--- In
cybalist@yahoogroups.com, "Abdullah Konushevci"
<akonushevci@...> wrote:
>
> --- In cybalist@yahoogroups.com, Sean Whalen <stlatos@> wrote:
> >
> >
> > --- Abdullah Konushevci <akonushevci@> wrote:
> >
> > > *wogwh-ni `ploughshare'. 1. Alb umb `small plowshare
> > > on a wooden
> > > plow', Tosk and Standard Albanian form, umi as Gheg
> > > variant, from
> > > nasalized zero-grade form *ungwh-u, cf. for same
> >
> > > Lat vo:mis:
> > > `ploughshare': OHG waganso `ploughshare', OPrus
> > > wagnis `coulter', Grk
> > > ophnis. (Pokorny wowing-s 1179.)
> >
> > If these words are related, I think *ungwhus would
> > be
> > unnecessary.
>
> First, why do you think or why do you doubt that they can't be
> related.
They may be related, they may not. I don't see enough evidence to
be sure either way. If umb has no PIE etymology there's nothing more
I can say about it. If umb has a PIE etymology then it could either
be from *wogWhnis or from some other related word without outside
connections but posited only on Alb evidence(such as *ungWhus). I'd
prefer to relate it directly to PIE *wogWhnis but this would require
several new rules without certain evidence. Reconstructing a proto-
form based only on Alb evidence(such as *ungWhus) which has a long
unknown history that could include many not-yet-discovered changes
seems to defeat the point of comparative linguistics: that several
words in several languages may go back to the same word in the same
language further back in time. Though PIE obviously had many words
from the same root there are several additional points specifically
against *ungWhus: why a u-stem comparable in form to an adjective
applied to people for an inanimate noun?; why initial un- not wn-?;
why would ngWh>mbh (or sim) in Albanian?
> > wegWhnis
> > wogWhnis
> > wugWhnis
> > wubhnis
> > wumbhis
> > wumbhus
> > umbhus
> > etc.
>
> But, if you take into consideration *supnos > Alb gjumë it will be no
> hard to assume that as well *-gWhn-/-ngWh- will end as -pn-/-np- >
> -mb-/-bm- > -mm- > -m-. I guess that Tosk and Standard form have
> homorganic /b/.
> To sum up, Alb form umi could be derived from *ugWh-ni- as well as
> from *ungWh-ni.
I needed metathesis in my derivation to reach a stage where final
i was preceded by a KW that could change i>u. Otherwise, as far as
I know, i would change the first u>y.
I had *wegWhnis > *wogWhnis because other PIE words show most ni-
stems without any possibility of o-grade in the first syllable. If
e>o between KW in Alb (as Greek kuklos) or similar rules for most
IE languages, that simplifies the nature of PIE ablaut (otherwise
there's anomalous o-grade forms that would just happen to be by P/KW
in PIE).