The Systematic Determination of Thought 243 



temporal, and logical signs — of the empirical data which 

 are utilized in the forms of knowledge. I do not know that 

 any one would be disposed to say that our knowledge of 

 the external world, of the characters of persons, of the 

 truths of history and natural science, are not attained 

 through experience bit by bit ; and the question to which 

 the a priori theory gives no answer is : How are these bits 

 found out ? Even given the * categories,' what sorts of 

 experiences fit the categories, and how is the fitting done ? 



§ 3. The Systematic Determination of ThoiigJit 



Leaving for a later section, therefore, the question of 

 the origin of the categories, and reverting to the only 

 remaining real alternative, the first thing to be said is that 

 two limitations confine us in finding the source of the 

 variations which are available for the determination of our 

 thinking, whatever the sphere or line of progress be. 

 First, the new thought-variations, to be candidates for 

 selection, are not mere stray products of fancy; yet sec- 

 ond, they are still not outside the problem of selection from 

 variations which arise somehow in the experience of the 

 individual thinker. Having these two limitations full in 

 mind, we find the third alternative — which in my own 

 opinion all the facts go to support — to be this: the 

 thought-variatio7is by the supply of which selective thinki7ig 

 proceeds occnr in the processes at the level of organization 

 which the system in question has already reached — a level 

 ivhich is thus the platform for further determinations in the 

 same system. 



Having stated this general position, we might examine 

 each of the lines or spheres of selective thinking already 



