From: elmeras2000
Message: 32730
Date: 2004-05-18
--- In cybalist@yahoogroups.com, enlil@... wrote:
[Glen, on the locative:]
>It doesn't conform to the other automatic
> declensional patterns and accentuation.
That is correct. That's why I have suggested an enclitic, as opposed
to a desinence. Enclitics move the accent to the position before
them, which explains the constant stem-final accent of the locative
and may also explain the absence of an overt ending if the original
form consisted of a short vowel which would be deleted when not
accented. If it were a desinence and the locative were a normal weak
case, the accent should move onto the vowel of the desinence in
hysterodynamic paradigms (as 'father'), which it does not do. That
would demand a special boundary between the stem and the lost marker
of the locative, a fact that would account for stem form of the
locative *-en/*-en-i of r/nt-stems.
Jens