A multicoloured mosaic of corncobs adorns the 19th-century ‘Corn Palace’ of Mitchell in South Dakota, celebrating the importance of this crop to the economy. But how did these plants, whose wild forms prefer the hotter, drier zones that are found in the highlands of Mexico, come to dominate the Northern Plains?
The answer is through the expansion of farming peoples up the river systems of North America – the Mississippi, the James, and the Missouri – over many centuries, accompanied by gradual selection of the most appropriate strains of crops to survive the changing environments and climates.
Though farming reached this region comparatively late, in about AD 1000, today the area is a major agricultural producer of commodities such as maize and beans. The people who lived at Mitchell Prehistoric Indian Village 1,000 years ago were among the first to grow maize here. Their descendants, tribes such as the Mandan, Arikara, and Hidatsa, still lived in similar style villages, and were using similar farming techniques along the banks of the Missouri when the first European settlers arrived.
Those early pioneers brought plants familiar to them, such as wheat, from home. But they swiftly adopted and showed an early preference for the tried-and-tested crops grown by the native farmers of the region, particularly maize. Indeed, the early seed catalogues of the late 19th century often cite their specific native source. Thus Mitchell tells the holistic story of the rise of agriculture, and corn in particular, in one of the major breadbaskets of the Americas.
A changing landscape
Mitchell Prehistoric Indian Village is located on the banks of Lake Mitchell in south-east South Dakota. The lake, however, is a modern reservoir. The site once stood on a bluff overlooking Firesteel Creek, a tributary of the James River. It belongs to the Initial Middle Missouri Culture and was, in its day, just one of many similar village settlements along the local river systems where the inhabitants made use of the flood plains to plant their gardens.
The village site is a National Register and National Landmark site, but what makes it special?
Many similar village sites have been badly disturbed by later farming activities, and eroded by ploughing. Many others along the Missouri have been lost to damming projects or suffered water erosion. The Mitchell site was lucky, however, because when the city was first laid out in the late 19th century, that particular area was included as city property, so it did not become farmland; on the other hand, it was on the very outskirts of town, and thus it was not built on either. When the significance of the site was recognised, it was turned over to a local citizens’ board to manage.
The undisturbed archaeology of this 2.5ha site begins at the grass roots, and continues in the oldest parts down through as much as 3m of cultural deposits. The village dates to around AD 1000, and was occupied for about 100-150 years before being abandoned. This was the typical pattern of settlement, with societies moving to fresh pastures after about a century as local resources, particularly the scarce timber, became exhausted.
Prehistoric village life
Previously, most work on plains village sites tended to concentrate on the houses. Today, we are working in an open area of excavation between two lodge depressions, investigating the deepest, and thus the oldest, parts of the site, sheltered beneath a permanent structure (see box on p.24) that allows us year-round access. One of the benefits of the open-area approach has been our ability to understand how activities were zoned within the inter-lodge spaces.
It is clear that outdoor spaces were highly organised, and particular activities repeated time and time again in the same places. In one area, we have found the remains of a roasting pit where hot rocks would have been used to slow-cook food. The clay-lined feature is full of fire-cracked rock and large animal bones. Stratigraphy in the clay lining, however, shows that the oven has been re-lined many times and repeatedly re-used.
Another zone of the site is full of cache pits that would have been used to store crops. Most of these cache pits were later re-used for the disposal of trash, and many were cut into by later pits dug into the same zone. Sometimes clean clay was used to line the new pits, perhaps to shield the contents from the trash material the new pit was cut though. Another area of the site appears to be a feature for rendering bone grease.
Bone fats were a very important resource to many prehistoric peoples. The fat was used as a high-calorie food, sometimes mixed with ground jerky or berries to make ‘pemmican’, which helped people through the long winters. However, grease was also used for lighting, fuel, waterproofing, as a base for paint, and for many other craft purposes.
Bone-grease is made by smashing up the spongy ends of long bones and other spongy bones such as vertebrae, and boiling the fragments to render out the fat, which floats to the top of the water, where it can be skimmed off. Before metal cauldrons, one way of doing this was to use large cooking pots, but if done on a larger scale, it could be made in a pit, perhaps lined with hide, with hot rocks added to heat the water.
We have found what we believe is just such a feature: a clay-lined shallow pit full of pulverised spongy bone and fire-cracked rock. Detailed analysis of fracture and fragmentation patterns and bone-type selection strongly suggest that this must be the remains of an in situ bone-grease rendering station, abandoned after the fat had been removed.
In general, it appears that the people at the village were producing bone-grease on a fairly large scale. In this particular case, given that food supplies seem to be abundant, this activity is most likely not related to dietary desperation, but the need for winter stores. It is also very possible that fat or pemmican was used by these people as a trade commodity in exchanges with people in the American bottom lands, where access to bison products was much more restricted.
