Norman Hinton wrote:
[In reply to:
> > The exact relationship between the two Germanic words is still in
>> dispute, though it certainly looks like an Indo-European ablaut-relation.
>> Some think that *burg- is related to Germanic *bergan- 'to protect'. And
> > the Greek pyrgos 'tower' and Pergamos 'the citadel of Troy' (assuming
> > they are related to the Germanic forms) make things interesting for the
> > etymologist .....]
>Give me an example of this dispute -- among whom ?
See, for example, Kluge, Etymologisches Worterbuch der deutschen Sprache,
23. Aufl., bearbeitet von Elmar Seebold, 1995, under "Burg" -- references
are given there. I used "in dispute" in the sense that alternative proposals
had been put forward (perhaps I should have said "uncertain"). Briefly, the
problems concern whether/how burh and beorg stand in ablaut relation to
each other, whether burh stands nearer to the Gmc verb *bergan-, and
how (and whether) Germanic *burg- relates to Greek words such as pyrgos,
Pergamos and phyrkos (which may go back to a pre-Indo-European substrate).
> I see no problem with either of those: "high place" seems to
> original notion of the I-E root.
I'm not saying there's any particular problem with the semantics of
root. It's clear, though, that semantic differentiation occurred later,
within Germanic, and certainly by the time of Old English. It's not
justified, as Paul Cullen has demonstrated, to say that OE beorg and
OE burh are the same word ("Why do you suppose that burh and beorg are
different words?" was the question at issue), when etymology and the
place-name evidence clearly show that they are different. It's true that
later, in post-AS times, the elements were sometimes confused as
the final element in place-names, but that's a different matter.
I've done a little more research on the matter and this is what I've
Barney (1985) says that the two words are unconnected
so that =
beorgan "to protect", gebeorg (n) "protection" and burh/burg (f) =
"stronghold" are from an IE root *bhergh- "to hide, protect" (Watkins, =
1985), while beorg (m) "hill, barrow" is from a homophonous IE root =
*bhergh- "high" (id.). The latter may be the source of the Latin fortis =
"strong" (although it may also be from IE *dher- "to hold firmly, =
support"). It's also the source of Burgundy (< PrG *burg-und-i- < IE =
*bhrgh-nd-i-) (Bammesberger, 1990).
So, it looks like we have two homophonous roots but
that's it. I =
can't connect Greek pyrgos (m) "tower" to *bhergh- "high" at all because =
assuming the root noun was thematised to *bhrgh-os/-=E2, the Greek would =
be **brakh-os(-=EA)/barkh-os(-=EA). In fact, I'd put good money on =
pyrgos being non-Indo-European since the consonantism is all to pot for =
an IE root.
Barber (1932) was of no use because I'd forgotten
that he was only =
concerned with words which indicate the position of the accent by =
Verner's Law unlike *bhergh- which doesn't.
ANSAXDAILY Digest - 3 Jul 2001 to 4 Jul 2001 (#2001-175)