"centum"/"satem" "exceptions" [was Re: Northwest IE attributes]

[Richard M. Alderson III]

On 5 Feb 2000, Peter Gray wrote:

>Miguel wrote:

>> For instance, I don't think Benveniste's root-theory allows *i and *u in the
>> V position (or does it?).  It would be nice to have statistics.

>Forgive my ignorance - I'm puzzled here.  *CiC and *CuC roots are plentiful,
>e.g. *digh goat, *bhidh pot, *k'ik strap, *knid louse; *trus reed, *k'up
>shoulder, *k'udh dung, *lus louse etc etc.

>Could you help me understand what you meant here?

Miguel was correct in his recollection of Benveniste's root theory, which was
based on the notion of _Wurzeldeterminativen_.  All roots are CVC, where V is
defined to be the ablauting *e/o; a further _'elargissement_ or determinative
can yield two series of stems (*not* roots), CVCC (I) and CCVC (II).

All roots are verbal, as are stems with a single determinative; stems with more
than one determinative are inherently nominal.

Everything you refer to as a root in your list would be called a stem by
Benveniste.

								Rich Alderson