236 INTRODUCTION TO PHILOSOPHY OF SCIENCE 



relation to his wife immediately becomes selected as essen- 

 tial. The problem of the analysis of the basic concepts of 

 the sciences is concerned with a similar selection of relation- 

 ships with regard to symbols in general. 



However, if the problem is thus formulated it can be for 

 the purposes of science greatly simplified. Science is defi- 

 nitely empirical in its ultimate reference. As a consequence 

 the essential reference of any symbol is to the realm of 

 events, and both truth and meaning must be determined 

 by this route. The problem of truth and meaning cannot 

 be solved finally by reducing symbols to other symbols, for 

 the resultant symbols are themselves inadequate unless their 

 empirical reference can be readily ascertained. Hypotheses 

 are both defined and verified in terms of data, as has already 

 been seen. No hypothesis can be clarified by defining it in 

 terms of another hypothesis which is equally vague, and no 

 proposition can be justified by deducing it from other propo- 

 sitions whose truths are equally doubtful. The only alterna- 

 tive to this principle is the admission of innate ideas and 

 self-evident propositions — both of which are contrary to 

 the empiricism of science. Hence extension predominates 

 over intension in the ascertainment of truth and meaning, 

 and the ultimate reference of any symbol is to the realm of 

 events. 



This is not, of course, to deny that the operations which 

 are employed in the determinations of both meaning and 

 truth may be very elaborate. In fact, it is the formulation 

 of these operations which is the core of the problem. An 

 important question to be asked with regard to every symbol 

 is, "To what events does it refer?' But a still more im- 

 portant question is, "How does it refer to these events?' 

 The meaning and truth of a symbol are a function not 

 merely of the events to which it refers but also of the manner 

 of referring to these events. Symbols are descriptions, 

 classifications, correlations, associations, generalizations, ab- 

 stractions, measurements, idealizations, limits, and interpola- 

 tions. All of these are descriptions of the way in which 



