Abstract

In his 1936 article, "Uber den Begriff der logischen Folgerung," Tarski claims to have provided precise definitions of the concepts of logical truth and logical consequence that remain "close in essentials" to the "common" or "everyday" concepts. In this dissertation, I examine Tarski's account of these notions, as well as the standard model theoretic definitions that are based on that analysis. I argue that Tarski's analysis, and hence in turn its model theoretic heirs, are fundamentally mistaken, that they do not capture the essential features of our ordinary understanding of those notions. I claim that the account's misdirection is, in a sense, quite obvious, but that it has been disguised in various different ways. First is a fallacious argument that seems to have been offered by Tarski in support of his analysis, second is the conflation of the model theoretic definitions with what I call "representational semantics," and third is a confusion of the principle underlying Tarski's account with a completely uncontroversial principle concerning the closure of logical truth under logical consequence