244 INTRODUCTION TO PHILOSOPHY OF SCIENCE 



that discourse about events on the empirical level is essen- 

 tially vague. This is due partly to the fact that all definition 

 must be in popular terminology, which is essentially obscure. 

 One must talk about time, for example, as the "going-on," 

 or the "passage," or the "flow" of events; one must describe 

 space as the 'spread-out-character' of events; one can 

 explain a class only as a "heap," "aggregate," "collection," 

 "pile," or "ensemble"; and one is obliged to characterize 

 matter by describing it as 'stuff," or "substance." But 

 the vagueness is due to a much more important cause. 

 Definition, in the strict sense, is not possible on the empirical 

 level. The clearness with which events are given is a per- 

 ceptual rather than a conceptual clarity, i.e., an extensional 

 rather than an intensional clarity. Definition is possible 

 only when symbols have been integrated into systems; but 

 on the empirical level this has not yet taken place. Symbols 

 must be clarified by the method of pointing, but not by the 

 method of definition. The function of discourse at this 

 stage is that of directing the attention to the place within 

 the realm of nature where the events in question can be 

 found, so that their characters can be determined by direct 

 inspection. Proper names, gestures, and pictorial symbols 

 are often more effective than abstract concepts. Discourse 

 is possible only in the sense that one may 'talk about'' 

 the symbols in question. One must proceed as did Poinsot, 

 according to Claude Bernard: x "'If anyone asked me to 

 define time, I should reply: 'Do you know what it is that 

 you speak of?' If he said, 'Yes," I should say, "Very 

 well, let us talk about it." If he should say, "No," I should 

 say, "Very well, let us talk about something else." 



But, in the third place, in addition to the ineffectiveness 

 of the method of definition and the consequent resort to 

 the method of pointing, one finds that the application even 

 of this latter method is not without difficulties. Pointing 

 is ambiguous. Events are highly complex entities and there 



1 A. J. Lotka, Elements of Physical Biology (Baltimore: Williams and Wilkins), 

 p. 19. 



