THEORY OF LANGUAGE. 399" 



ing a similar idea in others, than signs fully fitted to convey 

 our meaning. It is in this case that we are most certain 

 of having performed an act of understanding. This fact may 

 convince us that language is not the medium of private 

 thought. 



Another mistaken view seems in some degree to influence 

 the speculations of metaphysical writers On language, though 

 it is not expressly maintained : — That man is originally prone 

 to communicate all his thoughts to others, and that language 

 is nothing else than the spontaneous catenations of our 

 thoughts put in words. No analytical inquiry is necessary 

 to demonstrate the fallacy of such an opinion ; but it may be 

 useful to keep in mind some facts on the subject, which will 

 serve to put us on our guard against any unconscious leaning 

 to it, in estimating the justness of particular theories of lan- 

 guage. A person whose conversation consists of mere think- 

 ing aloud is always considered as betraying weakness by an 

 imprudent deviation from the original purposes with which 

 language is employed. We expect a speaker to have some 

 other object in view, than merely to gratify a proneness to 

 communicate his thoughts. Many thoughts which it is reckon- 

 ed allowable and laudable for him to entertain cannot be ex- 

 pressed to others, without subjecting him to the charge of im- 

 pertinence. 



There is one fact in the history of man, which is certainly 

 worthy of more attention than it generally receives, being too 

 often overlooked in its relation to some other views with which 

 we are occasionally occupied, — that thought is of much earlier 

 date in the life of the individual than language. Inattention 

 to this fact appears to be the reason, that those who have ex- 

 pected to find in etymology a developement of the first prin- 



3 E 2 ciples 



