> I agree here: so at least we have agreed that we have here a 'later'
> extension. Remain to detect when this extension took place...on my
> side I think that for sure was long after a/not-stressed>& and after
> dh>zero in intervocalic words....
>
> So má-dhësht was extended in -i: remain to compute the real
> timeframe (after sec X in my opinion) ...I will check based on the
> Slavic loans in Albanian
>
1. Unfortunately we don't have Slavic a around [700,900] but only
Slavic o, but anyway its phonetic value was much closer to a than to o
because the old Slavic loans in Greek and Albanian are reflected as a.
So we can consider that we are during the transition of a > o in
South-Slavic
Slavic a->o/unstressed > Alb a
Sl. koçub& 'crest, tuft' > Alb. kaçuba 'bush'
Conclusion: we can consider that a/unstressed > ë was for sure ended
around 700-900 CE (I considered that it appears somewhere in sec I-
IIICE)
2. In addition a/unstressed > & seems that remains active during
the Roman Period (until around 600-700 BC):
Arguments:
The Albanian loans having Lat o: > Alb o (so loaned after o: > e)
still reflect the a/unstressed > &
Lat. sanito:sus 'healthy' > Alb. shëndosh <-> Rom. sãnãtos 'id.'
We have also kërrutë < Latin cornu:ta even there we have o>ë (maybe
in the r-context) and not a>ë, is important to note that the passage
to ë took place after u:>y see Alb. u>Latin u: (by the way, I need to
move earlier the u:>y timeframe) so very late...
3. Regarding the lost of intervocalic dh, gh in tri-syllabic words
we have the attestation of Constantin Porphyr. (913-956) of Raoussa <
Illyr. Ragussium
So the intervocal dh, gh were already lost at that moment : I
think that we can well localize the lost of intervocalic gh, dh
somewhere between 600-900 BC (and of cause d>dh took place earlier but
not long before)
Conclusion: the -i: in má-dhësht'i: was added AnyTime After 700 CE
but for sure Not Before 700CE, because the a remain a, so 'at least'
in the same time with the lost of intervocalic dh...
This conclude that the extension of the final -i:, Very Probable
is 'Recent' : and it was attached after sec X as I initially said....
Sorry Piotr, but as you can see is 'very probable' that the final -
i: was added to má-dhësht after the lost of dh in try-sillablic
words....
I will keep my rule : the intervocalic dh was lost only in tri-
syllabic words (in this case there is no need for 3>D3 etc.....we can
keep directly dz > dh that merged next with the intervocalic d>dh)
Best Regards,
Marius