Re: Not "catching the wind " , or, what ARE we discussing?

From: dgkilday57
Message: 57440
Date: 2008-04-16

--- In cybalist@yahoogroups.com, "fournet.arnaud"
<fournet.arnaud@...> wrote:
>
> ----- Original Message -----
> From: "dgkilday57" <dgkilday57@...>
>
> [...]
>
> >Arnaud cited the Arabic word as <qaTu>, and derived it from PAA.
I
> >have only seen <qiTT->, indef. nom. <qiTTun>, def. nom. <al-qiTTu>
> >quoted. I find no reason whatsoever to refer an Arabic word
lacking
> >demonstrable ancient cognates in other Semitic languages to PAA.
> >DGK
> ========
> I misquoted the word. I took shadda for u. (written too small)
> qit.t.u is right.
>
> Now you have PAA words like *giraw- "lion, wild cat"
> well attested in southern PAA (omotic, chadic)
> Cf. Starostin.
> They are structurally the same as qit.tu
> Velar + Dental with i as main vowel.
> We may hypothesize that *k?itaw > *giraw

We may hypothesize until the cows come home, but you need more
examples of the operation of these particular soundlaws in order to
convince anyone. And words which are "structurally the same" need
not be related. I think you fail to grasp the extraordinary
difficulties of doing long-range reconstruction. You cannot simply
decree that two words which are "structurally similar" and roughly
equivalent semantically must be related. Your attention to details
of sound-correspondences must be at least as good as what IEists
practice in traditional PIE reconstructions.

> Hebrew also has a word with that structure : xa:tu:l
> The long u: suggests that the w- might be part of the root.

What word? If by <x> you mean h.et (usually transcribed <h.> or
<H>), the only thing close to it in the lexicon is
<h.ittu:l> 'bandage for a wound' (Ezek. 30:21), from the root
{h.tl} 'to bind, surround, swathe'. This cannot be related to Arabic
<qit.t.un>. Hebrew /q/ and /t./, not /h./ and /t/, correspond to
Arabic /q/ and /t./. You cannot blithely scramble consonants in
Semitic. If you are serious about doing long-range comparison
involving Semitic languages, you need to study comparative Semitic
philology first. And to avoid mistaking loanwords for cognates, you
need to be familiar with loanwords into Arabic and the other Semitic
languages. I just ran across a reference today to a paper by Irfan
Shahid, "Latin Loanwords in Arabic", in _Encyclopedia of Arabic and
Arabic Linguistics_, which unfortunately is not on the Web.

> The emphatic t.t. of Arabic could be from t+w.
> All these data are fairly coherent.

Again, you need to study Arabic philology. I don't know, since I
haven't studied it; perhaps a double emphatic can arise that way.
But if I expected to convince anyone else, I would have to cite some
actual forms and references to the literature. Merely guessing where
a double emphatic comes from doesn't cut it.

Douglas G. Kilday