Afro-Dravidian connection

[Michael Witzel]

At 14:08 +0200 9/4/99, Koenraad Elst wrote:

>3) The Afro-Dravidian connection: ...

This is a particular nice piece of "evidence". Sergent has reported what
the French, Bengali and Senegalese at Dakar  profess to have found out:

Its basis is the supposed relationship of Dravidian with West African
languages such as Wolof and Ful  as has been proposed and defended since
the Fifties,  cf. the resume in Sergent (1997: 52 sqq.) Quite differently,
Ruhlen (1987:  260) simply states:  "The Nilo-Saharan connection appears
the most promising."

These are two quite different proposals to begin with, as Ful and Wolof
belong to the West-Atlantic branch of the large Niger-Kordofian (or
'Niger-Congo') family which includes all West African, Central African and
the Bantus languages;
Wolof etc.  do *not* belong to the Nilo-Saharan family that stretches from
Niger, via S. Libya, Chad, Sudan, and Kenya to  Tanzania.

The combination of Wolof and Dravidian uses a set of  similarities between
Peul/Ful and Wolof, and Dravidian, and with various other African
languages, ranging from  Somali (Cushitic branch of Afro-Asiatic) to Mande
and Bantu, both belonging to two different major branches of Niger-Congo
(=Niger-Cordofian).
This amounts to comparing Dravidian to a slew of disparate data ranging
from modern Albanian (IE) to modern Chechen (N.E. Caucasian).

The old, pseudo-Voltaire adage that in etymologies  "consonants count
little, vowels nothing", is exemplified by  these data; note, among  the
Wolof/Drav. consonant correspondences:
-kk- = t;   -tt-   = t;    -t = l, etc.

and among the Wolof/Drav. vowels (first syll. only):

aa =  i,  ee,aa;
a=  e, o, a, u
i =  ee;
u= a, ee,  etc.

in addition, whole syllables disappear for no good reason (jaa : ceerai,
fab : pebga/perukku, xol : karal),  (but note the reverse: juurom : cem?!).
It should be obvious that lautgesetze are not involved but accidental
similarities.
Based on such flimsy 'sound laws', naively accepted by Sergent, the
proposed correspondences in grammar must be taken still less seriously. How
do we know that a Wolof k in -ku- represents Drav. k in -ku- and not a -t-
(cf. fukk, kit, x)?

No method...


At 14:08 +0200 9/4/99, Koenraad Elst wrote:

>3) The Afro-Dravidian connection: ... I have nothing against much more
>recent >connections between  Africa and a section of the Indian
>population, notably the >Dravidians.

 ==========================================================================
Michael Witzel                          Elect. Journ. of Vedic Studies
Harvard University                  www1.shore.net/~india/ejvs
---------------------------------------------------------------------------
my direct line (also for messages) :  617- 496 2990
home page:     www.fas.harvard.edu/~witzel/mwpage.htm