Re: [tied] Re: Lith. Acc.pl.

On Mon, 7 Oct 2002, Sergejus Tarasovas wrote:

> --- In cybalist@..., Jens Elmegaard Rasmussen <jer@...> wrote:
>
> > I'm afraid this will lead to misunderstandings. Richard asked about
> the
> > nom.sg., not the acc.pl..
>
> Mea culpa. My apologies to everybody for this unnecessary confusion.
> It was the keyword _Saussure(-Fortunatov)'s law_ (rather not
> applicable to i- and u-stems N. sg.) that acted as a red herring
> here.
>
> >    Acc.pl. -ìs, -ùs with non-acutes do indeed act like they were
> acute
> > themselves, since they attract the accent. This can hardly be
> phonetically
> > regular from IE *-ins, *-uns, but it may well be analogical on the
> endings
> > -ùs (-úos-ius) and -às (-áNs-ias) of o/a:-stems.
>
> Still asking (very prudently this time): if you assume that Acc.pl. -
> ìs, -ùs were historically short, why do we have North High Lithuanian
> Nom. sg. àkè.s (close [e.]) 'eye' (regularly < *-is) but Acc. pl.
> àkìs (regularly < *-í:s)? If you agree they were long and beared
> circumflex rather than acute accent (non-acutes means circumflexes,
> doesn't it?), how would you explain the shortening then?

This actually confirms what I said, although I did not take the time to
check for it (I see now that Z^inkevic^ius' maps 20 and 23 show the
difference, if only in a small area). I suggested that acc.pl. akìs be
derived from *akí:ns with analogical length-and-acute from *vilkó:ns and
*z^iemá:ns. The analogy is the same as in Sanskrit: acc.pl. -i:n, -u:n,
even -r.:n in masculines patterned on -a:n of a-stems (versus -i:s, -r.:s
in feminines patterned on a:-stems). I see that Stang speculated about the
same analogy.

Jens