Re: [tied] Re: Mi- and hi-conjugation in Germanic

On Wed, 16 Mar 2005 16:52:12 +0000, elmeras2000
<jer@...> wrote:

>--- In cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:
>
>
>I can surely follow you. The two words vydra and vêdro would be
>parallel if derived from *udráH2 and *wedróm via Winter's law and an
>accompanying event of accent retraction. Still, the short *-e- of
>the latter form is not motivated by anything other than the desire
>to make these two particular words parallel. And if they are, what
>are we to do with Latv. dubra and dzidrs? And of course also OCS
>dUbrI which does not have a jer between /b/ and /r/.

I don't get the impression that the word was terribly
frequent (it survives, barely, and as a pl.t., in R. and
Ukr., and is also obsolete in Cz. and Pol., unknown
elsewhere). A low-frequency word may not have been attested
with all its inner yers by accident. It depends on which
OCS manuscripts exactly have it, which I don't know and
cannot easily find out. I find the parallel with Lith.
dubury~s "'Einsenkung, Vertiefung, Grube, mit Wasser
gefulltes Loch', auch Flußn." striking, but it would require
a bit of O.C.S. philology to see if it can be true.

>You could be -
>accidentally - right if they can be given radical accent, which is
>not a problem with substantivized forms as *dhúb-ri- and *dhúb-raH2,
>but more of a problem with dzidrs 'clear' which generally has a
>short /i/, although a variant dzi~drs also exists. Also, idra could
>be *H2íd-raH2, but the adjective idrs is less appealing. It also
>leaves dubùs and pigùs to be explained only by some blocking
>influence from /u/ which would be really strange, but apparently
>without contradiction.
>
>I think one must say in all fairness that you have shown that we do
>not really know whether Winter's law was blocked by media + /r/.

We can agree that media + /n/ had a strong blocking effect.
/n/, /l/ and /r/ are all sonorants, which means that in some
cases they can be grouped together as to their effects on
the phonetic environment. But in other respects,
syllabification for instance, /n/ behaves more like a stop.
In principle, it's possible that /n/ had a strong blocking
effect on Winter's law, and /l/ and /r/ didn't. Or less so.

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