244 PHYSICAL INVESTIGATION. [PART I. 



language upon the ideal world ; for it is clear, that regulation 

 of the latter greatly depends upon the former. Entirely asyn- 

 thetic, monosyllabic languages, allow our individual ideas, 

 which correspond to individual words, to stand in independent 

 juxtaposition, merely indicating some rude distinctions between 

 chief and secondary ideas. In contrast with them, poly syn- 

 thetic languages force us to grasp the whole idea, and inti- 

 mately to connect the secondary ideas with the chief idea, to 

 take in at one glance the whole situation, not piecemeal and 

 successively. That they prevent the dismemberment of ideas 

 in a greater degree than the former, is proved by many sub- 

 stantive nouns in these languages, like "hand," "father," 

 " son," occurring, not separately, but always in connexion with 

 a possessive pronoun. It is of the greatest importance, for 

 the regulation of the mass of our ideas, how many and what se- 

 condary ideas our language induces us to connect as integral 

 parts with the chief idea to which they refer, or what may be 

 added as relatively independent parts. It is not less important 

 whether, as in inflected languages, the relations are expressed 

 by particles which, separated from the chief word, have no 

 distinct signification. These grammatical forms of the mother 

 language become habitual to us before we arrive at reflection ; 

 for what language presents combined in one sound, we conceive 

 together, and what it presents in a separate form, we conceive 

 as relatively separate. 



These elementary habits in connecting individual ideas, be- 

 long to the most important special laws to which the concep- 

 tion of man is subject ; and on account of the power which 

 they exert in the elementary construction of our ideas, an 

 essential change of the structure of a language in a people, 

 which continues as a people, is highly improbable. There is 

 no doubt that a gradual change in the grammatical structure 

 of a language is possible ; and if it be probable that all 

 merely grammatical words (forms) were originally words of 

 independent signification, and that even the syllables of inflec- 

 tion sprung from originally independent words, which were 

 merely added to the chief word, then there exists between the 

 types of language as little an absolute constant difference as 



