Re: Lïzxvööse Verbs I: ActiveTri-Consonantals

Marcus Smith wrote:
>Morphology can produce words that "should" produce geminates. For example,
>the 2pl object clitic is t-, and there is never an epenthetic vowel. Teach
>us begins as [tma...]. When you would expect to get a word beginning like
>[tta...], you seem to get a "pulse" to separate the consonants. (Don't ask
>for phonetic details, I'm not good enough at this stuff to say what is
>going on.) It is distinct from medial or final gemination.
Interesting.  In the Leti language (Eastern Lesser Sundas, Indonesia) this
is also seen, at least in the written dictionary forms (the work of a
skilled linguist, Jonker 1932).  Some presumably unpronounceable clusters,
e,g, ptuna 'star', tpirsa '...?', pniali 'anchorage' (I'm working from
memory ;-( )-- these generally due to V-deletion;  redupl. ppuata 'woman'
(actually <*/vata/); llena 'lightning'.  In the older wordlists on which he
drew, these are often written with an intervening "e" (probably for Dutch
schwa) or a harmonizing vowel (pniali:  "piniali" or "peniali").
My guess would be that here there's also a "pulse" separating the
consonants, probably containing a very brief schwa, voiced or voiceless
depending on environment.  A neat phenomenon.