Indian Antiquary Vol 60

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

Page 25

JANUARY, 1931)
DRAVIDIC MISCELLANY
13
The existence of these forms (purely Dravidian) with exactly similar formative or derivative endings would be enough to indicate -nal of anal also as being Dravidian. The origin of this -n- will be evident when we analyse the structure 188 of the analogical forms given above; -n- in all these cases appears as a hiatus-filler between two vowels, as in Tamil padi-n-áru (sixteen), nadanda-n-on (he walked), odi-n-a, the past relative participle of ôdu (to run), etc.
Anal (fire) may thorofore be considered to be a native Dravidian form. It appears to he neither a tatsama (as suggested in the Madras Tamil Lexicon) nor necessarily a reduoed form of lanal, with the loss of initial k. .
As for the alal-group mentioned at the beginning of this article, as being semantically related to the anal-group, the structural kinship also would be clear when we consider that I, the pecnliar retroflex continuative of Tamil expresses a definite set of associated connotations 19 like 'to be reduced to a low state,' to be mixed up,' etc. This sound appears to be of a secondary character in the Tamil-Kannada group, employed as it is in a number of forms where the above said connotations are conspicuous. A comparison of these Tamil-Kannada forms with !, with their cognates (with other sounds) in Tamil-Kannada itself and in other dialects would suggest (vide IHQ., June 1929) that the introduction of this sound I has conferred socondarily the particular connotations referred to above. In alal, the idea of reduction to a low state' is more or less apparent in tho different special meanings which it has, in addition to the basic significance which it sbares with anal.
We may now advert to the relationship, if any, of karial to anal. In view of what we have discussed above regarding the ancient character of anal, alal, we cannot straightway explain the question by postulating that anal is secondarily derived from kanal, with the disappearance of the initial kn. The question, we admit, is not susceptible of any other easy solution, in view especially of the ancient character of Dravidian kanal and its cognates in Dravidian. The following are some of the forms signifying basically 'fire,' 'beat,' etc. :Tam. kây (to be hot); ka- (to be ardently watchful); keittu (fire); kari (to be
burnt out); kanji, kdyol, etc. Tel. kicc- (fire); ciccu (fire); kägu (to be hot), etc. Kann. kitt (fire); kây (to be hot). Tuļu cicc- (fire); kây-uni (to be hot); katt-uni (to burn). Gôndi kahle (to foel heat, perspire); karv (to burn); kás (to become hot); kis (fire). Kui kúga (to warm oneself); kamba (to be burned); kanda, kara (to be hot); kapad
(to be dried up). Kurukh cicc (fire); kañji (boiled rice-water); kar. (to be heated); khaidna (to make
dry); khatná (to cook). Bråhûi khalhar (fire). Now, an examination of these forms would show here again that we have to deal with a base with initial - and a vowel whose character may have been somewhere near the value of . It is significant that, so far as we could see, the variations in the character of the vowel in the different forms are almost as large in these k- forms as in the forms derived from the vooalio bases mentioned above.
The basic form here does resemble the vocalic base mentioned above, except for the initial k-. We cannot, however, find out with our present materials the exact character of the relationship which, if it existed at all, should have arisen at a prehistorio stage.
188 pun-al (stream of flowing water) contains the base pu (to flow), -al, the suffix, and the hiatus. filler -.
www is constituted of ve- the base meaning 'to be hot,'sto., the histas-filler -- and the sud d
19 Vide my article in IHQ, June 1929, pages 336 e seg.: of. N.o A. Carnoy's observations ou page 28 of his rocent work La Science du Mot.