Last
year I took the opportunity to look at the papers of one James W. Lynd. You probably don't recognize the name—he's
not among the more famous people in American frontier history, though perhaps
he should be as he casts a very long shadow.
Born in Baltimore in 1830, James Lynd moved to Minnesota in 1853 in
order to study the language, culture, and history of the Dakota Indians. His intention had been to write a book about
their history, a project he was never quite able to bring to fruition. Somehow along the way he served a term as a
Minnesota state senator—everyone was a senator in those days, after all—but this
isn't why he didn't finish his book, nor why he casts a long shadow. Because James William Lynd was also the very
first person killed in the infamous Dakota Uprising of 1862.
On
the morning of 18 August 1862, the dawn of the massacre, it so happened that
Lynd was at the Nathan Myrick & Company trading post of the Lower Agency at
Redwood, Minnesota. At the fateful
minute, Lynd was standing at the doorway to the post, sipping from his morning
cup of coffee, when suddenly a Dakota man named Thawásuota ("Plenty
Hail") rushed toward him and announced that "Now I will kill the dog
who would not give me credit." He
then shot Lynd dead through the torso. Thus
began the massacre: what one historian calls "the most violent ethnic
conflict in American history".
The
sad irony for Lynd is that, in all likelihood, Thawásuota didn't even want to
kill him. His intended target—the person
who had denied him his credit—was the notorious Andrew Myrick, brother of
Nathan Myrick who ran the store. Andrew
Myrick has gone down in history as having once dared the starving Indians of
Minnesota to "eat grass" but it's doubtful if he actually said such a
thing, as is the story of his body being found with his mouth stuffed with grass. Nevertheless he was known to have been
"peculiarly obnoxious" according to one contemporary. He was killed only minutes after Lynd.
If
Thawásuota or one of the other Dakotas had recognized James Lynd, it's perhaps
possible that they might have spared his life.
Lynd had married two Dakota women à
la façon du pays and had sired by them three children. He spoke the Dakota language fluently, and
they in turn had given him the name Wičháwanap'iŋ or "Raccoon Collar". Many such white Minnesotans were saved from
the massacre by their Dakota friends or in-laws. But, then again . . . many weren't.
Over
600 settlers would eventually die in the ensuing massacre. In response to this giant bloodfeast, a
punitive military expedition was dispatched into Dakota Territory, whilst back
in Minnesota thirty-eight Indians were hanged on the gallows in the largest
single-day execution in United States history.
But more than that, one could argue that the tensions first strung in
Minnesota later snapped at the Sand Creek Massacre, the effects of which
ignited half a dozen Indian Wars between 1864 and 1869. By decade's end over a thousand soldiers,
settlers, Sioux, Cheyenne, Arapaho, Comanche, and Kiowa were dead. Could it all have been avoided without the
killings in Minnesota? Who knows.
The
manuscript for James Lynd's book only narrowly escaped from all this. The papers were being kept at Nathan Myrick's
trading post, and after killing Lynd, the Dakotas tossed the trunk containing
them into a nearby ravine where they languished for several months. The next Spring in 1863, the trunk was
recovered by a company of soldiers, but unfortunately the privates—who may have
been no more literate than the Indians—took to tearing away scraps and using
them as gun wadding. Much of the
manuscript was lost in this way, until the troop captain recognized its value
and ordered his men to stop. From here
the papers found their way to the Minnesota Historical Society where they
remain still. [1]
Aside
from the saga of the physical pages themselves, the Lynd manuscript is
interesting for being among the first "secondary source"-type
monographs written specifically about the history of the Sioux tribe. In fact it may be the very first such work,
at least that I know of—everything older are things like fur trader and
missionary journals, government reports, or studies on Dakota religion. Despite this, Lynd and his work has mostly
been ignored in favor of his more prolific contemporaries like Stephen Riggs,
T. H. Williamson, or the Pond brother—all of them missionaries.
One
reason for the neglect is that, actually, his book isn't very good. James Lynd shared your typical suite of 19th
century neuroses concerning Native American ethnography: page after page is
spent debating whether Indians descend from either the Japhetic or the Shemitic
races of mankind; there is a chapter titled "Early History" which is
actually mostly-empty speculation on whether the Sioux preserve some ancient
memory of the Bering Land Bridge migration (they don't); and very little is
recorded of what the Sioux of Lynd's time still remembered from more recent
eras.
But
there's another reason. After surviving
Indians, elements, and Army privates, the Lynd papers were put in the hands of
the Reverend Stephen R. Riggs, noted 19th century Dakota-ologist. Riggs would go on to publish one chapter of
the Lynd manuscript that survived intact (the chapter on "Religion")
in the journal Collections of the
Minnesota Historical Society (vol 2).
Along with this chapter, Riggs included a brief summary of the other
Lynd chapters, some of which weren't in a condition to be published. Unfortunately, this one chapter was published
under the title:
"History
of the Dakotas; by Hon. James W. Lynd; with introductory note by Rev. S. R.
Riggs"
This
was also published separately as a booklet, which you can find on archive.org.
Riggs
is clear in his summary that another of Lynd's chapters (titled "Dakota
Tribes of the N. West") also survived intact and in publishable condition,
as well as major portions of several others.
However, the misleading title of the one chapter he did publish has
apparently led others to assume—or misremember—that this amounts to the
entirety of the surviving Lynd materials.
It does not.
I
have been told by honest individuals that the full Lynd papers were all
published "somewhere" by the Minnesota Historical Society . . . but
after going through their entire catalog on the Internet Archive, I have found
nothing except for the aforementioned chapter edited by Riggs. If the rest of the manuscript was ever published,
anywhere, then I have no idea where. But
I think it's telling that whenever I have seen a scholar cite James Lynd's
writings, their citation has always been to the Riggs-edited chapter and never
anything else. (The one exception is
Albert Goodrich, in an article for the 1915 issue of CMHS.) From all this I can only conclude that people
generally aren't aware that the unpublished portions of the manuscript even
exist.
So
what does it say? Well as I said: not a
whole lot really. Lynd's book may be the
first of its kind, but I doubt he ever got to know the Dakotas personally as
well as the missionaries did. But the
reason I'm mentioning it here is because Lynd is the only writer to ever refer
to a particular "lost tribe" whom he calls the Unktoka. He mentions the Unktoka in two places:
"The Unktoka occupied the country south of
the Saint Croix river, east of the Mississippi.
The meaning of the word is "Our enemies". They were totally destroyed by the Iowas and Isanyati Sioux, before the commencement of the present
century."
(James
Lynd, manuscript, chapter titled "Dakota Tribes of the N. West")
"From
the time that the Sioux first occupied the head waters of Lake Superior to the
year 1640, it is impossible to say much of the Dakota tribes. The Sioux gradually moved south and west,
pressing upon the Iowas, Winnebagoes, and a tribe of Indians called Unktoka of Dakota extraction, which last
named tribe was afterward annihilated by the joint warfare of Sioux and
Iowas. This migratory movement must have
been very gradual, however; for the distance from Mille Lac to lake Superior is
but fifty miles in a straight line—a very moderate days journey for a Dakota
under any pressing circumstances. It is
probable that the Sioux, for many years after their removal to Mille Lac, were
in the habit of hunting in the lands immediately west of lake Superior; for
they were seen there in numbers by Father Alloüez in 1665 at the mouth of St.
Louis river, and subsequently by the French traders."
(James
Lynd, manuscript, chapter titled "Migration and Treaties") [2]
There
are a great number of rare, antique, and obscure tribal names in the annals of
frontier history, but "Unktoka" is of the very rarest kind. Aside from Lynd (and Riggs, who mentions them
in his summary) I can find no other writer whatsoever—no missionary, fur
trader, linguist, or historian—who has ever so much as written their name. And there is nothing online. Partly this is no doubt due to the
aforementioned neglect of the Lynd manuscripts, but it's still a rare day that
you find such a naked hapax legomenon of a tribonym like this. So I'm here to ask the question: did the
Unktoka exist, and if so, who were they?
