NOTES ON THE KANSA WORD LIST OF MAXIMILIAN,

This paper analyzes the Kansa Indian language vocabulary collected by Prince Alexander Philip Maximilian of Wied-Neuwied during his journey through the plains of North America between 1832 and 1834. Prince Maximilian's word list is likely the oldest existent vocabulary on this language. His vocabulary, published in 1843 as an appendix to his travel journals, is then compared to more recent observations of Kansa words by James Owen Dorsey in the late 1880s and by the author in the 1970s. Observations about the phonology of Kansa are offered which, in turn, clarify the ultimate source of Maximilian's list. (MDM) *********************************************************************** Reproductions supplied by EARS are the best that can be made from the original document. Ic *********************************************************************** NOTES ON THE KANSA WORD LIST OF MAXIMILIAN, PRINCE OF WIED1 U.S. DEPARTMENT OF EDUCATION Office ot Educational Research and Imp' Overlent EDUCATIONAL RESOURCES INFORMATION CE NTER (ERIC) document has been reproduced as .ece,ved from the person or organization originating it 0 Minor changes have been made to improve reproduction quality Points of view or opinions slated in this document do not necessardy represent official OERI position or policy Robert L. Rankin University of Kansas "PERMISSION TO REPRODUCE THIS MATERIAL HAS BEEN GRANTED BY

Observations about the phonology are offered which, in turn, clarify the ultimate source of Maximilian's list.
One of the earliest Kansa language vocabularies, if not the earliest, in existence is that collected by Alexander Philip Maximilian, Prince of Wied-Neuwied, during his voyage through the plains of North America between 1832 and 1834.The list was published as an appendix to Maximilian's journals in 1843, along with vocabularies of a large number of other plains languages.
In this paper the words on Maximilian's 1833 list are compared with the same Kansa words recorded by the rev.James Owen Dorsey, probably sometime between 1888-1890 and by this author in the 1970's.2From this comparison it is possible to make a few linguistic observations that seem pertinent, primarily about the phonology of the words on Maximilian's list, then, based on these and other observations, certain inferences can be drawn about the source of the list itself.
The Maximilian and Dorsey lists are presented below in their respective authors' original notation; the Rankin list is surface phonemic (with the exact status of vowel length still undetermined).
Comments offered by the several authors are placed beneath the pertinent entries in the proper columns.The Konsas, or Kansas Indians have always lived on the river of that name (Gallatin, p. 127).For the past thirty years they have lived at peace with the Osages; and the two tribes have intermarried.They still number fifteen hundred souls, and possess a tract of three thousand acres.They speak the Osage dialect, which belongs to the Dac6ta linguistic group.(Thwaites,vol. 24,p. 229) And although we find such delightful comments as 'Most of the... languages of the Missouri Valley can more easily be reproduced by the Germans and Dutch than by other nations; because,.., their own speech abounds in gutturals and regularly has hard endings,' in point of fact, Maximilian took special care to write the words he heard accurately.

Maximilian
He had read Duponceau and Pickering carefully and, citing both, discusses problems of orthography at some length.
He used what amounts to a transcription system ,3 based on a combinatinn of German and French orthograph- ic norms with the addition of a number of special dia- critics and liberal use of explanatory notes.His system utilizes primarily German spelling conventions, resorting to French only when German lacks a particular sound.Thus, for example, nasal vowels and [2] are written the French way, i.e., with postposed nasal con- sonants and the letter j, respectively.
Turning to Maximilian's Kansa list, a number of things can be said.
