92 WHAT IS SCIENCE? 



obvious and more prominent in theories than in laws. 

 One aspect of this difference has been noted already ; 

 the acceptance of a theory as true does involve a personal 

 choice in a way that a law does not. Different people do 

 differ about theories ; they can choose whether or no they 

 will believe them ; but people do not differ about laws ;* 

 there is no personal choice ; universal agreement can be 

 forced. Again, if we look at the history of science, we shall 

 find that the great advances in theory are more closely 

 connected with the names of the great men than are the 

 advances in laws. Every important theory is associated 

 with some man whose scientific work was notable apart 

 from that theory ; either he invented other important 

 theories or in some other way he did scientific work 

 greatly above the average. On the other hand there are 

 a good many well-known laws which are associated with 

 the names of men who, apart from those particular 

 laws, are practically unknown ; they discovered one 

 important law, but they have no claim to rank among 

 the geniuses of science. 2 That fact seems to indicate 

 that a greater degree of genius is needed to invent true 

 theories than to discover true laws. 



The same feature appears in the early and prehistoric 

 stages of science. Science, as we have seen, originally 

 took over from common sense laws which had been 

 already elaborated ; and although it has greatly refined 

 and elaborated those laws and has added many new 

 types, it has never wholly abandoned the laws of common 

 sense. Modern science depends as much as the crudest 

 common sense on the notion of a " substance " (a notion 



1 Except in so far as people may refuse to admit a law or to regard 

 it as anything but empirical, because it is not in accordance with 

 some theory. But then they admit that the laws describe the facts 

 rightly ; they only suggest that some other and equally accurate way 

 of describing them would be preferable. 



a In case this book falls into the hands of some expert physicist, 

 I would suggest that examples may be found in the laws of Stefan, 

 Dulong and Petit, or Bode, 



