Cahokia and the Archaeology of Power

3. The Cultural-Historical Contexts

3 The Cultural-Historical Contexts When I arrived at the foot ofthe principal mound, I was struck with a degree of astonishment, not unlike that which is experienced in contemplating the Egyptian pyramids.What a stupendous pile of earth! -Conmlent of Henry Brackenridge [1814: 187] upon viewing Cahokia's Monks Mound. MISSISSIPPIAN CULTURE AND CAHOKIA The prehistoric Mississippian cultures ofthe eastern United States represent the epitome of social, political, and religious development of the area's aboriginal inhabitants. While there is near universal acceptance of Mississippian as the premier cultural "climax," vigorous debates continue on "what is Mississippian?" James B. Griffin (1985) has extensively documented the changing archaeological concepts from the early ceramic definition ofHolmes (1903) to the more ecologically oriented definition proposed by Smith 0978a, 1984). Griffin (1985: 62-63) concludes his overview by portraying Mississippian in a broad new light-one that is not tied to artifact types, construction techniques, arbitrary categories of cultural evolution, specific habitats, or even geographic regions. Instead, Mississippian societies (Griffin 1985: 63) are characterized as those that (1) participated in a series of new cultural innovations between A.D. 700 and 900; (2) incorporated those disparate innovations into their culture through contacts with other groups; (3) con- structed planned permanent ceremonial centers, towns, and associated settlement hierarchy; (4) possessed various forms ofa hierarchical social, political, and religious system; (5) participated in a religious system that emphasized the interaction ofspirit world and man with a rich iconography expressed in marine shell, copper, ceramics, and stone; (6) participated in extensive trade networks; and (7) reached a cultural "crest" between A.D. 1200 and 1500. Cahokia One of the earliest and most dominant centers ofMississippian cultural development was in the Central Mississippi River Valley at Cahokia in the American Bottom (Fowler 1974; Pauketat 1994: 40-107; Milner 1990; Emerson and Lewis 1991; Stoltman 1991). The section of the American Bottom area that figured so importantly in Mississippian times was the northern floodplain zone containing about 450 km2 of the Mississippi RiverValley just south of the confluence of that river with the Missouri and Illinois Rivers. Recent research has provided a detailed description of the Late Woodland and Emergent Mississippian cultural development (Kelly et ai. 1984a, 1984b;Kelly 1987, 1990a, 1990b;Emerson andJackson 1987a) that preceded the appearance of the Middle Mississippian groups in the area. While it is clear that many of the traits that later were important in the Mississippian cultural matrix appeared about A.D. 600 (Emerson and Jackson 1987a; Fortier et ai. 1991; Kelly 1990a), the period from about A.D. 925 to 1050 was one of intense transformations of the social, political, religious, and economic organizations ofthe indigenous LateWoodland peoples (Kelly 1987). In the American Bottom, that period saw the full-scale adoption of maize horticulture, nucleated villages, and many material traits that would coalesce to form the Middle Mississippian culture. The greatest development and elaboration of prehistoric culture in the American Bottom occurred aboutA.D.1050 with the appearance ofMiddle Mississippian societies.The period was marked by the proliferation of Mississippian peoples and their settlements up and down the valley bottoms and the associated bluffs. One of those temple mound centers, Cahokia, far exceeded the rest in size and presumed complexity, covering more than 13 km2 (Fowler 1989). Estimates place about 120 mounds within the site boundaries , which also included approximately 1.8 km2 ofavailable residential and agricultural space (cf. Pauketat and Lopinot 1997).The dominant feature of this site was Monks Mound, a large, multiterraced platform mound located within the central ceremonial precinct. That massive mound and a number of smaller mounds and plazas, along with about 120 ha of the ceremonial 44 Cahokia and the Archae%,f{Y <2f Power areas, were surrounded by a bastioned palisade for at least part of the site's history. In addition to Cahokia, which was the major ceremonial or perhaps even urban center in the American Bottom, there were four other significant multimound centers (figure 3.1). These centers each included several hundred acres of habitation and ceremonial zones in addition to several