* *
*
Lynd
tells us four basic facts about the Unktoka:
- They were Siouan.
- Their name means "Our
Enemies" in Dakota.
- They were located south of the St
Croix river.
- They were destroyed in the early
19th century.
At
least three of these claims can be seriously questioned.
FIRSTLY,
that they were Siouan — Lynd lists the Unktokas as one of 19 Siouan-speaking
tribes (or "Dakota" in his nomenclature). The other 18 are listed as:
Sioux, or
Dakota
Assinaboines
Mandans
Apsarokas or
Crows
Winnebagoes
Osages or
Washashas
Kansas
Kappaws [Quapaw]
Ottoes
Missourias
Iowas
Omahas
Poncas
Arickarees [Arikara]
Minntarees
or Gros-Ventres [Hidatsa]
Arkansas [Quapaw]
Pawnees
Ahahaways [Hidatsa]
If
Lynd is correct that the Unktokas were a Siouan-speaking tribe and were
distinct from the others on this list, then they must have been a tribe unknown
to modern science, because there isn't any other Siouan tribe
"missing" from this list (except for the Ohio Valley Siouans whom you
wouldn't expect anyway). As you may have
noticed, though, we can't just accept Lynd at his word here. The list is overstuffed. The Quapaw are listed twice as both the
"Kappaws" and the "Arkansas", and the Pawnee and Arikara
are erroneously listed as if they are Siouan tribes (they are in fact Caddoan). Lynd does seem to be aware that the name
"Gros Ventres" can refer to two distinct tribes, however he still
mistakenly believes that both tribes are Siouan—in reality only one is (the
Hidatsa, or "Gros Ventres of the Missouri"), while the other (the
Atsina, Aaniiih, or "Gros Ventres of the Prairie") are Algonquians.
"Minnetarees
or Gros-Ventres : This nation includes both the Minnetarees or Gros Ventres
proper and the Gros Ventres of the
Prairie. The former number about
900, and occupy the country adjoining the Mandans; the latter are wanderers
between the head waters of the Missouri and Saskatchewan rivers, and number
about 2000."
(James Lynd, manuscript, chapter titled
"Dakota Tribes of the N. West")
A
tribe called the "Ahahaways" are listed, whom Lynd refers to as
another "lost tribe" like the Unktoka. In reality these people were also Hidatsas .
. . kinda . . . actually they were one of the three originally-independent
villages who later formed the tribe now known as the Hidatsa. [3] Based on the information that Lynd gives, his
source for the Ahahaway must have been Henry Schoolcraft, and Schoolcraft's
source was William Clark the explorer.
Clark lists the Ahahaway in his 1804 Fort
Mandan Miscellany as if they were an entirely separate tribe, which at the
time they may have been. But all of this
shows that James Lynd was working from multiple printed sources for his
book—some of which he didn't entirely understand. In other words, not everything he writes was
necessarily an unfiltered record of Dakota oral tradition as told to him
directly.
SECONDLY,
that the name Unktoka (Uŋkthóka)
means literally "Our Enemies" in Dakota. The construction of this word seems a little
strange to me. In fact I'm not even sure
if it's grammatically correct, though I can't say with certainty. Nouns that are marked for possession like
this usually have a tha- element
prefixed (as in the name of Crazy Horse: Tȟa-šúŋke
Witkó, "His-Horse-Is-Crazy"). There is a class of nouns referring to family
members and body parts which can be marked for possession without the tha-, but I can find no indication in
any of my Lakota and Dakota references that thóka
"enemy" is such a word.
Furthermore, even if you were to allow that, then phonologically the
full uŋk- prefix isn't typically used
before a consonant-initial stem: in those cases it's shortened to uŋ- or sometimes lengthened to uŋkí-.
But
I don't really know a lot about Dakhóta grammar. For what it's worth, Stephen Riggs seemed
unconcerned by the name, and he is supposed to have spoken excellent Dakota—in
fact his grammar and dictionary are still the works usually cited for the
Mdewakanton dialect of Santee Dakota. So
maybe this is fine...?
A
name meaning literally "Our Enemies" may sound a little vague to be
the name of just one, particular group of enemies, but that's actually not a
problem. In Assiniboine, the sister
language of D/Lakota, the same root tóga
is used to mean both "enemy" in general and "Gros Ventre"
specifically. In the Stoney language
(the redheaded stepchild of the Dakotan family, and I mean that with love) the
Blood tribe of the Blackfeet are called the Togabi.
[4]
And
THIRDLY, that the Unktoka were destroyed at the beginning of the 19th century,
or half a century prior to when Lynd was writing — It's a little unbelievable
that an entire Native American group would have existed in Iowa/Minnesota so
recently without anyone noticing. If you
push the date back to the 17th or early 18th century then the idea is more
plausible, but how can someone be so wrong about something from only fifty
years prior? I don't know, but it can
happen—and it happened in the case of the Assiniboine. There is a persistent tradition which states
the Assiniboine split away from the Sioux at some point in the past. Usually it's supposed to have been the
Yanktonai band from whom they broke off: this particular detail isn't supported
by the linguistics, but the two tribes are nonetheless related.
As
for when the schism happened, that's
another matter. The historian Arthur J.
Ray in Indians in the Fur Trade makes
the point that writers kept saying it was "fifty years ago" for the
entire duration of the 1700s. In other
words, for a century nobody ever bothered to update the tally (and I'm sure I
once found an example Ray missed from the 1800s, too). And in fact they were wrong even in 1700: a
tribe called the "Assinipour" was first noted by the French in
1640. Whenever the Sioux-Assiniboine
split happened (linguists estimate 1000 A.D. give or take a century or two), it
was much earlier than when people were saying, and maybe the same is true for
the destruction of the Unktoka.
* *
*
Suffice
it to say then that James Lynd was a lil' sloppy in what information he gave
about the Unktoka. We probably can't
trust all of it; maybe we can't trust any of it. A person might still wonder, though, who were
they . . . were they even anybody at all?
There are a couple of possibilities, if we assume for argument's sake
that most of what James Lynd told us
about them is vaguely correct.
One
candidate is a group of Hurons (or maybe Petuns) and Odawas who refuged on Isle
Pelée for a few years during the 1650s.
Isle Pelée or "Bald Island" (so called because it was bare of
trees), also known as Prairie Island, is a large eyot in the midst of the
Mississippi River about ten miles downstream from the St Croix River
embouchure. These people were part of
the great diaspora of former-Michigander tribes who were fleeing beyond Lake
Michigan to escape the ravages of the Iroquois raids. The incident was described by Nicholas
Perrot:
"The
Outaoüas finally decided to select the island called Pelée as the place of
their settlement; and they spent several years there in peace, often receiving
visits from the Scioux. But on one
occasion it happened that a hunting-party of Hurons encountered and slew some
Scioux. The Scioux, missing their
people, did not know what had become of them; but after a few days they found
their corpses, from which the heads had been severed. Hastily returning to their village, to carry
this sad news, they met on the way some Hurons, whom they made prisoners; but
when they reached home the chiefs liberated the captives and sent them back to
their own people. The Hurons, so rash as
to imagine that the Scioux were incapable of resisting them without iron
weapons and firearms, conspired with the Outaoüas to undertake a war against
them, purposing to drive the Scioux from their own country in order that they
themselves might thus secure a greater territory in which to seek their
living. The Outaoüas and Hurons
accordingly united their forces and marched against the Scioux. They believed that as soon as they appeared
the latter would flee, but they were greatly deceived, for the Scioux sustained
their attack, and even repulsed them; and, if they had not retreated, they
would have been utterly routed by the great number of men who came from other
villages to the aid of their allies. The
Outaoüas were pursued even to their settlement, where they were obliged to
erect a wretched fort; this, however, was sufficient to compel the Scioux to retire,
as they did not dare to attack it."
