> Zero-grade forms seem to imply such a sense. Unfortunately, my German
> dictionary's not good enough for me to fuly understand Pokorny's
> statement,
> "tiefstuf. slav. z^ürdü in aksl. z^rüdü `Holz', russ. z^erdü `lange,
> du"nne
> Stange', poln. z`erdz/, sloven. z^r^d `Wiesbaum'". Baltic does have a
> fence
> word, "lit. [...] gardìs f. `Gatter, Gitter'", meaning open fencework as
> in
> railings or a trellis.
In Wahrig's (a German-German dict., similar to Oxford's or
Webster), the etymology for Garten in a nutshell is given like
this:
<<ahd. garto, got. garda "Hürde", daneben ahd. gart "Garten", got.
gards "Haus, Hof" < germ. *garda- "Haus als umzäunter Besitz"
< idg. *ghorto-, *ghordho-; zu *gher- "fassen"; verwandt mit Gurt,
Gürtel, gürten>>
[ahd. = althochdeutsch; got. = gotisch; germ. = germanisch;
idg. = indogermanisch = PIE] ["fassen" above: in the sense "um-
fassen", i.e. "to comprise, enclose"]
I'd mention in this semantic context for the German language
Hecke, vb. (um)hegen, Gehege, Hag/Haag [the latter: a frequent
toponym as den Haag in NL and throughout Germany & Austria];
these are in semantic close relationship both with Zaun "fence"
(cf. town) and with Garten (and, as we see above, with "girdle").
> OIr du:n, Welsh din 'fort, castle, fortified place', seen
> in Celtic place names such as London, Lyons and the
> alternative name, Dunedin, of Edinburgh.
Then also in ancient toponyms along the Danube (up to its
delta), e.g. Singidunum, Noviodunum...
> Richard.
George