Oneota sites are widely distributed throughout the deciduous forests and prairies of southern Minnesota (Figure 18.1). Regional variations of Oneota pottery, called the Ogechie series, are also found at sites in the northeastern prairie zone and in the Northwoods (Ready 1979d). Radiocarbon determinations and the presence of historic trade goods in some sites indicate that this archaeological tradition appeared about AD 1050 - 1100, and probably earlier in eastern Wisconsin, and persisted into the early historic period (Table 18.1; Dobbs 1982, 1984a; Overstreet 1997).
The origins and history of the Oneota tradition in Minnesota remain clouded for a variety of reasons, including disagreements among archaeologists over the correct interpretation of radiocarbon dates, possible mixing of components at some sites, the persistence of earlier migration models, and a general similarity in material culture that has blurred the internal developmental dynamics of the cultural tradition.
The Oneota archaeological record in Minnesota, however, is not entirely obscure. In this chapter, we summarize the material culture and settlement-subsistence patterns of the Blue Earth phase of that tradition, the only widespread Oneota phase in pre-historic Minnesota (sites of the related Orr phase appear in southeastern Minnesota during the proto-historic period).1 Interpretations of the origins and developmental dynamics of the cultural tradition are reviewed in the second half of this chapter, and the nature of the Oneota presence in the northern part of the state is discussed in Chapter 21. Historic Oneota components associated with the Orr phase are described in Chapter 22.
As with other ceramic-containing archaeological complexes in Minnesota, the presence of distinctive ceramics is the trait most frequently used to identify a Blue Earth Oneota component at a site. Blue Earth ceramics are described in the following section. Other, less common items used to identify Oneota components include the occasional presence of catlinite disc pipes, paired sandstone abraders, fresh-water mussel shell fish lures or decoys, bison and elk scapula hoes, and copper beads and ornaments (e.g., stylized serpents, maces, and raptorial birds) (Overstreet 1997:251). Blue Earth settlements are typically open (unpalisaded) village farming communities that contain the remains of corn, beans, and squash, and numerous circular subsurface storage pits, except for some pre-A.D. 1250 settlements (e.g., Bartron) that may be palisaded. Evidence that hunting and gathering were important subsistence pursuits is present in these settlements and at special activity camps. In Minnesota, Blue Earth components are present south of the Minnesota River, except for a few sites in the St. Croix Valley.
Although Oneota components of all phases tend to share many of the traits mentioned above, they are variable in other traits. Examples are house form, ceramic attributes, and site location.2
Artifacts and Features. The most distinctive and diagnostic trait of Blue Earth assemblages, when compared to other Mississippian complexes in the state, is the ceramic repertoire. In fact, the identification of a component as Oneota anywhere nearly always depends on the presence of Oneota potsherds, for, as Dale Henning (1983b:4.68) succinctly expresses it, “our definitions do not fully comprehend similarities and differences in other technological endeavors by these [Cambria, Mill Creek, etc.] late pre-historic occupants of the Midwest.”
Vessels are shell-tempered, round bottomed, globular jars with rounded lips and generally high straight to slightly out-curving rims that are slightly everted (Dobbs 1984a Gibbon 1973, 1979, 1983). Vessel capacities in a sample of jars from the Center Creek locality range from 0.5 to 5 gallons. Short, double strap handles and loop handles are common. Vessel surfaces are smooth, except for occasional patches of smoothed-over cordmarking. Nearly all vessels are decorated on the shoulder and, in decreasing order of occurrence, on the rim interior, rim exterior, and lip with trailed lines, tool impressions, punctuates, or a combination of these decorative techniques. Decorative designs are usually rectilinear, with concentric circles the only curvilinear motif; patterns of decoration incorporating chevrons of one kind or another are most common. Since ceramic assemblages in southern Minnesota differ from one another in motif popularity and, perhaps, jar shape and size, no one assemblage can be considered typical of the phase. However, most assemblages appear to be spatial and temporal variations within one phase.
The non-ceramic material culture of Blue Earth sites in Minnesota is similar to that of other Mississippian complexes in the state. Table 18.2 presents lists of artifacts from the Bartron site (21GD2) and from Vosburg (21FA2), a Center Creek locality site, for comparison. Chipped stone tools at village sites consist of simple unnotched triangular projectile (Madison) points, end and side scrapers, knives, drills, wedges, gravers, choppers, blade tools, and expedient flake tools (Figure 18.4).
