IE & Semitic roots

[Dominique.Thillaud]

At 10:42 +0100 10/11/97, Luis Gonzalez-Reimann wrote:

>I'm not sure I understand what you mean by IE roots not having an intrinsic
>vowel.  When you state an IE root, it has a vowel.  Say, for instance, *paa
>(Skt. paal); *men (Skt. man); *deik (Skt. dRz); *do (Skt. daa); *dus (Skt.
>duS).  There are also IE roots that consist solely of vowels; such as *ai,
>*ei, *o.  These can then have different grades.  (I am using Calvert
>Watkins' simplified notation).

That's sure that Eurindian had vowels, but the problem of their status and
origin in the roots is not simple:

        1) we can't consider 'i' and 'u' as pure vowels, because they are
just vocalic forms of semi-vowels 'y' and 'w' and able to alternate with
consonantic or diphthongued forms (the same is true for 'm', 'n', 'r' and
'l').
        2) some of Greek vowels 'e', 'a', 'o' (short or long) can be
considered as evolutions of laryngeals 'H1', 'H2', 'H3' who all can give
Sanskrit 'i' or 'aa'.
        3) the only 'true' vowel who can remaind in an Eurindian root is
the alternance '' / 'e' / 'o' (Sanskrit '' / 'a' / 'a'). Because the degree
'empty' can occur, we can't considere it as an 'intrinsic' vowel.
        4) it's true that some 'a' seems to be intrinsic in certain
Eurindian words, but this words are frequently expressive or popular ones
and we know too few about this level of language.
        5) in your examples,
*men- is (3) (a-mn-esy, men-tal, Greek pf. me-mon-a),
*paa (to protect) is (2) (Greek pOma "lid", perhaps pitR ?),
*paa (to drink) make difficulties,
*do is (2, H3) (-di-ta),
*dus is (1) (Greek deuomai "to lack" < *deusomai)
*deik is (1) (adikSi, Greek deiknumi "to point to") and without rapport with:
*derk is (1) (dRz-, Greek derkomai, pf. dedorka "to see with intensity",
degree empty in Greek drakOn "snake, dragon" (here, Greek 'ra' = Sanskrit
'R')).

It's therefore possible to say that, in a certain sense, Eurindian *roots*
are without vowels. Some problems are left unresolved, but this
presentation is commonly accepted as a good working base.
        Hoping to help,
Dominique

Dominique THILLAUD
Universite' de Nice Sophia-Antipolis, France