it seems natural to assume that realism must involve more than a belief in such truthaptness. Should we accept Wright's recasting of the question? I am prepared to do so, though I should record misgivings about the assumption that aptness for truth-even for minimal truth-is just a matter of satisfying certain constraints of assertoric syntax and discipline (see Holton 1993). Even though truth is minimal, aptness for such truth may involve substantial requirements: it requires that the discourse in question is capable of expressing bona

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 28 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.