Re: [tied] Re: Decircumflexion, N-raising, H-raising: Slavic soundr

On Mon, 26 Apr 2004 22:29:24 +0000, Sergejus Tarasovas
<S.Tarasovas@...> wrote:

>--- In cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:
>
>> It's in itself reasonable to think that -aí was changed to
>> -ai~ due to contraction (-ijaí > -jai~), as was the fem.
>> nom. sg. -ijá: > -e:~.
>
>I wonder how one would explain collectives in -ijà (vilkijà vs.
>vìlke:, perkú:nija) then?

Something with *-ih2-, I guess.. The -ija:- (> e:) in the
ja:-stems (e:-stems) is the result of Sievers (-CC-ja: >
-CC-ija: after heavy syllable).

>> After all, contractions are one of
>> the major sources of circumflex accentuation.
>
>But for adjectives you posited that contraction leads to circumflex
>under stress only. Why gerì vs. z^e~me: then? Analogy?

Well, the jo- and ja:-stems behave differently. In the
jo-stems, there's a difference between stressed and
unstressed *-ijos (nom.sg. -is vs. -y~s, and apparently also
nom.pl. [adj.] -ì vs. -iai~), but otherwise there's no
difference between them or with the normal *-jo stems
(nom.sg. -ias). The nom.sg. -is is short, so cannot be
circumflex, in the other forms *-ij- (regardless of accent)
was reduced to /j/, without contraction. The only form I'm
not so sure about is the nom.pl. What we have in the
adjectives is -ì (didì, didelì), apparently from *-ijaí with
deletion rather than contraction of */ij/.

In the -ja: stems the Sievers forms, regardless of accent,
always contract /ija:/ to /e:/ throughout the paradigm. The
accentuation is circumflex, except in the ins.sg. and
acc.pl.

>> The problem
>> is explaining why that didn't happen in the ins.sg. and
>> acc.pl. (of both genders).
>
>And in the historical L.sg. (Old Lith. -è, later replaced with -yjè).

Doesn't -è come from suffixed *-en?

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