VERIFICATORY TECHNIQUES 217 



expansion rests upon insight rather than rules, upon pene- 

 tration rather than superficial awareness, and upon genius 

 rather than techniques. Whether this activity is considered 

 to be "creative" involving the "invention" of explanatory 

 entities, or "penetrative" involving the "discovery" of ob- 

 scure features of nature, depends upon one's general theory 

 of science, and need not be decided — if it can be decided at 

 all — at this point. It may be that scientific genius is like 

 artistic genius, founded upon the activity of imagination 

 which freely transforms and modifies the given; if this is the 

 case the third stage would be differentiated from the fourth 

 in the fact that in the latter the given has been definitely 

 increased by contributions which are the product of human 

 inventiveness. On the other hand, it may be that scientific 

 genius is characterized only by a peculiar insight which 

 enables it to explore the hidden recesses of nature, i.e., it may 

 be that the mental act which is responsible for a theory is 

 nothing more than the vague awareness of the existence of 

 certain data, heretofore undiscovered. If this is the case the 

 third stage would be differentiated from the fourth only in 

 the fact that the data to be explained by the theory are 

 clearly recognized in the former and only vaguely grasped 

 in the latter. At any rate, the techniques of the third and 

 fourth stages are not differentiated; so far as they can be 

 formulated at all they are such indeterminate operations as 

 serial extension and interpolation, cause and effect infer- 

 ences, and analytic and synthetic operations. In all of these 

 operations, however, the hypothetical expansion is a matter 

 which is not susceptible of formalization; it reduces ulti- 

 mately to a matter of insight or genius, which is to say that 

 it is essentially a mystery. 



A theory which has developed through the four stages is 

 ready for verification. Whether the verificatory operations 

 are considered as contributing to the development of the 

 theory and hence as constituting a fifth stage in its growth, 

 or whether they are considered to be techniques of testing 

 or of corroboration, is a matter of no great concern. It is 



