Some South Dakota Prehistory

In the previous post, I briefly summarized part of the prehistory of the Pawnee, Arikara, and Mandan.  I was already digressing away from the main topic, so I tried to make what I wrote quick and simple—there were details that I chose not to get into for reasons of space—but I'm afraid it just ended up being rushed and unclear.  Also I made a few small errors, of omission and of commission.  So I'm gonna try to cover part of that again, and try to be clearer and more detailed (and accurate) this time around.  As always, remember that I am especially liable to get things wrong when talking about archeology.

The Pawnee and Arikara tribes are closely akin, and seven centuries ago were still one people.  Relatively, at least—the archaeological record perhaps doesn't tell us how unified they were socially or politically, but they presumably still spoke one language.  They lived in sedentary village communities in southern Nebraska and northern Kansas.  As before, I'm using the term "Panaic" to refer to this ancestral population, for the sake of convenience.  The Panaic people are represented by the "Central Plains Village" archaeological tradition.

Meanwhile, South Dakota was inhabited by another group of village-dwelling people, unrelated to the Panaics, represented by the archaeological "Middle Missouri tradition".  There are two areas to the Middle Missouri tradition: the core area along the Missouri River in central South Dakota, roughly between the mouths of the Bad and Cheyenne rivers (henceforth: "the Pierre area"); and the so-called "Eastern Periphery" extending thence across southeastern South Dakota and into northwestern Iowa.

In the previous post, I made two errors regarding the Middle Missouri tradition: 1) I limited it to the Pierre area, not including the Eastern Periphery, and 2) I labeled it "proto-Mandan".  Error #2 might not strictly be wrong per se, but the evidence is ambiguous: the Middle Missouri tradition certainly included the proto-Mandan, yes... but it likely also included the proto-Crow-Hidatsa as well, and possibly other tribes as well.  Specifically who was in the Pierre area, and who in the Eastern Periphery, is not known.  The only hard evidence we have are archaeological remains, and it just so happens that the Mandan and Hidatsa have always been extremely similar in material culture, so it's next to impossible to tell where they were located relative to each other in prehistoric times.

"Mih-Tutta-Hangkusch: A Mandan Village" by Karl Bodmer, 1841.

The Crow, Hidatsa, and Mandan constitute a dialect group known as "Missouri River Siouan".  However, because the Crow and Hidatsa did not become separate from each other until probably the 1700's, and because they and the Mandan are nigh-indistinguishable in the archaeological record, I'm going to employ another term-of-convenience to refer to their collective forebears: "Minnetaric".  Remember: "Panaic" and "Minnetaric" are both names I have invented for convenience's sake—they are not terms used by scholars, or anyone else.  In the 13th century, Minnetarics almost certainly occupied the Pierre area.  The Eastern Periphery was probably Minnetaric as well—perhaps proto-Crow-Hidatsa, perhaps other Minnetaric tribes that got lost to history.

By the middle of the 14th century, the Panaics had migrated from the central Plains and established themselves along the Missouri River, downstream from the Pierre area.  These South Dakota Panaic settlements belong to the archaeological "Coalescent tradition".

In my last post, I surmised that this migration was what split the Panaics into the Arikara and Pawnee:

"The 1350-1400 period is reasonably close to the estimated date of separation for the Pawnee and Arikara languages (Parks 1979), so it seems plausible to me that this was when the "Panaic" tribe(s) fissioned: the ancestors of the Arikara emigrating to South Dakota, and the ancestors of the Pawnee presumably going somewhere else."

This is still possible, I suppose, but most archaeologists prefer to say that a later Panaic migration (in 1550) represents the Pawnee-Arikara split (Logan 1996).  Parks' actual estimate for the Pawnee-Arikara schism was around 1450 A.D., give-or-take—or in other words, right between the 1350 migration and the 1550 migration.  So the linguistic evidence is ambiguous as to when the schism occurred.  However, I have no reason to doubt the archaeologists on this, so let it be said: the early Coalescent tradition in South Dakota represents the Panaic people, not just the proto-Arikara.

South Dakota was already something of a war zone when the Panaics decided to move in—settlements along the Missouri River tended to be much more defensive in nature, compared with the Central Plains villages that the Panaics were accustomed to.  These Missouri River settlements were typically surrounded by a ring (sometimes two) of defensive earthworks, topped with palisade walls, and positioned at strategic locations atop cliffs or on hillsides.  For a century prior to the Panaics' arrival, peoples from the Initial Middle Missouri (IMM) and Extended Middle Missouri (EMM) phases had been fighting each other for possession of the Pierre area (Wood 2001, 192).

