80 DOGMATISM AND EVOLUTION 



mal, and which themselves contain no content. If it be argued, 

 that this does not involve the ultimate separation of form and 

 content, since these purely formal propositions are valid only 

 with reference to experience, we may at once reply that to con- 

 ceive pure forms which apply indifferently to all content, is to 

 conceive their applicabilty as not dependent on the nature of 

 that content, and hence logically unrelated to any content. 



A more important modification of the Kantian position than 

 the one we have just discussed is the elimination of the thing-in- 

 itself, by those successors of Kant regarded as most truly carrying 

 out the critical principles. This doctrine too, it is urged, is in- 

 consistent with the implied logic of the critical philosophy. The 

 concept of an unknowable thing-in-itself , lying beyond the limits 

 of possible experience, is an utterly useless conception, playing 

 no real part in the system. Its only alleged connection with 

 experience lies in the assumption that it is the source of the un- 

 formed "matter of sensation." But since no such bare matter 

 is to be discovered in experience, the positing of a source for 

 it becomes simply gratuitous. Indeed, so it is argued, if sensa- 

 tion did yield an utterly unorganized matter, its organization in 

 experience would be entirely impossible. In short, criticism is 

 concerned only with the conditions of possible experience, and 

 what lies beyond experience is altogether outside its scope. 



If the argument of the preceding pages has carried any weight 

 to the reader's mind, it must appear evident that Kant's assump- 

 tion of the thing-in-itself is by no means gratuitous; that, on 

 the contrary, it is a conception of vital importance not only to 

 the Kantian theory, but to the critical philosophy generally. 

 For if it is said that the relations we find in experience, the 

 terms in which we think, are relations essential to the nature of 

 thought, this must mean that these relations constitute the 

 nature of thought as such, and must hold of experience univer- 

 sally. These relations, then, in belonging to the nature of thought 

 as such, are not inherent in, or constitutive of, the elements 

 which they connect. If they were, the claim that they owed 



