issues as may directly interest a student devoted to language for its own sake.  The formal method of investigating language, in the meantime, can hardly supply the needed spur.  Analysis is all very well so long as its ultimate purpose is to subserve genesis—­that is to say, evolutionary history.  If, however, it tries to set up on its own account, it is in danger of degenerating into sheer futility.  Out of time and history is, in the long run, out of meaning and use.  The philologist, then, if he is to help anthropology, must himself be an anthropologist, with a full appreciation of the importance of the historical method.  He must be able to set each language or group of languages that he studies in its historical setting.  He must seek to show how it has evolved in relation to the needs of a given time.  In short, he must correlate words with thoughts; must treat language as a function of the social life.

* * * * *

Here, however, it is not possible to attempt any but the most general characterization of primitive language as it throws light on the workings of the primitive intelligence.  For one reason, the subject is highly technical; for another reason, our knowledge about most types of savage speech is backward in the extreme; whilst, for a third and most far-reaching reason of all, many peoples, as we have seen, are not speaking the language truly native to their powers and habits of mind, but are expressing themselves in terms imported from another stock, whose spiritual evolution has been largely different.  Thus it is at most possible to contrast very broadly and generally the more rudimentary with the more advanced methods that mankind employs for the purpose of putting its experience into words.  Happily the careful attention devoted by American philologists to the aboriginal languages of their continent has resulted in the discovery of certain principles which the rest of our evidence, so far as it goes, would seem to stamp as of world-wide application.  The reader is advised to study the most stimulating, if perhaps somewhat speculative, pages on language in the second volume of E.J.  Payne’s History of the New World called America; or, if he can wrestle with the French tongue, to compare the conclusions here reached with those to which Professor Levy-Bruhl is led, largely by the consideration of this same American group of languages, in his recent work, Les Fonctions Mentales dans les Societes Inferieures ("Mental Functions in the Lower Societies").

If the average man who had not looked into the matter at all were asked to say what sort of language he imagined a savage to have, he would be pretty sure to reply that in the first place the vocabulary would be very small, and in the second place that it would consist of very short, comprehensive terms—­roots, in fact—­such as “man,” “bear,” “eat,” “kill,” and so on.  Nothing of the sort is actually the case.  Take