On Tue, 08 Mar 2005 14:40:01 +0000, elmeras2000
<
jer@...> wrote:
>--- In cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:
>>
>> Isn't it obvious? haita haitis haitiþ vs. hait, haist,
>> hait. OS, OHG io (not iu) confirms this.
>
>No, not to me. Is Gothic taken to have *added* reduplication?
Did I say that?
>Is the
>reflex of a putative pre-Gmc. *-ei- taken to be something which
>never appears as the reflex of *-ei- in present stems?
Yes.
The equivalent Umlaut in the *eu-present is still seen in
OHG (everywhere else the eo ~ iu alternation was levelled):
inf. biotan
1 biutu
2 biutis
3 biutit
1 biotume:s
2 biotet
3 biotant
(we have *eu > iu before u, i, j, *eu > *eo > io before a,
in the same way that we have *ei > ii before i/j/u, and *ei
> *ee before a; the rule of course follows automatically
from the Umlaut of PIE *e > e/i, *i > e/u and *u > o/u under
the same circumstances).
For *ei-verbs, we would expect an original distribution with
Gmc. *e:2 in the inf., 1 and 3pl., but *i: elsewhere. This
was levelled to *i: everywhere (sti:gan).
In the preterite, we would expect *ei > *e:2 in the singular
(*-h2a, *-tha), and we have levelling from there (he2:t).
=======================
Miguel Carrasquer Vidal
mcv@...