A Relic of Proto-Siouan *rp/np 'one' in Mississippi Valley Siouan

Proto-Siouan "one" is reconstructed in two versions from two separate cognate sets, both of which are defective in that each has been entirely lacking from one or another of the major Siouan subgroups. One of the sets for "one" is found in Mississippi Valley Siouan, and it contains the same root as the indefinite article that is found in the same subgroup. Mississippi Valley Siouan, in Quapaw, also provides confirming evidence for the older of the two reconstructible words for "one," which had been found only in Ohio valley (Southeastern) Siouan with a probable cognate in Hidatsa. This discovery adds substance to the cognate set from which the proto form was posited. The Mississippi Valley Siouan subgroup is the connecting link between the two subgroups in which "nq-" is otherwise attested. "Nq-" also has cognates in the more distantly related Catawba. Contains 16 references. (LB) *********************************************************************** Reproductions supplied by EDRS are the best that can be made from the original document. *********************************************************************** "PERMISSION TO REPRODUCE THIS MATERIAL HAS B. N GRANTED RY

Proto-Siouan 'one' is reconstructed in two versions from two separate cognate sets, both of wLich are defective in that each has been entirely lacking from one or another of the major Siouan subgroups.1One of the sets for 'one' is found only in Mississippi Valley Siouan (and possibly Mandan).
It contains the same root as the indefinite article that is found in the same subgroup.The reconstruction *wf(y)p is based on the following cognates: Dak.NW, Wit wida; Winn.hilti; Chiw.
Dheg. *wfp-xti: and perhaps Mand.máxana C hwixrp.Except for Mandan, the sound changes and other alterations that yield the attested forms are relatively well understood.: In the same languages the indefinite articles have a form based on *wi (Winn., Chiw:, Dheg.) or *w@ (Dak.), depending on whether or not the language has metathesized the vowels of *wf(y)p.
The older of the two reconstructible words for 'one' is the less well attested, found until now only in Ohio Valley (Southeastern) Siouan with a probable cognate in Hidatsa.
The Hidatsa cognate retains only the first syllable (morpheme?),(compare Crow hawdta 'one'), and it undergoes the usual Crow/Hidatsa denasalization.At the moment, it seems best to reconstruct the root as *iv-or *rp-, that portion which is attested in all of the above mentioned languages.
It would be nice to find confirming evidence in Mississippi Valley Siouan for this reconstruction, and we find it in Quapaw, hidden away in the construction for 'once, one time'.The term is hi npxtf.Iterative numeral words in Quapaw are formed by proposing hi 'times, amount' to the number and suffixing a nasal vowel, -4, which often collapses with the stem-final vowel.The result was probably a compound at one time composed of the numeral stem and the verb 'do, use', whose form is 1,7g (cf. the Osage cognates with -9, below).
The expected Quapaw form, if it were based on mfxti 'one', would be *hi mIxti-4 contracted to hi mixti.Compare Quapaw hi nppti-f 'twice', hi dabn$ 'thrice', hi towl 'four times', and the Kansas Working Papers in Linguistics.Vol. 13, 1988. pp.122-126.Osage cognate ixeratives (120) widxci-9 'once' (and the morphemically similar Omaha term), (MD Sppl-0 'twice', (120) 66104-p 'thrice', (120) tdpi-p 'four times', etc.The actually occurring Quapaw form, hi npxti is clearly composed of preposed hi 'times', suffixed -4 (probably from the earlier ?p'do, use', cf.Osage), and the numeral stem np-xti 'one', which is in turn composed of the intensifier -xti 'real, very' (attested throughout Siouan, cf.Quapaw mf-xti 'one') and the root np-, which can only mean 'one' in this context.'As far as I can determine n9-, then, is a relic, isolated in this unique Quapaw derivational construction, although now that it is known to occur in the Mississippi Valley subgroup, perhaps other instances will be discovered.4If this discovery adds nothing to the phonological shape of our reconstruction, it does at least add considerable substance to the cognate set from which the proto forr was posited.The Mississippi Valley Siouan subgroup is the connecting link between the two subgroups in which *up-is otherwise attested.
I mentioned above that st,up-is probably the more-archaic of the two reconstructible forms.One justification for this view is the fact that *nip-is the more widely distributed of the two forms; it is found in the subgroups that are most distant from each other, both linnuistically and geographically.Another is the fact that it is np-that has cognates in the more distantly related Catawban.
Our knowledge of Catawban comes from the remnants of several languages once spoken in the Carolina piedmont.
In early historical times these Catawban speaking remnants banded together with the Catawba proper forming a poly-dialectal 6ommunity whose linguistic divisions are only poorly known.The only other independent Catawbau language of which we have knowledge is Woccon.
Catawba 'one' is transcribed in a variety of ways by the various researchers who attempted field work on the language.All show variation between initial oral and nasal stops however.
Catawba: Carter 1980) (Lawson in Carter 1980) Taken together, the Catawba and Woccon forms reproduced above (there are numerous other variant transcriptions) illustrate two important points.First, the initial syllable of the Catawban forms appear to be the proper match for Proto-Siouan *ND-'one' reconstructed above.
Second, the alternation or (more probably) variation between initial oral and nasal stops preceding nasal vowels is as common (and ancient) in Catawban as it is in Siouan, making this phonological characteristic of Siouan-Catawban very old indeed.The only suffix with the shape that I am aware of is the iterative suffix (discussed above) used with numerals in Osage and Omaha-Ponca.It, in turn, may be derived from the PS verb 'do' (although we would expect sto rather than *f).If we reconstruct *wi # , we may, in fact, be reconstructing an iterative form with the meaning 'once' rather than 'one'.
2The phonologically similar Biloxi word with the translation 'eldest, elder', noxti, found in Dorsey and Swanton (1912, 238) probably exemplifies a different but homophonous PS root *rp-'to grow, age', cf.Ioway ny 'grow up' or Osage nphi 'reach mature age', nOhp 'older person'.*Koontz has suggested that the common Siouan birth name for the first born daughter (usually written Winona in Dakota) should be analyzed wi 'female', -np 'first' and nf 'diminutive'.
Yet another relic may occur' in Quapaw dn.(e 'that' + np 'one') meaning 'only that' as opposed to oinfpa (e that + Nips 'two'  meaning 'both'.La Flesche (1932) has enom'xtsi 'he, she, it   alone.'This means that the Catawban forms for the numerals 'one', 'two' and 'three' all match corresponding Proto-Siouan forms well.
have been made to improve reprOduchon (lushly Potnts of view or opinions Slated in MIS docu ment do not necessarily represent official OERI position or policy