A study of raw materials at sites in the Center Creek locality found that four basic types of stone were used: oolitic chert, fine gray chert (Rapid/Grand Meadow), white chert, and quartzite (Dobbs 1984a:77-78). Obsidian and Knife River flint were absent. Approximately half of all chipped stone in samples in the study was oolitic chert from nearby gravel deposits or from bedrock quarries along the Blue Earth River, 30 miles to the north. Fine gray chert probably quarried from outcrops near Grand Meadow, 80 miles east of the locality, ranged in popularity in samples from 15 to 40 percent, and white chert and quartzite in ranges less than 15 percent. Most end scrapers and flake knives are made of gray chert, and most large tools, such as choppers, of quartzite. Oolitic chert was generally used in making projectile points, and most expedient flake tools are made of oolitic or gray chert.
Sandstone abraders, hammerstones, anvils, celts, manos, and grinding stones are found, too, at village sites, as well as bison scapula hoes, picks of antler and bone, bird and mammal bone awls, split beaver incisors, punches, tubes, and occasional harpoons and ornaments. Shell tools and ornaments are rare, but a few shell ‘fish’ lures and notched mussel shells have been found during excavation or are in local collections. Copper and catlinite are rare in most sites, although the amount of catlinite, especially geometrical lumps, increases in more recent components (Figure 18.5)
Detailed descriptions of Blue Earth artifact assemblages from a variety of site types in Minnesota can be found in Dobbs (1984a) and Gibbon (1973, 1979, 1983).
Oneota village sites are present in the Red Wing locality (Bartron: 21GD2), along the St. Croix River north of Stillwater (Sheffield: 21WA3), in several localities along the Blue Earth River (Center Creek and Willow Creek), and along the Upper Minnesota River (Fort Ridgely: 21NL8). Small amounts of Oneota pottery are also present in the upper levels of many sites from the St. Croix and Mississippi river valleys in the east to the South Dakota border in the west (Figure 18.1). A cluster of early historic (Orr phase) sites in the Root River drainage in western Houston County is discussed in Chapter 22. The map on the right shows the broader distribution of Blue Earth-like Oneota pottery in the western Great Lakes (Anfinson 1987).
Bartron is a 7 to 10 acre, possibly palisaded, village site on a low island in the Mississippi River floodplain (Gibbon 1979). Radiocarbon determinations indicate the village may have been occupied between A.D. 1050 and 1200, although some archaeologists believe the village dates to ca. A.D. 1250 (Table 18.1). Features and artifact density and distribution suggest that the village was occupied for less than 50 years. Storage/refuse pits are numerous, and houses are represented by nearly square to rectangular posthole outlines surrounding black, ‘greasy,’ depressed, trampled floors (Figure 18.6). The two most extensively excavated structures measured 29.5 by 31 feet and 13 by 18 feet. Animal remains from the site, which is essentially single component, are listed in Table 18.3. Maize kernels and one grain of wild rice were found during dry screening. A small number of Middle Mississippian traits are present at Bartron and will be discussed below.
Bartron is not the only Oneota village site in this region, for a series of villages on the Wisconsin side of the river extends from just north of Red Wing to the outlet of Lake Pepin in the south (Figure 18.7). This series includes the Double (47Pi81), Adams (47Pi12), and Armstrong (47Pe12) sites (Penman 1980, 1984a, 1984b; Hurley 1978; Gibbon and Dobbs 1991). Maize has been found at all of the excavated Lake Pepin villages, which have MASCA corrected radiocarbon dates ranging from A.D. 1010 to 1440 (Dobbs 1982; Crawford and King 1978; Zalucha 1985; Penman 1988). Preliminary analyses of the ceramics from Adams suggest that the sample from this site is more similar to that from Armstrong than either is to Bartron. Whether these samples represent comparable analytical units and just what relationships existed between components at these sites remain to be determined.
