58 SYSTEMS OF CONSANGUINITY AND AFFINITY 



of these stock languages with each other was then made, to find how far the root 

 forms of their vocables were identical ; and also to discover another class of affini- 

 ties Avhich the grammatical structure of these stock languages might reveal. It 

 was early ascertained that grammatical structure was the ultimate criterion by 

 which the admission of a doubtful language must be determined, since the number 

 of constant vocables became smaller in the extreme branches of a family ethnically 

 connected, and the subtile process of naturalization might explain their presence m 

 each without being indigenous in either. In this manner a true family of lan- 

 guages was bound together by common grammatical forms, and by the more simple 

 and conclusive bond of common vocables. The Turanian dialects, so called, have 

 been much less investigated, and are less thoroughly known than the Aryan or 

 Semitic, in consequence of their great numbers, their inaccessible position, and the 

 vast extent of the areas over which they are spread. It is not claimed that the 

 same coincidences in grammatical forms, or identity of vocables exist in the several 

 branches of the Turanian speech. A limited number of common words and of 

 common roots, running, not through all the branches of the Turanian speech, but 

 here and there through certain portions, furnished some evidence of original unity, 

 but not enough, standing alone, to sustain the classification. These dialects also 

 agree with each other with respect to their articulation. They are agglutinated in 

 their structure, and this common feature has entered, to some extent, into the basis 

 upon which they have been organized into a family of languages. If, however, 

 agglutination is a stage of growth or development through which all languages 

 must pass after emerging from the monosyllabic and before reaching the inflectional, 

 which is the received opinion, it does not furnish any basis for the organization of 

 these dialects into a family of speech. Beside this, the use of this common feature 

 of agglutination, as a ground of classification, forces the Chinese and its cognate 

 dialects into a position of isolation, and interposes a barrier between them and the 

 proper Turanian dialects where none such may exist. For these reasons the reJuc- 

 tion of this great body of languages, under a Northern and Southern division, into 

 one common family, the Turanian, could not be other than a provisional arrange- 

 ment. The science of language is impeded rather than advanced by raising to the 

 rank of a family of languages such an incongruous assemblage of dialects as are 

 now included in the Turanian. The Aryan and Semitic standard is much to be 

 preferred. 



Upon the basis of the systems of consanguinity and affinity of the Asiatic 

 nations, they divide themselves into at least two distinct families, each of which, 

 it seems probable, will ultimately become as clearly distinguished from the 

 other as the Aryan now is from the Semitic. A comparison of the systems of a 

 limited number of these nations has led to singular and rather unexpected 

 results. The system of the Turanian family proper, which Avill be presented in 

 a subsequent part of this work, separates it from the Arj-an and Semitic by a 

 line of demarcation perfectly distinct and traceable. Such a result furnishes no 

 occasion of surprise. On tlie other hand, it excludes from the Turanian connec- 

 tion, by a line not less distinct and unmistakable, the Ugrian and Turk stocks, 

 which are the principal members of the Northern division of the family, as now 



