>I'm not sure about the development glacia > gheatza, but it should
>have been something like glacia > *glatza > *g^atza > *g^etza >
>gheatza. (I've reordered my rules - haven't checked Miguel's
>version - so I could be wrong, but certainly making gl > g^ earleir
>feels right because of the Italian parallel.) Then all we need is
>for an > în to precede the start of the change a > ea in this word
>
>Richard.
There ain't no "ea" there. The spelling "gheatza" has been
since 1954 in use. Prior to that, the K-L and G-L transfor-
mation phenomena were rendered by attaching the diphtong
[ya], not the diphtong [ea], to "ch" and "gh" (e.g. ghiatza,
lighian, a veghia, a ingenunchia, ochian & al. / Surprinsingly,
chiar has stayed as such, i.e. they didn't make *chear of it).
Just look 'em up in: Sextil Pu$cariu & Teodor Naum, Indreptarul
ortografic si ortoepic..., the editions 1932 up to the reform in
1954.
These diphtongs are very akin - in some subdialectal areas of
the Romanian language, esp. in Western and Northern ones,
some people even tend to perceive them as being one and
the same diphtong. A typical feature: most foreigners who
learn Romanian rarely can pronounce a genuine [ea] diphtong,
the'll tend to utter [ya].
But actually the K-L and G-L initially was (and still *is* in the
NW half of the Romanian-speaking territory as well as at least
in the Northern part of ci-Pruth and trans-Pruth Moldavia)
thoroughly palatal K and G followed by the vowel (in this case
[a]) without any diphtong there. One has opted for the diphtong
*in written* only because, for some historic reasons, the SE
subdialects got the upperhand in the forming of the Romanian
literary standard language. The SE subdialects (including Bucharest)
have lost these palatals forever. Hence, people from those
regions cannot imagine how the word <gheatza> can be
uttered without a diphtong. But they can get on a train to
travel 100-200 km to the North, and just... listen.
Therefore, there is no need for an asterisk before g'atza, because
it's there, "alive", in the vocabulary of roughly one half of the DR
native-speakers *today* (and for many decades to come). Only
for the earlier gl- and gliding g(l)y- variants is the asterisk
justified -- well, at least, AFA the DR dialect is concerned, coz
in Aromanian these forms might be there as living fossils, so
to speak.
George