Ancient South Asian word for 'cereal'

[David Stampe]

kalyans at ix.netcom.com (Srinivasan Kalyanaraman) writes
>   Let me cite from Southworth: 
>
>   "For the Munda languages, we are fortunate to have 
>   two recent articles by Arlene and Norman Zide which 
>   deal with reconstructions of cultural vocabulary, 
>   including cereals. They courageously suggest a date 
>   of 3500 BP as the date for proto-Munda, and claim 
>   that the vocabulary of the speakers of proto-Munda 
>   included terms for 'husked (and uncooked) rice'... 
>   *ru-kug 'uncooked husked rice (cf. Korku rum 'to 
>   husk')(Zide, 1973:7)
>   [other unrelated forms dispensed with - DS]

As I pointed out, you cited only the second syllable of the Munda form.
The <kug> syllable does not appear independently with the meaning `rice'.
Despite the Korku form */rum/ there is no evidence that */ruGkug/ (where
G is the velar nasal) is a compound form.  Here is the Munda evidence:

proto-Munda */ruGkug/ `uncooked husked rice': Sora /roGko/, Gorum
/ru~k(u)/, Gutob /ruku:/ (Ramamurti 1938), Remo /ru~ku/ (Fernandez),
/ruGku/ (Bhattacharya), Kharia /romku'b/, Juang /ruGkub/, Mundari
/roGko/.  [The velar to labial changes in Kharia after a labial vowel
are paralleled in other forms, and do not point to Korku /rum/.]

The forms you cited from IA and Dravidian are of the general shape
kaNk, as you noted, not kug.  

>   I am not able to access many Dravidian forms readily; 
>   but if my recollection serves me right, konku (Tamil) 
>   also refers to a 'cereal' [cf. also konku-nATu]. 

If you mean to compare the ku syllable of the Tamil form, isn't the
vowel simply the automatic enunciative vowel?

>   Bengali (contiguous to Oriya) prefers the vowel sound 
>   'o' to replace 'a'. 
>
>   kaNku, konku are permissible phonetic variants in 
>   South asian. 

But none of this gets closer to Munda */ruGkug/.

>   Is it erroneous to link phonetic forms such as -kug 
>   and konku within a semantic cluster of contiguous 
>   linguistic sub-areas?

Obviously, it's erroneous to compare only the second syllable of the
Munda word */ruGkug/ `rice' to the second syllable of the Sanskrit
form /kanku/ `millet', and then compare the remainder of the IA and
Dr. forms you cite to the first syllable of the Sanskrit form.  You
might as well compare English Here-ford `species of cattle' to English
horse and German Pferd, and note that cattle and horses are all farm
animals in the Germanic Kulturkreis.

David Stampe, Linguistics, Hawai`i