Methods of Ethics. 17 



tion, and that no crack would have occurred had 

 the heat been equally diffused as in thin glass 

 vessels through which it passes rapidly. The 

 illustration suggests that deductive science, hav 

 ing apprehended the reasons of phenomena, may 

 be able to predict their occurrence ; and every 

 body is acquainted with the sublime prophetic 

 achievements of astronomy. This power of pre 

 diction clearly marks off the deductive from the 

 experimental sciences. And so much being 

 premised, we are now prepared for the inquiry 

 whether ethics belongs to either division ? If it 

 be of the same general type as the sciences of 

 nature, it must be either a deductive or an experi 

 mental science. 



In assigning ethics to either of these classes, 

 however, one assumption is made too significant 

 to pass without distinct mention. The sciences of 

 nature all rest upon the presupposition that events 

 follow one another in a fixed and regular order, 

 that the same cause under the same circumstances 

 always produces the same effects, that the entire 

 realm of natural phenomena is subject to the 

 reign of inexorable law. Deny the principle of 

 universal causation, and natural science is smitten 

 with paraly sis. You may be in doubt about the 

 proof of the principle ; you may attempt to for- 



