DESCRIPTIVE SCIENCE 129 



bryonic science, and should reveal many of the features of 

 more highly developed knowledge. It may thus be worth 

 while to make an analysis of such a system with a view to 

 determining its significant features. 



Furthermore, even the sciences which are characterized 

 as non-empirical, or rational, contain aspects which sug- 

 gest that they are not to be sharply divorced from the 

 descriptive sciences. Jevons argues that every science con- 

 tains facts of four different kinds. "(1) We may be ac- 

 quainted with facts which have not yet been brought into 

 accordance with any hypothesis. Such facts constitute 

 what is called Empirical Knowledge. (2) Another extensive 

 portion of our knowledge consists of facts which having 

 been first observed empirically, have afterwards been brought 

 into accordance with other facts by an hypothesis concerning 

 the general laws applying to them. This portion of our 

 knowledge may be said to be explained, reasoned or general- 

 ized. (3) In the third place comes the collection of facts, 

 minor in number, but most important as regards their sci- 

 entific interest, which have been anticipated by theory and 

 afterwards verified by experiment. (4) Lastly, there exists 

 knowledge which is accepted solely on the ground of theory, 

 and is incapable of experimental confirmation, at least with 

 the instrumental means in our possession." 1 If one admits, 

 with Jevons, that every science contains facts of the type (1), 

 he is obliged to conclude that it is, to that extent, empirical 

 or descriptive. Furthermore, if one admits that the lines 

 separating these four kinds of facts cannot be sharply drawn, 

 he is obliged to conclude that the difference between a 

 descriptive and a rational science is itself a relative matter. 

 This argues, at least, for the importance of clarifying the 

 distinction as much as possible. 



Granting, then, that there are descriptive sciences, one 

 may attempt to determine their features. Such a science, 

 according to Jevons, is any body of symbols which is 

 "derived directly from the examination of detached facts, 



1 Principles of Science, p. 525. 