- Nicholas Perrot
(in: Emma Helen Blair, The Indian Tribes of the Upper Mississippi, pp 163-4)
The
occupation of Isle Pelée lasted from 1657~8 to 1660, after which the Hurons and
Odawas turned back east toward Chequamegon Bay.
Aside from a couple inconsequential details, Nicholas Perrot is our only
source for this episode—it makes you wonder how many other such episodes we
don't know about. Some of the details
match what Lynd says: the location of Isle Pelée is more or less "south of
the Saint Croix river" where the Unktoka were supposed to have lived, and
they were driven out by the Sioux.
Perrot tells us nothing about the Ioway being involved, however, nor do
the Odawa (Algonquian) or the Huron-Petun (Iroquoian) speak a Siouan
language. And even though the Odawa and
the Chippewa are two somewhat distinct groups, I still am not sure if the
Dakotas would have forgotten that there were Ojibwes involved in this incident,
given how important Dakota-Ojibwe warfare was to their history.
* *
*
There's
another possibility, which allows the Unktoka to have been a Siouan-speaking
tribe, though it also doesn't follow Lynd's criteria exactly (probably nothing
does). And though the Isle Pelée theory
might be the more "sober" explanation, I actually favor this second
alternative—but explaining what it is and why requires a bit of
backtracking. It involves the history of
the Ioway and their relatives: the Oto, Missouria, and Winnebago.
On
either side of Isle Pelée are the towns of Diamond Bluff, Wisconsin, and Red
Wing, Minnesota. Archaeological remains
from both these locations show that from 1000 to 1300 A.D. the culture here was
in transition away from the older "Effigy Mound" type and toward the
"Oneota" type. Oneota is the
archaeological culture associated with the earliest Chiwere-speaking peoples:
the Ioway, Oto, and Missouria. Much
later, other tribes would take to crafting Oneota wares—mainly the Dhegiha—but
according to Dale Henning, these other groups didn't arrive until around 1500
A.D. [5]
To
the east in Wisconsin lived the Chiwere's sister tribe, the Winnebago (also
called the Ho-Chunk). When the French
first met the Winnebago in the 1600s they had been ravaged and immiserated by
warfare with the Illiniwek, but there's reason to believe that they were once a
more vast and powerful nation. They were
the ones who most likely built the mound center of Aztalan—which bears some
uncertain relationship to Cahokia—and they are the tribe most probably
connected to the Effigy Mound culture, which ended in the early 2nd millennium. I call the Winnebago the "sister"
tribe to the Chiwere, but the Chiwere themselves bestow upon them a different
metaphorical kinship. According to old
and enduring tradition, the Winnebago are the "grandfathers" from
whom the other Chiwere people descend.
This surely is an echo of their medieval moundbuilding civilization.
Around
the year 1300, the settlements at Red Wing and Diamond Bluff were abandoned,
and at least some of the emigrants must have moved downstream to the La Crosse
Terrace region, where two rivers flow in from the west: the Root River and the
Upper Iowa River. Three centuries later,
when Europeans first began probing this area, these tributaries were inhabited
by Ioway. A second bloc of Ioway was
located further west in the "Great Lakes" region of Iowa, accessible
via the Des Moines River from the east or via the Little Sioux from the west. They likely also controlled the all-important
Pipestone Quarry in southwest Minnesota.
In the 1680s the Ioways abandoned the eastern bloc under pressure from
Algonquian tribes (the Ioway were still on good terms with the Sioux at this
time) and they concentrated at the western bloc. But in time, the northern Iowa prairies would
all be taken over by the Sioux.
The
Ioway were a mobile people. The early
Frenchmen encountered temporary Ioway settlements along the Zumbro River and
along the Blue Earth tributary of the Minnesota. Mildred Mott Wedel, whose work is fundamental
to early Ioway history, readily ascribed to them a ranging territory that
extended to the Minnesota river. But as
far as actual, permanent settlements go, the Ioway are not known to have ever
lived further north than this. At least,
not since their ancestors left the Red Wing locality in 1300.
However,
there was another story current among the scholars of the late 1800s. According to them there had once been Ioway
settlements on the north side of the Minnesota River, near where it empties
into the Mississippi by the Falls of St Anthony (modern day Minneapolis). This was based—so they said—on tribal oral
histories which claimed that the earthen burial mounds visible near the Falls
had been built by ancient Ioways, before they were driven south by the Sioux. Variants of this story were repeated by
several authors including James Lynd, Gideon Pond, Thomas Williamson, Doane Robinson, and
others. The first example I know of is
by Alexander Ramsey in 1849, and the last is by Newton Winchell in 1911. [6] Gideon Pond's version is the most detailed,
and the only one which cites a named Dakota individual as its source:
"Takoha, the old war prophet, says that
the Iowa Indian never occupied the country around the mouth of the Minnesota
River. He affirms that it once belonged
to the Winnebagoes, who were long ago driven from it by the Dakotas—a few
others of the Dakotas agree with Takoha. But Black
Tomahawk, who is by some of the most intelligent half-breeds, considered
the best Mdewakantonwan traditionalist, says, that in the earliest years of the
existence of the Dakotas, they became acquainted with the Iowa Indians, and
that they lived in a village at the place which is now called Oak Grove, seven
or eight miles from Fort Snelling, on the north side of the Minnesota
river. The numerous little mounds which
are to be seen about Oak Grove, he says, are the works of the Iowa Indians. [ .
. . ] They were the enemies of the Dakotas, who used occasionally to make a
war-path from Mille Lac, where they then resided, down to the Iowa village, and
carry off with them scalps, which made glad the hearts of their wives and
daughters. The strife between the two
nations eventually became desperate, and the gods, who are always deeply
interested in Indian wars, espoused the cause of the Dakotas. [ . . . ] The old
man from whom we gather the substance of what has gone before, says that these
mounds are the remains of the dwelling houses of the ancient Iowas. Some say that they are not the remains of the
dwellings of the Iowas, but those of some other people with whom tradition does
not acquaint them; and others, still say that they are ancient burial
places."
Already
there's some uncertainty here about what exactly was going on. The war prophet Takoha maintained that the Winnebago, not the Ioway, were the
people involved—while other unnamed individuals said it was "some other
people with whom tradition does not acquaint them". A Winnebago presence here seems less likely
than an Ioway. But Pond wasn't the only
one who reported... confusion... about what tribes formerly dwelt by the
Falls. Williamson (1856) wrote that besides
the Ioway, "the Winnebagoes, Otoes, and Omahas have been named among the
nations driven by the ancestors of the Dakotas from the Minnesota
valley." He later (1880) added that
the Cheyenne were first on the lower Minnesota river, then the Ioway, and then
the Dakota. Robinson called the people
driven from the Minnesota "the Hohas, or Iowas". This is possibly just a mistaken
transcription of Ayúȟwa (Ramsey had
called them "Ho-wahs"), but it rather looks like the word Hóhe which is normally the Dakota name
for the Assiniboine tribe. Winchell said
that the Ojibwe attributed the mounds not to the Ioway but to the Gros
Ventres—by which he probably meant the Hidatsa.
Some
of the above can perhaps be salvaged: it's plausible that the Oto, Omaha, and
Cheyenne once inhabited parts of the Minnesota River valley—or near enough to
it, anyway. But for the most part there
doesn't seem much basis anymore for believing in Ioway villages at St Anthony
Falls. Modern archaeology doesn't link
any of the mounds there to either the Oneota or the Chiwere, as far as I know. And the historical/ethnological people no
longer repeat the story of the "Minneapolis Ioways". George Hyde
didn't bother mentioning it 26 years after Winchell (he spends literally one
sentence on the Dakota-Ioway conflict), nor do Gary
Clayton Anderson and Pekka Hämäläinen speak of such a thing.
Nor
apparently had the Ioways ever heard of it.
