Re: [tied] Re: Metathesis in sonant+ labiovelars, Greek

--- In cybalist@yahoogroups.com, "Joao S. Lopes" <josimo70@...> wrote:
> I have doubt if this shift is Proto-Greek. It occurs in a bunch of
> few words. If such shift was regular we would have phygos instead
> of phobos, and pik- instead of pep- (*pekW-).

You mean of course **pip-, don't you?

> I think that these words can be traced to some "Graecoid" substratum
> or adstratum.

With Graecoid sub-/adstratum you mean an Indo-European dialect that
corresponds with Greek in virtually all isoglosses except for this
feature? Thereby the sacred "Ausnahmslosigkeit der Lautgesetze" is
saved, which is of course an ideal for all comparativists.

In reality, however, "sound laws" will often not apply in all cases.
The laws do not influence all instances in a magic split of a second,
but work themselves through the linguistic system from word to word and
pattern to pattern (lexical diffusion). A couple of words may be
forgotten, before the laws recede, or the laws may be suppressed early
in the process, so that only few words are affected. The "Old School"
explained such contradictory evidence as a result of dialect mixture,
but it may be unnecessary and at any rate uneconomical, as long as one
accepts the concept of incomplete sound laws.