Re: *wogwh-ni ‘ploughshare’

--- In cybalist@yahoogroups.com, "Abdullah Konushevci"
<akonushevci@...> wrote:

> First of all, there is nothing unutural then zero-grade forms to be
> nasalized, so *wegWh-ni in zero-grade form would look *ungWh-; -ngWh-
> as could be testified from lembus < *lengWho-s have yielded Illyrian -
> mb- as have Lating -ngu- in lingua > Rom limba etc. Such change,
> therefore, is arealy justified.

If the word shows C-N in most languages but N-C in just one I'd
prefer to assume metathesis is the cause in that one language. I
can't be completely sure of that or that they're related from PIE
source(s), it's just the most likely possibility.

> Possibility the /o/ to be
> bilabialized to /u/ preceded by bilabial are evident.
> So, phonetically speaking, Alb umi< umbi 'small plowshare on a wooden
> plow' could be as well an o-stem noun, as could be an u-stem too.
> Your point that derivatives of same root must be derived from same
> form and same suffix, to my view, is not correct and this could be
> proved with many examples.

Not "must", but could be. If every difference between languages
is taken to be a result of different words in the proto-language a
truly comparative approach couldn't be developed. That is, a word
like wheel vs. kuklos would be described as coming from variant or
optional forms like *kWekWlos and *kWukWlos with some specifically
unique or rare type of reduplication with u instead of i or e. Of
course, use of the comparative method would indicate that Greek is
one of a group of languages that rounded e > o in certain round or
labial environments, and even more evidently o > u in other places
later.

> Lets take just one example: Alb gjumë 'sleep' and Greek hypnos are
> derived from zero-grade form *sup-no- of the root *swep- 'to sleep',
> till sopor 'a deep sleep' is derived from *swep-os- with regular
> rhotacism and somnum from *swep-no with assimilation -pn- > -mn-.
> I think also that basic form *swep-o have yielded Alb djep 'cradle',
> for *sw- have yielded Alb d-, if followed by stressed vowel,
> otherwise v-.

But *swepn.os has e surrounded by both KW and P in the same syl-
lable. Therefore hupnos and somnus are likely to be from the same
PIE word.