associated mounds. There was such a center in the present location of St. Louis, one in East St. Louis, one to the north at Mitchell, and one at Pulcher to the south near Columbia. Pauketat (1994: 80-92) has most usefully visualized the central portion ofthis massive concentration ofmonuments and habitaFigure 3.1 Location of Sites in Text. Cultural-Historical Contexts 45 tion areas as Cahokia's Central Political-Administrative Complex, noting the extreme difficulty of archaeologically separating the Cahokia, East St. Louis, and, perhaps, even the St. Louis "sites." A number of single temple mound towns and numerous small villages, hamlets, farmsteads, and camps were scattered throughout the area. The picture ofa densely populated floodplain appears, with a subsistence base of corn in addition to heavy utilization of the wild floral and faunal resources. Archaeological research at Cahokia and the surrounding area suggests the presence ofa highly stratified society with a developed religious and political leadership. The presence of large-scale retainer sacrifice and differential distribution ofstatus goods suggest that we are dealing with at least a chiefdom level of political organization, although some (e.g., Gibbon 1974; O'Brien 1989, 1993) have argued that it may represent an actual example ofthe state. Chronology and Taxonomy In the summer of 1971, the Cahokia Ceramic Conference was held to integrate and interpret more than one-half century of information that had accumulated since W K. Moorehead's first excavations took place in 1921 (Fowler and Hall 1972, 1975; Hall 1972b). A sequence of eight phases was established, incorporating ceramic, architectural, and social information in their definitions. Considering the size of Cahokia, the intensity of occupation in the American Bottom, and the obvious cultural pluralism that existed in the area, these phases were proposed as representing the current state ofknowledge rather than as definitive.The phases included a Patrick phase, pre-A.D. 600-800; an Unnamed phase, A.D. 800-900; Fairmount phase, A.D. 900-1050; Stirling phase, A.D. 1050-1150; Moorehead phase, A.D. 1150-1250; Sand Prairie phase,A.D. 1250-1500; and a final Unnamed phase, A.D. 1500-1700. This sequence has survived intact for the most part, although refinements have been added by subsequent research, especially that of the FAI-270 Archaeological Mitigation Project (Bareis and Porter, eds. 1984).The initial attempt to clarify the Cahokia ceramic sequence was by Gregg (197Sa) who, on the basis of his work at the Horseshoe Lake site, suggested that the Unnamed phase (A.D. 800-900) be defined as the Jarrot phase.This suggestion has not been accepted generally. Later research by Kelly (1980) on the Merrell Tract in Cahokia provided evidence that the Lloyd phase should replace the Unnamed phase and, furthermore, that the Fairmount phase should be divided into early and late portions. The most dramatic alteration of the phase sequence, however, was produced by the FAI-270 research. Kelly et al. (1984b) suggested the delinea46 Cahokia and the Archaeology ifPower tion of an Emergent Mississippian period that subsumed all of the phases between A.D. 800 and 1000. In addition, they were able to define northern and southern American Bottom cultural variations within this time frame to create two sets of phases. Kelly (1987, 1990a) formalized this north-south division as the northern Late Bluff and the southern Pulcher co-traditions. The Lloyd (A.D. 800-900), Merrell (A.D. 900-950), and Edelhardt (A.D. 950-1000) phases are in the north, while the Dohack (A.D. 800-850), Range (A.D. 850-900), George Reeves (A.D. 900-950), and Lindeman (A.D. 9501000 ) phases are in the south. In the American Bottom, the concept of an Emergent Mississippian period is appropriate to describe the series of cultural transformations that led to the emergence ofa fully Mississippian culture at about A.D. 1000 (cf. Kelly 1987). Most recently, Fortier et al. (1991: 5-10) have put forward a number of additional modifications to the sequence, suggesting a new Sponemann phase (ca.A.D. 750-800) and a subsequent Collinsville phase (ca.A.D. 800-850) in the Late BluffEmergent MississippianTradition. Kelly (1990a: 117-124), on the other hand, assigns Sponemann to the Late Woodland period and dates it ca. A.D. 700-750. At question in this discussion is the relative importance ofvarious defining traits-Fortier emphasizes those that resemble the Emergent Mississippian period, Kelly those of the Late Woodland period. The FAI-270-related research can be seen