76 SCIENCE AND METHOD. 



should have no hesitation in saying that this was 

 chance. 



Our frailty does not permit us to take in the whole 

 universe, but forces us to cut it up in slices. We 

 attempt to make this as little artificial as possible, 

 and yet it happens, from time to time, that two of 

 these slices react upon each other, and then the effects 

 of this mutual action appear to us to be due to chance. 



Is this a third way of conceiving of chance? Not 

 always ; in fact, in the majority of cases, we come 

 back to the first or second. Each time that two 

 worlds, generally foreign to one another, thus come 

 to act upon each other, the laws of tliis reaction 

 cannot fail to be very complex, and moreover a very 

 small change in the initial conditions of the two 

 worlds would have been enough to prevent the 

 reaction from taking place. How very little it would 

 have taken to make the man pass a moment later, 

 or the slater drop his tile a moment earlier ! 



VI. 



Nothing that has been said so far explains why 

 chance is obedient to laws. Is the fact that the 

 causes are small, or that they are complex, sufficient 

 to enable us to predict, if not what the effects will 

 be in each case, at least what they will be on the 

 average? In order to answer this question, it will 

 be best to return to some of the examples quoted 

 above. 



I will begin with that of roulette. I said that the 

 point where the needle stops will depend on the 

 initial impulse given it. What is the probability that 

 this impulse will be of any particular strength? I 



