Sanskrit and Tamil question

[George Hart]

I find it hard to believe that it has been "demonstrated" that  
Dravidian and Australian languages have "certain words" in common --  
though apparently some inhabitants of South India share a y-chromosome  
with the aborigines (no doubt they were part of a population that  
emigrated from Africa to Australia 50,000 years ago).  After all, it  
is well known that completely unrelated languages have many words that  
look related but are not.  I would be very interested if Prof. Witzel  
could share this paper or make it available if at all possible.   
Certainly any relationship that could be convincingly established  
would be important.  50,000 years is a very long time -- certainly,  
genes might persist, but it's hard to believe that any element of  
language could be preserved all those years.  And of course there is  
the question of whether Dravidian even originated in South India.

Of course, nadii is IE.  The Tamil word for "walk" is naTa, with a  
retroflex T.  George Hart

On Oct 6, 2008, at 7:29 PM, Michael Witzel wrote:

> What a mess, and where to begin? Thus, a few points interspersed  
> below:
>
> On Oct 6, 2008, at 5:24 PM, Loriliai Biernacki wrote:
>> Can any Tamil scholars or Vedic Sanskritists help with the query I  
>> received
>> below?
>
>>> If you know someone who knows, could you find out the etymology of  
>>> the word
>>> nadi. This word is used for rivers in the early chapters of the word
>>> Mahavamsa.
>
> nadii is attested since the earliest Indian text, the Rgveda. It  
> means 'stream,  river', see Turner's CDIAL #6943, Mayrhofer's EAWia  
> II p. 8-9, who derives it from Indo-European *ned  (Sanskrit nad)  
> 'to sound loudly' (as also of mountain streams). No Austric here, as  
> the root is well attested in Indo-European, also as river names.
>
>>> Offhand, the root seems to be nada - or walk denoting motion in  
>>> Tamil and
>>> perhaps other Dravidian languages.
>
> For Dravidian see Burrow-Emeneau's DEDR  # 3590 Ta. natu (-pp-, - 
> tt-) to extinguish, quench, eclipse (as rays), destroy; nantu  
> (nanti-) to be extinguished, put out (as a lamp), etc. etc.  Thus,  
> no "walking"
>
> Both CDIAL and DEDR are online at our SARVA (substrate language)  
> project :
> <http://www.aa.tufs.ac.jp/sarva/entrance.html>
>
>>> In Vedic Sanskrit it is used for the now
>>> lost Saraswati River in the Sind, but is used mostly in connection  
>>> with the
>>> nervous system. There is also a possibility of nadi having an  
>>> Austric origin.
>>> Ganga for river is very likely of Austric origin.
>
> It is used for rivers in general; the Sarasvati is not yet lost in  
> the Rgveda but only in the Brahmana texts a few centuries later.  
> Sindh is unknown to the Rgveda; the southernmost area is Bhalaanas  
> (probably the modern Bolan area/pass in N. Baluchistan, near Quetta).
>
> As for nadii and (unlikely) Austro-Asiatic, see Mayrhofer, EWAiia II  
> 8-9.
>
>>> I have a strong hunch that the pre-history of Ceylon was dominated  
>>> by the
>>> Austrics - people who did practice agriculture and not all hunter- 
>>> gatherers as
>>> is now represented. Magama (Hambantota) and the Pura in  
>>> Anuradhapura are are
>>> certainly Austric and have a long history before Buddhism. The  
>>> Archaelological
>>> Dept. has shown a long reluctance to go beyond the 3rd Century BC.
>
> Archaeology is politicized both  India as well as in Sri Lanka. All  
> publications to be read with caution.
>
> -pura is a clear Sanskrit and Indo-European word: Vedic pur = Greek  
> polis (also  Lithuanian pilis) Originally "(filled in) fortification  
> (wall)." See Mayrhofer, EWAia, s.v. pur.
>
> Austric presence in the south and in Sri Lanka is unknown, so far.  
> The relevant studies simply have not been done for the Dravidian  
> speaking areas of S. India and for Sri Lanka.
>
> We need someone go through the DEDR and check out all words that are  
> not likely as Dravidian by sound and structure,  and compare obvious  
> substrate words in the Nilgiris (cf. Witzel 1999, see my website:  
> Early Sources for South Asian Substrate Languages.Boston: Mother  
> Tongue, extra number 1999 pdf) as well as those transmitted by the  
> Veddas of Sri Lanka (see de Silva 1962). Only then can we be sure  
> who was present in these areas before the speakers of Dravidian (and  
> Sinhala, in Sri Lanka).
>
> Vaclav Blazek (Czech Rep., at the Harvard Round Table, Oct. 2006,  
> see my website) has recently shown that Dravidian languages contain  
> words that are preserved in Australian languages (that incidentally  
> also has Dravidian-like retroflex sounds, just listen to them!), --  
> indicating that speakers of the Australian (not Austric!) languages  
> had been present there, long ago. As genetics also indicate (recent  
> paper: Nilgiri Kurumbas, W. Bengal/Nepalese Rajvamshis and  
> Andamanese share an old NRY marker, 50-40,000 years old).
>
> Cheers,
> MW
>
> (refs. to CDIAL, DEDR, Veddas, Nilgiris, etc. in my 1999 paper in  
> MT, on substrate languages).
>
>
> Michael Witzel
> witzel at fas.harvard.edu
> <www.fas.harvard.edu/~witzel/mwpage.htm>
>
> Dept. of Sanskrit & Indian Studies, Harvard University
> 1 Bow Street,
> Cambridge MA 02138, USA
>
> phone: 1- 617 - 495 3295 (voice & messages), 496 8570, fax 617 - 496  
> 8571;
> my direct line:  617- 496 2990