240 PHYSICAL INVESTIGATION. [PART I. 



of vocabularies, formerly so much in vogue, are now deemed 

 insufficient, as these comparisons were without method, pro- 

 ceeded from no fixed principle, and the estimation of the pre- 

 vailing differences of words was quite arbitrary. Even in lan- 

 guages where the resemblances of many words are unmistake- 

 able, it cannot be decided from them alone, whether they have 

 been transmitted or merely borrowed, whether they are the 

 consequence of a genealogical relationship, or of an exchange 

 or communication from one language to another originally dis- 

 tinct, but which subsequently came in contact. Many of these 

 vocabulary resemblances may also be accidental, and produced 

 by identical original invention. The extent and degree of the 

 resemblance of individual words must also be taken into con- 

 sideration, as well as their signification, especially such as 

 designate common things indispensable to peoples even in 

 a low state of cultivation ; for these are not so much altered 

 in an exchange of languages as those belonging to things found 

 only among peoples more civilized. But all this does not in- 

 validate the general principle, that word comparisons alone are 

 insufficient to decide upon the affinity of languages. 



It seems, then, that, in judging of the affinity of languages, 

 greater importance is attached to grammatical structure than 

 even to the resemblance of the roots of a language. Equality 

 of original invention with regard to the structure of language 

 among perfectly distinct races is, on account of the great com- 

 plications and great variety of grammatical proportions so 

 improbable that it nearly amounts to an impossibility. It 

 appears, therefore, requisite that two languages should, on 

 comparison, agree to some extent in both the chief elements 

 before we can decide on their affinity. We shall endeavour to 

 point out the reasons which justify us in inferring the consan- 

 guinity of peoples from the quality of the grammatical struc- 

 ture of their languages. 



When a language is transmitted from one generation to an- 

 other, not merely the words, but their mode of connexion, as 

 sentences, is also transmitted and appropriated by unconscious 

 imitation, and thus becomes fixed. The latter point is not 

 sufficiently appreciated, though it is quite clear that we do not 



