Professor BtfHLER on the Sanscrit Linguals. 117 



rowed by the Dra vidians from the Sanskrit, Tamil at least 

 would have altered them. 



4. Tamil, which of all Dravidian tongues came least 

 into contact with Sanskrit, uses them most frequently — a 

 circumstance incompatible with the supposition of the deri- 

 vation of the cerebrals from Sanskrit. 



Dr. Caldwell, I think, has proved two things beyond all 

 doubt; lstly, that the so called cerebrals — or as we shall term 

 them henceforth, Unguals — of the Dravidian dialects are 

 not derived from the "Sanskrit ; and 2ndly, that they did 

 not belong to the original sounds of the primitive Indo- 

 European tongue. 



But I cannot allow, that he has established sufficiently 

 his own assertion, viz., that the Sanskrit linguals owe their 

 origin to Dravidian and Scythian influence. For he has 

 forgotten to prove, that sounds can be or are introduced 

 from one language into another, and to show, how the 

 migration of the linguals from Dravidian into Sanskrit was 

 effected. As the case stands, a stickler for the purity of 

 Sanskrit might start another hypothesis — that the Sanskrit 

 might have developed independently the sounds in question 

 just as the English has done with its palatals. Likewise a 

 full discussion of the origin of the laws according to which 

 the greater number of the Sanskrit linguals is produced from 

 dentals, through the influence of a preceding r, ri, 11 or sh, 

 would have been highly desirable. Lastly, Dr. Caldwell's 

 statements contain a little error in point of fact. He says 

 " none of the cerebrals (linguals) have ever been discovered 

 in any of the primitive languages which are related to 

 Sanskrit (Indo-European)." This is perfectly true in regard 

 to t, th, dh, n (b. ) But the Sanskrit r, ri, ri and sh are accord- 

 ing to the testimony of the grammarians * and of the 

 pronunciation of the "modern Pandits, likewise lingual; 



