noun physical_somatic_body_part

Proto-Siouan-Catawba *i-pa(t) (?)

Proto-Siouan *i-ré•tpa

Proto-Crow-Hidatsa *-réhpa

Crow ištéhpa RG

Hidatsa réhpa ‘navel’ J , néhpa

Pre-Mandan *rétpa

Mandan répta ‘navel’ H:179

Proto-Mississipi-Valley *i-ré•tpa

Proto-Dakota *čhekpá

Lakota čhekpá ‘navel’ RTC , čhekpá ‘navel, twins’ EJ

Dakota cekpaḣdoka , †čhekpa ‘navel’

Proto-Hoocąk-Chiwere *ré•twa

Chiwere retwa ‘navel’ LWR:24

Hoocąk reečáwa KM , reecawa(šų)

Proto-Dhegiha *réhtašǫ

Omaha-Ponca thétashoⁿ , †ðéttašą F&LF:108

Kanza/Kaw yéttašü ‘navel’ RR

Osage théṭashoⁿ , †ðéhtašǫ ‘navel, umbilicus’ LF:143b

Quapaw déttašǫ ‘navel’ JOD


Catawba hipat ‘his navel’ FS

General comment

The options in reconstructing Proto-Siouan: *i + é•tpa > yé•tpa would leave Chiwere/Hoocąk, DH unexplained as they have reflexes of Proto-Siouan *r. The better solution *i + é•tpa > i + r + é•tpa (with *r epenthetic) accounts for all but the OVS forms (that have numerous other irregularities). Lakota typically shows čh < *y < *r following this prefix. Here *i- is 3sg. inalienable possessor; the rest is ‘navel’. The Biloxi/Ofo forms Biloxi †čįpǫ́ , tciⁿpóⁿ ~ -yaⁿ (DS-265a), Ofo †čhį́•pu , tché̄mpu (DS-329b, Ssf) may be related, but it is still uncertain how. There are several other problems with this set. Apart from the initial resonant, Crow and Hidatsa should normally inherit a long, accented vowel. The Catawba form, assuming it is cognate, suggests that the word is morphologically complex with a boundary between *-re•t- and *-pa. Any Proto-Siouan-Catawba reconstruction is always extremely tentative. This entry also shows that in DH, Proto-Siouan *tp > *pt > *ht just as *tk > *kt > *ht. In Lakota *tp > kp. Cf. also ‘tongue’, ‘kettle’. Inherited Crow/Hidatsa e > i unless the magic of glottalization preserves it. The fact that we’ve got an (irregular) short e here could be explained by shortening.

Details Language Word Source