Permissable Indo-european roots

[Harry Spier]

Dear list members,

Dr. Witzel's recent article in Electronic Journal of Vedic Studies 5-1
"Substrate Languages in Old Indo-Aryan" gives the following formula for
permissable indo-european roots:

prefixes +/-{(s)(C)(R)(e)(R)(C/s)}+/-suffixes

with the restrictions:
1) sounds inside a root arranged in order of preference C/s-R-e
2) non-permissable consonent groups, the types: bed, bhet, tebh, pep,
teurk/tekt (Skt. *bad, bhad, tabh, pap, tork, takt)
3) b is very rare in IE;
4)only S (but not) s is allowed in Vedic after i,u,r,k

Examples kIn, kIk, mag, balb, bRs are given as examples of verbal/nominal
roots that do not exist in IE as not conforming to the formula
{(C)(R)e(R)(C)}.

Can someone explain to me why "mag" which appears to be form (ReC) is not a
permissable IE/IIr root?  Dr. Witzel gives among the examples of a
permissable root the form (CReC)zrath, and the form (sReC)svap, but no form
(ReC).

Also Robert Beekes in his book "Comparative Indo-European Linguistics" in
12.1.1 and 11.10.6 gives as not permitted in IE roots:
1) two voiced consonents.
2) unvoiced + (voiced)aspirated in same root.


Is Dr. Witzel's list of non-permitted consonent groups i.e. bed, bhet, tebh,
pep, teurk/tekt a further refinement to the above criteria from Dr.Beekes
book?

Yours,


Harry Spier
371 Brickman Road
Hurleyville, New York
USA 12747

______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com