The precise status of vowel length in Kansa is still a problem today.Recordings made in the 1970's show it sporadically in words where it occurs regularly in other Siouan languages such as Winnebago or Crow.So, while undoubtedly significant at one time, it ap- parently became variable as use of the language decreased and presented difficult problems for the transcriber in later years.Maximilian took special note of vowel length in all of his word lists stating, 'I have tried to indicate the length of a syllable by adding an h,...' (Thwaites,vol. 24,p. 206) Maximilian adopted the device used in German orthography for noting long vowels, i.e., his use of h does not parallel attempts English speakers sometimes make to indicate continental vowel values.We find length indicated in the following items: ARROW, HEAD, MOUTH, SUN, WATER, EARTH, FIRE, GOD, PIPE, WOMAN..This list suggests two generalizations.
First, Maximilian notes the vowels in all monosyllables as long.Monosyllables are indeed all long in a number of Siouan languages.White Eagle and Miner (personal communication) both report this independently for Winnebago, and it also seems to be true of at least Chiwere and Crow in addition.Dorsey's (ca. 1888) normalized Kansa transcriptions do not give clear indication of vowel length, so they are of little help to us, and length is subject to considerable variability in today's rapidly obsolescing Dhegiha languages (Rankin, field notes; Richard T. Carter, John E. Koontz, personal communication).So we have not known how closely Dhegiha followed common Siouan lengthening rules, and it would be nice if Maximilian's transcriptions provided the evidence.
That Maximilian wrote length for all of the Kansa monosyllables on his list (ARROW, HEAD, MOUTH, SUN, WATER), appears to confirm that the rule did apply generally in Kansa in the 1830's, but we shall have to return to this problem presently.
Second, in Chiwere, and with considerable variation in Dhegiha today, all accented initial syllables appear to be long.For the most part Maximilian's list shows this also (EARTH, FIRE, PIPE).The only exception is CHILD, which should have length but where none is indicated.
GOD and WOMAN perhaps should not have the long vowels attributed to them in the word list.
Voicing of obstruents is another phenomenon that is important as it is one of the phonological features that differentiates Kansa from Osage.
In the Kansa of the 1970's all instances of the lax (i.e., unaspirated, unglottalized and ungeminated) series of stops had voiced (except, of course, in clusters with voiceless fricatives), i.e., common Dhegiha *p t k > b d (1) g.
In the century-old Dorsey materials there are excr.ptionsto this voicing.
Dorsey wrote numerous instances of [k] for modern [g].There are fewer instances of [t] for modern [d], and I was able to find no instances at all of [p] for modern [b].
This parallels the situation in other Dhegiha languages that I have discussed in greater detail elsewhere.Within Dhegiha the lax series has voiced most completely in Omaha-Ponca and Kansa, partially in Quapaw and scarcely at all in Osage.The philological and comparative evidence indicates that the voicing began with the bilabials and proceeded through the dentals affecting the velars last.
The Maximilian list shows no voiced stops where any later source has voiceless stops, but it does have a few apparently conservative cases of voiceless stops where the later sources have innovated voicing in the words for HAND and MOUNTAIN.Interestingly, these involve labials, all of which had voiced in the later sources available to us.All are pretonic; in the Maximilian list all posttonic lax stops are written voiced.
Among the fricatives there is one substitution of g for 1 in CHILD, which is written with sch; in TOMAHAWK the same morpheme, ii.ga SMALL (unstressed) is written with the letter j representing 1.3 Beyond the voicing discussed above, a few additional observations regarding Maximilian's transcription of obstruent consonants are possible.
There is no graphic sign of the glottalization in WOMAN, but the word, written as it is with k, shows at least that common Dhegiha *x? (preserved still in Qua- paw) had already merged with k7 by the 1830's, i.e., *wa306 > Kansa wak76.Lack of a graphic representation of glottalization is not surprising however, as even the most talented amateur phoneticians who attempted to record American languages often had trouble deciding what to do with it.
Only one aspirate is present in the data.RIVER carries Maximilian's cryptic comment 't often like h', but while he could have been referring to the aspira- tion of the t, which is phonemically th throughout the Dhegiha Siouan subgroup, it seems just as likely that he was referring to affrication instead (Kansa th (611) regularly preceding front vowels).
