Re: [tied] Re: Albanian-Romanian Concordances

Piotr Gasiorowski wrote:

> ----- Original Message -----
> From: "alex_lycos" <altamix@...>
> To: <cybalist@yahoogroups.com>
> Sent: Monday, April 14, 2003 10:47 PM
> Subject: Re: [tied] Re: Albanian-Romanian Concordances
>
>
>> well Piotr, please be a bit patient with a poor soul:-)
>> 1) Xavier says "and maybe too (peut etre aussi)" albanian 'sorrë'
>> 2) the Romanian form is with "c^". Would that means that the
>> satemisation process was kind of k^>c^>s ?
>
> The development of *k^ in most positions was into Albanian <th>
> (voiceless dental fricative), presumably via an affricate of the *c
> [ts] type. The cluster *k^w, however, developed into Proto-Albanian
> *c^ [tS], which eventually yielded Modern Albanian <s>. The sound
> must have been a *c^-like affricate during the time the early Slavic
> loans entered Albanian, since Slavic *c^ is reflected as <s> in that
> layer of loanwords. The "Albanoid" substrate words in Romanian
> preserve the old pronunciation, just like English preserves Old
> French affricates (rather than Modern French-style fricatives) in
> words like <chance> or <gentle>

OK, this is nice and good explanation, thank you. Here I have a trouble
with the alteration of /k/ and /g/.
It is said that the "k" has the following steps: 1) became palatalized
(ki) and after this became africatized ( c^)
These should be the process of alteration for the velars /g/ and /k/. So
we will have:
k > k'i > c^ ; g > g'i < g^
Is this true or rather the change are k > c^> k'i respectively g > g'i

> g^?

One more question: g > d= satemization or not?