In 1837 a delegation of Ioway chiefs journeyed to Washington D.C. to
settle a territorial dispute with the U.S. government. As part of their case, the chiefs drew a map
of the Ioway's ancestral territory, encompassing all the places they had dwelt
since an original departure from a homeland near Lake Michigan (a memory of
their Effigy Mound ancestry I guess).
The map they drew still exists, and it doesn't show any settlements near
Minneapolis, nor is the Minnesota River even depicted.
You
could choose to believe the story if you wish.
True, the burial mounds were not made by the Ioway as Black Tomahawk
claimed . . . but many folks also believe that Stonehenge was built by the
Celts. The fact that they're wrong
doesn't mean there were never Celts living near Stonehenge. But I think that some of these stories were coming from the same place as James Lynd's tale about the Unktoka. Here's what Lynd himself said about the Ioway:
"The
planting grounds of the Ayuḣba or Iowas are said to have been
visible as late as 1830 upon the St. Croix lake near Hogan Wanke."
(Lynd, ms, chapter titled "Early History")
[8]
The
[Upper] St Croix lake is about a dozen miles up the St Croix river from its
confluence with the Mississippi. This is
a slightly different location than Mendota or the north bank of the Minnesota,
which is interesting—but it's also not far from the Red Wing locality and where
Lynd places the lost Unktoka tribe.
In
general though, there seems to have been a vague memory that southeast
Minnesota was once inhabited by a tribe who (probably) spoke a Siouan language,
only they couldn't agree exactly on who that tribe had been. Some said that they were Ioways—but in a way
which conflicts with the known history of the Ioway tribe. Others that they were Winnebagoes, or even
Hidatsa. Others said they were a
forgotten people "with whom tradition does not acquaint them" and
referred to only by the vague and meaningless signifier of Our Enemies.
* *
*
This
part of the country—Wisconsin, south Minnesota, Iowa, north Illinois, and upper
Michigan—endured quite a few changes in the 16-17th centuries. Most of the tribes famous to this area were
relative newcomers: only the Sioux, Winnebago, Menominee, and Chiwere are what
you might call the aboriginal inhabitants.
The Fox, Sauk, Kickapoo, Mascouten, and Potawatomi came as refugees in
the 1600s fleeing from the aggression of the Iroquois and Neutral confederacies. Huronians like the Wendat and Petun came as
well and, as we've seen, some of them briefly settled along the
Mississippi. The previous
century-and-a-half or so had already seen the Wisconsin Ojibwe migrate westward
from the Great Lakes nexus point, and the Omahas come up from their ancient
home on the Ohio River. Groups of Miami
and Illiniwek also arrived from the east, coming either in the 16th or the 17th
centuries.
Having
a dozen tribes move in over the course of a handful of generations can only
have been a massive disruption to the human geography of "Greater
Wisconsin". The status quo ante is
very difficult to work out—probably impossible in its details—but we should not
be surprised if a tribe or two we don't know about got lost in the scramble.
I am not just speaking in hypotheticals.
We already know of one "lost" Siouan language that was spoken
not far from here—where exactly is hard to say—which came within a gnat's toe
of being lost completely and forever without anybody noticing it ever
existed. I am referring to Michigamea.
The
Michigamea tribe were one of the subtribes of the Illinois or Illiniwek
confederacy, along with the Peoria, Cahokia, Kaskaskia, and others. The Illiniwek shared a language with the
Miami which belongs to the Algonquian family, and so naturally it had always
been assumed that the Michigamea were Algonquians as well. Then in the 1990s, the linguist John Koontz
examined the brief and hitherto-inscrutable fragments of Michigamea language
that had been recorded by Jean Bernard Bossu in the 1750s, and he concluded
that the language was actually Siouan.
More precisely: it belonged to the Mississippi Valley Siouan subfamily
(MVS) which also includes Chiwere-Winnebago, Dhegiha, and Dakotan.
I'm
not suggesting that the Unktoka were the Michigamea; that seems unlikely.. But they do show that there were once more
Siouans "out there" than the written histories had always said. What's more interesting is that, according to
statements Koontz made ont he now-defunct Siouan Mailing List, not only can
Michigamea not be any of the known MVS languages,
but it might not even have belonged to the three known MVS subfamilies . . . in other words, it was neither Dakotan nor
Chiwerean nor Dhegihan.
Unfortunately
we can't quite hang all our hats on this rack.
Koontz himself didn't seem fully convinced that Jean Bernard Bossu's
data was reliable. On the Mailing List
he raised the possibility that Bossu was misrecollecting various bits and bobs
from a few other Siouan languages he could imperfectly speak, and glommed them
all together into a frankenstein's monster of Mississippi Valley Siouan—hence
Michigamea's apparently eclectic mix of features. It's also the opinion of some that, actually,
the Michigamea were perhaps an Algonquian group turned Siouan, rather than the
other way around. I personally disprefer
that hypothesis, but even if it's true it doesn't explain where they picked up
their new Siouan language from. [10]
* *
*
That's
all well and good, but a lot of it's based on older scholarship which tends to
get superceded—occasionally even with good reason. And I still haven't said who I think the
Unktoka were. But there's one more
thing, a very interesting article from a few years ago, written by Colin M.
Betts in the Midcontinental Journal of
Archaeology: "Paouté and Aiaouez: A New Perspective on Late
Seventeenth-Century Chiwere-Siouan Identity" (2018). In this article the author makes the argument
that, up until the very late 1600s, there had more or less been two different
Ioway tribes.
The
English name for the Ioway tribe—and for the state of Iowa, which used to be
pronounced "Ioway" (I'm told my grandfather said it like this)—is
ultimately derived from the Sioux word Ayúȟwa. This was borrowed into the languages of the
Algonquian tribes living east of the Ioway and the Sioux, from whom it was then
borrowed by the French who spelled it like "Aiaouez". The French version of the name was probably
said with 4 syllables "A-ia-ou-ez", reflecting the syllabification in
the Algonquian versions: Miami-Illinois Aayohoowia,
Fox A:yohowe:we, Menominee Ayo:ho:wɛ:w, Shawnee Ha:yawʔhowe. But the name for the Ioway in their own
language is entirely different: Báxoje
in the practical orthography, or Páxoče
in Americanist notation. This name was
also adopted by the French who referred to the Ioways as the
"Paouté". [11]
Betts
argues that these Aiaouez and Paouté from the French writings were originally
two different groups living in two different places. Recall the two Ioway "blocs" I
mentioned. Betts says that the Paouté
were those living in northwestern Iowa—on the upper Des Moines and Little Sioux
rivers, and near the famous pipestone quarry of Minnesota—while the Aiaouez
lived in northeastern Iowa along the Root and Upper Iowa rivers. At some point prior to 1697 (probably around
1685) the Aiaouez left the La Crosse tributaries and joined their cousins the
Paouté in northwestern Iowa. The two
peoples eventually merged into one, and their descendants are the historic and
modern Ioway tribe.
I
came across Betts' article while already in the midst of thinking about the
Unktoka problem. Before I read it I was
much more inclined to think that James Lynd's "lost tribe" was just a
loose strand that probably doesn't reward much tugging: any of a dozen ways he
might have misunderstood the Dakota oral traditions, or that those oral
traditions could themselves have been wrong.
I probably wouldn't have written this then. But I think that Betts' thesis gels
suspiciously well with what Lynd wrote about the Unktoka: a lost tribe living,
more or less, where the Aiaouez had once been, and who were remembered as being
kind of maybe Ioways or Winnebagoes but also kind of maybe not. And of course they weren't exactly destroyed, as the modern Ioways are still their descendants.
Betts
bases his argument upon a close tabulation of the earliest French references to
the ancestral Ioway: specifically which
of the two names are used when to
refer to people living where. He makes no reference to the Lynd writings,
and presumably isn't aware of them.
