On 2009-02-21 06:04, david_russell_watson wrote:
> "... A similar rule operates in Pawnee,
> a Caddoan language spoken on the other side of the world,
> though the details are different; see Douglas R. Parks,
> A grammar of Pawnee (New York 1976: Garland), pp. 14, 42-3)."
See a detailed description of this rule in Wichita (related to Pawnee):
https://www.indiana.edu/~iulcwp/pdfs/01-deguchi02.pdf
I would interpret the affrication of dental stops in *TT (as well as
*TK) clusters as phonetic reinforcement, preventing their realisation as
unreleased stops and so increasing their contrastivity ([CORONAL] being
the least salient place of articulation in terms of auditory cues such
as vowel-consonant transitions). Note t > ts in the High German Shift
and the affricated allophone of /t/ (in "strong" positions) in many
accents of English.
Piotr