80 WHAT IS SCIENCE? 



molecules. We may apply once more our fundamental 

 test of universal agreement which serves to distinguish 

 the objects concerned in laws from any others. If some- 

 body denied the existence of molecules, how could we 

 prove him wrong ? We cannot show him the molecules ; 

 we can only show him the gases and expound the theory ; 

 if he denied that the theory proved the existence of the 

 molecules, we should be powerless. We cannot prove 

 by his actions that he is perverse or deluded ; for his 

 actions will be affected only by the properties of gases, 

 which are actually observed, and not by the theory intro- 

 duced to explain them. Actually the dynamical theory of 

 gases has been denied by men of science of high distinc- 

 tion. Usually the denial was based partly on the asser- 

 tion that the laws of gases could be deduced accurately 

 from the theory, but it has often been accompanied by 

 the contention that, even if they could be deduced 

 accurately, the theory was not true, and not worthy of 

 acceptance. No denial of that case would be possible 

 if the theory were indeed a law. 



We conclude therefore and the conclusion is vital 

 to the view of science presented here that a theory is 

 not a law, and consequently, that the explanation afforded 

 by^tEeory cannot simply be the explanation by generaliza- 

 tion which consists in the exhibition of one law as a par- 

 ticular instance of another. It does not follow that 

 theories have nothing to do with laws, and that it is 

 immaterial for the theory that the laws of dynamics 

 are true, and of very great generality. We shall see 

 presently that this feature is of great importance. But 

 it does not involve that the theory is itself a law. 



THE VALUE OF THEORIES 



After this protest against a dangerous misunderstand- 

 ing, let us return and develop further our view of theories. 



