--- In cybalist@..., "P&G" <petegray@...> wrote:
> > I was speaking purely formalistically. The problem was: how
> > to get rid of the "if initial" part of the formulation of Verner's
> > law. We can do that by inserting glottal stop *signs* between
words,
> > let them unvoice the following fricative and then let them
disappear
> > in non-voiced environments.
> >
> > Torsten
>
> We could also insert Santa Claus, but it still wouldn't be an
acceptable
> explanation. To explain given facts, we need a process based on
what we
> suggest really happened, not just something "purely formal".
>
> Peter
In a sense you are right, since the ontological status of Santa Claus
and of glottal stops are both problematic. Not that glottal stops
don't exist, but are just that: stops. They are not somethings, but
cessation of something (voice) that was already there.
One could express the glottal stop in pseudo-MIDI as
1: +voice for 1 time unit
2: +voice for 1 time unit
3: -voice for 1 time unit
4: +voice for 1 time unit
5: +voice for 1 time unit
(pianola style)
or
1: +voice for 5 time units
3: interrupt voice for 1 time unit
("functional" style)
The latter style is perhaps how you describe a prosodic feature.
So, to let the glottal stop match something that "really happened",
we could let that ting be "the intention to keep words separate",
present in some languages but not in others. It's not a thing that
happened, but it's an intention to make it happen if necessary.
Torsten