Re: [tied] Some questions

From: Andrew Jarrette
Message: 40106
Date: 2005-09-18

Thanks, Miguel.  Pardon my ignorance, but what is the "allative"?  Is it a to-case (Latin ad)?  I infer that it corresponds to the instrumental in other IE languages, is this correct?
I really don't understand the need for a separate ablative case if it was the same as the genitive in the singular and the same as the dative in the plural!  At least Avestan had separate ablative singular forms for more than just the o-stems, according to my source.  I found these Avestan endings in a book part of which I photocopied (the part containing all the declensional and conjugational paradigms of the various branches of IE), unfortunately not the part with the author!  I believe, however, that it may have been an English translation of Brugmann.
Beekes hinted that there might have been an ablative plural ending *-yos.  I think he must have derived this from the IE collective ending *-yos (Skt. gavyah, s^unyah, dantyah, divyah, etc.).  I tend to hold to the belief that if there was at least one separate ablative singular ending, there must probably have been regular ablative singular and plural endings for all declensions.  The o-stem ablative singular would then be the only relic of this earlier state of affairs.  Of course, I know there is really no evidence at all of a separate ablative plural, but I still would like to believe there was, especially from a semantic point of view (i.e. to avoid misunderstanding between dative and ablative meaning in the plural).
Andrew
Miguel Carrasquer <mcv@...> wrote:
On Sat, 17 Sep 2005 18:59:00 -0400 (EDT), Andrew Jarrette
<anjarrette@...> wrote:

>1.  What is the origin of -au in Sanskrit dvau and the classical Sanskrit dual nom. and acc. -au?  Is there any sort of "dualizing" or "binary/bipolar" particle *u or the like?  Such as in *we(:) (*u + *e(:)?) "or" or Gothic thau, jau ( tha-u, ja-u?)?

This was discussed here recently.  There is a theory by
Jasanoff which states that final -óHe > -ó:u (but '-oHe >
'-o:).  The rule, as desribed by Melchert in Anatolian
Historical Phonology (I haven't seen Jasanoff's paper),
would explain:

them. 1sg. '-o-h2e > -o:   
them. du.  '-o-h1e > -o:   
them. all. '-o-h2e > -o:    (Lith. -ù, Hitt. -a)

NA dual    dwó-h1e > dwo:u
them. all  -ó-h2e  > -ó:u   (Lat. illu:c, hu:c)
pf. 3sg.   -..óH-e > -ó:u   (Skt. dadha:u)
pf. 1sg.   -..óH-h2e > -ó:u (Skt. dadha:u)

Against this, one can argue that the 1sg. thematic ending
cannot have been -oh2e (the thematic vowel is -e- > -a-
before -h2-), and that -oh2e would have given Lithuanian
circumflex -uõ, not -uó: > -ù (cf. a:-stem gen.sg. -ah2os >
-õs).  Neither is the thematic dual ending -/h1e/ : it's -e
or -ye or -ih1, and the quality of the thematic vowel (*o
not *e) excludes the possibility of initial *-h1-, while
Lithuanian acute -ù also speaks against a contraction from
*-o-h1e.  The Lithuanian instrumental in -ù cannot be from
*-o-h2e for the same reasons as given above for the 1sg.,
and the instrumental ending is in fact -eh1 (thematic -o-eh1
> -oh1 > Lith. -ù). (Where the Hittite allative comes from
is a different matter, but I doubt it contains a laryngeal).
Latin -u: in hu:c and illu:c admits a number of different
explanations (e.g. *-oi).

I would explain the (optional) -u in Sanskrit dual -a:(u) as
a leftover from the labialization of *h3 (-xW), and I would
reconstruct the ending as *-o-h3: that immediately produces
the right quality of the thematic vowel (whether *h3 was
voiced or not), and the right intonation in Lithuanian
(acute, not circumflex).  The theory is confirmed by the
Kartvelian borrowing *ok^toh3 => os^txw, which retains the
PIE /xW/.


>2.  Why did the IE dual have the same form for nominative and accusative? (I realize this is probably unanswerable.)

The two forms probably merged.  I think that in the thematic
dual, the dfference is still there: animate *-o-h3
representing the former nominative, and inanimate *-o-ih1
the former accusative/oblique.

>3.  Is there any evidence at all for an ablative plural/dual ending in Proto-IE?  Was it always identical to the dative plural/dual?  Does this state of affairs not lead to confusion?

The ablative plural/dual is always the same as the dative.
In the singular, the ablative is always the same as the
genitive, except in the o-stems.

>4.  Are the Avestan ablative singular endings -oith/-aedha (i-stems), -aoth/-wath (u-stems), -aya:th (a:-stems) an innovation or inherited? th stands for thorn or theta.

Beekes' Gathic grammar doesn't mention these forms, which
might mean that tehy are later innovations, after the o-stem
abl. sg.

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