as a refining rather than an altering influence on the Mississippian sequence. When the concept of the Emergent Mississippian was actually implemented, it subsumed the "early" Fairmount phase that had been proposed previously by Kelly (1980). Once this clarification was made, it became evident that the "late" Fairmount was essentially Mississippian in content, and it was redefined as the Lohmann phase (A.D. 1000-1050).A somewhat similar event occurred on the opposite end of the chronology with the recovery of Oneota structures and features on FAI-270 sites. The Sand Prairie phase was shortened from A.D. 1500 to A.D. 1400, and a Vulcan phase, representing an Oneota cultural presence, was placed at A.D. 1400 to A.D. 1600.The remainder of the Mississippian phases remained intact and were strengthened by the addition of further information gained from excavation and analysis. A description ofthe Mississippian cultural phases now extant in the American Bottom area are selectively outlined here, based on the original data presented in Fowler and Hall (1972, 1975) as modified by Milner et al. (1984) and other, more recent work by regional scholars (for a fuller treatment see Emerson 1995: 66-70).The chronology has been adjusted here to take into account the modifications suggested by Robert Hall (1991) to bring the previous arbitrary cultural time line into closer conjunction with calendar years. Cultural-Historical Contexts 47 Lohmann Phase--A.D. 1050-1100. The Lohmann phase represents the first Middle Mississippian phase in the American Bottom.This phase, which represents the later portion of the Fairmount phase as originally defined in Fowler and Hall (1972) and which equates with Hall's (1966) Pre-Ramey or Pulcher phase, was delineated by the research ofthe FAI-270 project (Milner et a1. 1984). Kelly (1990b: 77) has proposed a southern contemporary of the Lohmann phase that he has named Lindhorst and views as the terminal phase of the PulcherTradition. The most important phenomenon during this time period is the implied transformation that took place in the political, religious, and social aspects oflife during the late Emergent Mississippian period and the early Lohmann phase. Since the more detailed definition of the transitional phases has been of recent origin, it has not been possible to determine how this change is reflected in the Cahokia internal sequence. Consequently, the events discussed here could date to either the Late Emergent Mississippian Edelhardt phase or the following Lohmann phase. The ceramic assemblage most frequently includes jars, but seed jars and bowls are common, with beakers, bottles, hooded water bottles,juice presses, and stumpware also present. Diagnostic jar rim forms were most typically angled and everted rims with angled shoulders.The majority of the ceramics were shell tempered although grog, grit, and limestone were sometimes used. Surface colors on slipped or polished vessels are predominantly lightslipped reds and tans, but brown or black does appear. Monks Mound Red (a ceramic type originating in the southern American Bottom PulcherTradition sites) is common in Lohmann phase assemblages. The lithic assemblages include frequent microdrills, corner- or side-notched, triangular points, hoes, discoidals, and the more common flake tools. It was during this period that the construction of Monks Mound began (Reed, et a1. 1968; Collins and Chalfant 1993; Skele 1988; Emerson and Woods 1990), and the alignment of the civic and religious center of Cahokia was laid out. By the late Lohmann phase, the sun circles or woodhenges were being constructed (Pauketat 1991 ;Wittry 1969; Smith 1992).The evidence for a stratified society appeared in the form ofelaborate status burials and retainer sacrifice in Mound 72 (Fowler 1991).Also contained with these burials were exotic items, which suggested to Fowler (1972: 88) a high degree of control over the exchange of exotic resources. This material was present in raw form (i.e., a roll of unworked sheet copper and sheets of unmodified mica). Caches offoreign projectile points also were found with the burials, indicating contacts with both the Caddoan and the Great Lakes regions. The utilization of Mound 72 apparently was confined entirely to 48 Cahokia and the Archaeology of Power the Lohmann period and may have represented the mortuary complex ofa specific elite lineage. The increased house size, the beginning of large public construction projects (e.g., mounds), as well as the presumed appearance of large Emergent Mississippian-Early Mississippian towns