Chronology and history of Germanic Consonant Shift

[Douglas G Kilday]

petegray (9 Jun 2001) wrote:

>As for the first sound shift, I'm not up to date with the most recent stuff,
>but I quote the following from the dtv-Atlas:

>    "For the date of the 1st sound shift, we can use the word Hanf, which
>comes from the Greek word kannabis.  This word is a loan word out of Scythian,
>which did not enter Greek till the 5th century BC.  In Germanic we meet the
>word in its shifted form *hanap-.  Since Germanic could not have borrowed this
>word very early, we can assert that at this time the rules *k > h and *b > p
>were still in force.  But it does not tell us how long this rule had existed.
>That it no longer was in force in the 3rd and 2nd centuries before Christ can
>be concluded from loan words from Latin, none of which have shifted forms.
>(The first contact of Germans with Romans as in this period.)"

A Scythian source for the 'hemp'-word is reasonable enough, given Old
Persian <kanab> and alleged cognates in Finno-Ugric. But why on earth would
Greek intermediation be necessary to get a Scythian word into Germania?
Darius I bridged the Danube in his costly campaign against the Scythians,
and that puts a large part of them much closer to the Germans than the
Greeks were.

>Walker-Chambers supports this.  He says:  "It is estimated that the First
>Sound Shift was completed by c 500BC; but we only know that it was finished
>before the Germanic peoples established contact with the Romans in the1st
>century BC, since none of the words borrowed from Latin were affected by it."

There's something wrong with this analysis, since the ethnonym
Catti/Chatti/Hesse and the hydronym Vacalis/Vachalis/Vahalis/Waal show that
the consonant-shift was _not_ completed everywhere by the 1st cent. BCE.
Both authors have assumed that "first contact" is relevant to the form of a
borrowed word. In the case of borrowings from Latin into Germanic, this is
probably false. The first borrowing of <coquus> 'cook' might well have
undergone shifting of initial [k] to [x] in some dialects, but the resulting
forms would have been superseded by fresh borrowings with [k], as
Romano-German contacts continued for several centuries. In situations like
this, "last contact" may be more important than "first contact".

DGK