Three Sites. Although archaeologists tend to associate the term “Oneota” with large horticultural village sites, Oneota components range from a handful of sherds in a site, to small and medium size camps, to large villages, as we saw in the preceding section. We highlight a small horticultural village (Bartron), a large horticultural village (Vosburg), and a small hunting and fishing base camp (Sheffield) in this section.
Bartron (21GD2) is a small village at the southern end of Prairie Island in the Red Wing locality (Figure 18.1). Radiocarbon dating suggests that it was initially occupied by ca. AD 1050, slightly before Silvernale phase sites like Bryan, Silvernale, and Mero, though this date is controversial (see Chapter 16). Its occupation coincided at least in part with the Silvernale phase, for materials of that phase were found in a structure at the site. Bartron has a material culture and subsistence remains similar to that found at other Blue Earth village sites, such as numerous ceramics and subsurface features, and evidence of maize horticulture. Parts of two houses were found and the possible remains of a palisade. Unlike large Silvernale villages, earthen burial mounds did not surround Bartron.3
Vosburg (21FA2) is a large horticultural village in the Center Creek locality. The site was excavated by Lloyd Wilford (19xx) in 19xx and by Clark Dobbs (1984a) in 1976. Typical of Oneota village sites, numerous storage/refuse pits were within the village, some of which contained three or more bison scapula hoes at the bottom. Maize, some of which is Northern Flint, common beans, sunflower, hazel, wild plum, and hawthorn were found during dry screening. A cemetery was located across the Blue Earth River from the village and some human bones were in pits in the village.
Sheffield (21WA3) is a small summer hunting and fishing base camp on the flood plain of the St. Croix River. Unlike horticultural villages like Vosburg, Sheffield has few storage pits, an absence of large scapula hoes, a low frequency of end scrapers, few fabricating and processing artifacts, and a larger percent than usual of hunting and fishing implements. Agriculture was either minor or absent as an activity. The marked homogeneity of the ceramic assemblage suggests that the base camp was occupied for a relatively brief period, perhaps for a decade or two. The component has an associated radiocarbon date of A.D. 1300 (uncal) (Wilford 1961d; Gibbon 1973).
Oneota Origins in Minnesota. The abrupt appearance, entirely new ceramic assemblage, and new subsistence-settlement pattern (e.g., large villages, a horticultural lifeway) of Oneota sites in the Upper Mississippi River valley have puzzled archaeologists for many years. In general, explanations for the abrupt appearance of Oneota sites in the Upper Mississippi River valley have favored either a Middle Mississippian stimulus after ca. A.D. 1050 or a slightly earlier (A.D. 950 - 1050), more general stimulus that resulted in the appearance of tribal lifeways and maize horticulture in the hinterlands and the Middle Mississippian emergence itself along the Mississippi River to the south. Today, a few archaeologists believe they have solved the puzzle. However, the very diversity of these models and their general level of abstractness indicate that the origin of Oneota sites in the Upper Mississippi River valley remains unresolved.
In the early 1960s, James B. Griffin proposed that the Oneota were descended from one or more Middle Mississippian communities that had moved northward into the Upper Mississippi River valley. In adjusting to the shorter growing season of a northern deciduous forest and perhaps because of deteriorating climatic conditions (the on-set of the cooler, drier Pacific climatic episode), their heirs became the socially less complex Oneota by ca. A.D. 1300 (Griffin 1946, 1960b, 1961; Hall 1962). The mixture and blend of Middle Mississippian and Oneota traits at Silvernale phase sites could be interpreted as evidence supporting this sequence.
Thomas E. Emerson (1988) has argued that an elite community from Stirling phase Cahokia occupied the Red Wing locality by A.D. 1000-1050.4 This was just one of a number of communities that fled northward to remove themselves from conflicts among rival chiefdoms at the time. Emerson points to the presence of “wall-trench architecture, platform mounds, and traditional Lohmann and Stirling phase ceramics” as well as the river valley location of the settlements as evidence for a site-unit intrusion into the Red Wing locality. “Its lack of visibility in the archaeological record,” he argues, “is partly due to the utilization of the same sites by later Mississippian peoples that has often badly mixed the components and partly to the lack of familiarity of many researchers with traditional Middle Mississippian utilitarian ceramics.” According to Emerson, then, pure Middle Mississippian components should be present in the area and other Mississippian components, that is, Oneota, are later.
