114 INDO-IRANIAN LANGUAGES 



has been almost exclusively directed to the study of the earliest 

 linguistic phase of India. In consequence of this, coupled with the 

 fact that the study of Sanskrit in Europe began in a scientific age, 

 we have not only long possessed a Sanskrit Dictionary which in 

 comprehensiveness and accuracy surpasses that of any other dead 

 language; but also a grammar dealing on historical principles with 

 the Sanskrit language by that great American scholar, W. D. Whitney, 

 which stands unequaled in a similar way. As having led to the 

 foundation of comparative philology, Sanskrit long maintained 

 an exaggerated preeminence in that science. This was followed 

 about thirty years ago by a reaction which, starting from the dis- 

 covery that the vowel-system of Sanskrit is less primitive than that 

 of the European languages, tended to assign quite a subordinate 

 position to Sanskrit. Though I have from my student days at the 

 University of Gottingen given a good deal of attention to compar- 

 ative philology, I do not consider myself entitled to express an 

 authoritative opinion on the details of this science. I nevertheless 

 venture to make the general assertion that Sanskrit still occupies 

 and will continue to occupy a dominant position in comparative 

 grammar. By this I mean that, if all the linguistic material supplied 

 by Sanskrit were eliminated, the lacunaB in comparative philology 

 would be immeasurably greater than if the linguistic material of any 

 other Indo-European language were lacking. This seems to me to 

 be evident in the great authoritative work of Professor Brugmann, 

 the leader of comparative philology at the present day. It will, I 

 have little doubt, be still more clearly established on the comple- 

 tion of the comparative Sanskrit grammar of Professor Wacker- 

 nagel, of Gottingen, the second volume of which was passing through 

 the press before I left England and has since (1905) appeared. This 

 work will, I think, surpass all other comparative grammars of indi- 

 vidual Indo-European languages hitherto published, both in fullness 

 of detail and scientific trustworthiness. 



Since in the early days of Sanskrit studies, European scholars 

 became acquainted only with that later phase of the ancient lan- 

 guage of India which is familiar to the Pandits, and is commonly 

 known as Classical Sanskrit, research remained almost entirely 

 limited to that dialect till about the middle of the nineteenth century. 

 Since then the earlier language of the Vedas has been assiduously 

 investigated in Europe and America. All the four Vedas have long 

 been accessible in thoroughly scientific editions, and much progress 

 has been made in the study of their language, their matter, and their 

 mutual relations. The Vedas have been proved by internal evidence 

 to be considerably anterior to the rise of Buddhism, that is, to have 

 been composed long before 600 B.C. The language of the three lesser 

 Vedas has further been shown to be posterior to that of the most 



