BOAS] HANDBOOK OF AMERICAN INDIAN LANGUAGES 7 



istics — traits which are considered as characteristic of certain groups 

 of mankind; while still others are based primarily on the stud}^ of 

 the languages spoken by people representing a certain anatomical 

 type. 



The attempts that have thus been made have led to entirely differ- 

 ent results. Blumenbach, one of the first scientists who attempted 

 to classify mankind, first distinguished five races — the Caucasian, 

 Mongolian, Ethiopian, American, and Mala}". It is fairly clear that 

 this classification is based as much on geographical as on anatomical 

 considerations, although the description of each race is primarily an 

 anatomical one. Cuvier distinguished three races — the white, yellow, 

 and black. Huxley proceeds more strictly on a biological basis. 

 He combines part of the Mongolian and American races of Blumen- 

 bach into one, assigns part of the South Asiatic peoples to the Austra- 

 lian type, and subdivides the European races into a dark and a light 

 division. The numerical preponderance of the European types has 

 evidently led him to make finer distinctions in this race, which 

 he divides into the xanthochroic and melanochroic races. It 

 would be easy to make subdivisions of equal value in other races. 

 Still clearer is the influence of cultural points of view in classifica- 

 tions like those of Gobineau and Klemm (who distinguishes the 

 active and passive races), according to the cultural achievements of 

 the various t3"pes of man. 



The most typical attempt to classif^^ mankind from a consider- 

 ation of both anatomical and linguistic points of view is that of 

 Friederich Miiller, who takes as the basis of his primary divisions the 

 form of hair, while all the minor divisions are based on linguistic 

 considerations. 



Relations between Physical Type, Language, and Customs 



An attempt to correlate the numerous classifications that have 

 been proposed shows clearly a condition of utter confusion and con- 

 tradiction. If it were true that anatomical form, language, and cul- 

 ture are all closely associated, and that each subdivision of mankind 

 is characterized by a certain bodily form, a certain culture, and a cer- 

 tain language, which can never become separated, we might expect 

 that the results of the various investigations would show better 

 agreement. If, on the other hand, the various phenomena wliich 

 were made the leading points in the attempt at classification are not 