James B. Stoltman (1983, 1986b) has supported a somewhat similar view, though he would place the site-unit intrusion in the A.D. 1050 - 1200 interval. He also suggests that Oneota culture in the area emerged from a Late Woodland base as a consequence of interaction with these Middle Mississippian intruders, that “the first fully Oneota assemblages post-date A.D. 1075 or 1100,” and that Oneota did not rise “to prominence over most of the Upper Mississippi Valley region” until the waning of Cahokia influence in the region by about A.D. 1200 (Stoltman 1986b:31-33).
A more recent variant of this class of models is Boszhardt’s (2004) suggestion that the presence of contagious diseases in the American Bottom and transported from there into the Upper Mississippi River valley was not only responsible for the movement of Middle Mississippian peoples up the valley, but for the emergence of Oneota culture after ca. A.D. 1200. According to Boszhardt (2004:77), “The ultimate transition to the Oneota Culture may have … been in response to Middle Mississippian transmitted diseases.” In his model, Silvernale is a transitional phase between regionally resident Late Woodland Lewis phase peoples and emergent Oneota. Again, the mixture and blend of Middle Mississippian and Oneota traits at Silvernale phase sites lends some support to this model.5
An alternative interpretation argues that there has been a failure to distinguish between two widespread and equally important transformations that occurred in the northern tier of the eastern United States and in adjacent portions of the Plains and Canada during the A.D. 950 - 1050 interval. The first involved the emergence of what might loosely be called tribal lifeways (see Chapter 2), and the second a process of ‘mississippianization.’ The early presence of the Oneota tradition in eastern Wisconsin (by A.D. 950), Great Oasis in southwestern Minnesota and northwestern Iowa (by A.D. 950), Fort Ancient in Ohio (by A.D. xx), and the Cahokia cultural system itself (by A.D. 1050), for example, are regarded in this interpretation as products of the first transformation. The second transformation, of which the Cahokia cultural system was a florescence, was centered within the Upper Mississippi River valley and can be identified by the appearance of Powell Plain and Ramey Incised ceramics, platform mounds, wall trench architecture, and other classic (Lohmann and Stirling phase) Middle Mississippian traits. Each of these transformations requires a separate if somewhat intertwined explanation.6
This does not mean that Mississippian influences of one kind or another were not important in the first transformation. What it does suggest, however, is that these influences may have their origins further south in areas such as the Central Mississippi Valley. Interaction with these areas, apparently, was also important in the emergence of classic Mississippian cultures in the American Bottom. As we have seen in earlier chapters, trends toward greater reliance on produced foods, greater population sizes, and sedentism have a long history in the Midwest. This second interpretation suggests, then, that the earlier transformation was in part a culmination of indigenous trends and in part a product of interaction with and migration from more complex societies to the south. Although the Middle Mississippian stimulus period (A.D. 1050—1200) is most often considered a period of extraordinary social change in the Eastern Woodlands, it can be argued that the preceding, largely neglected A.D. 800 – 1050 period was equally dynamic but in a more subtle way.
Regardless of the correct answers to the above questions, the major events in southern Minnesota between A.D. 1200 and the historic period are the fading of Middle Mississippian influences after A.D. 1200, Oneota expansion after A.D. 1200, and the attendant readjustments that accompanied these events. The suggested causes of Oneota expansion are many and varied. Penman (1988), for example, attributes their southward movement from Lake Pepin to the La Crosse area to the Neo-Boreal climatic deterioration, while Emerson (1988) explains their expansion in the thirteenth century and fourteenth centuries as the result of the availability of northern adapted maize in an otherwise virtually identical agricultural system.
According to Benn (1988), the driving force of Oneota expansion was the structure of their production system. In his model, the Oneota possessed an inherently expansive “tribal” mode of socio-economic structure. Each tribe was a political and territorial entity with a hierarchical network of leaders, sodalities, lineages, and totems. “As the community membership grew and leadership positions multiplied and conflicted, segments of the community inevitably fissioned. The leadership of newly emergent segments had to produce its own socio-surplus to attract adequate supplies of labor.” As a result, the system “was compelled to expand as each segment reproduced the mode of production.” Expansion was rapid, leading to the disappearance of Woodland peoples from the central Prairie peninsula by A.D. 1200 and to the assumption of a multi-ethnic character by the Oneota, as they “displaced competitors from their territories and incorporated new supplies of labor by capturing women and adopting children to clan affiliations.” This aggressive spirit, he argues, is symbolized on their ceramics in stylized hawk forms (Figure 18.8).
