Re: sound change question

Thomas Wier wrote:

> Quoting Sylvia Sotomayor <kelen@...>:
>
> > So, Kélen has five stops:
> > p (labial)    t (dental)    ts* (alveolar)    c (palatal)    k (velar)
> > (*ts is often just [s] but is still counted as a stop.)
> > I've been contemplating some sound change with the goal of getting rid
> > of [p]. Which would be more reasonable?:
> >
> > p    t   ts   c   k     becoming
> > k    t    ts   c   q     or becoming
> > kw  t   ts   c   k
>
> I think probably the first set, where *p becomes /k/.  Arapaho
> underwent a similar process on its way from Proto-Algonkian.
> It's reasonable to think that a pull-chain effect might have
> started with *k becoming [q], leaving an opening for *p to fill.
> It's an unusual change, but I know of no natural language that
> would take *p and turn it into [kw] -- quite the opposite change
> would be expected.
Agreed.  Evidently you want to keep it as a stop; my first thought was for
it to > fricative [P] or [f], then/or > h, then/or 0. All widely attested.

Reply