COSMIC PHILOSOPHY 



fied the problem. His classification well enough 

 expresses the order of development of the sci- 

 ences, in so far as their development has de- 

 pended merely on the relative simplicity or 

 complexity of the phenomena with which they 

 have had to deal. It rests upon the assumption 

 that, with few and unimportant exceptions, the 

 progress of generalization has been from the 

 simple to the complex. Now this is not the 

 case. The progress of generalization has indeed 

 been partly determined by the relative simpli- 

 city or complexity of the phenomena to be 

 generalized (and this fact accounts for the con- 

 siderable amount of truth which the Comtean 

 doctrine contains) ; but it has been also deter- 

 mined by several other circumstances. In the 

 chapter on " Laws in General " to be found in the 

 first edition of " First Principles," but omitted 

 in the revised edition, 1 Mr. Spencer has called 

 attention to some of these circumstances. He 

 reminds us that not only are phenomena early 

 generalized in proportion as they are simple^ 

 but also in proportion as they are conspicuous or 

 obtrusive. " Hence it happened that after the 

 establishment of those very manifest sequences 

 constituting a lunation, and those less manifest 

 ones marking a year, and those still less mani- 

 fest ones marking the planetary periods, astro- 



1 [Published in the Essays, Library Edition vol. ii. pp. 

 144-160.] 



34 



