Non-IE roots in Germanic/@, a, e, i, j, o, u

[Yoel L. Arbeitman]

While I cannot comment on any "Vasconic"/Greek comparison, the conceptual
analogy between that proposed there for "Iron" and "star" does not work by
IE Lautgesetz in the parallel Greek "iron": Latin "star" word. The Latin
word is generally considered to derive from an IE root *sweid- "shine".
This initial consonant sequence would become in Greek *h(e)id-. as the w
disappears in "Standard Greek" (remains in some dialects, written with a
digamma, is restorable in Homer by meter and is mostly extant in
Mycenaean). But even in +w dialects, we would then expect *hwid= in Greek
as initial *s- > h in Greek. The alternation sid-us/ sid-er- (nom. nt. vs.
stem) in Latin means that the suffix is IE *-os/-es- and that
intervocalically the -s- in sideris (genitive) is the product of rhotacism.
The same -r- in the Greek word for "iron", sideros, cannot be the product
of rhotacism. Thus (1) the initial s- is wrong and the "suffix" -r- is
wrong if we are to deem the Greek and Latin words as cognates. The only
remaining possibility is to conceptualize the Gk. word as a Wanderwort
(from Latin????) or to consider both Gk. sideros and Latin sider- as
borrowed from elsewhere. But this is impossible for Latin sid-us.
	Yoel

At 09:15 AM 3/10/99 GMT, you wrote:
>Rick Mc Callister wrote:-
>> *i:sarno [Celtic, Germanic] > iron, Eisen n.
>> [< ?Vasconic *isar "star";
>> see Basque izar "star"] [mcv2/98, tv2/98]

>Larry Trask <larryt at cogs.susx.ac.uk> replied:-
>> No comment.

>Also, Greek `side:ros' = "iron", Latin `sidus' (gen `sideris') = "star".  This
>semantic association was quite possible in early times when Man had not yet
>found how to smelt iron and iron was a precious rarity available only as
>natural nickel-iron alloy in meteorites.