On Thu, 1 Feb 2001 15:50:45 -0600, Eric Christopherson <[log in to unmask]>

>> >> (no less than) two additional long vowels were still
>> >> distinct from the original long ones: [O:] < awa and [E:] < aja.
>> >
>> >Are those phonemic in modern Arabic? I thought that *awa and *aya both
>> ended
>> >up simply as /a:/.
>> Correct - for open syllables; in closed ones they get shortened and
>> narrowed to [i], [u] (cf. the paradigm of 'empty' verbs like k(w)n 'to
>Ok, but the gist of my question was whether [O:] and [E:] from *awa and
>are still separate in Arabic. I was under the impression that although
>are allophones such as those from /a:/, they're conditioned by factors
>than morphology.


>> >Lateral release? Where did that come from, a local dialect of Arabic?
>> >(Proto-Semitic supposedly had tl' and hl which correspond to Arabic
>> emphatic
>> >d and Hebrew sin.)
>> Spanish words like alcalde < al-kad.ij- seem to evidence the lateral
>> quality of [d.] in the Andalusian dialect (which appears very archaic
>> in some other respects, too - cf. the treatment of the article before
>> 'solar' consonants in Gibr-al-tar, Al-taire, etc.).
>So these were stops with lateral release, not lateral affricates?

Oops... I'm afraid I used the two terms as synonyms (or rather,
understood the latter as a subset of the former). Am I wrong here?

I think, however, that nobody can determine such details for a
reconstructed system. It seems that extant Semitic langs preserve only
lateral fricatives.