I recently had occasion to look into the etymology of the
Danish place-name in the subject line. In OWN it appears in
a number of forms; <Sjóland> and <Selund> seem to be the
most common, <Selund> apparently occurring earlier. I've
also run across mention of an ODan. <Siâland>. Digging
around on the net, I found this at
<
http://sprogmuseet.dk/ordhistorier/sj%C3%A6lland/>:
Den almindeligste accepterede opfattelse af oprindelsen
til navnet <Sjælland> er imidlertid at den intet har at
gøre med hverken 'sjæl', 's(j)ø', 'land' eller 'lund'.
Den gammeldanske form <Siâland> stammer tilbage fra en
fællesgermansk form *<Selha+wundia-> som er sammensat af
et ord *<selha-> og en afledningsendelse *<wundia-> 'som
er forsynet med, som ligner'. Ordet *<selha-> kan være to
forskellige ord: (a) ét der svarer til og betyder det
samme som dansk <sæl>, og (b) ét der betyder 'fure,
indskæring'.
The most generally accepted view of the origin of the name
<Sjælland>, however, is that it has nothing to do with
<sjæl> 'soul', <s(j)ø> 'sea', <land> 'land', or <lund>
'grove'. The ODan. form <Siâland> derives from a Common
Gmc. form *<Selha+wundia-> that is composed of a word
*<selha-> and a derivational ending *<wundia-> 'that is
provided with, that resembles'. The word *<selha-> can be
either of two different words: (a) one that corresponds to
and means the same as Danish <sæl> 'seal', and (b) one
that means 'furrow, incision, cut, notch'.
It goes on to explain why (b) is generally preferred.
The idea is clear enough, but this does leave me with a
couple of questions about the details. First, why *-ia- in
*wundia-? The <-und> suffix that appears in a number of
island names looks to me like the outcome of something like
*-wnt-ih2 > *-wund-i: > *-undi > -und. Secondly, *selha-
'seal' is no problem, but is there any Gmc. evidence for
*selha- 'furrow, cut'? OE has <sulh> 'a furrow, a gully',
but that appears to be zero-grade (and athematic). Is this
minor carelessness in a popular presentation, or am I
missing something?
Brian