254 Journal of the Asiatic Society of Bengal. [May, 19 10. J 



and aw, which have the check, in Sonepur form their genitive 



mi 



the predicative infix i is changed into j. So this j represents 

 rather a euphonical change, than proving the presence of a 

 semi-consonant as the compiler would have it. All this goes to 

 show that the Mundari equivalents of the Santali semi-consonants 



/ 



Santali. 



differ widely f\ 



An unbiassed examination of the grammars and original 

 language specimens of Mundari must have led to the acknow- 

 ledgment of this difference. But as it has been looked at 

 through Santali spectacles and no justice was done to its 

 phonetical peculiarities, the language has been misrepresented 

 and distorted in the records of the Linguistical Survey. 



As a principle, I think, in every language only that is 

 written what is really spoken. Nobody will find fault with 

 the French, because they have no d in their jour (Latin dies), 

 which is still preserved in the pronunciation of Italian 

 giorno ; or with the English, because they write no g in lay 

 (Saxon leggan) as the Germans do in legen ; or with Hindi- 

 speaking people who write the Tadbhava word ^i^if (Sanscr 

 ^TRV ) instead of the Tatsama word ^rfa^ used in Bengali ; 

 or with somebody transliterating the Urdu word tarjuma 



(Arabic *+^J*) into Roman characters without h, because Per- 

 sian has that h. So Mundari words should, and I am sure, 

 will also in future not be written with a, a, k', ch', t\ and p', 

 which letters represent Santali sounds not spoken in Mundari. 

 The letters a and a are to be dropped altogether. For the dental 

 and labial semi-consonants d' and b' respectively should be 

 used and for the vowel check the sign ('), as Father Hoffmann 

 used, them. The apostrophe is preferable to the colon (:) 

 (visarga in Nagri), because the latter with Roman characters is 

 an interpunction only, while the apostrophe serves to denote 

 elisions and similar checks. I have written the Mundari words 

 in this paper as suggested. In Devanagri for the vowel check 

 the visarga is used ; for the dental and labial semi-consonants, 

 at least of the roots, ^ and ^[ (with virama) should be written. 



I am afraid, the section on Mundari phonetics in the 

 Linguistical Survey will disappoint all who live among the Mun- 

 das and know the language, and will be misleading for beginners 



use 



this 



the Linguistical Survey, which should be corrected in a publi- 

 cation of such prominence and authority. 