Interestingly, the sound is nonetheless written by Maximilian as t, not the perhaps to be expected tsch.That affrication of dentals was proceeding apace however, is shown by FIRE where *ppé-te has already become ppé.fe.
Tense or geminate stops are always written by Maximilian with letters representing voiceless stops, but otherwise they are not distinguished from the other stop series.In other words, Maximilian did not distinguish aspirates from tense (geminate) or glottalized stops, hardly surprising since only the one aspirate and one ejective were represented in his Kansa list.
Maximilian (Thwaites,vol. 24,p. 206) states, 'As the French acute accent seemed to me entirely adequate for indicating the correct intonation of the Indian words, I have chosen it to show on what letter or syllable the emphasis is to be placed.Occasionally, owing to haste or lack of time, it may have been forgotten or omitted.' The translator mentions (ibid.p. 209) that 'The vertical accent seems to be used to indicate secondary, or less emphatic stress....' Little can be said about accent in this word list other than that it generally seems to correspond to the notation of stress in the other sources.
FIRE and MAN appear mistranscribed, or misrepresented by Maximilian's source, as stress falls on the initial syllable of these words in Kansa and throughout Dhegiha.
The spelling of EARTH suggests that the development of modern Kansa y from common Dhegiha *r in a nas- al environment may not have been complete and that some occlusion may still have been present.
In this word Ponca and Quapaw have n while Osage has 5 and Kansa has y.
The earlier stage in Kansa probably corresponded closely to the Osage.Maximilian's comment 'n nearly like h' may refer to lack of occlusion, i.e., a [6]like phone preceding a nasal vowel.
The final vowel of HAIR is written i despite the fact that there was a readily available symbol for LI (cf.MOUNTAIN).Common Dhegiha ü is preserved yet today in both Kansa and Osage, but for at least one hundred years there has been a certain amount of phonetic fluctuation when it comes to rounding.HAIR, if rendered correctly to, and recorded by, Maximilian, could be interpreted as evidence for similar fluctuation as early as the 1830's.
Finally, there are several words that are either unrecognizable or appear quite different from their later forms.
MOUTH should not have a rounded vowel and cannot be explained except perhaps as a copying error.It does not have a rounded vowel in any Siouan language, so this is one instance in which access to the original manuscript would clearly be very useful.
There must be an editor's error or misprint here.
PIPE also almost certainly contains a copying or printing error.It was probably recorded nah-ni-ba by Maximilian with later misinterpretation of the second n as h (Kansa 'tobacco' is ngni varying with n4n6).
RIVER shows oral I rather than the aberrant .1". re- corded by Dorse:1.
This not only agrees with what this author found in the 1970's but with the oral vowel found in this word in all the rest of the Dhegiha languages also.
Here it was probably Dorsey who erred.
We may now consider briefly the source of Maximilian's Kansa vocabulary.
In most instances he indicated the source of each vocabulary in a footnote; unfortunately Kansa is one of the very few cases in which no source is given, at least in Thwaites.
About his sources generally, Maximilian writes: I have written these vocabularies, in part from the pronunciation of the Indi.ansthemselves; in On this list I have boldfaced the three suspect items.
ISLAND, Kansa rumätschi, Otoe rd-mi-tschi or rumaetschi cannot be completely reconciled with any later occurring Kansa form.
In addition, [r] is not a Kansa sound, not even allophonically.The recorded form is virtually identical to Maximilian's Otoe entry however, which he says was 'written from the pronunciation of Major Dougherty,...who understands the language thoroughly.' (Thwaites,vol. 24,p. 285) SUN should have initial m.This nasal is present in every Dhegiha language, but in Otoe it has the modern form bi, where b is phonetically either an unaspirated [p] or a [b].Maximilian's Otoe entry for SUN is written pih, exactly like the suspect Kansa entry.
TOMAHAWK in Kansa and Osage is based on the word that is usually translated 'knife', m414 (thence m4hispe).
