augmentative/diminutive shifting

[VOORHIS at BrandonU.CA]

> Thanks.  Perhaps then it is extremely old in Siouan languages. If there are
> any scholars of contiguous families (Algonkian, Muskogean, etc.) on the list,
> would you say that shifting was similarly lexicalized in these families?
>
> Jess Tauber
> zylogy at aol.com

In Algonquian, in Cree, a change of t to c is required (= a productive rule)
throughout any noun that acquires the diminutive suffix -is: atim 'dog',
acimosis 'puppy', awaasis 'child', nicawaasimisak 'my children' -- in the
latter example, -t- normally follows possessive prefixes, like ni- 'my', before
a vowel, but even that affixed t has to change to c because the word has -is,
cf. astotin 'hat', nitastotin 'my hat'.

Other Algonquian languages don't don't do as Cree does, but t~c correspondences
in cognates suggest they may have done something similar formerly: Kickapoo
mecemooha, Mesquakie metemooha 'old woman', Kickapoo and Mesquakie cahkw-,
Ojibwe takkw- 'short'.  The -VVh- in Kickapoo and Mesquakie is another
diminutive suffix, and 'short' is obviously diminutive in meaning.

	Paul