Re: Question: 'mperie' < lat. 'imperium'

Hi!

Philip Newton writes:
>...
> my informant informs that that "nkiteb" has a weak "i" or "e" at the
> beginning ("li nkiteb" being presumably something like [li In.'ki.tEp]
> or possibly [li n=.'ki.tEp]). This written form is only used after a
> vowel. Also, it seems that "nkiteb" and "inkiteb" are pronounced
> identically, so that distinction in Maltese is apparently solely
> orthographic.
Ah, I see.  My language will definitely not have a prothetic vowel,
that's decided already -- I don't fear a syllablic consonant, it
happens everwhere.  But those are only details, of course.

I came to the point where I thought of resurrecting that lost vowel
after a consonantal preposition n < IN (so n + mperie > *ni'mperie),
but decided againts it (instead, it will be 'ne'mperie', in analogy
with 'le', the article, which collapses to 'l' often and forms 'nel' <
in + ille, so 'in' has a 'ne' allomorph already).  The old initial
vowel will be ultimately lost (and even quite early in this branch of
Romance), and what I will think about is only what consequence it has
on the language.

>...
> However, she says that "huwa mportanti" is ['u.wam.por.'tan.ti], i.e.
> without syllabic [m=] or a prothetic vowel. She guesses that this is
> due to the [m] vs [n] but isn't sure.
>...
Aha.  Intuitively, that felt natural to me, too (but for all
nasal+consonant clusters), and it's interesting to really find it that
way.

BTW, is that a loan from Sicilian or otherwise Southern Italian?
It's 'mpurtanti' in Sicilian IIRC.

>...
> Just because one form has a syllabic "n" doesn't mean related forms
> necessarily have it, too -- for example, I asked about "ntuza" vs.
> "jintuza" (perfect and imperfect, respectively); they both have three
> syllables. So while the first has [In] (or something like that), the
> second merely has [jIn] as the first syllable, not [jI.n=] or
> something like that.
>...
Ah, yes.

Nice examples from Maltese, thanks a lot for asking your informant! :-)

> So, what does that mean for your language?
>
> Nothing, ...
Hehe. :-)

As for feasibility of either possibility, I know that Suahili has
phonemic initial /mb/- vs. /m=.b/-, so indeed anything would work, I
guess...

> unless you decide to do things the same way. However, they're a
> datapoint for nasals in a syllable of their own (syllabic or with
> prothetic vowel), rather than forming "mpe" or "nta" syllables.
>
> As for "kum.pe.rje" vs. "ku.m=.pe.rje", Maltese has both behaviours,
> so take your pick.
Thanks for the data!  I will consider everything again.  Because I
could not convince myself of one thing or the other, I had defined
dialects. :-) But I am not really satisfied with that.

And does anyone have data on Sicilian?? :-)

**Henrik

Reply