Here, instead of containing m414, it is compounded from mgze-'metal' and the second element, -spe, from the later attested form, the word for 'axe'.This follows neither the Kansa nor Otoe pattern, but it is close to the modern Omaha mqz9spe 'axe'.4Recall that Dougherty's assignment involved both the Otoes and Omahas centrally.
The only other possible source of Maximilian's Kansa list would have been Osage.He remarks more than once on the similarity of Kansa and Osage.For example on Kansa he writes (Thwaites,vol. 22,P. 252): 'Their language is entirely the same as that of the Osages, and the language of these two people is only a dialect, originally not different from that of the Omahas and Puncas, being distinguished only by the pronunciation, and not by its roots.'Maximilian's Osage sample was 'written from the pronunciation of Mr. Chardon, who had lived a long time among the Osages and understood the language perfectly.'(Thwaites,vol 24,p.296)Osage was spoken far to the South in southwest Missouri and adjacent areas of Oklahoma and southeast Kansas.The fact that the non-Kansa substitutions on Maximilian's Kansa list are taken from Otoe and Omaha (of Nebraska) rather than the more closely related but geographically distant Osage all but rules out Chardon as a source for the Kansa list, leaving Dougherty the only likely candidate.
In spite of the fact that Maximilian's 1833 Kansa word list is clearly second-hand, most of the vocabulary is obviously Kansa and even contains examples of the features that separate Kansa from Osage.Observations on the progression of consonant voicing, affrication of dentals, nasalization and accent can be at least tentatively advanced.
Finally, it can be stated with some certainty that Major John Dougherty was the source of the Kansa vocabulary and that Dougherty's primary linguistic competence in Siouan was in Chiwere (Ioway-Otoe-Missouria) rather than Dhegiha (Kansa, Osage, Omaha-Ponca, Quapaw).The vowel length shown for all monosyllabic Kansa words then may be misleading however, since in Englishlmonosyllables cannot end in short vowels and Dougherty would probably have pronounced them long no matter what he actually heard.Vowel length aside though, judging from the Kansa and Otoe lists, Dougherty appears to have had a fairly good ear for Siouan phonologies.NOTES 1 This preliminary note is based on my analysis of the published versions of Maximilian's Kansa, Otoe and Osage word lists (Thwaites 1906, vol. 22-24) and so may include inaccuracies introduced by the editor and/or publisher.I would have preferred to work from primary sources of course, but these are not yet available to me.
Over the long run it might also be quite useful to consult the papers of Maj.John Doustherty, who was the source of several of Maximilian's lists.Maj.Dougherty evidently had a good ear for the phonetics of the Siouan languages and may have recorded some of them more extensively himself.Accoz-ding to the Fort Leavenworth Archivist, Steve Allie (personal communication), Dougherty's papers may be stored at Carlisle Barracks in Pennsylvania.
2 I am grateful to the American Philosophical Society and to the Graduate Research Fund of the University of Kansas, both of which supported my field work on the Kansa language.I am also especially grateful to Mrs. Maude Rowe and her family, without whose help the Kansa project could never have been undertaken.Mrs. Rowe worked patiently with me for several years beginning in 1974, and most of my transcription is based on her pronunciation of Kansa.

3
In the modern language the unstressed allomorph serves mostly as a diminutive marker.
In this form it has lost both its friction and oral stop, ap- pearing in the variant higa [hilja].

4
The Omaha term itself is evidently a loanword from Santee Dakota, where it has the same shape.So, in fact, either language could have been the source.
document has been reproduced as .ece,vedfrom the person or organization originating it 0 Minor changes have been made to improve reproduction quality Points of view or opinions slated in this docu-the duty of every traveller in distant, little-known lands to make contributions, according to his ability, to the knowledge of their languages....' Maximilian zu Wied-Newied Abstract: The Kansa word listed collected in ca.1833 by Prince Maximilian of Wied is compared with two more recent transcriptions of the same terms.