41 8 MENTAL EVOLUTION IN MAN. 



passed from the root-stage upwards. Having done this, I 

 proceeded to consider the question touching the origin of 

 these roots themselves. 



First, as to their number, we found that the outside esti- 

 mate, in the younger days of philological research, gave one 

 thousand as a fair average of the roots which go to feed any 

 living language ; but that this estimate might now be safely 

 reduced by three-fourths. Indeed, in his latest work. Professor 

 Max M tiller professes to have reduced the roots of Sanskrit 

 to as low a number as 121, and thinks that even this is 

 excessive. Regarding the character of roots, we saw that 

 some philologists look upon them as the actual words which 

 were used by the pre-historic speakers, who, therefore, "talked 

 with one another in single syllables, indicative of ideas of 

 prime importance, but wanting all designation of their rela- 

 tions." * On the other hand, it is now the generally accepted 

 belief, that "roots are the phonetic and significant types 

 discovered by the analysis of the comparative philologist as 

 common to a group of allied words," f — or, as it were, composite 

 phonograms of families of words long since extinct as 

 individuals. We saw, however, that this difference of opinion 

 among philologists does not affect the present inquiry, seeing 

 that even the phonetic-type theory does not question that the 

 unknown words out of the composition of which a root is now 

 extracted must have been genetically allied with one another, 

 and exhibited the closeness of their kinship by a close 

 similarity of their sounds. 



A much more important question for us is the character 

 of these roots with respect to their significance. In this con- 

 nection we found that they indicate what Professor Max Miiller 

 calls "general ideas," or " concepts ; " bear testimony to an 

 already and, comparatively speaking, advanced stage of social 

 culture ; are all e::pressive either of actions or states ; and 

 betray no signs of imitative origin. Taking each of these 

 characters separately, we found that although all the 121 

 roots of Sanskrit are expressive of general ideas, the order of 



* Whitney. t Saycc. 



