OF ARTS AND SCIENCES .* MAY 14, 1867. 291 



diate conception), and its applicability to the latter is hypothetical. 

 Take, for example, the proposition, " This stove is black." Here the 

 conception of this stove is the more immediate, that of black the more 

 mediate, wliich latter, to be predicated of the former, must be dis- 

 criminated from it and considered in itself, not as applied to an object, 

 but simply as embodying a quality, blackness. Now this blackness is a 

 pure species or abstraction, and its application to this stove is entirely 

 hypothetical. The same thing is meant by " the stove is black," as by 

 " there is blackness in the stove." Embodying blackness is the equiv- 

 alent of black.* The proof is this. These conceptions are applied 

 indifferently to precisely the same facts. If, therefore, they were dif- 

 ferent, the one which was fii'st applied would fulfil every function of 

 the other ; so that one of them would be superfluous. Now a super- 

 fluous conception is an arbitrary fiction, whereas elementary concep- 

 tions arise only upon the requirement of experience ; so that a super- 

 fluous elementary conception is impossible. Moreover, the conception of 

 a pure abstraction is indispensable, because we cannot comprehend an 

 agreement of two things, except as an agreement in some respect, and 

 this respect is such a pure abstraction as blackness. Such a pure 

 abstraction, reference to which constitutes a quality or general attribute, 

 may be termed a ground. 



Reference to a ground cannot be prescinded from being, but being 

 can be prescinded from it. 



§ 8. Empirical psychology has established the fact that we can know a 

 quality only by means of its contrast with or similarity to another. By 

 contrast and agreement a thing is referred to a correlate, if this term 

 may be used in a wider sense than usual. The occasion of the intro- 

 duction of the conception of reference to a ground is the reference 

 to a correlate, and this is, therefore, the next conception in order. 



Reference to a correlate cannot be prescinded from reference to a 

 gi-ound ; but reference to a ground may be prescinded from reference 

 to a correlate. 



§ 9. The occasion of reference to a correlate is obviously by com- 

 parison. This act has not been sufficiently studied by the psycholo- 

 gists, and it will, therefore, be necessary to adduce some examples to 

 show in what it consists. Suppose we wish to compare the letters 



* This agrees with the author of " De Generibus et Speciebus," Ouvrages Inedits 

 d'Abelard, p. 528. 