Betts
points out that technically both Aiaouez and Paouté are exonyms: according to
James Owen Dorsey, the Ioways' original name for themselves was Chékiwére. I believe the significance of this is
misplaced. Betts uses this to disprove
the idea that Aiaouez and Paouté constitute an exonym-endonym pair for the same
tribe, but the names both being exonyms doesn't necessarily mean that they
refer to two different groups. He also
cites Mildred Mott Wedel in support of the idea that Báxoje was originally the Otoes' name for the Ioway—and not a name
the Ioway used for themselves—but I don't think this is what Wedel says at
all. The question of how and why the
Ioways abandoned their self-designation Chékiwére
in favor if Báxoje remains, in my
opinion, unresolved. Dorsey's
explanation—that people used different names depending on whether they were on
home turf—just doesn't seem plausible to me.
In any case though, there seems to be a parallel situation with the
Otoes, who call themselves Jiwere but
evidently used to call themselves Watótta
[12].
Putting
the Oto issue aside, it would make sense why the Ioways have had two different
names for themselves—"Baxoje" and "Chekiwere"—if in fact
they descend from two different founder populations. According to this view, the Báxoje were the "Paouté" of
the upper Des Moines river, and the Chékiwére
were the "Aiaouez" of the La Crosse tributaries. After 1685, when the two tribes moved in
together, they were thenceforth referred to using either one or the other of
their original two names.
This
would explain the confused nature of the Dakota traditions. The Dakota had originally known the
Aiaouez/Chekiwere as the Ayúȟwa, and
after 1685 they continued using this name to refer to the new consolidated
Ioway tribe. Later generations
remembered that another tribe had once lived in southeastern Minnesota, but
they weren't sure whether that tribe still existed or not. Some said they were Ioways, others said they
were a lost tribe—and they were both kinda right. Those who remembered the Aiaouez as a lost
tribe could hardly have used the original name Ayúȟwa to refer to them, since they still used that name to refer
to a different tribe which was still very much alive. And so the Dakotas coined a new name: Uŋkthóka. Maybe this name never attained wider currency
among the Sioux, but the last few who used it—and who knew whom it referred
to—managed to tell one person, and that person managed to write it down,
shortly before being shot in the chest by a man who was angry at Andrew Myrick.
[13]
* *
*
Is that the answer then? Well, maybe, but who knows. I should emphasize that I'm making three separate points:
1. More attention should be paid to the
unpublished James Lynd materials.
2. You should read Colin Betts' article and
make up your own mind re the supposed Paouté vs. Aiaouez distinction.
3. Lynd's "Unktoka" were the same
as Betts' "Aiaouez".
Points
#1 and #2 I am absolutely certain of. #3
not so much. I wouldn't wager much on this theory, and I really don't know who the Unktoka were at the end of the day—if they even were anybody. But the idea is intriguing.
There
is also a third possibility, which occurred to me very late in writing this so
I don't have much on it. But I recently
read Mark Walczynski's book Jolliet and
Marquette, about the French expedition of discovery to the Mississippi in
1673. In it Walczynski has this
footnote:
"Recently,
[David] Costa discovered that the Kaw referred to the Ottawa as
"Indokah". [Michael]
McCafferty speculates that "Indokah" ~ *inohka might have been a word
used by Siouan speakers south of the Great Lakes to refer to the Great Lakes
Algonquians in general. Miami-Illinois
had extensive prehistoric contact with Siouan speakers[...]"
(endnote 5:22)
The
word inohka here (technically ī̆nō̆hka) is the name which the
Illiniwek used to refer to themselves. I
can't find anything more about this "Indokah" name that the Kansa
used for the Odawa, but in another publication David Costa also mentions a name
"Intuka" which the Quapaw used to refer to the Peoria Illiniwek. McCafferty thinks this name is related to Inohka, and perhaps it is. But it's hard not to notice that these
Dhegiha words also look very similar to Uŋkthóka.
Would
that require the "Our Enemies" translation to be abandoned? It might . . . though I'm not sure. I would note that the thóka meaning "enemy" has no known cognates outside of
Dakotan; I would also note that there is no known, recorded word in Dakota that
refers to the Illiniwek, though they must have had one, since the two tribes
were deadly adversaries in the 17th century.
But by the mid-1800s the Illiniwek had long ceased being major
players. The Sioux at that point might
not have remembered much about them anymore, and the word they once used to
refer to them—or for "the Great Lakes Algonquians in general"—could have gotten whittled down to something else, a word for any enemy. [14]
NOTES
[1] – My sources for James Lynd's
life and death are Dakota Dawn by
Gregory Michno (2011), Massacre in
Minnesota by Gary Clayton Anderson (2019), Sketches Historical and Descriptive of the Monuments and Tablets
erected by the Minnesota Valley Historical Society (1902, no author given),
and "Memoir of Hon. Jas. W. Lynd" by Rev. S. R. Riggs (Coll. Minn. Hist. Soc., Vol 3, 1880), as
well as the introduction by S. R. Riggs in "History of the Dakotas: James
W. Lynd's Manuscripts" (Coll. Minn.
Hist. Soc., Vol 2, 1889, originally published 1865). Riggs is also the one who called Myrick
"peculiarly obnoxious".
Michno's book discusses the
mythology surrounding Andrew Myrick's death.
Anderson's goes along with the popular, maybe-apocryphal version. Michno also suggests that Thawásuota maybe did in fact recognize and intend to
murder Lynd, and that it was Lynd who had made the "eat grass"
comment. I find this unlikely.
I don't know exactly what Lynd was
doing at Myrick's store, aside from drinking coffee. Monuments
and Tablets says that he was "serving temporarily as a clerk", Dakota Dawn says he was "in
charge" there, but S. R. Riggs ("Memoir") merely says that he
was "then stopping" there. The
modern books both name Lynd's killer as "Much Hail" and
"Tawasuota", but Monuments and
Tablets give his name as "Waukon Wasechon Heiyadin" or "One
Who Travels Like a Sacred White Man" (i.e. like a preacher). I don't know if this is a different guy or
just another name for the same person . . . the book has a photo of Mr. WWH,
and I think it looks like the same
guy, but I'm not great with faces. The
exact forensic details of Lynd's murder also aren't agreed upon. Riggs ("Memoir") says he was shot
in the back by "two Indians, with double-barrelled guns" but most
sources only name one killer, and two at the same time sounds a little Agatha
Christie to me.
Riggs gives Lynd's Dakota name as
"Raccoon Collar" or "We-cha-ha-na-pin". Wičháwanap'iŋ
is my interpretation of that spelling (wičhá
"raccoon" + wanáp'iŋ
"necklace, pendant breastplate"), but don't quote me on it. It's interesting that he writes the name in
shitty-English-approximation and not the precise phonetic orthography he uses
in his linguistic works.
The specific abuses inflicted
upon the manuscript by the American soldiers is another matter with no
agreement. Reverend Riggs in
"Memoir" says they were used as "gun-wadding", but in his
introduction to the published "Religion" chapter he quotes their
commanding officer Capt. L. W. Shepherd as saying they were used for
"cleaning arms", and in the preface to the Lynd manuscripts
themselves (written by J. Fletcher Williams) it's said that the soldiers
"used some of the leaves for waste paper, and the rest were kicked about
the floor of the apartment, trampled on and stained with tobacco
juice". So yeah... they might have
been wiping their asses with it.
[2] – The state of the manuscript
is such that I can only give chapter titles, not numbers. At some point the chapter "Dakota Tribes
of the N. West" was removed by Stephen Riggs so he could make a
handwritten copy. Riggs apparently never
returned this chapter to the Lynd collection.
Whoever later bound the manuscript and added a title page, index, page
numbers, table of contents, and introduction (J. Fletcher Williams, I assume)
didn't include the "Dakota Tribes of the N. West" chapter, and I have
no idea where the original is—I couldn't find it with the Lynd papers, but
luckily Riggs' copy still exists. Riggs
called the missing chapter "the first chapter" in the introduction to
the published "Religion" chapter, but on his handwritten copy it is
called "Chapter II". The table
of contents by J. Fletcher Williams has different chapters for both I and II.
