Indian Antiquary Vol 61

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

Page 156

142
THE INDIAN ANTIQUARY
August, 1932
It will be seen that in all cases the vowel following the initial k is either e ori. We shall take a few words and examine all the forms in the cognate languages and try to reconstruct the primitive form on their basis.
The ear: Kan, kivi, Tel. chevi, Tam. sevi, Mal. chevi, Tul. kebi. Prim. form : * kevi. Small : Kan. kinna, Tel. chinna, Tam. sinna, Mal. sinna, Tul. kini. Prim. form : * linna.
To do: Kan. géy, Tel. chey, Tam. sei, Mal. chey, Prim. form : * key. (Initial voicing in Kan.)
Sandal: Kan kera, keravu, kerpu, Tel. cheppu, Tam. seruppu. Prim. form : * kerupu.
I suggest, therefore, that the primitive Dravidian vowels & and i had a palatalizing effect over the preceding consonant k and changed it into ch. If this theory is correct every primitive Dravidian form with an initial syllable ki or ke should be represented by 8 or ch in Tam. and Tel. If there are exceptions they must be accounted for in the following ways. The original vowel following the first consonant may not be e or i. It might be a vowel of guttural quality. A case of softening only in Tel. with the retention of the hard sound in Tam. is due to this: the k is preserved in Tam. on account of the influence of the succeed. ing consonant, which is lingual or cerebral, i.e., !, 2, , 1, or !. This rule applies only to Tam. This will explain Tam, ketta + Tel. chedda, Tam, kili + Tel. chiluka, and forms like Tam. kira (old) + 0. Kan. Kerava, Tam. kér, kel, kéd, + 0. K. kelu.
The word for 'to do' and its semantically related root for hand' present some difficulty. The Dravidian forms for 'to do'are: Kan. gêy, Tel. chéy, Mal. chey, Tam. sei, Kud. key, Gond. ki, Kota. kl. The forms for 'hand' are: Kan. keyyi, key, kayi, kayyi; Tel. cheyyi, kelu; Tam. kei ; Tul. kai. The root shows palatalization in some forms and not in the others.
Palatalization is not an uncommon change in the history of language. It is found in the Indo-European languages. Sanskrit regularly palatalizes the tack-stops when they are followed by the palatal vowel e. Though this vowel was later on changed to a, a comparison with related forms in other languages shows that e is older. Cf. Skr. pañca+Grk. Trévre (<* penkwe), Skr. catváras + Lith. keturi. On the other hand, if a oro follows the consonant the back consonant is preserved, as in Skr. kakşa+Lat. coxa. It is interesting to find a similar development in the Dravidian group.
The rule in Dravidian, therefore, is as follows:
Where a primitive k is followed by the vowels e or i, it is preserved in Kanarese, but is changed into 8 in Tamil and ch in Telugu. This change does not occur in Tamil where the vowel is followed by a lingual consonant, i.e., t, d,,, or !.
It is quite possible that in Tamil k was first palatalized into ch and later simplified into 8, as the latter sound is considered more refined.
I offer this explanation tentatively, but I believe that it covers all the cases in a satisfactory manner.