2,s2 ANTIENT METAPHYSICS. Book III. 



from the actions of body, fuch as ivclgh^ ponder^ rejiedl^ pi-of- 

 pe£l, retrofpe&^ &c. And In this way M. Gebelin thinks that the 

 letters, or a certain compofition of them, have an analogy to the 

 ideas and fentiments of the mind, as well as to the objedls perceived 

 by our fenfes. Nor is M. Gebelin fingular in this opinion ; for 

 Plato, in the Cratylus, has maintained the fame dodrine ; but for 

 my part I think that there is fo much fancy and conjedure in this 

 matter, and fo little certainty, that I do not chufe to fay any tiling 

 more concerning it, except that I think it fliews language not to be 

 a natural operation, as many fuppofe it to be, and among others M. 

 Gebelin himfelf, but a work of the greateft art, fo great, that we 

 are not able to give a latisfadlory account of the firft principles of 

 it, I mean the formation of the radical founds, or firft words of it. 



And here I return again to the Shanfcrit language, which, accord- 

 ing to the account given by Father Pons the Jefuit, who was a mif- 

 fionary in India, is a moft wonderful piece of art, and of fcience 

 too ; for it analyfes all the particular ideas exprefled by the words, 

 into the general ideas from which they arlfe : And thefe are exprelTed. 

 by monofyllables, which are the roots of the language ; for M. Ge- 

 belin is certainlv \n tlio right, when he fays, that the roots of the 

 primitive language mvift have been all monolyllables, which are the 

 fimpleft words, and therefore beft fitted for derivation and compo- 

 fition. From thef: roots are deduced, in long order and with great 

 variety, but afcevt:''ned by fixed and determinate rules, ail the ftve- 

 ral words of ^he language, exprefling particular ideas, wliich fall un- 

 der the general ideas denoted by the roots. And he gives an ex- 

 ample of this in the monofyllable km, which denotes the general 

 idea of aHion ; and from it are formed, by derivation and compofi- 

 tion, all the words, of number no doubt veiy great, exprefling the 

 different kinds of action. And in this way, he fays, if you are 

 pofTefTed of the roots of the language, and the rules of derivation and 



com- 



