This is gonna sound like a weird question and you'll all
be wondering what gLeNny is up to but here it goes.
Let's hypothesize that you have a language with stops that
contrast between aspirated and non-aspirated. Let's say
that the language evolves over time such that aspirated
and non-aspirated stops merge together as non-aspirated
stops *p, *t, and *k in initial position only. The aspiration
contrast remains medially.
Let's also say that this merger takes place only for initial
_velar_ and _labial_ stops while a distinction between *t- and
*tH- remains.
Is this typologically credible? If not, how might such
an idea be modified so that it is more plausible?
- gLeN
_________________________________________________________________
The new MSN 8: advanced junk mail protection and 2 months FREE*
http://join.msn.com/?page=features/junkmail