470 THE POPULAR SCIENCE MONTHLY. 



ine, we find the course uniformly pursued to be that of setting 

 out with propositions of which the negations are inconceivable, 

 and advancing by successive dependent propositions, each of 

 which has the like character — that its negation is inconceivable. 

 In a developed consciousness (and of course I exclude minds of 

 which the faculties are unformed) it is impossible to represent 

 things that are equal to the same thing as being themselves un- 

 equal ; and in a developed consciousness, action and re-action can 

 not be thought of as other than equal. In like manner, every 

 because, and every therefore, used in a mathematical argument, 

 connotes a proposition of which the terms are absolutely coherent 

 in the mode alleged : the proof being that an attempt to bring 

 together in consciousness the terms of the opposite proposition is 

 futile. And this method of testing, alike the fundamental propo- 

 sitions and all members of the fabric of propositions raised upon 

 them, is consistently pursued in verifying the conclusion. Infer- 

 ence and observation are compared ; and when they agree, it is 

 inconceivable that the inference is other than true. 



In contrast to the method which I have just described, distin- 

 guishable as the legitimate a priori method, there is one which 

 may be called — I was about to say, the illegitimate a priori 

 method; but the word is not strong enough: it must be called 

 the inverted a priori method. Instead of setting out with a 

 proposition of which the negation is inconceivable, it sets out 

 with a proposition of which the affirmation is inconceivable, and 

 therefrom proceeds to draw conclusions. It is not consistent, 

 however : it does not continue to do that which it does at first. 

 Having posited an inconceivable proposition to begin with, it 

 does not frame its argument out of a series of inconceivable 

 propositions. All steps after the first are of the kind ordinarily 

 accepted as valid. The successive tlierefores and hecauses have 

 the usual connotations. The peculiarity lies in this, that in every 

 proposition save the first, the reader is expected to admit the 

 logical necessity of an inference drawn, for the reason that the 

 opposite is not thinkable ; but he is not supposed to expect a like 

 conformity to logical necessity in the primary proposition. The 

 dictum of a logical consciousness which must be recognized as 

 valid in every subsequent step, must be ignored in the first step. 

 We pass now to an illustration of this method which here con- 

 cerns us. 



The first sentence in Kant's first chapter runs thus: "Noth- 

 ing can possibly be conceived in the world, or even out of it, 

 which can be called good without qualification, except a Good 

 Will." And then on the next page we come upon the following 

 definition : 



