l8 SCIENCE AND METHOD. 



(Which, then, are the facts that have a chance of 

 /recurring? In the first place, simple facts. It is 

 , evident that in a complex fact many circumstances 

 are united by chance, and that only a still more 

 improbable chance could ever so unite them again. 

 But are there such things as simple facts ? and if there 

 are, how are we to recognize them ? Who can tell 

 that what we believe to be simple does not conceal 

 an alarming complexity? All that we can say is 

 that we must prefer facts which appear simple, to 

 those in which our rude vision detects dissimilar 

 elements. Then only two alternatives are possible ; 

 either this simplicity is real, or else the elements 

 are so intimately mingled that they do not admit of 

 being distinguished. In the first case we have a 

 chance of meeting the same simple fact again, either 

 in all its purity, or itself entering as an element into 

 some complex whole. In the second case the intimate 

 mixture has similarly a greater chance of being re- 

 produced than a heterogeneous assemblage. Chance 

 can mingle, but it cannot unmingle, and a combination 

 of various elements in a well-ordered edifice in which 

 something can be distinguished, can only be made 

 deliberately. There is, therefore, but little chance that 

 an assemblage in which different things can be dis- 

 tinguished should ever be reproduced. On the other 

 hand, there is great probability that a mixture which 

 appears homogeneous at first sight will be reproduced 

 several times. Accordingly facts which appear simple, 

 even if they are not so in reality, will be more easily 

 brought about again by chance. 



It is this that justifies the method instinctively 

 adopted by scientists, and what perhaps justifies it 