Aerial photograph from 1965 of the ruins of a fortified Middle Missouri settlement.  After four centuries of neglect, and years of getting literally plowed over, the walls and "mural tower" sections are still clearly visible in outline. (from Wood 2001)

After the Panaics arrived on the scene, things just got worse, and a war broke out against the Minnetarics over control of South Dakota (one can only assume that the IMM and EMM forgot their quarrel in the face of a common enemy).  This war is famous among archaeologists because of an archaeological site known as Crow Creek—a spectacular testament to human violence.  At some point in the early 1300's, a Minnetaric war party surrounded a Panaic settlement and—after apparently laying siege to it for some time—attacked, killing nearly 500 of its inhabitants:

"This truly was a massacre rather than a battle; most villagers appear to have been clubbed to death while fleeing.  There is not an embedded arrow point in any of the bodies.  Men, women, and children were indiscriminately killed.  Their noses, hands, and feet were sometimes cut off, teeth smashed, and heads and limbs cut from the body.  The victims, from babies to elders, were universally scalped and mutilated.  The scale of the deaths suggests that most of the inhabitants were killed." (Emerson 2007)

Crow Creek might be the bloodiest massacre committed by a non-state civilization known to all of history (Pinker 2011, 49).  Clearly, then, the Minnetarics didn't take this Panaic invasion lying down.  But, however many teeth they smashed and noses they cut off, it evidently wasn't enough, and by mid-century they had abandoned the entire Eastern Periphery and withdrawn to the relatively-constricted Pierre area.  The Iowan portion of their former territory was taken over by the Chiwere (a group that includes the Ioway, Oto, and Missouria tribes... though I don't think the Missouria were involved here); whether there was a theater of war here like that in South Dakota, I honestly have no idea.

Wherever the proto-Mandans had been before, they were certainly in the Pierre area by ≈1350.  If there had ever been other Minnetaric tribes (apart from the proto-Crow-Hidatsa), they were either destroyed or had their tribal identity subsumed under someone else.  As for the proto-Crow-Hidatsa, they're conventionally supposed to have lived in eastern North Dakota prior to ever encountering the Mandans, but it's not impossible that some may have been in South Dakota with the Mandans at this point.

This, by the way, is why I've been reluctant to use the word "Mandan" when referring to the Middle Missouri tradition.  I don't want to say that, e.g. "the Mandans lived in the Pierre-area" in the 14th century, because maybe they didn't—maybe the Mandans were those people living in the Eastern Periphery, and some other tribe lived in the Pierre-area, were swamped by the Mandan refugees, and lost their tribal identity.  Or maybe it was the other way around.  Or maybe they were both Mandans.  Maybe there's something to the oral tradition that says the Hidatsa Awaxawi band was originally south of the Mandans (Wood 1993).  Or maybe there isn't.  Probably no one knows.

What is known is that: whoever was present in the Pierre-area in 1350, spent the next two centuries or so being gradually driven northward by the expanding power of the Panaics.  By approximately the 16th century or so, the Panaic advance had driven the Mandans and Hidatsas upriver, into central North Dakota which they inhabited in the historical period.  Meanwhile, the Panaics had expanded southward as well, until their northern and southern borders corresponded more or less with the present north and south borders of South Dakota.

Sometime around 1550 A.D., or 200 years after first moving to the Missouri River, a segment of the Panaics migrated back south into Nebraska and settled along the forks of the Loup River.  Whether they had hung on to fond memories of "the old home" for two centuries, or were motivated by other concerns, I don't know.  As mentioned earlier, this was probably the migration that split the Panaics into two tribes—with those remaining in South Dakota becoming the Arikara, and those living in Nebraska becoming the Pawnee.  The Pawnee later split into two geographical divisions: Skiri and South Band.  It was once thought that this split predated the Arikara schism, and that the latter tribe originated as a sub-band within the Skiri—that is now known not to be the case.

There were probably also developments within the "Chiwere" bloc shown on the map, but I haven't yet looked into that.  For now, just take that portion of the "1550-1600 AD" map with skepticism.

Cited sources:

Thomas E. Emerson, "Cahokia and the Evidence for Late Pre-Colombian War in the North American Midcontinent". In Chacon & Mendoza eds., North American Indigenous Warfare and Ritual Violence (2007).

Brad Logan, "The Protohistoric Period on the Great Plains". In Jack L. Hofman ed., Archeology and Paleoecology of the Central Great Plains (1996).

Douglas R. Parks, "The Northern Caddoan Languages: Their Subgroupings and Time Depths" (1979).

Steven Pinker, The Better Angels of Our Nature: The Decline of Violence in History and its Causes (2011).

W. Raymond Wood, "Hidatsa Origins and Relationships" (1993).

— "Plains Village Tradition: Middle Missouri". In Raymond J. DeMallie ed., Handbook of North American Indians, Vol 13: Plains, Pt 1 (2001).

Uncited sources:

Duane Anderson, "Ioway Ethnohistory: A Review, Part I" (1973).

Dale R. Henning, "The Oneota Tradition". In W. Raymond Wood ed., Archaeology on the Great Plains (1998).

— "Plains Village Tradition: Eastern Periphery and Oneota Tradition". In Raymond J. DeMallie ed., Handbook of North American Indians, Vol 13: Plains, Pt 1 (2001).

Richard A. Krause, "Plains Village Tradition: Coalescent". In Raymond J. DeMallie ed., Handbook of North American Indians, Vol 13: Plains, Pt 1 (2001).

Douglas R. Parks & Waldo R. Wedel, "Pawnee Geography: Historical and Sacred" (1985).

W. Raymond Wood & Alan S. Downer, "Notes on the Crow-Hidatsa Schism" (1977).