Re: Danaans [was Poseidon]

--- In cybalist@..., tgpedersen@... wrote:

>
> Torsten (talking to himself) writes:
> A humble question *dx > *t'(glottalic) (> PGerm. t)?

This would mean (gLeN Gordon's idea) Nom. *daxn- > *da:n-, Acc. *dxan-
*tan-. By paradigm levelling we could get a *dan-, too.
Suppose *t/*d develops into a shibboleth between two peoples (eg.
Skiri *t/Bastarni *d). Then the Skiri might want to apply a rule *d >
*t elsewhere in their language. See

www.angelfire.com/rant/tgpedersen/Shibbolethisation.html

where I try to explain what I mean by "shibbolethisaion" (I can't
pronounce it, sorry!).

Saxo mentions five kings of the Danes named Dan. Kings like to remain
in the nominative (because they are mostly subjects, grammatically,
not politically!), hence their names are Dan-, not Tan-. But you
can't have a king Dan ruling a people of *Tan-'s, hence Danes.
(a similar example: the masculine, oblique form of the Dutch definite
article used to be "den". It was abolished also in the written
language in the 1930's, except for two cities: Den Haag(s'Gravenhage)
and Den Bosch(s'Hertogenbosch), since cities rarely are (grammatical)
subjects (expressions like "Den Haag says..." are metaphorical).

> Suppose PGerm. was a pidginized (verb inflection simplification)
> trade language used first at the transfer point between the

Northern

> (Wistula) and Southern (Dniestr) traders by Southern river traders,
> we expect a "t/d insecurity" to spread along all the Southern

rivers

> (if controlled by those Southern traders). So much for Grimm.
> As that route for political reasons must be moved east

(Duna/Dniepr)

> we find a Baltic Finnic substrate at the "transfer point"
> (Orsha/Vitebsk). Hence stress moved to first syllable.
>
> Sorry for the looseness. I can't put it more precisely.
> On the other hand, I think perhaps we need to see a trade language

as

> not being bound to particular area (which presupposes people living
> there most of their lives, as in the famous dialect maps of the

19th

> century) but as bound to a route.
>
> Torsten