We see many items of material culture that are strongly influenced by the Mississippian Culture of Cahokia, and it is clear that at least some items of material culture have been traded in, either directly or indirectly. Some fragments of seashell have clearly been traded all the way up from the Gulf of Mexico, and small amounts of copper have been recovered that were brought from the Great Lakes region. The people of Mitchell Village may well have been supplying furs, hides, fat, and jerky in return for such materials, though the full picture of trade is hard to understand as such organic items rarely survive.
Our latest season’s work was highly successful, and added to our understanding of the site in a number of different ways. First, we had an excellent year for finding archaeobotanical remains: indeed, we probably recovered more such evidence this year than in the previous 11 years combined.
We excavated an ashy feature that may well have been a hearth with ovens next to it. The ash was very rich in carbonised plant material, which we extracted through flotation sieving. The hearth ash contained many carbonised corn kernels and some sunflower seeds, while some of the possible oven areas to the side were much richer in charred corncobs.
From studying the cobs we can establish the nature of the crop at the time, using criteria such as how many rows of kernels the cobs had around their circumference. The kernels themselves are similar in size to modern-day examples, but the cobs are clearly much smaller and narrower.
We also learned much more about cache-pit structure. One cache pit we excavated belled out impressively from a fairly narrow neck. At the bottom, where it was quite damp, we found a thick mat of semi-decomposed organic material. Though this was in an advanced state of decay, it was possible to see that the material consisted of bundles of grass. There is sufficient structure left that we stand a good chance of finding out much more about the contents of this material.
Early 20th-century accounts of surviving Hidatsa village horticulture tell us about the structure of storage cache-pits. Gilbert Wilson’s 1917 account of Buffalo Bird Woman’s garden tells us that cache pits were lined on the base and sides with tied bundles of grasses held in place with willow sticks. So our find at Mitchell Village may well be evidence that this practice dates back to the earliest villages in the region.
Thus we are learning a great deal about the nature of agriculture at this site, and have uncovered considerable evidence for the very wide range of wild food exploited by the inhabitants of the village. As well as farming maize, beans, squash, sunflower, and amaranth, vast numbers of bison were also hunted. Bison probably formed the majority of the animal protein consumed, there was also ample exploitation of deer, small mammals, freshwater shellfish, fish, and turtles.
Material culture
This year’s material culture finds were also interesting. We recovered two miniature artefacts, including a tiny arrowhead and a tiny, but perfectly formed, pot. The pot is in the standard form of a cooking vessel, but is less than 4cm across. Is it a child’s toy, a practice piece, or did it hold some substance kept in small quantities, such as paint?
We also found many examples of agricultural implements, such as bison-scapula hoes and a portion of a scapula sickle. Another fascinating item was a mother-of-pearl fishing lure, which would have been used alongside bone fishhooks that look every bit like modern ones but for the lack of a barb.
Items showing signs of Mississippian influence were once again in evidence, including Cahokia-style stone points. We need to carry out further analysis, particularly on the ceramics, to establish the extent to which the Mississippian-type artefacts are local reproductions or actual imports – probably they are a little of both.
Discover Mitchell Village
As well as being beautifully preserved, there are several features that make Mitchell Prehistoric Indian Village a very special archaeological resource. One is an interpretative museum that has been established alongside the site. The other is the Archeodome.
This specially designed structure was erected over a portion of the site. It was designed to have minimal impact on the archaeological deposits, while providing a protective cover for a large open area where excavation is continuing. Digging with air conditioning is very pleasant, but there are more serious reasons for this special construction: the winters in South Dakota are very harsh, and the summers see some heavy storms. Without the Archeodome, outdoor excavation would be limited in area, and opened and closed seasonally. The dome is also equipped with finds-processing laboratories and curation areas. Augustana University and the University of Exeter have run joint field schools in the facility for the last 12 years. The dome also gives members of the public an excellent view of ongoing excavations from a ramped gantry that leads up and around the excavation floor. This path leads to the laboratories, where visitors are encouraged to talk to the archaeologists processing artefacts. As such, the Mitchell Prehistoric Indian Village represents a unique scientific resource and centre for presenting archaeology.
The Mitchell Prehistoric Indian Village is regularly open for visits from April to October, and during the winter months groups can book to see the site by appointment. Please visit the website mitchellindianvillage.com for details.
All images: Alan Outram, exeter.ac.uk/archaeology/, unless otherwise stated.
Source: Alan K Outram, University of Exeter, UK; L Adrien Hannus, Augustana University, Sioux Falls, USA.
