THE ORIGIN OP LANGUAGE 55 



lutely no trace in the roots of early Sanskrit. These 

 concepts were not in the language, and we must 

 naturally infer that they did not exist in the minds of 

 the people using the language. 



Comparative Study of Living Languages. — Sanskrit is 

 evidently far from a primitive language, and we may 

 naturally turn to languages of the lower savage races 

 to see if among them we may not find something 

 more primitive. Such we certainly do find. Among 

 the various tribes whose languages have been studied 

 we find some that are on a par with the imaginary 

 early man when his speech consisted of a few words 

 only. It is sometimes a marvel to us who have such 

 difficulty in mastering one or two foreign tongues to 

 read of African explorers visiting tribe after tribe, 

 each with its own language, and learning quickly to 

 converse in them all. But the surprise somewhat dis- 

 appears when we learn of the real dearth of words 

 and the simi^licity of speech. Savage languages are 

 totally lacking in abstract words, showing, of course, 

 that the savage thinking must be in concrete rather 

 than abstract ideas. Frequently such a language 

 will have names for concrete objects, but no general 

 name for classes of objects. For example, one Zulu 

 tribe has ten names for different kinds of cows. It 

 has a name for a red cow, a white cow, etc., but it lias 

 no name for "cow." The Tasmanian language has 

 several words rejoresenting different kinds of trees, 

 but no word for "tree." Some travelers have sup- 

 posed that this represents a very complex language ; 

 but the reverse is the truth, for it represents a lan- 

 guage too crude to have formed any class objects, 

 and it certainly represents a type of thinking so dis- 

 tinctly in the concrete as never to have felt the need 



