Indian Antiquary Vol 59

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

Page 195

SEPTEMBER, 1930 ]
TAMIL ARISI (RICE) AND GREEK ORUZON
179
bear to the Greek form. This aspect, as we have mentioned above, has been cursorily adverted to by Prof. Bloch, as the main object of his paper was to establish the derivation of Greek oruzon from an old Iranian form. Prof. Bloch admits that the Dravidian forms are native but observes that neither Greek oruzon nor Sanskrit vrihi could have had any connection with the Dravidian forms. It is my purpose to show in this paper that, viewed from the standpoint of Dravidian, the relationship of the Dravidian forms to the Greek word cannot be dismissed so easily.
The known Dravidian forms are the following:
Tamil: arifi, cri, vari, paddy.' Telugų: vari, paddy.' Kannada: akki, rice.' Tulu: ari, rice,' akli, baru. Malayalam: ari, 'rice'; vari, paddy.' Kai: urgi. (Another Küi word kudi or kali, rice,' paddy,' is different and
probably allied to Tamil kúlu.) Toda: ask, 'rice.' Gondi: vanji, 'paddy,' also rice'; pari, rice.'
Kurukh does not evidently show allied forms: both khess, paddy,' and tikhil, rice,' are different. Kurukh arkha (culled shoots) contains the Dr. base ar, but the meaning is not restricted.
To begin with, we have to consider if we can isolate the forms with initial v- from those with initial vowels. Prof. Bloch seems to suggest that they are different. I venture to suggest that the two sets of forms are olosely connected, and that those with initial vowels should be regarded as primary, inasmuch as they represent the Dravidian radical ar or ar, to 'cut.'
(a) ari in Dravidian, as a verb, means to remove,' and is derived from ar or ar, to "cut off.' or separate.' (Cf. the alternative forms arikkirai and a raikkirai, potherb,' for the interchange of r and r); ar or ar has given us a number of forms like aruval, arakku, ara, eto. The most ancient of the forms for rice (husked and not husked) is ari, found in Tamil meaning
paddy' or any 'handful of grain. This meaning apparently contains the idea of something cut off. The word was subsequently applied to paddy and rice alike.
Malayalam fixed the meaning of rice' for ari, while late old Tamil and Kannada used derivative forms with the suffix k.
It is clear, therefore, that the forms with an initial a are primary, and that Tamil ari, paddy,' represents the most ancient form directly derived from the root.
(6) -ki is a derivative ending very common to Dravidian, and is employed, with its variants -ke, -Je, -ku, -gu, etc., to form derivative nouns from verb roots; though Kannada mostly employs -ke or -ge, and Tam. - kai or .gai, the central Dravidian dialects show -ki (6.g., Kai gippki, etc.)
Kannada akki and Tamil aridi were such derivative nouns, formed with this affix; in Kannada ariki gave alks, just as iriki gave rise to ikki, "house.' Tamil & was the palatalized resultant of the original -k-, as in elçi (<e?).
ari in Tamil has the ollowing meanings : 'paddy,' .ear of paddy,''rioe.' The semantic developments offer no difficulty. The primary meaning, 'paddy,' is retained in ari, vari; the derivative meanings are exclusive for akki, aridi and ask.
Prof. Bloch has noted the close resemblance in form and meaning between the group: Tel. vari, Tulu bari, and Gondi vaflji, on the one hand, and Malay vari, bari and Javanese padi, on the other. While Prof. Bloch refrains from making any comment on the relationship of the Dravidian and Austric forms, Mr. Tuttle in his article suggests that "probably Tamil vari and Tel. vari, having the sense of Malay padi, were derived from Austric."
It will be seen from the following discussion that the Dravidian forms with initial v. could be explained as native Dravidian, derivable as they are from the forms without the initial -
5 See my paper on Dravidian in the Indian Historical Quarterly, September 1928.