THE CRITICAL PHILOSOPHY 77 



framed, that is not in itself already partly determined a priori, 

 Kant has passed beyond rationalism. But his treatment of uni- 

 versals is by no means free from rationalistic assumptions. That 

 is, while he maintains that no universal can have meaning or valid- 

 ity except through reference to contingency, he supposes that 

 there are universal propositions which in their own content con- 

 tain no contingent element whose meaning is wholly independent 

 of their application. While such universal propositions are not in 

 themselves knowledge are, in fact, mere "cobwebs of the brain" 

 we may use them as premises and deduce consequences from 

 them, which become valid knowledge through their reference 

 to possible experience, or rather to the possibility of experience. 

 Our present concern is, not to refute this position, but to point 

 out its inherent rationalism. For to suppose that universals, 

 which do not in themselves contain their reference to experience, 

 do yet have such reference, is to assume that there are relations 

 wholly external to the terms which they relate. Furthermore, 

 to suppose that essentially meaningless propositions are yet capa- 

 ble of standing in logical relations which possess formal cogency, 

 is to assume that validity of logical relationship is wholly external 

 to the meaning of the terms related. In one sense it is true that 

 Kant transcends this; namely, in his insistence that it is because, 

 and only because, universals do bear a relation to experience 

 that they have significance and validity. Yet in conceiving that 

 this relation is not constitutive of their very universality, he fails 

 to give any logical ground for their reference to experience. 



Again, what is the 'experience' to which universals must refer 

 to gain validity? To suppose that universals may, in abstraction 

 from their experiential filling, perform logical functions, is to 

 suppose that one may abstract the universal as such. But if the 

 universal as such is abstracted, the 'experience,' apart from the 

 universal thus abstracted, must be the contingent as such. Fi- 

 nally, we wish to call attention to the essential rationalism of the 

 conception of the nature of abstraction, which is implied in this 

 treatment of universals. It is conceived that the process of ab- 



