Re: [tied] The Slavic imperfect

On Tue, 14 Jan 2003 19:55:33 -0000, "Sergejus Tarasovas
<S.Tarasovas@...>" <S.Tarasovas@...> wrote:

>--- In cybalist@yahoogroups.com, "Sergejus Tarasovas
><S.Tarasovas@...>" <S.Tarasovas@...> wrote:
>
>> Xaburgajev et al. argued the imperfect is a late areal (south-
>western
>> Balkans) innovation, though unfortunately I don't remember the
>> details.
>
>
>I still can't provide you with the details, but at least I can give a
>reference.
>
>Georgij Xaburgajev in "Pervyje stoletija slavjanskoj pis'mennoj
>kul'tury" (Izdatel'stvo Moskovskogo universiteta, 1994) writes:
>
>"Tol'ko balkanskim slavjanskim govoram byl svojstven i
>kodificirovannyj Konstantinom [i.e., St. Cyril -- S. T.] imperfect
>(javljavs^ijsja, vidimo, praslavjanskim dialektnym novoobrazovanijem,
>otsutstvovavs^im v govorax za Dunajem i v Vostoc^noj Jevrope [sm.
>Xaburgajev G. A. _Drevnerusskij i drevnepol'skij glagol v sravnenii
>so staroslavjanskim (k rekonstrukcii praslavjanskoj sistemy
>preteritov). // Issledovanija po glagolu v slavjanskix jazykax.
>Istorija slavjanskogo glagola. -- Moscow, 1991, s. 37-49])... suffiks
>imperfekta obrazovalsja v rezul'tate soc^etanija aoristnogo affiksa -
>x-/-s^- (no uz^e ne *s [he probably means the obviously analogical
>nature of *x in that case -- S. T.]) i davnego iterativnogo affiksa -
>a- (sr. _prosi-ti_ --> _pras^-a-ti_), kotoromu fakul'tativno [thus he
>argues that so called "contracted" forms are at least in some cases
>the froms without this "optional" affix -- S. T.] mog preds^estvovat'
>affiks -e^- (iz *-e:-), vyraz^ajus^c^ij ideju sostojanija [tam z^e,
>s. 48-49]".

My theory for now is that the imperfect is based on a verbal form in
-e: or -a: (iterative or stative), suffixed with the past tense
(imperfect) of "to be" *e:s- [which survives in Slavic as be:xU, be:,
be:, be:xomU, be:ste, be:s^eN, i.e. with prefixed b- from the root
*bhuH-]. This is largely parallel to the Latin imperfect in -e:-ba:-
or -a:-ba:- (only using a different past tense of "to be").

As such, the hiatus breaker -j- must be old in these forms, otherwise
we would have /e^/ not /a/ (e:je: > e^ja), and the forms without -j-
must have lost it (e^ja, aja > e^a, aa), contracted forms being more
recent still.

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