> >Besides, et is a very specific case. Any examples of regular words?
>
> Lots of examples. BEN ("son") has the same alternation. So does SHEN
> (sometimes).
Again, monosyllabic words are special case. Their tzere should reduce to
schwa in constructus, but that would leave the word without a vowel at all;
thus, segol is expanded schwa, not contracted tzere.
> >Again, this is a specific case. Why, by the way, there couldn't be
> >epenthesis at the word end? cotevt - cotevet.
>
> Epenthesis only occurs between two consonants. (Your example of
> cotevt/cotevet is not at the end of the word, but immediately
> preceding the end.)
uh? cotevt - cotevEt
> >The important thing about seghol is that it occurs either in specific
> >epenthetic positions, or with specific consonants. No other vowel is
relates
>
> I don't thinnk so, but segol is certainly is not limited to
> epenthesis, as we now seem to agree.
Strictly speaking, segol appears in epenthesis and as allophonic variation
of short vowel.
Allophonic variation obviously uses and existing vowel, but doesn;t invent
one. Segol appeared in epenthesis, and was incidentally employed in
allophonic variations.