94 Darwinism and Other Essays. 



be consistent in principle with their past testi- 

 mony. Whatever system of forces we estimate 

 or measure in support of our implicit belief in the 

 constancy of Nature, we must sooner or later ap- 

 peal to some fundamental unit of measurement 

 which is invariable. Without some such constant 

 unit we cannot prove that the order of Nature is 

 uniform : but we cannot prove the constancy of 

 such a unit without referring it to some other 

 unit, and so on forever ; while to assume the con- 

 stancy of such a unit is simply to assume the 

 whole case. 



It would seem, therefore, that our belief in the 

 trustworthiness of Nature is not properly described 

 when it is treated simply as a vast induction. It 

 should rather be regarded as a postulate indispen- 

 sable to the carrying on of rational thought, a 

 postulate ratified in every act of experience, but 

 without which no act of experience can have any 

 validity or meaning. It is for taking this view of 

 the case that Mr. Spencer is charged with rearing 

 a system of philosophy upon " undemonstrable 

 beliefs assumed to be axiomatic and irresistible." 

 Considering that the undemonstrable belief in 

 question is simply the belief in the constancy of 

 Nature, one would be at a loss to see what there 

 is so very heinous in Mr. Spencer's proceeding, 



