Indian Antiquary Vol 62

Author(s): Richard Carnac Temple, Charles E A W Oldham, S Krishnaswami Aiyangar, Devadatta Ramkrishna Bhandarka
Publisher: Swati Publications

Page 62

56
DRAVIDIC PROBLEMS
[MARCH, 1933
vehku (to desire ardently) is connected with the Dravidian base vě. (to be hot) which has produced numerous forms with the help of affixes. Here one does not see any absolute necessity to trace the form with the aydam to a base with final -1 or !, thougb one way con ceivably connect it with vei (to desire).
[C] Common instances of words with dydam in combinative positions are the following:
kal+tidu > kahd'idu (the stone is bad) mu! +tidu > muhdidu (the thorn is bad) pal+tuli > pahd'uli (many drops)
al+tinai > ahd'inai (inferior group) In the first three instances, alternatively we may have respectively also kat't'ridu, mullidu and pal'l'ruli. The following points are significant in connection with this combinative change
(a) The aydam appears only in connection with l or ?+the dental t. The surd involved is only the dental.
(b) The first word always has a short radical vowel ; if this vowel is long, no change takes place (cf. eritras 370 and 37). of Eluttadigaram, Tolkappiyam), and not even the assimi. lative conversion happens, e.g., pál+tidu would be retained as pål tidu (the milk is bad).
The process whereby the aydam is generated is here again similar to that in ahlu, ihtu mentioned above. When the components merge into each othor intimately, the higher accent falls on the syllable containing the surd (which becomes alveolar or retroflex on account of the influence of alveolar l or retroflex !, as the case may be) and the dydam is generated through the intermediate stage of the mouth-fricative corresponding to alveolar t' or retroflex . The alternative forms kat't'rídu and mutridu with geminated surds instead of the group dydam+surd, confirm the existence of the higher accent in this syllable. In pál tidu, there is no merging of the components in view of the long vowel in pál; and, therefore, neither assimilation nor the generation of the dydam is possible.
[F] SECONDARY -H- OF GôŞDI IN CONNECTION WITH VOICELESS PLOSIVES. (a) Gôndi causative stems, formed with the afix f- show a seoondary R-G immediately before -- in instances like the following:
tiri. (to be turned round) tiriht. or tiruhl- (to cause to turn round). vari- (to fear)
variht., varhut, varist- (to gause to fear, to frighten). kari. (to learn)
kareht. (tc teach); mei- (to grazo)
a mohs(to cause to grase); lind- (to eat)
tiht. (to feed); und- (to drink)
uh - (to give to drink); karé (rig). (to be shaken) karhut-, karuht. (to shake).
. The alternative forms with before -- were explained by me as probably due to the influence of Indo-Aryan instances, like the so-called "reversion of h > sibilant in ni kama, ete. Since >. in Judu. Aryan is a rare change, and since the cases of "ruversion " referred to above may not have involved a real
change at all, a better explanation for the alternative -- of Gondi would be that hare the fricative [] which we have postulated as an intermediate stage (in connection with t) in the production of the aspirate, changed into the sibilant in some cases, side by side with the convorsion of [0] to .h. It is significant that there is no alternativo -- in connection with tho uspicato appearing before the plural ending -k of Gondi words. (See below.)