Also: I lied. Lynd actually mentions the Unktoka in three places: the first and second are
in the two passages quoted. The third is
on a random piece of scrap folder where the Unktoka are labeled
"extinct" and the Ahahaway are speculated to be Mandan.
[3] – Once upon a time, the
Hidatsa-speaking peoples occupied a far greater area than they did in the
classical frontier era, and their territory overspilled the western, northern,
and possibly also eastern borders of what is now North Dakota. The western segments later recrystallized as
a separate tribe—the Crow—but this reconfiguration might not have happened
until the 18th century. Prior to then,
there were at least five groups: the Awaxawi, the Awatixa, the Xiraca (or
Hidatsa-proper), and the ancestors of the Mountain Crow and the River
Crow. But to call the first three groups
"Hidatsa" may be teleological, since they may not have originally
formed a solid group vis-a-vis the two proto-Crow bands. The Awatixa and Xiraca are said to have been
more closely akin to the Crow than to the Awaxawi—more: the Awatixa and the
Mountain Crow may have originally been closer to each other than to the Xiraca
or River Crow, and likewise for the latter two.
This is why the Awaxawi were named as their own tribe by Clark: because
at the time, they were.
I find it useful to use the
spelling "Xiraca" for the one of the original three villages, and
"Hidatsa" for the modern tribe which they made a part of—they are the
same word: the x is a guttural fricative, the r is tapped like in Spanish, and
the c is a "ts" sound. I'm not
exactly clear on the histories and relationships of the proto-Hidatsa
bands. W. Raymond Wood says that the Awaxawi
spoke one dialect and the Xiraca and Awatixa spoke another, but according to
oral tradition it was both the Awaxawi and Xiraca who once lived in eastern
North Dakota (near the Red River) before they moved west and joined the
Awatixas on the Missouri. In other words
it's not clear which of the other two bands the Xiraca were most similar
to. It would help if we knew more about
the dialects spoken by these original bands—it might not necessarily be the
case that the language we call Hidatsa is even all that similar to what was
spoken by the Xiraca.
[4] – This is assuming that 19th
century Dakota worked the same as 20th/21st century Lakota (and that I'm
understanding the latter correctly). I'm
basing this off of the New Lakota
Dictionary and the Lakota Grammar
Handbook by Jan Ullrich and Ben Black Bear, which are the most thorough
analyses of Lakota by far. No resource
for Santee Dakota even comes close, unfortunately. Sioux words cited in this post are given in
the NLD orthography, which differs slightly from the orthography used by the University
of Minnesota. Note in particular that
Santee Dakota lacks the "rough aspirate" series of stops (pȟ tȟ kȟ)
which so characterize Lakota and Yanktonai.
The Assiniboine form cited tóga
is basically identical to the Lakota tȟóka
and Dakota thóka once you account for
spelling conventions, and the fact that Assiniboine's stops have generally
undergone a "softening" no doubt under influence from Plains
Cree. The Stoney name, which ends in the
plural suffix –bi, is taken from the
Stoney Mobile Dictionary App.
I couldn't tell you how common generally
it is in Native American languages for a word meaning "enemy" to also
mean a particular tribe, but as another example: Clark's The Indian Sign Language mentions that Hidatsa, Mandan, and Arikara
signers used the same sign for both "enemy" and "Sioux".
Regarding the questionable
grammaticality of *Uŋkthóka: it is
hypothetically conceivable that words for "friend" and
"enemy" could be considered extensionally as kinship terms for
grammatical purposes. But, again, there
is no such indication of this for either thóka
"enemy" or kholá
"friend" in any source I could find.
I also couldn't tell you why a word like Uŋkthóka, lacking the pi
suffix, appears to be possessed by a first person dual inclusive.
Stephen Riggs had this to say of
Lynd's proficiency in speaking Dakota: "I have heard a great many white
men talk Dakota, but I have yet to hear one, in all respects, talk it 'with the
fluency and idiomatic intonations of the natives' [as a writer for the Louisville Journal had claimed of Lynd]. Mr. Lynd, previous to his death, spoke the
language too well to have made such a claim for himself. But it is proper for me to say, that he did
speak the Dakota language very fluently, and doubtless understood its
grammatical construction better than most white men in the country" ("Memoir"). So Lynd could definitely speak the language,
but at the same time, Riggs here sounds a bit like he's trying not to speak ill
of the dead.
[5] - The Dhegiha are also associated
with Oneota, but Henning (1998) implies that they didn't show up until the
1500s. Dhegiha migration legends
recorded in the 19th century vaguely agree with this. There are also a very small number of Oneota
sites attributed to the Santee and Miami.
My sources for Oneota archaeology are: Dale R. Henning, "Plains
Village Tradition: Eastern Periphery and Oneota Tradition" (in Handbook of North American Indians: Plains,
2001); Dale R. Henning, "The Oneota Tradition" (in Archaeology of the Great Plains, 1998);
Thomas D. Thiessen, "Traditional and Historical Summary" (Plains Anthropologist, Vol 49, 2004);
Thomas E. Emerson, et al., Late Woodland
Societies (2000); and Shirley J. Schermer et al., Oneota Historical Connections (2015). The Oxford
Handbook of North American Archaeology I found to not be very helpful at
all.
Several contributors to Oneota Historical Connections—Eric
Hollinger, Robert Hall, John Staeck, Patt Murphy—are either skeptical or
agnostic that the Winnebago were the Effigy Mound people.
The La Crosse Terrace may have
been a dawning of a kind for the Ioway in particular, but my take from the
archaeology (particularly Oneota
Historical Connections) is that the ancestors of the Oto and Missouria also
fanned out to northwestern Iowa and southwestern Iowa/northwestern Missouri at
around this same time, ca. 1300 A.D.
Maybe these were all people who had come from Red Wing/Diamond Bluff, I
don't know. The southeastern quadrant of
Iowa state also had Oneota that can't positively be identified with a known
Chiwere tribe; maybe they too were lost.
Mildred Mott Wedel's monograph Peering at the Ioway Indians Through the
Mist of Time (1986) is fundamental for 17th century Ioway history, but she
is a bit more literalist in her read of the sources than I would be. According to her, there was only one locale
of Ioways at a time during the 1600s: first they were in northeastern Iowa,
then they moved to northwestern Iowa and subsequently zigged and zagged a
bit. This differs from Betts'
reconstruction, both on the map shown (from Oneota
Historical Connections) and in his later article I discuss later. It was written in the 1680s that the Ioway
were accessible from the Mississippi via a certain river, which may have been
the Des Moines or the Iowa River. Betts
identifies it as the Des Moines, which leads to northwestern Iowa. Wedel seems to
favor the Iowa River—specifically the Cedar River by way of the Iowa River,
whose headwaters are a stone's throw away from the upper Upper [sic] Iowa River
in northeastern Iowa.
Also, a side observation about
Dhegihan and Chiwerean history. Rory
Larson analyzed the cognate sets with glottalized consonants in Mississippi
Valley Siouan, and he reconstructs Proto-MVS *xʾ to have become *kʾ in
Proto-Dhegiha (and then ʔ~Ø in
Omaha). He then reconstructs Proto-MVS *xk to have become kʾ in the Dhegiha languages, and to have become xʾ in Ioway-Oto but NOT in Winnebago
(where it became šg). The implication is that Proto-MVS *xk first became *xʾ in Proto-Dhegiha, which then became *kʾ as per the usual shift.
This would mean that the somewhat unusual sound change xk > xʾ was an areal development that
happened in Dhegiha and Chiwere specifically—i.e. in the states of Iowa and
Missouri—and that the Chiwere tribes were perhaps "angled" more
toward their western and southern neighbors than they were to the Winnebagoes
(their "grandfather" tribe).