al
The discovery of a Quapaw reflex of *np-'one' strengthens our etymology for Proto-Siouan-Catawban 'one' appreciably.NOTES 2The cognate sets that I refer to here were assembled in part at the Comparative Siouan Workshop organized by David Rood and held in the Summer of 1984 at the University of Colorado.Specifically, data were contributed by Randolph Graczyk (Crow), A. Wesley Jones (Hidatsa), Richard Carter and Patricia Shaw (Dakota), and the author's field notes (Kansa, Quapaw, Osage, Omaha and Ponca).John Koontz also provided Omaha data, read a draft of this paper, and provided a number of very useful commo:ats and suggestions.Other (i.e.published or otherwise available) sources are listed in the bibliography and include Hollow 1970 (Mandan), Vantine 1982 (Chiwere), Miner 1984 (Winnebago), Swetland 1977 (Omaha), Dorsey and Swanton 1912 (Biloxi and Ofo), and Dorsey 1882, Hale 1883, Sapir 1913 and Frachtenberg 1913 (Tutelo), Shea 1984 (Catawba) and Carter 1980 (Woccon).Zeyrek 1982 discusses the Siouan and Catawban counting systems generally.The 1984 Comparative Siouaa Workshop was supported by the National Science Foundation under great #BNS-8406236 and by the National Endowment for the Humanities under grant #RD-20477-84.21do not mean to imply that the sound correspondences for this lexical set are all regular.The reflexes of what Koontz and I have reconstructed tentatively as *(y) are, in fact, irregular, and there are some problems with the initial labial.Koontz suggests that the ify may be epenthetic.Presumably it might then postdate common Mississippi Valley Siouan, and this might explain its phonetically plausible but historically irregular development.Since sequences of VIV: are not normal in MVS, *Pli(y)f is probably bimorphemic, f.