Hello Miguel,
Your arguments are "false friends" here:
1. a) ep-Romanians "has problems" to pronounce /di/ IN SOME
CONTEXTS , BUT NOT IN ALL. Here are other contexts too where
lat. /di/ -> ep-rom. /di/:
Lat. adiliare -> rom. adia
Lat. landica -> rom. lindic
Lat. eradicare -> rom. ridica
Lat. radicula -> rom. ridiche
b) also ep-Romanian /di/ is obtain from other Latin Loans (and not
only) so its existance (LIKE IN in your ALBANIAN examples too) cannot
be put in doubt :
Lat. credentia -> rom. credinþã
Lat. manitia -> rom. dimineaþã (rom. de + manitia )
In conclusion :
You wanted to make a FALSE /di/ generalization in order to exclude
the inherited /3/ at ep-Romanians, BUT it is not the case.
( Also a Romanian antroponym, that seems to be very old is: Dinu
not to add Dacian ones : Diurpaneus , Diegis, Dinogetia
: this paranthesis is outside of the argumentation, is only an
indicator that /di/ /dj/ well existed at ep-Romanians or at Dacians)
2. " This is an extremely common development cross-linguistically,
by the way(Greek *dje:us > Zeus /dzeus/, Polish *d'ecko > dziecko,
*dikU > dzik, etc.), so it in no way depends on the prior existance
of substratal /3/ in order to work."
The fact that is an "extremely" common development cross-
linguistically, doesn't EXCLUDE anything. It ONLY ADDS the
possibility to be a "Latin internal development" (as M. Iacomi think
too), but its phonetism don't sound at all as Latin. For this reason
an external influence of Lat. dies -> Rom. dzi is more probable...see
again my message that explain 3 related to Albanian-Romanian "common
words" too.
Next , you said "so it in no way depends" ....but saying this
without any arguments, it's just an affirmation like many 1000
others...
Best regards,
marius alexandru
P.S. : As regarding your Albanian examples :
For Albanian dimër we can take into account an intermediary *3
(for PIE *g^H) (in Dacian, sorry I mean, in Latin times),
I mean that is not sure 100% that we have :
*g^H -> *D
*g^Heimon -> Geg dimën, Tosk dimër
we could also had :
*g^H -> *3 -> D
*g^Heimon -> *3imër -> Geg dimën, Tosk dimër
In dieg we have no problem because he have an original D and not a g^
*dHegWHo: > *diäg- > dieg 'burn'
for sure using PIE *D -> alb. d you obtain a /di/ in Albanian, that
of course exists.
But both observations above has no relevance regarding the
presence of /di/ at ep-Romanians.
In any case it would been better to use some Latin loans in
Albanian, in order to detect the Albanian situation during Roman
Times.
--- In cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:
> On Tue, 03 Feb 2004 01:35:45 +0000, alexandru_mg3
<alexandru_mg3@...>
> wrote:
>
> > 2. Early Proto Romanians had some difficulties when they
learned
> >Latin (simply because they wasn't Latins), regarding how to
> >pronounce Latin /di/ in some contexts. They cannot pronounced
> >Latin /di/ correctly so they are using their OWN sounds: 3 /dz/ or
> >3i /dzi/.
>
> In fact, that proves that the "early Proto Romanians" were *not*
the same a
> the "early Proto Albanians". Albanians had no problem with the
cluster
> /dj/ (e.g. djeg "burn") or the sequence /di/ (e.g.
dimër "winter"). In
> Romanian, on the other hand, we have /dj/ > /3/ > /z/, /di/ > /3i/
> /zi/.
>
> This is an extremely common development cross-linguistically, by
the way
> (Greek *dje:us > Zeus /dzeus/, Polish *d'ecko > dziecko, *dikU >
dzik,
> etc.), so it in no way depends on the prior existance of
substratal /3/ in
> order to work.
>
> =======================
> Miguel Carrasquer Vidal
> mcv@...