This would apply to the era after 1500 A.D. when Henning estimates the
Dhegiha first entered Oneota territory.
(source: Rory M. Larson (2019) "Glottalized consonants in
Mississippi Valley Siouan", in Proceedings
of the 38th Siouan and Caddoan Languages Conference, ed. Ryan M. Kasak)
[6] – The citations are: Alexander
Ramsey in the Annual Report of the
Commissioner of Indian Affairs for 1849-50; Ramsey, "Our Field of
Historical Research" (orig. 1851, reprinted in Collections of the Minnesota Historical Society, Vol 1, 1872); the
James W. Lynd manuscripts (1868-62); Gideon H. Pond, "Ancient Mounds and
Monuments" (orig. 1853, reprinted in Coll.
Minn. Hist. Soc. Vol 1, 1872); T. H. Williamson, "Who were the first
men?" (orig. 1856, reprinted in ibid),
and "The Sioux or Dakotas, of the Missouri River"(Coll. Minn. Hist. Soc. Vol 3, 1880);
Doane Robinson, A History of the Dakota
or Sioux Indians (1904); Dorsey & Thomas, "Iowa" article in Handbook of American Indians North of Mexico
(ed. F. W. Hodge, 1907); Thomas Foster & William Harvey Miner, The Iowa (1911); and Newton Horace
Winchell, The Aborigines of Minnesota
(Minn. Hist. Soc., 1911). Not all of
these are equally explicit on the point of exactly where the Dakota first met
the Ioway—including the first reference, Ramsey 1849/50—but the similarities in
language show that they're all repeating the same essential information. I believe they may all have been based on
Pond's information, including Ramsey who maybe had heard it from Pond
personally before it was published.
[7] – According to G. Malcolm
Lewis' analysis the Ioway map does include the Minnesota River, with a sequence
of village sites moving westward just north of either the Root, Zumbro, or
Cannon River. This seems to be inferior
to the analyses by William Green and Mary Kathryn Whelan, who both say that the
moving villages are between the Upper Iowa and Root rivers (more or less as the
archaeology and history would agree) and that the Minnesota River is not
depicted on the Ioway map (sources: G. Malcolm Lewis, "Indian Maps: Their
Place in the History of Plains Cartography" (Great Plains Quarterly, 1983); William Green "Description of
the 1837 Ioway Map" (in Atlas of
Early Maps of the American Midwest: Part II, 2001); Mary Kathryn Whelan, The 1837 Ioway Indian Map Project
(Master's thesis, U. of Redlands, 2003)).
This is the Lewis version:
This is the Green version:
(Whelan doesn't have any handy
diagram like these.)
[8] – In a footnote to this
passage Lynd writes "Hazen Moors et alia". At first I thought he was saying that the
area of the Ioway planting grounds was known as the "Hazen Moors",
but no: Hazen Mooers [sic] was a person (although there is also a "Mooers
Lake" not far from Redwing). Mooers
had previously worked at the Lower Sioux Agency at Redwood where Lynd was later
murdered, so presumably the two men had met.
[9] –The details of these
migrations aren't my main point here, but the prior locations of the
Potawatomi, Sauk, Fox, Kickapoo, Mascouten, and Miami are discussed in various
publications by Timothy Abel and David Stothers and in the analyses of the
Huron Novvelle France map by Conrad
Heidenreich and John Steckley, among others—I'll probably dig into this in a
later post; the relative positions of the Sauk and Fox are sometimes flipped,
and the locations of the Miami and Potawatomi are particularly hard to pin
down. The ca. 1500 arrival of the
Dhegiha is mentioned by Dale Henning in Archaeology
of the Great Plains), and the westward movements of the Miami and Illiniwek
are mentioned by Robert Mazrim in Protohistory
at the Grand Village of the Kaskaskia (ed. Robert F. Mazrim, 2015) and by
Emerson, Emerson, & Esaray in Palos
Village (ed. Duane Esarey & Kjersti E. Emerson, 2021); these are also
vague as to the prior location of the Miami and even moreso the Illinois. The color key might be off for the
Miami-Illinois migration a bit, as I'm not sure if they came in the 16th or the
17th century. Also, not indicated on the
map, but there may have been something else going on with the Menominee and
with another tribe called the Noquet.
I tend to like using the name
"Illiniwek" rather than "Illinois" because it's easier that
way to distinguish from the state.
Despite looking "Indian-y" this is not the tribe's own endonym,
it's a respelling of their Old Odawa name.
If you know a little Algonquian then you probably think this means
"the people", but appearances are deceiving: the Old Odawa [Ojibwe]
form ilinwe:k is a borrowing of
Miami-Illinois irenwe:wa which is a
verb that means "to speak in the regular way". (In this sense it's an interesting
counterpart to the Algonquian names for Iroquoians and Siouans, which come from
*na:towe:wa "to speak in an
abnormal way".) David Costa
discusses this in his article "Illinois" for the Society for the
Study of the Indigenous Languages of the Americas (2007), which you can find by
googling Michael McCafferty's article "Peoria" on academia.edu. McCafferty's chapter "Illinois Voices"
in Protohistory at the Grand Village of
the Kaskaskia also discusses this, but he differs on a few minor details.
[10] – Robert Mazrim (in chapter
10 of Protohistory at the Grand Village
of the Kaskaskia) comments that some elements of archaeological ware from
Kaskaskia have been proposed to have come from the north, around western
Michigan. No comment is made regarding
the Michigamea, but there might be a connection.
The two sentences of Michigamea
provided by Bossu are "indagé ouai panis" translated "je suis
indigne de vivre, je ne mérite plus de porter le doux nom de père", and
"tikalabé, houé ni gué" translated "nous te croyons, tu as
raison"—both obviously overtranslations.
Koontz interprets these as:
"indagé
ouai panis"
įdaǰe wé bnįs
his-father neg
I-am-neg
"I
am not his father"
"tikalabé,
houé ni gué"
htíkdąbe wé nįge(s)
you-think neg
it-lacks(-neg)
"Your
thinking is not lacking."
For some absurd reason, the
Seymour Feiler translation respells these as "Indagey wai panis" and
"Teekalabay, houay nee gai".
Maybe they were already spelled that way in John Reinhold Forster's old
1771 translation, but that's no excuse.
Walczynski's book Jolliet and Marquette (p 97) cites
Michael McCafferty as saying that "linguistic information" shows the
Michigamea to be Dhegiha. I consider
McCafferty to be trustworthy—and I don't know how much behind-closed-doors
unpublished research has been done on Michigamea since then—but John Koontz at
least was very specific in the 1990s that the language was not Dhegiha.
(Sources: John Koontz,
"Michigamea: A Siouan Language?" (from his website); John Koontz,
"Michigamea As A Siouan Language" (1995, handout for the 25th Siouan
and Caddoan Languages Conference); the Siouan Mailing List, September 2005;
Jean-Bernard Bossu, Travels in the
Interior of North America (1962) pp 73, 109.)
[11] – The usual etymologists
cite the Dakota form Ayúȟba as the
word which the Algonquians borrowed, but it's quite obvious that they actually
borrowed a form more like Ayúȟwa
which currently exists only in the Teton (Lakota) dialect. It seems less likely that the Algonquians
borrowed this name specifically from the plainsgazing Tetons than it is that Ayúȟwa is closer to the original form as
once spoken by all seven divisions of the Sioux, and that this "Old
Sioux" version was what the Algonquians borrowed. The Comparative
Siouan Dictionary doesn't reconstruct any cognate sets for Lakota ȟw ~ Dakota ȟb, but the similar cluster of Lakota sw ~ Dakota sb is found
in words like "comb" and "rattle", where it goes back to
Proto-Dakotan *sw. On the other hand,
David Rood said that he did "not know whether [b] or [w] is older" in
sets of sw ~ sb. (in: Advances in the
study of Siouan languages and linguistics, ed Rudin & Gordon, 2016).
