374 EEV. 



speech " The mode in \vh: 

 stitutes one of the most di 

 character, and one which materially contributes to the determin- 

 ation of its relationship."* Tamil forms its preterite by adding 

 'd' which for euphony is sometimes preceded by 'n,' owing to the 

 Tamil fondness for nasalization says Caldwell. This may or may 

 not be the reason for the appearance of the 'n,' but the conmion 

 form of the preterite in Kabi^ Wiradhuri and other Australian 

 In tlie Dravidian the accent is on the 

 This is commonly the case in Australian and is 

 by the agglutinating character of both 

 languages. 



It is a most formidable obstacle to the theory of the relationship 

 of Dravidian and Australian speech that so distinguished a philo- 

 logist as Dr. F. Miiller, who was on the scientific staff of the 

 " Novara," should have declared emphatically against it. He says 

 that, viewed even apart from the racial difference the glossarial 

 affinities are too weak to support the affirmation that the languages 

 are genealogically related. There are he adds, certain points 

 observable which lead to the conclusion that such connection is 

 impossible (unmoglich). Now for his arguments. He asserts that 

 if a genealogical relationship existed, it would receive fullest expres- 

 sion in the speech of the West of Australia which is geographically 

 nearest the Dravidian languages. But this is an unwarranted asser- 

 ption that affinity of speech depends upon 



:'o7: 



pretty well proven in this essay, that migration was from the 

 north not from the west, and that the west was one of the corners^ 

 into which the purer Papuan race was forced. Further, he affirms 

 that the 'nan-nin' type of pronoun prevails more or less in Thibet 

 China and elsewhere as well as in Central India. A good argu- 

 ment, but the likeness is not generally so close. He further 

 objects to the rules of class-marriage being introduced as evidence 

 of relationship, because similar regulations are found in other parts. 

 I think however, that the likeness between those of India and 

 Australia is most marked. Besides the reasons already adduced 

 for affirming a genealogical connection between the people of 

 Central India (called Dravidians here be it observed only for con 

 venience, aborigines of India would be more appropriate), I have 

 fallen upon a cluster of glossarial analogies which joined with the 

 other evidence should outweigh Dr. Miiller's awful " uninnglich.' 

 Hitherto philology has been baffled in the attempt to establish 

 relationship between the Australian numerals and those of 

 languages to the north. A conmion Australian term for Uvo 



* Caldwell, Dravidian Grammar, p. 300 



