Abstract

The dissertation deals with two questions having to do with the concept of truth. The first is what it is to say that a sentence is true, and the second is how to treat the liar paradox. Even though the dissertation begins with the two famous traditional theories of truth , Tarski's semantic theory is its real starting point. His theory is not only the most influential challenge to the traditional theories but is also the springboard from which all subsequent works on the subject come. The debate between the deflationary views and the substantial theories is over the philosophical significance of Tarski's theory. The deflationary views argue that Tarski's theory captures every important property on the concept of truth, whereas the substantial theories say that truth is a substantial concept beyond what Tarski specifies. I claim that both views fail to account for the concept completely. ;Davidson suggests an interesting way in which Tarski's theory can be understood. Davidson argues truth is a primitive concept that cannot be defined in terms of any other concept. So this primitive concept, taken with Tarski's theory, enables us to understand the meaning of the object language. Although Davidson's view is preferrable to any other, I think it does not give an adequate account of the concept of truth. I claim that no single theory can explain the concept completely. This is not due to the deficiency of the theory but to the concept of the truth itself. ;Tarski's theory is presented in the context of a puzzle about the concept of truth, which concerns the possibility of constructing paradoxes. Solutions to the paradox, roughly speaking, are made either by means of a hierarchy of languages or by rejecting the principle of bivalence. I criticize such solutions, appealing to Herzberger's naive semantics and Gupta's revision theory of truth. I argue their views are the most acceptable solutions to the paradox and also throw a new light on the concept of truth.