There is a feature common to most
Siouan languages which regularly alternates between final –a and –e on
verbs. I've seen this invoked to say
that the Algonquian forms were actually borrowed from something like Ayúȟwe rather than Ayúȟwa, but I'm not sure if this quite works.
The Miami-Illinois form is from
David Costa, "Miami-Illinois Tribe Names" (31st Algonquian Conf,
2000). The Fox, Shawnee, and Menominee
are from Douglas Parks' synonymy for "Iowa" in the Handbook of North American Indians, Vol
13 (2001). Mii Dash Geget tells me the
details are a bit hard to work out for the Ojibwe name cited by Parks, but the
older Ojibwe form may have been Aayo'oowe.
[12] – The Dorsey cite is: Rev.
J. Owen Dorsey, "On the Comparative Phonology of Four Siouan
Languages" (Annual Report of the
Board of Regents of the Smithsonian Institution, 1885). Here is his explanation for the name Chekiwere:
"Ȼegiha
means, "Belonging to the people of this land," or, "Those
dwelling here," i.e., the
aborigines or home people. When an Omaha
was challenged in the dark, if on his own territory, he usually replied,
"I am a Ȼegiha." So might a
Ponka reply, under similar circumstances, when at home. A Kansas would say, "I am a
Ye-gá-ha," of which the Osage equivalent is, "I am a Ȼe-ʞá-ha." These answer to the Oto "Ʇɔi-wé-re"
and the Iowa "Ʇɔé-ʞi-wé-re." "To speak the home dialect" is
called "Ȼegiha ie" by the Ponkas and Omahas, "Yegaha ie" by
the Kansas, "Ʇɔiwere itcʿe" by the Otos, and "Ʇɔeʞiwere
itcʿe" by the Iowas. When an Indian
was challenged in the dark, if away from home, he must give his tribal name,
saying, "I am an Omaha," "I am a Ponka," etc."
The Wedel cite is: Mildred Mott
Wedel, "A Synonymy of Names for the Ioway Indians" (The Iowa Archaeological Society, Vol 25,
1978). The only statement of hers I can
see which might be interpreted as saying the word "Baxoje" comes from
the Oto dialect and wasn't originally used in Ioway, is this:
"There
is another Siouan name for the Oto, watóta (Whitman 1937:xi) from which
the modern name stems. It was recorded
by Euro-Americans as being used not only by the Oto as a self-designation (e.g.
Bradbury 1811 in Thwaites, ed. 1904-07, 5:80) but also by the Ioway-Missouri
and Dhegiha speakers in early Euro-American contact situations. This has continued in the shortened form,
Oto, to the present day. It parallels,
in a sense, the possibly descriptive designations for the other Chiwerans, as paxóche
for the Ioway and ni-u-t'a-tci (Dorsey NAA 4800:307) for the
Missouri. According to Indian tradition,
the three terms all relate to the episodes of separation of these peoples from
one another after having formerly existed as a single group, all of whom spoke
a proto-"Chiwere" language.
Dorsey was told these were the self-designations used when these people
were away from home."
In fairness, the issue of which names were used by which group to refer to which other group, and how these may
have changed over time, is one which requires very careful and precise
description . . . and neither Dorsey nor Wedel were very precise or
careful. Dorsey elides over the
difference between "which language are you speaking" and "where
are you speaking it"—to be honest he sounds a bit like an alien who only
recently learned the concept of there being multiple words for things. Wedel says that paxóche "parallels" watóta
but isn't clear on how: Betts apparently thinks she's saying that they're both
words used by the Oto, but I think she's saying they are both endonyms. I'll add that if the former, then that
wouldn't very well explain why the French were using it, since they and the
Illiniwek were even further away from the Oto than they were from the
Ioway. I may be misinterpreting
something here.
"Chékiwére" is the
spelling Betts uses in place of Dorsey's "Ʇɔé-ʞi-wé-re". Since Dorsey's upside-down letters appear to
mark tenuis stops, that would be Jegiwere
in the Goodtracks orthography, or Čekiwere
in Americanist notation. I'm ignoring
Dorsey's accent marks because I don't know whether it makes sense analytically
for words in Ioway-Oto to have multiple stresses. The modern Oto self-designation is given by
Goodtracks as both Jiwére and Jíwere so I'm ignoring the accent there
as well (Wedel, using Dorsey's notes, also has the "wrong" stress for
Báxoje. Evidently Chiwere accent is difficult). Wedel's Watóta
is given by Goodtracks as Watóhda and
by Robert Rankin (in the HNAI) as watótta.
[13] – Colin Betts, to be clear,
doesn't say in his own words that the Paouté and the Aiaouez were different
"tribes". He prefers terms
like "locality" and is apparently of the view that both localities housed
people belonging to the same overall Ioway nation.
This might be the case: the evidence isn't exactly conclusive either way. The Aiaouez may even have been closer to the Winnebago for all we know (though I doubt it). Part of this is just semantics: what exactly one wishes to use the word "tribe" to refer to. To what extent the Aiaouez thought of themselves as distinct from the Paouté, you need a time machine to find out.
[14] – The Walczynski endnote is
on page 255. David Costa refers to the
Intuka in the article "Illinois"—part of the superarticle Three American Placenames—in the Society
for the Study of the Indigenous Languages of the Americas Newsletter XXV:4,
January 2007. More about Miami-Illinois
nomenclature is in Michael McCafferty's article in Protohistory at the Grand Village of the Kaskaskia. The macron-breves are because we don't know
whether the first two vowels are long or short.
Both Inohka and Indokah/Intuka are unanalyzable in their respective
languages. I, not being a Siouanist, am
trepid about declaring that thoka has
no non-Dakotan cognates or that no Dakota name for the Illiniwek is
attested—but the Comparative Siouan
Dictionary, at least, has no entry for thoka,
with one possible cognate in Dhegiha (as part of a compound word meaning
"antelope, goat") having a note saying it might be a Dakotan
loanword; and neither the Handbook of
North American Indians nor the New
Lakota Dictionary list any word for the Illinois or its component tribes.
One can imagine a scenario where
James Lynd was told a name like "Indokah" or "Intuka" and
construed it himself as "Unkthóka". This at least would explain why it looks so
grammatically unusual.
The uŋ(k)- prefix in L/Dakota and Assiniboine actually has a front
vowel in Stoney: į(g)- in Morley
dialect and įgi- in Alexis
dialect. And in Chiwere-Winnebago the
cognate prefix is hį-. So one might be tempted to say that in
Proto-Dakotan the word for "our enemy" could have been *iŋ(k)thóka. That looks like something that could easily
have been borrowed to or from the Dhegiha forms, regardless of what those words
meant in Dhegiha ("our" in Dhegiha tends to have an ą or ǫ vowel). If intuka/indokah was borrowed into Sioux then
it may have been they, rather than Lynd, who reinterpreted it as a possessed
noun (if it even is one at all).
Unfortunately, the cognates
outside of Mississippi Valley Siouan do make it look like uŋ(k)- is the older form, and it's generally good practice to assume
fewer conservatisms in Stoney anyway. If
thóka is in fact a borrowing then it
must have happened during the Proto-Dakotan period since the word exists in
Stoney and Assiniboine, and so at that time it likely wouldn't have referred to
the Illiniwek who probably weren't in contact with the Sioux until at least the
16th century. It might have referred to
another Algonquian tribe, per McCafferty's idea that it referred to the Great
Lakes Algonquians in general? You can
see that while "intuka" ~ "unktoka" looks
impressionistically promising, it is very hard to make any of the details
actually work for this theory.
(Morley source: Elias
Abdollahnejad, "Verb Conjugation in Stoney Nakoda: Focus on Argument-Marking
Affixes" (2018); Alexis source: Corrie Lee Rhyasen Erdman, Stress in Stoney (Master's thesis, U. of
Calgary, 1997))
No comments:
Post a Comment