Again, whichever explanation is correct, Blue Earth phase Oneota sites dominate southern Minnesota after ca. A.D. 1200. The Blue Earth River localities rise to prominence, there is a strong Oneota presence along the St. Croix River at such sites as Sheffield, and Oneota sites and ceramics spread westward into the northwestern prairie zone and into adjacent portions of the Dakotas. As we will see in Chapter 21, Oneota ceramics in the form of Ogechie ware become widespread at this time, too, throughout the woodlands of central Minnesota.
Settlement-Subsistence Models. The most ambitious study of Oneota settlement patterns in southern Minnesota was conducted by Clark Dobbs and Orrin Shane at the Center Creek and Willow Creek localities along the Blue Earth River (Dobbs and Shane 1982). According to their study, each settlement cluster is known to contain about 50 Blue Earth phase sites. Dobbs (1984a) identified six settlement types for the Center Creek locality at the confluence of Center Creek and the Blue Earth River (Figure 18.9). The types were defined by comparing the extent, density, and content of surface scatters of artifacts. Small sites with low artifact densities and relatively large numbers of scrapers were considered hide processing stations (Type 1) and butchering/lithic workshops (Type 5). Sites of variable size and artifact density with relatively large amounts of pottery and scrapers were interpreted as habitation areas where hide processing was important (Type 2), and moderate to large sites with low proportions of pottery, projectile points, and scrapers and high proportions of utilized flakes were regarded as butchering stations (Type 4). Moderate sized sites (2.5 to 7.5 acres) with high proportions of non-utilized flakes, projectile points, and end scrapers and low proportions of other lithic tools were considered lithic workshops (Type 3). Very large sites with dense artifact scatters and high proportions of ceramics, projectile points, and knives were thought to be large horticultural villages (Type 6).
In general, major habitation sites and cemeteries in the Center Creek locality are on elevated patches of outwash sediment in the Blue Earth River floodplain, and smaller habitation sites where animal and hide processing apparently took place are just outside the floodplain at slightly higher elevations. The highest and most distant sites from the floodplain are assumed to be small special-use stations where lithic reduction and animal processing took place.
An interesting aspect of this settlement pattern is the near absence of Blue Earth sites in this region east of the Blue Earth River (except for cemeteries), even in the Center Creek and Willow Creek localities. The river may have served as a boundary between social groups or settlement west of the river may have provided easier access to bison on the prairies to the west or to closely related social groups in northwestern Iowa. Dobbs and Shane (1983:68) have suggested that the Center Creek and Willow Creek localities may represent tandem villages, a pattern found among the Oto and Ioway, historic groups generally considered descendents of the Oneota in this region of the Upper Mississippi River basin.
Animal remains from 1976 University of Minnesota excavations at Vosburg (21FA2), a Center Creek locality large horticultural village site, are listed in Table 18.3 (Dobbs 1984). This list needs some interpretation. While bison scapula hoes are fairly numerous at sites in the locality, often occurring in groups of three or more at the bottom of storage/refuse pits (Dobbs and Shane 1983:65), only a single identifiable non-tool bison element was found in the sample; therefore, bison tools in the assemblage have been excluded from the list. Fish, bird, and turtle remains are underrepresented in the list, for remains from only a single feature were identified. Dobbs does not mention the presence of mussel shell in his report, but shells were recovered during the salvage excavation of a pit in 21FA97, a nearby site (Anfinson 1986b).
Maize, common beans (Phaseolus sp.), and sunflower were found at Vosburg in 1976 during dry screening, along with hazel, wild plum, and hawthorn. Flotation samples obtained during the excavation have not been processed. The maize from Vosburg is highly variable, but at least two distinct types are present, one of which is Northern Flint (Dobbs 1984a:66). Since the site may have been repeatedly reoccupied over a 700-year period, these data are not necessarily representative of any one particular component.
