46 INTRODUCTION TO PHILOSOPHY OF SCIENCE 



data consist of the behavior of the scientist as he manipulates 

 telescopes and microscopes, as he applies burners to flasks 

 containing liquids, as he dissects organisms — but here again 

 the significant aspect is not so much what the scientist is 

 doing in a physical way as what he is "thinking about." 

 And this can be determined only by asking the scientist, 

 which involves all of the difficulties associated with the 

 formulation of the question in language, the possible mis- 

 understanding on the part of the scientist, his own doubtful 

 analysis of his mental processes, and the obstacles in the 

 way of communicating the results of this analysis to another. 



However, the elusive character of the data for the study 

 does not compel one to abandon the empirical method. It 

 suggests, rather, that one should ascribe to his results no 

 more certainty than they are entitled to claim. It suggests 

 that he should recognize the dependence of his conclusions 

 on the results of psychology, biology, linguistics, and math- 

 ematics. It suggests, in other words, that the logic of science 

 is highly complex, and subject to revision as the more ele- 

 mental sciences of which it is composed progress and develop. 

 When psychology and biology are able to tell the logician of 

 science more about the symbolic process, when linguistics 

 is able to tell him more about the nature of symbols and 

 meaning, when mathematics and logic are able to tell him 

 more about the structure of deductive schemes, then he 

 will speak with greater authority in the field of the logic of 

 science. Until then he must recognize that his study is, so 

 to speak, twice removed from reality, and that its pursuit, 

 while empirical in character, must be subject to limitations 

 and inadequacies by virtue of the obscurity of its data. 



When one turns to the actual analysis of scientific cogni- 

 tion, though he finds it to be complex, he can readily dis- 

 cern certain characteristic elements and certain dominant 

 features of structure. The elements are four in number. In 

 the first place, there is the scientist — a self or person who 

 instigates and controls the activities of investigation and 

 formulates their results in a system of symbols. In the 



