--- In
cybalist@yahoogroups.com, "alex" wrote:
> a propos "ied" versus "brad". Here is the most funny analysis. For
> "brad" it is supposed that the singular is made analogical from the
> plural "brazi" because of the Albanian word which request an "z".
The substratum word "brad" (`fir-tree`) is indeed supposed to have
analogically rebuilt the singular in /d/ from the plural form. The
reason for reconstruction is still not "Albanian word" but the very
weak oposition between a sg. /braz/ and a pl. /braz'/ (unvoiced
final "i" hardly audible) correlated with canonical alternance of
consonants /d/ - /z/ before palatal glide. Analogy is probably
Common Romanian since Aromanian and Megleno-Romanian exhibit the
very same singular form.
Albanian word suggests only what could have developped if there
were no analogical "milieu" - a /z/, from an initial substratal
consonant yet to be determined.
> For "ied" there is not supposed anymore that there is a sg.
> remade too analogical
There is no reason to do that. The word "ied" < Lat. "haedus" is
the perfect model for analogical reconstruction since it exhibits
naturally the canonical alternance.
> because of Alb. word, but this should derive direct from "haedus".
I fail to see what's so funny about that. It's exactly what
analogy is about: two originally different paradigms are reduced
to the one of them which looks more "regular" in the language.
In Romanian the alternance /d/ - /z(i)/ is well in place, there
is nothing to wonder about.
Marius Iacomi