364 INTRODUCTION TO PHILOSOPHY OF SCIENCE 



accomplishes its end by mere fiat. Any law may acquire 

 universality by a simple act which denies the possibility 

 that the law should have exceptions. This does not deny 

 the existence of the so-called exceptional cases, but merely 

 that they are properly to be called exceptions; they are 

 asserted as lying outside the scope of the law. The law 

 that unsupported bodies fall is not violated by balloons, 

 since these are supported bodies and hence do not come under 

 the scope of the law. A law which has been observed to be 

 true in a large number of cases may be asserted as legislative 

 over all future cases, and incapable of violation. Such a law 

 may be said to be nomically necessary, i.e., true by an arbi- 

 trary fiat. In empirical laws the events are first character- 

 ized, and the association is then asserted to hold between 

 the characterized events; but in nomically necessary laws 

 the characterization of at least one of the events is deter- 

 mined by the existence of the association. If it is an empirical 

 law that water boils at 100°, there must be some means for 

 identifying a substance as water independent of its boiling 

 point; but if this law is nomically necessary then whatever 

 boils at this temperature must be water, and whatever boils 

 at some other temperature cannot be water, for water is 

 identified by this property. Nomically necessary laws func- 

 tion as definitions, since they cannot be refuted. 



It seems clear that the necessity which is a feature of such 

 laws is obtained only at the cost of objectivity. Though one 

 can know that the law is incapable of refutation, he can 

 never know whether it is any longer descriptive of events. 

 Considerations of the kind raised in Chapters VII and X 

 become relevant. The intensional features dominate over 

 the extensional. Meaning and truth are determined not 

 by illustration but by definition. Interest has shifted from 

 events to symbolic systems. Knowledge is passing from 

 description to explanation. Hence the appearance of nomic 

 necessity in a scientific law is a signal for the passing of the 

 science from the descriptive to the rational or explanatory 

 stage. Nomic necessity can always be achieved through the 