While our knowledge of Oneota settlement and subsistence patterns in southern Minnesota remains largely speculative, several tentative conclusions can be drawn. First, Oneota groups seem to have had a mixed hunting-gathering-horticultural subsistence base. The presence of large storage pits, scatters of carbonized maize, and agricultural implements indicate a greater degree of dependence on cultigens than was the case among earlier Woodland societies. Second, settlement and subsistence patterns seem to have varied somewhat over space and time in adjustment to local environmental and socio-political conditions. Finally, the presence of bison scapula tools in village sites and the extensive scatter of small components throughout the southwestern prairie zone suggest that seasonal bison hunts were an integral part of Oneota lifeways, as it was among historic Native Americans in the region.
A departure from this pattern is found at the Sheffield site (21WA3), which seems to be a summer base camp, where fishing and hunting activities were dominant pursuits and horticulture was either minor or absent as an activity (Wilford 1961d; Gibbon 1973). The site is located on the floodplain of the St. Croix River adjacent to open deciduous forests and some grasslands. Unlike the more common horticultural villages, Sheffield has few storage pits, no large scapula hoes, relatively few fabricating and processing artifacts, including end scrapers, and a high frequency of hunting and fishing implements. If the ratio of ceramic vessels (female) to chipped stone artifacts (male) has any validity as an indicator of female-male composition at a settlement, there was an approximate even balance of males and females at the site. Sheffield could be a special activity camp associated with a larger, more southern horticultural village, but the ceramics differ somewhat from those in Oneota sites along the Blue Earth River.
Mortuary Practices. Oneota mortuary practices in the Red Wing locality between ca. A.D. 1000 - 1200 remain unclear. Oneota pottery has been found with Middle Mississippian ware in some earthen burial mounds in the locality, one of which was a “panther” effigy mound (Johnson et al. 1969; Rodell 1991), but mound burial does not seem to have been a trait of the Emergent Oneota horizon in eastern Wisconsin. Instead, burials were found in pits below house floors or in extended positions in cemeteries (Overstreet 1997:257). Possible burial mounds at the later Walker-Hooper site in central Wisconsin may have formed by accretion on natural knolls (Brown 1993; Overstreet 1997:266).
Later burials attributable to the Blue Earth phase in the Prairie Lake region have been encountered accidentally during gravel removal operations in the Center Creek locality. 21FA84, a cemetery on the east side of the Blue Earth River, for example, may have contained as many as 100 burials, and another cemetery probably once existed on the same side of the river about 3.7 miles north of 21FA84 (Dobbs 1984a:75). A few individual human bones and an infant burial were found in pits at Vosburg. Some of these remains are charred. Gravel removal at Humphrey and Vosburg also exposed burials. In general, largely incomplete secondary burials are found in small numbers in major villages. Large cemeteries are located near these villages in similar settings but on the opposite side of the river. Burials in the cemeteries appear to have been extended primary interments. No mound burials are known to be associated with this locality.
Possible Blue Earth phase burials have been excavated at the Judson (21BE6), Camden (21LY2), and Great Oasis (21MU2) sites west of the Center Creek locality. Judson Mound #1 near the Cambria site contained a shell-tempered mortuary vessel and five extended primary burials in a subsurface pit (Wilford 1956b). The Camden Mound, which was excavated by amateur archaeologists, contained a shell-tempered vessel associated with what was probably a primary burial (Chamberlain 1936). A secondary bundle burial was found in a trash pit at the Great Oasis site with shell-tempered sherds.
A peculiar situation exists at the Sheffield site, where human bones were found in the midden and a refuse pit during excavation in 1959 and 1960. An adult left talus and a tooth were in the feature, along with deer, turtle, and other animal bones. An adult left mastoid process and a small portion of a burned and possibly cut cranium were together in one excavation square, and an additional small part of a cranium, also possibly cut, was found in another square. The presence of human bone in the village debris may indicate disruption by animal activity of a burial mound located some 30 to 50 feet south of the village site or, according to Wilford (1950a), the possibility of the practice of cannibalism.
Contributors to this chapter: Guy Gibbon
Date of last contribution: December 2008