DESCRIPTIVE SCIENCE 131 



peculiar meaning properties which are attached to them, to 

 refer to events in a less direct way. If one knows the mean- 

 ing property of any word symbol, he knows in advance the 

 kind of event which is being referred to, though he need 

 not be aware of the event at the moment. The symbol 

 enables one to ascertain whether any event which may be 

 given is or is not the referent of the symbol. This selective 

 feature of the symbol permits it to be correlated with its 

 referent. Ideally there is a one-to-one correlation between 

 the elements of a system of symbols and the elements of a 

 system of events. For every event there should be one and 

 only one symbol, and for every symbol there should be one 

 and only one event. Actual language is far from ideal in 

 this respect. There are words which are ambiguous in 

 meaning, and there are diverse linguistic forms which are 

 practically synonymous in meaning. Descriptive science 

 aims to be representative only in the sense that it tries to 

 set up a one-to-one correlation between events and sym- 

 bols. Hence such a science is inadequate either if there 

 are clearly given events for which there are no symbols, 

 or if there are symbols which refer to no events or to events 

 ambiguously. The former inadequacy would be illustrated 

 by a botanical science which has not yet identified and 

 classified all known plant forms, and the latter by a physics 

 which talks about perfect levers and gases or by a chemistry 

 which has only one symbol to designate the various isotopes 

 of a substance. 



The essential difficulty in determining precisely what is 

 meant by a descriptive science is the fact that data are 

 given with varying degrees of clarity and obviousness. One 

 of the most significant conclusions to be drawn from the 

 history of thought is that investigators disagree among 

 themselves as to what is and what is not clearly given. 

 The problem, what are the ultimate data for knowledge, 

 has never itself been satisfactorily solved. What is clearly 

 given for one individual may not be so for another; and 

 what is clearly given for an individual at one time may not 



