08-07-03 01:48, etherman23 wrote:
> I'm trying to get a better handle on the laryngeal theory, but
> something bothers me that I hope someone can explain. Consider the
> word *hdont, meaning tooth (here h is a laryngeal). Why is it that in
> this environment the h > 0 instead /@/?
That's a branch-specific thing. A "prothetic" vowel does appear in
Greek, Armenian and Hittite in this context (Gk. odont-, with the schwa
coloured by assimilation to the root vowel, Arm. atamn, both from
*h1dont-). Similarly, we have Gk. aster-, Arm. astL, but Lat. stella,
Eng. star, Skt. str.bHih. (ins.pl.), all from *h2ster-. It's a matter of
cluster simplification versus anaptyxis as alternative resolution
strategies in dealing with complex onsets containing laryngeals (which
were primarily consonants, after all).
Piotr