[215] The neo-Kantian epistemology of the nineteenth century often characterized this definition of "truth" as an expression of a methodologically retarded naive realism, and declared it to be irreconcilable with any formulation of this question which has undergone Kant's 'Copernican revolution'. But Kant too adhered to this conception of truth, so much so that he did not even bring it up for discussion; this has been overlooked, though Brentano has already called our attention to it. 'The old and celebrated question with which it was supposed that one might drive the logicians into a corner is this: "what is truth?" The explanation of the name of truth—namely, that it is the agreement of knowledge with its object—will here be granted and presupposed . . .xxxi.
'If truth consists in the agreement of knowledge with its object, then this object must thus be distinguished from others; for knowledge is false if it does not agree with the object to which it is related, even if it should contain something which might well be valid for other objects.'xxxii And in the introduction to the "Transcendental Dialectic" Kant states: 'Truth and illusion are not in the object so far as it is intuited, but in the judgment about it so far as it is thought.xxxiii
Of course this characterization of truth as 'agreement', adaequatio, OJLolwcns, is very general and empty. Yet it will still have some justification if it can hold its own without prejudice to any of the most various Interpretations which that distinctive predicate "knowledge" will support. We are now inquiring into the foundations of this 'relation'. What else is tacit(» posited in this relational totaliry of the adaequatio intellectus et rei? And what ontological character does that which is thus posited have itself?
What in general does one have in view when one uses the term 'agreement'? The agreement of something with something has the formal character of.a relation of something to something. Every agreement, and therefore 'truth' as well, is a relation. But not every relation is an agreement. A sign points at what is indicated.1 Such indicating is a relation, but not an agreement of the sign with what is indicated. Yet manifestly not every agreement is a convenientia of the kind that is fixed upon in the definition of "truth". The number "6" agrees with "16 minus 10". These [216] numbers agree; they are equal with regard to the question of "how much?" Equality is one way of agreeing. Its structure is such that something like a 'with-regard-to' belongs to it. In the adaequatio something gets related; what is that with regard to which it agrees? In clarifying the 'truth-relation' we must notice also what is peculiar to the terms of this relation. With regard to what do intellectus and res agree? In their kind of Being and their essential content do they give us anything at all with
1 'Ein Zeichen zeigt auf das Gezeigte.'