Re: [tied] The disappearance of *-s -- The saga continues

On Sun, 11 Apr 2004 14:25:21 +0000, elmeras2000
<jer@...> wrote:

>--- In cybalist@yahoogroups.com, enlil@... wrote:
>
>> Well, maybe we're both wrong because with accusative plural
>> *-ns and *kWi-s "who" nothing seems to work, not even your
>> rule.
>
>For what it's worth, I think this is a place where we have to
>distinguish the s's. The sibilant of the nominative was oroginally
>different from the one marking the plural in the acc.pl. ending *-ns
>or *-ms. I think it shows that difference here: It was *-é:nz, *-
>é:rz and *-o:nz, *-o:rz, *-o:yz that developed into PIE *-é:n, *-
>é:r, *-o:n, *-o:r, *-o:y. The other sibilants neither lengthened nor
>disappeared.

I thought you thought the *-s in the acc.pl. _did_ lengthen,
at least it was my impression that you reconstructed o-stem
acc.pl *-o:ms (*-o:ns).

In any case, if *-ms doesn't lengthen a previous /o/, /i/ or
/u/ at the PIE level, another possibility to explain the
non-loss of *-s in the acc.pl. is that the -s was added
_after_ the lengthening rule _and_ after the
loss-of-/s/-after-sonorants rule. I at least think that the
*-s in the acc.pl., dat/abl.pl., ins.pl. and loc.pl. is
secondary, although it's hard to establish exactly _when_ it
was added. Could have been pretty early.

=======================
Miguel Carrasquer Vidal
mcv@...