Re: [tied] The Magic Mountain

--- In cybalist@yahoogroups.com, Piotr Gasiorowski wrote:

> 26-06-03 09:45, g wrote:
>
>> Thank you; I'm not sure Rum. <mare> will shine in a new
>> light, but <mogilla> and <magure> seem to be precious
>> hints with respect to other words: Rum. <moghila/movila>
>> "heap, hillock," <mãgura> "hill, mountain, peak." Yet the
>> Romanian dictionary recommends the reader to compare
>> the latter word with Alb. <magullë>. Does this one belong
>> to the same Alb. mog-/mag- group? (As for Rum. <movila>
>> or <moghila>, acc. to the same source, < Old Sl. <mogyla>.)
>
> The derivation of <madh> from *meg^h2- (with either an o-grade
> or a secondary weak grade, *m&g^h2-) rules out any connection
> with <magullë> and <mãgura>, since these show a non-Satem velar.
> I think a very early Slavic loan must be assumed here, since
> *mogyla (phonetically *[magu:la:] at the time of the initial
> Slavic expansion) is a common Slavic word meaning 'kurgan, burial
> mound'; the retention of /g/ between vowels in Albanian would be
> hard to explain in an inherited word without a transparent
> morphological base.

That doesn't fit Romanian <mãgura> since there is *no* Slavic
loanword to exhibit rhotacism of VlV. Either the word would be
a strange singular case of *very* early Slavic loanword, or the
word is pre-Slavic. At this point, I would pick by far the second.

> Romanian <moghila ~ movila> is of course a post-rhotacism
> borrowing from a later Slavic source, [...]

That's obviously correct, adding that Romanian linguists haven't
discovered yet any clear pre-rhotacism Slavic loanword. It makes
even less sense to loan the same word twice, in two forms having
not so different meanings and different stress patterns (paroxytone
for "movilã", proparoxytone for "mãgurã").

> As regards its etymology, the word *mogyla is as enigmatic as a
> steppe kurgan itself. An extremely speculative (and therefore
> suspect) Iranian etymology has been proposed (Sarmatian *magu-ula-
> 'magic hill'), from *magHu- as in OP magus^ 'magus', Eng. <may,
> might>, Slavic *mog-ti 'be able to', similar but unrelated to
> *meg^h2-.

Voila. If the word exists in Satem Slavic, with that apparently
non-Satem phonetics, one can think at least at two other different
ways to explain Romanian <mãgura> as substrate and Albanian
<magullë> as stratum through a Satem language Y having the word:
1. derived from a Satem specifical root, or a diferent PIE root
in agreement with Satem phonetics (if possible for Slavic, should
be possible for Y);
2. loaned from some other language having contact with them (this
would still fit since Sarmatians had geographical contact also
with Daco-Moesian speaking population).
I would rather pick 1 or 2 instead of
3. Romanian <mãgura> and Albanian <magullë> are very early loans
from Slavic.
... due to rhotacism and stress pattern.

Regards,
Marius Iacomi