90 WHAT IS SCIENCE? 



found ; we might find theories to satisfy the first two 

 conditions by the exercise of sufficient patience in a pro- 

 cess of trial and error ; but how can we possibly be sure 

 that they will satisfy the third ? The answer that 

 must be given will be clear to the reader if he has accepted 

 the results of the previous discussion. There cannot be, 

 from the very nature of the case, any kind of rule whereby 

 the third condition may certainly be satisfied ; the mean- 

 ing of the condition precludes any rule. Actually theories 

 are always suggested in view of the first two conditions 

 only ; and actually it turns out that they often fulfil 

 the third. And once again they most often turn out to 

 fulfil the third when they are suggested by those excep- 

 tional individuals who are the great men of science. It 

 is when the theory seems to them to explain the laws, 

 when the ideas introduced by it appear to them acceptable 

 and satisfactory, that nature conforms to their desires, 

 and permits to be established by experiment the laws 

 which are the direct consequence of those ideas. 



The form in which that statement is put may appear 

 rather extravagant, and we shall return later and consider 

 some questions which it seems to raise. But the general 

 view that true theories are the expression of individual 

 genius will probably seem less paradoxical than the view 

 put forward in the previous chapter, that true laws also 

 contain a personal element. The difference is indicated 

 by the words that are used ; we speak of the " discovery " 

 of a law, but of the " invention " of a theory. A law, 

 it is implied, is something already existing which merely 

 lies hidden until the discoverer discloses it ; a theory, on 

 the other hand, does not exist apart from the inventor ; 

 it is brought into being by an effort of imaginative 

 thought. I do not think that distinction will bear 

 examination ; it seems to me very difficult to regard laws 

 and not theories as something existing independently of 

 investigation and wholly imposed by the external world, 



