PROBLEMS IN INDIAN LANGUAGES 115 



important, the Rig- Veda, through the application of the statistical 

 method by Professor Whitney, to whom the historical investigation 

 of Sanskrit owes more than to any other scholar, as well as by Pro- 

 fessor Lanman and others. Moreover, in the hymns of the Rig- Veda 

 itself, the existence of chronological strata has been discovered, 

 and some important general results have been arrived at, chiefly 

 by the labors of Professor Lanman, Professor Oldenberg, and the late 

 Abel Bergaigne. The most prominent problem which here confronts 

 Vedic scholarship is, by means of the minute investigation of all the 

 available evidence, phonetic, grammatical, lexicographical, metrical, 

 to ascertain the lines of demarcation dividing these literary strata. 

 The solution of this problem is of the highest importance for the 

 history of the Sanskrit language and literature. The work on Vedic 

 Meter recently (1905) published by Professor E. V. Arnold, of Ban- 

 gor, Wales, contributes valuable material to its solution as far as 

 the metrical evidence of the Rig-Veda is concerned. 



Another unsolved problem, which partly depends on the one just 

 mentioned, is the approximate age of the Vedic language and litera- 

 ture, and the approximate date of the Aryan migration into the north- 

 west of India. Its solution appears to me to have made no advance 

 during the last forty-five years. Indeed, the question seems to be 

 invested with more doubt now than it was then. For there is, at the 

 present time, a difference of more than 3000 years between the 

 lowest and the highest estimate of the beginning of the Vedic age. I 

 cannot help thinking that this enormous divergence will, by patient 

 investigation, be reduced to one of a very % few centuries. Professor 

 Jacobi's astronomical theory based on the doubtful interpretation 

 of a Vedic word, which would indicate that the rainy season in the 

 early Vedic period began under astronomical conditions different from 

 those of later times, is ingenious, but has in my opinion been refuted 

 by Professor Oldenberg. According to this theory, the Vedic period 

 would begin about 4500 B.C. It seems to me quite incredible that 

 the comparatively small divergence between the language of the 

 earliest Vedic period and that of Panini (who dates from about 

 300 B.C.), a divergence hardly greater than that between Homeric 

 and Attic Greek, should have required more than 4000 years to 

 accomplish. Considering how very closely the language of the oldest 

 part of the Avesta, the Gathas, estimated to date from about 600 B.C., 

 approximates to that of the oldest Veda, I find it hard to believe 

 that very many centuries could have elapsed from the time when the 

 Indians and Persians were still one people. In fact, 1500 B.C. seems 

 to me to be rather a high estimate for the approximate date at 

 which the Indo-Iranians separated and the Indians invaded the 

 northwest of India. More definite knowledge of the chronology of 

 the Rig- Veda, coupled with all the evidence which Iranian philology 



