Actions That Make Us Know

The knowability paradox is usually formulated as a problem about the static propositions which express the knowledge that we can achieve in principle. In this paper, I propose to put these issues in a more ’dynamic’ light, by shifting the emphasis to the epistemic actions that produce knowledge, or sometimes even ignorance. The very notion of ’knowability’ seems mainly an existentially quantified residue of knowledge-producing actions, just as ’provability’ is the static property of propositions that remains when we suppress their live proof and its production. In particular, can every static proposition which is true trigger a dynamic action of announcing that it is true, or of learning that truth? Keeping track of what actions do over time is notoriously difficult, as the truth values of relevant propositions keep changing in processes of computation, physical movement, games, or communication. We discuss some basic issues that arise when we place ’knowability’ in a setting of one or more epistemic agents performing a possible variety of epistemic actions. 1 The problem: verificationism incurs the Fitch paradox Verificationism is an account of meaning and truth whose origins lie in logical proof theory, especially, in its constructivist versions. The idea is that ’truth’ can only be assigned to propositions for which we have evidence. This view is found with logical authors like Dummett and Martin-Lšf since the 1970s, but it has also penetrated since into general philosophy. Stated as a sweeping claim, this take on truth implies the general verificationist thesis that what is true can be known: