THEORIES OF SCIENTIFIC CONCEPTS 157 



his nose into the end of a blank passage, and we would then 

 formulate a set of verbal rules, instructing us at a certain 

 stage of operations to switch our procedure and begin oper- 

 ating in a new way which we had found appropriate by the 

 method of trial and error." 1 



The position here defined is clearly positivistic. It insists 

 that concepts be defined in terms of data which are them- 

 selves operationally determined, and that the mind proceed 

 very cautiously when it attempts to explore the realm of 

 less clearly given objects. "Since our concepts are con- 

 structed of operations, all our knowledge must inescapably 

 be relative to the operations selected." 2 Hence we cannot 

 impose a concept defined in one realm upon another; this 

 expresses the attitude of caution which is characteristic of 

 positivism. Furthermore, a large number of our so-called 

 concepts prove to be pseudo-concepts because they are, 

 strictly speaking, meaningless. 'If a specific question has 

 meaning, it must be possible to find operations by which 

 an answer may be given to it. It will be found in many 

 cases that the operations cannot exist, and the question there- 

 fore had no meaning. For instance, it means nothing to ask 

 whether a star is at rest or not." 3 



Both Pearson and Bridgman, therefore, insist upon the 

 basic importance of data. For Pearson the data are merely 

 given; for Bridgman they must often be produced opera- 

 tionally, and, having been so obtained, take on their opera- 

 tional origin as part of their content. For both writers, 

 symbols are simply classificatory and descriptive. Pearson 

 insists upon this fact even in connection with such apparently 

 remote notions as atoms; Bridgman argues that concepts 

 must be kept empirical by defining them in terms of their 

 proper range of application, and by re-defining them in 

 case they are to be applied to novel spheres. It seems safe 

 to characterize both positions as positivisms, though one 

 must admit that Bridgman' s variety is somewhat less nar- 



1 Nature of Physical Theory (Princeton: Princeton University, 1936), p. 28. 



2 Logic of Modern Physics, p. 25. 3 Ibid., p. 28. 



