SCIENTIFIC DISCOVERY 183 



The best approach to this analysis of the act of discovery 

 is to be found in a distinction which has been made fre- 

 quently in the philosophy of science. It is usually formulated 

 as the distinction between the abstractive and the hypothetical 

 methods. In a passage already quoted * Bavink expresses an 

 equivalent opposition in terms of the " elaborative " and the 

 "hypothetical'' methods. Two other formulations of the 

 distinction may be given. According to Cassirer we can 

 proceed in two ways in science. "We can, by a pure * abstrac- 

 tive' method, separate from a group of given things or 

 phenomena that group of determinations which is common 

 to all members of the class, and which belongs to them di- 

 rectly in their sensuous appearance; or we can go behind the 

 phenomena to certain hypotheses for the explanation of the 

 field of physical facts in question. Only the first procedure 

 strictly corresponds to the demands of scientific and phil- 

 osophic criticism. For only here are we sure that we do not 

 falsify the observations by arbitrary interpretation; only 

 here do we remain purely in the field of the facts themselves, 

 for while we divide the facts into definite classes, we add no 

 foreign feature to them." 2 Dingle's terminology is essen- 

 tially the same. ''Abstraction is the detection of a common 

 quality in the characteristics of a number of diverse observa- 

 tions: it is the method supremely exemplified in the work of 

 Newton and Einstein. Newton, for example, gave us 'laws of 

 motion.' Now motion is not an experience; what we observe 

 are moving bodies. Motion is an abstraction, a quality con- 

 ceived to be possessed by all moving bodies, however much 

 they may differ in size, shape, color, beauty, virtue, or any- 

 thing else. The laws of motion express the characteristics of 

 this common quality, and they are therefore a rational 

 means of correlating a vast body of common experience. . . . 

 A hypothesis serves the same purpose, but in a different way. 

 It relates apparently diverse experiences, not by directly 

 detecting a common quality in the experiences themselves, 



1 Chapter VIII, pp. 168-169. 



2 E. Cassirer, Substance and Function (Chicago: Open Court, 1923), pp. 193-194. 



