EARLY HISTORY OF MANKIND. 219 



elaborate system of inflection and agglutination is shown to 

 be simply a farther development of the language-forming 

 principle, as it may be called or the Chinese system may 

 be described as an arrestment of this principle at a particular 

 early point. It has been fully shown, that between the struc- 

 ture of the American and other families sufficient affinities 

 exist to make a common origin or early connexion extremely 

 likely. The verbal affinities are also very considenfble. Hum- 

 boldt says, " In eighty-three American languages examined 

 by Messrs. Barton and Vater, one hundred and seventy words 

 have been found, the roots of which appear to be the same ; 

 and it is easy to perceive that this analogy is not accidental, 

 since- it does not rest merely upon imitative harmony, or on 

 that conformity of organs which produces almost a perfect 

 identity in the first sounds articulated by children. Of these 

 one hundred and seventy words which have this connexion, 

 three-fifths resemble the Manchou, the Tongouse, the Mongol, 

 and the Samoyed ; and two-fifths, the Celtic and Tchoud, the 

 Biscayan, the Coptic, and Congo languages. These words 

 have been found by comparing the whole of the American 

 languages with the whole of those of the Old World ; for 

 hitherto we are acquainte4 with no American idiom which 

 seems to have an ^clusiv.e correspondence with any of the 

 Asiatic, African, or European tongues.( 89 ) Humboldt and 

 others considered these words as brought into America by 

 recent immigrants ; an idea resting on no proof, and which is 

 much discountenanced by. the common words being chiefly 

 those which represent primary ideas ; besides, we now know, 

 what was not formerly perceived or admitted, that there are 

 great affinities of structure also. I may here refer to a curious 

 mathematical calculation by Dr. Thomas Young, to the effect, 

 .that if three words coincide in two different languages, it is 

 ten to one they must be derived in both cases from some 

 parent language, or introduced in some othe* manner. " Six 

 words would give more," he says, " than seventeen hundred 

 to one, and eight near 100,^00; so that in these gases the 

 evidence would be little short of absolute certainty." He 