suggest to some that there were quantum increases in population (from an Edelhardt population of 1,300 to 2,700 to a Lohmann population of 10,000 to 15,000 at Cahokia [Pauketat and Lopinot 1997]) and in the degree of social and political control. Harn (1971: 34-35) identified six outlying towns, each with a minimum of one temple mound, as dating from this transitional period. These towns may have developed in situ out of previous Emergent Mississippian settlements or may have represented the actual movement of people out from Cahokia proper. There is no doubt that by the Lohmann phase these temple towns were a permanent part of the American Bottom scene. This phase also was the transition from single-post to wall-trench house construction, with both forms occurring, and an increase in the overall size of the residential structures. This gradual increase in size is a phenomenon that continued throughout the Mississippian sequence, and it has been documented by research ofthe FAI-270 Project for small sites outside ofCahokia proper (Milner et al. 1984, figure 61). These outlying small sites are important sources of information on change in the American Bottom, andYerkes (1983) has argued for evidence of at least household specialization during the Lohmann phase. There is some evidence that at least minimal contacts to the north were established during this phase. Early Mississippian traits have been identified at the Rench site in Central Illinois (McConaughy 1991), the Collins site in northeastern Illinois (Douglas 1976), the Trempealeau site in southwestern Wisconsin (Green and Rodell 1994), and the Chapman site in northwestern Illinois (Emerson 1991a). Stirling Phase--A.D. 1100-1200. This phase is the time ofmajor Cahokian expansion. Stirling influences, to various degrees, were present at the East St. Louis, Mitchell, Divers, and Pulcher sites in the American Bottom, north into the Lower Illinois RiverValley and the Spoon River sites ofthe Central Illinois RiverValley, the Apple River sites of northwestern Illinois, and as far north as Aztalan (cf. articles in Emerson and Lewis 1991; Stoltman 1991). An interesting phenomenon that may be associated with the Stirling phase seems to be the establishment of large nucleated towns at some distance from Cahokia and, at the same time, the apparent abandonment and absorption of closer towns. Both Harn (1971: 36-38) and Munson (1971: 1416 ) posited a reduction oflarger sites during this time period, based on the evidence from surface collections.This apparent reduction, however, clearly Cultural-Historical Contexts 49 reflected an increasing population nucleation into fewer centers rather than any absolute population decreases (cf. Milner 1986).The nucleation ofpopulation had begun during the Emergent Mississippian and Lohmann phases with the appearance ofsmall temple towns and their associated hamlets (cf. Kelly 1992). The desertion of these small towns during the Stirling phase, while Cahokia expanded, suggests that this was due to a population shift from the satellites to the major population center. Winters (in Harn 1971: 36) points out that this nucleation ofpopulation and abandonment of the surrounding area was a common pattern in the process of urbanization. Previous population estimates computed for the Cahokia site suggest that from 10,000 to 43,000 people occupied the location during the Stirling and Moorehead phases (cf. Gregg 1975b; O'Brien 1972a; Reed et al. 1968). Recent new population calculations by Pauketat and Lopinot (1997) suggest Stirling Cahokia populations of5,000 to 7,000. This means a density of 2,800 to 4,000 people per km2• Fowler tends to accept the higher figures, given the large amount ofpublic construction that occurred at this time (1972: 91). Milner (1986) has explored the question ofAmerican Bottom Mississippian population density outside the main Cahokia site, based on the extensive excavations of the FAI-270 Project. In his sample areas to the south of Cahokia, he found that relative population density was greatest during the Lohmann and Stirling phases, decreased during Moorehead times, and sharply declined in the Sand Prairie phase. Lohmann and Stirling phase population density estimates ranged from 9 to 23 individuals per km2. The Moorehead figures were from 5 to 9 individuals per km2, while the Sand Prairie density varied between 1.5 and 3 per km2• It is evident that the rural portions of the American Bottom, as represented in Milner's data, contained a small, very dispersed, portion of the population. Unfortunately, while we have fairly