I have to correct what I said about
<(h)yll>. The opinion that it derives from PIE *sh2ul- (or rather
*sh2u:l- to get the right vowel quality) was cautiously supported by
the great Albanologist Eric Hamp in an article written some time ago, which is
why I accepted it, but there are problems with it. The initial *sh2- should
have been reduced to *s-, but the expected Albanian reflex of that is
<gj->. (Note by the way how
strange-looking the regular Albanian developments are. For example, PIE supnos
'sleep' > gjumë).
In order to obtain <h-> or zero
(there is some dialectal variation in Albanian in this respect) one would need
to assume a special (and hard-to-verify) development of the rare cluster *sh2-
(perhaps an early merger with *sk-, which does yield Albanian <h->) or
irregular loss of s- in the onset (very suspect). Then, there are semantic
difficulties (the semantic shift 'sun' > 'star' via a generalised meaning
like 'light in the sky' is not impossible but too ad hoc to be accepted without
reservations).
However, the most problematic aspect of
this etymology is that the Albanian "sun" word <diell> is regularly
descended from a PIE form -- in fact, it derives from *sh2wel- (the
full-grade variant of *sh2ul-) > *swel- > diell. Believe it or not, PIE
*sw- is regularly reflected as <d-> in Albanian -- e.g. *sworg-eje-ti >
dergjet 'lies ill'. The diphthong <ie> is the regular development of
stressed *e before a sonorant, not a contraction of *-ihy- or *-ihe-. It
looks, then, as if the "sun" root had already been taken. (As for your
question whether <diell> survives outside Albanian, nearly all words for
the sun in the various IE languages -- sun, Sonne, so:l, soleil, su:rya-,
he:lios, haul, saule:, slUnIce, etc. -- are derived from the same protoword as
<diell>.)
For these reasons alternative etymologies
of <yll> must be considered. One often-mentioned possibility is that it
derives from the PIE root *h1eus- 'burn, singe', found in Latin (u:ro: <
*eus-o:, u:stor 'burner of corpses'), Germanic (OE ysle < us-l-j-o:n-
'glowing ashes', ON ysja 'fire'), Greek (heuo: < *euho: < *eus-o:) and Old
Indic (os.ati 'singe' < *eus-e-ti). The "ashes" word represented by the
Germanic derivative *us-l-i- would work rather well as the prototype of
<yll>. The semantics (stars = "glowing embers") is perhaps parallelled by
the PIE "star" word (*h2ste:r) itself -- see the Cybalist discussion of this
word a few months ago.
[A tantalising thought for Glen and
others: Etruscan usil 'sun' could have a rather straightforward (non-metathetic)
connection with an IE root: *h1us-el-, *h1eus-elo- 'the scorcher'? If no-one
buys it, I won't insist :)]
Piotr
----- Original Message -----
Sent: Tuesday, January 30, 2001 6:14 PM
Subject: [tied] Re: Albanian connection
As what regards the deriving of <Hyll> from
<Diell>,<Di-hell>, <Di-hyll> (Sun) it is not the first time I
hear that suggestion but what makes me wonder is much a philosofical
question:
How did these anscient people know that the stars were nothing more but
Suns? Keeping in mind that this knowledge is very recent and in Middle Ages
nobody would dare to think it.
And another thing I wonder about is:
Has <hyll> or
<dihell> survived in any other language but Albanian
??