--- In
cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...> wrote:
> alexandru_mg3 wrote:
>
> >
> > Also je is well attested in Decebal's Times see Diegis, Tsierna
etc...
> > so je is old enough
>
> The question is, what is it old enough for? As far as I can see, even
> -rr- < *-rn- counts as a single consonant from the point of view of
the
> je ~ ja alternation, and the _only_ obstruent cluster before
which /ja/
> is clearly evidenced is *st.
> Piotr
Alb. rr < rn is well present in the Latin loans (see Alb. kërrutë) ->
so this transformation is for sure more recent then je/ja (in fact
ie,je,ja could have different waves based on the alternance of ie and
je in Alb. words)
My opinion is also, that there are some je-s cases for which we can
hardly imagine an open syllable (I don't have these examples now, but I
I have already checked this last night)
I think that this "open/closed syllable" rule for je/ja should be
reviewed...
Best Regards,
Marius
Note linked to rn > rr timeframe
=================================
Due to rn > rr timeframe : we cannot have sorrë < *c^(w)a:rna: <
*kWe:rsna:.
Is not possible to have an n inside, because the Romanian form is
without n : c^wa:ra and c^ > s is very old.
So because rn remained rn in Romanian and rn timeframe is after c^<->s
timeframe, in other words n should be also present in Romanian, we can
conclude that there is no n in the PIE root of Albanian sorrë.
In this situation the single alternative is that the PIE Root was *kWer
(s)n.a: => with n. > a
(also I'm not sure about the s inside, but being between r and n.
should be lost earlier)
Best Regards,
Marius