Kant insists that the use of concepts must be subject to empirical conditions if they are to have objective validity. This article analyses Kants distinction between empirical and transcendental truths. Since transcendental concepts are pure concepts without spatio-temporal content, their objective validity is of the same second-order kind as that of unschematized categories. This characteristic of transcendental concepts implies that the cognitive powers picked out by them are not particular psychological mechanisms, but rather abstract functional structures. Transcendental concepts owe their (...) objective validity to the realizability of the functional structures by empirical cognizers like humans. This relation in turn helps to explain the nature of transcendental truths. (shrink)

This paper compares Kant's transcendental idealism with three main groups of contemporary anti-realism, associated with Wittgenstein, Putnam, and Dummett, respectively. The kind of anti-realism associated with Wittgenstein has it that there is no deep sense in which our concepts are answerable to reality. Associated with Putnam is the rejection of four main ideas: theory-independent reality, the idea of a uniquely true theory, a correspondence theory of truth, and bivalence. While there are superficial similarities between both views and Kant's, I find (...) more significant differences. Dummettian anti-realism, too, clearly differs from Kant's position: Kant believes in verification-transcendent reality, and transcendental idealism is not a theory of meaning or truth. However, I argue that part of the Dummettian position is extremely useful for understanding part of Kant's position - his idealism about the appearances of things. I argue that Kant's idealism about appearances can be expressed as the rejection of experience-transcendent reality with respect to appearances. (shrink)

It is often claimed that anti-realism is a form of transcendental idealism or that Kant is an anti-realist. It is also often claimed that anti-realists are committed to some form of knowability principle and that such principles have problematic consequences. It is therefore natural to ask whether Kant is so committed, and if he is, whether this leads him into difficulties. I argue that a standard reading of Kant does indeed have him committed to the claim that all empirical truths (...) are knowable and that this claim entails that there is no empirical truth that is never known. I extend the result to a priori truths and draw some general philosophical lessons from this extension. However, I then propose a re-examination of Kant’s notion of experience according to which he carefully eschews any commitment to empirical knowability. Finally I respond to a remaining problem that stems from a weaker, justified believability principle. (shrink)

Kant's writings on logic illustrate the comparison argument about truth, which goes as follows. A truth-bearer p is true if and only if it corresponds, or it agrees, with a portion of reality: the object(s), state(s) of affairs, or event(s) p is about. In order to know whether p agrees with that portion of reality, one must check if that portion of reality is as p states. Using the terms of the comparison argument, one must compare p with that portion (...) of reality. This is impossible, because the only knowledge of reality we can have is in the form propositions, beliefs, or judgments, whose agreement with reality is as much in need of justification as the agreement of p with reality. Therefore, it is impossible to know which truth-bearers are true. This paper reconstructs Kant's version of the comparison argument. It is argued that, according to Kant, the argument is sound only under the assumption of transcendental realism. Transcendental idealism avoids the sceptical consequences of the comparison argument. (shrink)

Ernst Cassirer claimed that Kant's notion of actual object presupposes the notion of truth. Therefore, Kant cannot define truth as the correspondence of a judgement with an actual object. In this paper, I discuss the relations between Kant's notions of truth, object, and actuality. I argue that's notion of actual object does not presuppose the notion of truth. I conclude that Kant can define truth as the correspondence of a judgement with an actual object.