BOAS] HANDBOOK OF AMERICAN INDIAN LANGUAGES 53 



not be denied that this point of view has certain elements in its favor; 

 but in the present state of our knowledge we can hardly say that it 

 would be possible to prove or to disprove it. 



We meet the same fundamental problem in connection with simi- 

 larities of languages which are too vague to be considered as proofs 

 of genetic relationship. That these exist is obvious. Here we have 

 not only the common characteristics of all human language, which 

 have been discussed in the preceding chapter, but also certain other 

 similarities which must here be considered. 



Influence of Environment on Language 



It has often been suggested that similarities of neighboring lan- 

 guages and customs may be explained by the influence of environ- 

 ment. The leading thought in this theory is, that the human mind, 

 under the stress of similar conditions, will produce the same results; 

 that consequently, if the members of the same race live in the same 

 surroundings, they will produce, for instance, in their articulate speech, 

 the same kind of phonetics, differing perhaps in detail according to 

 the variations of environment, but the same in their essential traits. 

 Thus it has been claimed that the moist and stormy climate of the 

 North Pacific coast caused a chronic catarrhal condition among the 

 inhabitants, and that to this condition is due the guttural pronuncia- 

 tion and harshness of their languages; while, on the other hand, the 

 mildness of the California climate has been made responsible for the 

 euphonious character of the languages of that district. 



I do not believe that detailed investigations in any part of the 

 world would sustain this theory. We might demand proof that the 

 same language, when distributed over different climates, should pro- 

 duce the same kind of modifications as those here exemplified; and 

 we might further demand that, wherever similar climates are found, 

 at least a certain approach to similarity in the phonetics of the lan- 

 guages should occur. It would be difficult to prove that this is the 

 case, even if we should admit the excvise that modifying influences 

 have obscured the original similarity of phonetic character. Taking, 

 for instance, the arctic people of the Old and New Worlds as a unit, 

 we find fundamentally different traits in the phonetics of the Eskimo, 

 of the Chukchee of eastern Siberia, and of other arctic Asiatic and 

 European peoples. The phonetics of the deserts of Asia and South 



