Re: [tied] Germanic nominal declensions

On Wed, 16 Jul 2003 15:35:21 -0400, "Brian M. Scott" <BMScott@...>
wrote:

>> Dat. *-o:i *-ai
>> Loc. *-oi *-ai
>
>> North and West Germanic -e could come from either Dat. or
>> Loc. Gothic -a cannot come from either (we'd expect -ai),
>> so it's a instrumental.
>
>Kienle takes North and West Gmc. -e to be from the Dat. and
>offers three possible sources of Goth. -a, namely:
>
>(1) On the basis of Goth. <hWamma> 'wem?' and <hWammeh>
> 'jedem' he suggests PGmc. * -é: from a PIE Ins. *-é: --
> why?
>
>(2) PIE Ins. *-ó:

*-é: and *-ó: both would give Gothic -a, so both are possible. The normal
o-stem ending was *-oh1 (this is based more on faith in the regularity of
the paradigm and the soundlaws underlying it than on actual evidence, which
consists of Lithuanian -ù), but since Gothic had replaced *-oso with *-eso,
I suppose it maight als have replaced *-oh1 with pronominal *-eh1.

>(3) He sees no problem with a derivation from PIE Loc. *-oi.
> It's not clear why he considers this a possibility but
> not derivation from PGmc. *-ai < PIE *-o:i. Any idea
> what he has in mind?

Not really. The Loc. in *-oi~ [with circumflex accent] would have given
-ai in Gothic, just like the Dat. in *-õ:i. A Loc. in *-oí [acute accent]
_would_ have given Gothic -a (bairada < *bheretoi), but Slavic Loc.sg. -ê
(not -i) proves that the PIE form was *-oi~ (c.q. bisylllabic *-oï).

>> pl.
>
>> The nominative in *-oi is unattested in Germanic. PIE
>> *-o:s would have given PGmc. *-o:z, which explains Goth
>> -o:s and ON -ar, but not OE -as, OS -os. The
>> reconstruction is thus:
>
>> PIE *-ó:ses *-o:siz
>> *'-o:ses *-o:ziz
>
>> which explains all the forms (Goth. -o:ss > -o:s, ON
>> *-o:ziz > -arr > -ar, OE/OS *-o:siz > *-as). OHG -a is
>> the acc. form.
>
>Kienle doesn't mention OE -as but gets all the rest from
>PGmc. *-o:z < PIE *-o:s < *-o-es (or as he would have it,
>PGmc. *-õz < PIE *-õs < *-o-es). He takes the OHG to be -a:
>(which he says is certain in Notker, though he allows that
>there may also have been -a from the Acc.); this he derives
>from PGmc. *-õz. He derives OS -os from PGmc. *-õs and
>says that Goth. -os could be from either PGmc. *-õz or *-õs.
>OS -as is then supposed to be the result of vowel weakening.

I agree that *-õz could have given OHG -a:, but I don't think PGmc. [before
it dropped short final vowels] allowed nominal forms to end in *-s (always
*-z) [in verbal forms we _do_ have 2sg. *-s, which I would explain as I did
the alternation *-z ~ *-s [etc.] in pre-PIE, by invoking SOV word order and
distinguishing word-final position from sentence-final position].

>Kienle would derive the lot from PGmc. *-amiz. He first
>says that Goth. -am and OHG -um point to PGmc. *-o-m..., for
>which runic ON <borumR> 'den Bauern' establishes a final
>*-z. For the intervening vowel he points to OE Dat.pl.
><ðæ:m> with i-umlaut from *þaimiz. This PGmc. *-amiz he
>takes to be from a PIE *-o-mis, mentioning Lith. <rankomis>
>'mit den Händen'. The vowel in ON -om from runic -umR may
>be misleading: at least one source says that /u/ (written
><o> or <u>) is the only vowel that occurs before retained
>/m/ in weakly stressed syllables (apart from some analogical
>optative forms). And now that I dig a bit further, Kienle
>appears to make a similar claim for OS and OHG: PIE *o
>appears as /u/ before /m/. As additional evidence he
>adduces the 1.pl.ind.prs. of the strong verbs, e.g., Goth.
><nimam>, OHG <nemume:s>, from PIE *-omes.

I don't believe that. For starters, the OHG thematic present has -ame:s
(berame:s "we carry") from *-ome[+ something], and a form -ume:s is clearly
secondary after 1pl. -um (Goth. ne:mum), from PIE perfect *..C-me >
*...C-m. > ...C-um. In my opinion, a development of PIE *o to Gmc. *o ~ *u
(instead of *a) before retained /m/ in unstressed syllables (and similarly,
the development of *o: to /u:/ before /m/ mentioned by Piotr[*]), are
problematical. We don't see it in Gothic Dpl. -am (*-omiz), nor in Goth.
1pl. -am, OHG -ame:s. In the o-stem Dpl., ON -om and OE -um are irrelevant
because ON and OE have -om and -um in _all_ declensions. OS and OHG -um
require an explanation, but I'm not sure *-om- > *-um- is it.


[*] Hogg (1992) discusses the NWGmc. raising of *[o(:)] (including
instances of *[o] considered to have been allophones of */a/) when
followed by *u in the following syllable, or directly by *m. In the
acc.sg./pl. of weak feminines we'd expect *-o:n-u- > *-u:n-u- > *-u:n,
spreading analogically in some dialects, while e.g. OE acc.sg./pl.
tungan (rather than *tungon) is the result of levelling out in the other
direction.

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