518 LOGIC OF THE MORAL SCIENCES. 



speak; including its own notions respecting the 

 sources of all those sensations and impressions. It is 

 not only impossible to do this completely, but even to 

 do so much of it as should constitute a tolerable 

 approximation. One apparently trivial circumstance 

 which eluded our vigilance, might let in a train of 

 impressions and associations sufficient to vitiate the 

 experiment as an authentic exhibition of the effects 

 flowing from given causes. No one who has sufficiently 

 reflected on education is ignorant of this truth; and 

 whoever has not, will find it most instructively illus- 

 trated in the writings of Rousseau and Helvetius on 

 that great subject. 



Under this impossibility of studying the laws of the 

 formation of character by experiments purposely con- 

 trived to elucidate them, there remains the resource 

 of simple observation. But if it be impossible to 

 ascertain the influencing circumstances with any 

 approach to completeness, even when we have the 

 shaping of them ourselves, much more impossible is 

 it when the cases are further removed from our obser- 

 vation, and altogether out of our control. Consider 

 the difficulty of the very first step of ascertaining 

 what actually is the character of the individual, in each 

 particular case that we examine, There is hardly any 

 person living, concerning some essential part of whose 

 character there are not differences of opinion even 

 among his intimate acquaintance : and a single action, 

 or conduct continued only for a short time, goes a 

 very little way indeed towards ascertaining it. We 

 can only make our observations in a rough way, and 

 en masse; not attempting to ascertain completely, in 

 any given instance, what character has been formed, 

 and still less by what causes ; but only observing in what 



