force, instrumental

physical_contact_impact

Proto-Siouan-Catawba

Proto-Siouan *raka-

Proto-Crow-Hidatsa *raka-

Crow dak-; bak-, dáak-, dak-

Hidatsa raka-; wa•ka-, ra•ka-, raka- , naka- etc.

Pre-Mandan *:ka- < **rka-

Mandan ka- RTC

Proto-Mississipi-Valley *ka- ~ *ra-

Proto-Dakota *ka- ~ *ra-

Lakota ka-; waka-, yaka-, ka-; ki+X = gla-

Proto-Hoocąk-Chiwere *ki-

Chiwere gi-

Hoocąk gi-, hai-, rai- , gi- (hai-, rai-)

Proto-Dhegiha *ka- ~ *ra-

Omaha-Ponca ga-; áa-, ðáa-, ga- RR

Kanza/Kaw ga-; áa-, yáa-, ga- RR

Osage ga- , †ka- LF

Quapaw ka-; áa-, dáa-, ka- RR

Proto-Southeastern *ra-ka-

Proto-Biloxi-Ofo *raka-

Biloxi daka-; ndáka-, ídaka-, dáka- , †daka- D&S

Ofo ka-, (a)baka, (a)tcaka , †ka- D&S

Proto-Tutelo-Saponi

Tutelo láka- , †láka- ‘cut with axe’ [cf. lákatkúca] JOD

Proto-Catawba

Catawba kąʔ ~ Ka- ‘cut’ MAS, KS, SS

General comment

In DH, the *k of *ka- is normally lost in 1Act and 2Act. E.g., in Kanza/Kaw, galü ‘shave’ is conjugated áalü, yáalü, galǘ-. The full form of ga- is preserved in the 1st and 2nd persons if a locative prefix is also present. Loss of -g- in prefixes may have something to do with stress placement, (e.g., 1Act á-a-ga-lü; 2Act á-ya-ga-lü, 3Act ágalü ‘shave off hair from the head’).

In Dakotan, Mandan, and DH, relic traces of Proto-Siouan *ra- are preserved in irregular dative-possessive constructions, as noted above. In Mandan, at least, this r cannot be epenthetic. Mandan ki+X = kara-.

*ra- may have been a morphemically distinct instrumental that could be compounded with *ka. Forms like Tutelo o-la-cpewa ‘cut’ suggest this, particularly when placed beside forms like lákatkúca ‘cut with axe’.

Chiwere, Hoocąk have simply replaced *ka- with ki- everywhere for reasons unknown.

Details Language Word Source