Indian Antiquary Vol 62

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

Page 163

AUGUST, 1933)
INITIAL FRICATIVES AND AFFRICATES OF DRAVIDIAN
151
(i) The correspondences of t. forms to others with initial sibilants or affricates do not appear to be very extensive or widespread inter-dialectally.
(it) 1. forms are retained extensively in large numbers in all dialects except in Tuļu, where t-alternates with s. or h-in a large number of instances.
(iii) The problem of the relacionship of t- to the initial sibilants and affricates is one beset with many difficulties. Few as are the instances that raise this question, the chronology of the change will have to be determined separately in each instance. This, however, is not now possible owing to lack of materials, and so we have to content ourselves with a few general perspectives.
t-in connection with front vowels in medial positions is known in the dialects to change into the sibilant or affricate (cf. Tamil adittu, adiccu, 'having beaten,' etc.) on account of the influence of the vowel which raises the point of articulation of the tongue from the dental region to the alveolar position. A similar change (i.e., of palatalization) may safely be postulated in at least a few cases for the correspondences of words with t- followed by front vowels on the one hand, and their cognates with initial sibilants or affricates on the other.
(iv) So far as the parallels with immediately following dorsal vowels are concerned, two sub-groups may be distinguished (pointed out as (b) (i) and (ii) above, viz., one, comprised of in ancient group of instances occurring in all dialects; and the second, consisting of a few instances in Küi and Tuļu chiefly, and rarely in Kannada; (b) (iii) is an exclusively Tuļu group.
Is it possible for us to envisage the view that Dravidian intitial t- may here have been secondary to 8. ?
(1) Tamil appears to have adopted and assimilated some Sanskrit words having initial fricatives, by changing these into t-, e.g., Skt. brî Tamil tiru ; send (army) tånai. (11) Tuļu changes initial 8. or C- of some Sanskrit words into t-, e.g.,
Skt. sanci - Tuļu tanji
sangati , tannati. » candana
tannana. Besides, a few cases of secondary t-(taisai 'to die,' teļ- <sett.) occur in native Tuļu words sub-dialectally.
Do these facts in any way warrant the postulate that t- in the instances given here is secondary to the sibilant-fricative ? An answer to this question should take into consideration the following facts :(1) Native t- forms are very widespread in the dialects, and the corresponding forms
with the sibilants or affricates appear largely only in sub-dialectal forms of Tulu
and in connection with a few forms (comparatively speaking) in the other dialects. (i) The few cases of the change of 8. to t. in Tamil adaptations of Sanskrit words
noted above could be explained as being due to different phonetic influences. (i) Tuļu adaptations with initial t- of Sanskrit words with initial s- are probably
due to the influence of the numerous sub-dialectal alternant forms with
initial t- and 8-.
(iv) In none of the native instances with 8-, can we prove the sound to be original ; on the other hand, the corresponding t- forms are so widespread as to suggest t. to be original.