ANALYSIS OF SCIENTIFIC CONCEPTS 243 



various stages of the process are great. The remainder of 

 the chapter will be devoted to a consideration of the most 

 important of these difficulties. 



DIFFICULTIES IN ASCERTAINMENT OF EMPIRICAL 



FOUNDATION 



The first difficulty is one to which reference has already 

 been made in connection with the difference between de- 

 scriptive and explanatory sciences, viz., the fact that the 

 empirically given is a variable, and fails to exhibit precise 

 boundaries limiting the range of "hard v data. Russell 

 himself admits that the distinction between "hard" and 

 "soft" data is one of degree. 1 The given exhibits a strict 

 continuity with reference to the clarity or obviousness of 

 its elements. Some data are more clearly given than others, 

 but may be less clearly given than still others. Hence one 

 must admit something like levels in the empirically given. 

 For example, it seems certain that the private perceptual 

 space of the individual is more clearly given than the one 

 universal geometrical space presupposed by science. But 

 the application of analysis reveals the fact that the private 

 perceptual space of the individual is itself a construction 

 out of visual, tactile, and motor spaces; hence the private 

 space which was presumed to be psychologically primitive 

 turns out to be psychologically derivative. Similarly, the 

 scientific concept of order seems to have its empirical foun- 

 dation in the perceived relation of " betweenness " ; yet this 

 relation may itself be considered as a construction out of 

 more primitive data constituting the merely directional 

 features of events, e.g., right and left, above and below, etc. 

 The consequence of these facts is that the description of the 

 empirical foundation of a concept or proposition implies always 

 a conscious recognition that the level which is being chosen 

 as basic is only relatively so, and may prove upon further ex- 

 amination to be reducible to one still more primitive. 



In the second place, one must be reconciled to the fact 



1 Our Knowledge of the External World, p. 70. 



