Re: [tied] Laryngeal theory as an unnatural

Abdullah Konushevci wrote:

>> In Catalan, we have agermanat "brother (of a brotherhood"), afillat
>> "adoptive son, godson", cunyat "brother-in-law", etc.
>>
>> =======================
>> Miguel Carrasquer Vidal
>> mcv@...
> **********
> As far as I know, the Latin word <cognatus>, is derived from prefix
> cog- + natus (cf. nomen, cognomen < cog-nomen) and here we have
> nothing to do with the suffix -at(a). Albanian, seems, was aware that
> here we have to deal with prefix, because the loan <kunat> 'brother-
> in-law' and <kunatë> 'sister-in-law' didn't undergoes palatalization
> of -gn- > -nj-.
>
> Konushevci

Abdullah, when exactly the Latin /o/ become /u/ in Albanian Loans? I got
one idea seing your assumption about perception as prefixes but first I
need to verify some things. This is why I ask.

To avoid people searching about the rules Lat > Rom, I give here the
rule exposed by Miguel for Latin /o/ & /o:/ and its situations when
become /u/ in Rom.:

1) /o/ > /u/ before /n/, /nC/ ( bun, munte), but not before /m/, /mn/
( omn, somn). Exception : nume > no:men ( perhaps due numeru > num.r)

I would like to add now that even the word "cumnat" shows that /o/ > /u/
before /mn/ thing which was excluded initialy by Miguel.

Alex