> 3) What is the evidence that these two words were borrowed from just
> Germanic and no other language?
>
Now for the "ash" word:
This is what Jouppe has:
"
Fi. kaski 'burnt-over clearing' < PreF *kaski / *kaśki
Sw. aska 'ashes' < ↑ PIE/PreG *ħæsk'- > Gmc. *askōn 'ashes'
Møller (Vi-sW) has:
(as)g^h-n-, got. azgo: :
(as)k^-n-, ahd. an. aska 'cinis', s. k^-n- 'cinis'.
...
k^-n- (< voridg. k^-n-) 'cinis' in
germ. ask-n-, Nom. an. ahd. aska 'Asche'
(as-, in an. o,s-gru:a 'das aschgraue' (vgl. Noreen Lautl. 35),
Reduktion von
a:s- in
sanskr. á:sa- 'Asche', lat. a:reo 'bin trocken');
+ y- s. k^-ny-
[: idg. ĝh-n- (< voridg. K.^-n-) in
asgh-n-, got. azgo:- 'Asche';
dieses g^h-n- = semit. s.-n- in arab. s.anan 'cinis', s. k^-ny-].
...
k-ny- 'Asche' (< voridg. intr. k^ánay-,
< k^-n- [: ĝh-n-], s. d., + y-),
gr. konía:, koníe: 'Staub, Asche', kónis 'Asche, Staub',
+ s- in lat. cinis (cinerēs) 'Asche',
: semit. s.-n-y- (< vorsemit. K.^-n-y- = K.^-n-
[= idg. g^h-n-, s. d.]
+ y- in arab. s.anan (-an < -a:n < -ayuN) 'cinis'.
"
In other words, a composite *xas-k^n-. As for the laryngeal, LIV
posits *h2éh1-s- and quotes Hittite has(s)- "Asche" and (derived)
ha:ssa-/hassa: "Herd".
How come this one also 1) has a 'Møller-cognate' in Semitic, 2) has
a/a: in Germanic and Italic?
BTW, there are similar problems with the supposed Latin loans in
Basque, which Trask was so dismissive of alternatives to: many of them
also have suspect PIE etymologies.
Torsten