reliable information for the areas outside of the main towns, we still have little for the actual towns themselves . Ofall the outlying satellites in the American Bottom, it is only about Mitchell that we have archaeological evidence to indicate that the site flourished at this time. Porter (1969: 151-152) has postulated that this is due to the town's critical position at the mouth of the Missouri River, which allowed it to control this area. However, recent limited excavations at the East St. Louis Mound Group byJohn Kelly (personal communication 1994) did reveal what appears to be an active utilization ofseveral mounds during the Stirling phase. It appears, however, that there is some resistance to any form of"assimilation " or "consolidation" because, during this period, stockades began to be erected around the central precinct at Cahokia. The earliest palisades 50 Cahokia and the Archaeology ofPower seem to have been started about A.D. 1200 and possessed circular bastions along the strongly built wall. This was followed by at least two additional rebuildings, including a palisade with square, closed-back bastions and a final fortification, including square bastions with an open back (Anderson 1969; Iseminger et a1. 1990). I have argued in other contexts (Emerson 1991b) that this early period of American Bottom Mississippian political consolidation was inherently unstable as various elite groups vied with one another for control of resources and power.That the eleventh century is marked by the dramatic appearance ofcultural homogeneity in the American Bottom suggests the probable political and cultural dominance of a single polity. To achieve such a dominance must have entailed a previous period ofpolitical strife and internecine warfare characterized by and encouraging population and settlement disruptions , elite migrations, and general instability in the locality. At Cahokia itself, a flurry of construction activities dated to the Stirling phase. Monks Mound was completed, and extensive mound building occurred in other parts of the site. Sun circles were still present, as well as large circular components in Tract 15B.The first accepted evidence for craft specialization on the site seems to be from this time period. Mason and Perino (1961) suggest, on the basis of the association of microdrills and marine shell, that an area to the west of the Kunnemann Mound Group may have been a center for the production of shell beads. This concentration of microdrills has been confirmed by recent extensive surface collections from the KunnemannTract (Holley 1995) and the elite manufacture ofshell beads identified at the Kunnemann Mounds by Pauketat (1993). In addition, through microwear analysis,Yerkes (1983) has demonstrated the close association of microdrills and shell working at a number of other American Bottom sites. Stirling phase structures were larger than those of the previous phase and were almost all constructed using the wall-trench method. For the first time, different functional types of Middle Mississippian structures can be recognized at the outlying sites, including sweat lodges, storage structures, and perhaps seasonally specific residences (cf. Emerson 1995). Virtually all of the ceramic assemblage is shell tempered, although other materials appear infrequently. The typical jar forms have angled rims, with everted and rolled shapes becoming common. Shoulders are still mostly angular. Plain surface treatments are usual, although black and brown filming often occurs. Ramey Incised jar forms appear for the first time in the assemblage. Forms such as bottles, hooded bottles, and seed jars continue. Stumpware declined and disappeared before the end of this phase. By the end of the phase Tippets Bean Pots, Cahokia Cordmarked, and globular Cultural-Historical Contexts 51 juice presses are found in the assemblage, but fabric-impressed pans have not been recovered. Lithic items continue virtually without change from the preceding Lohmann phase. Limited contacts to the south are indicated by the presence of exotic ceramics and represent a continuation of this contact from earlier periods. Southern Cult motifs in the form of the forked eye/weeping eye motif are common. Exotic materials such as red ocher, galena, quartz crystals, marine shell, mica, and flint clay are often recovered from Stirling contexts. Frequent evidence for fertility symbolism is associated with this phase, as are large stone figures and pipes and symbolic ceramics (Emerson 1982, 1989; Pauketat and Emerson 1991). Moorehead Phas~A.D. 1200-1275. Although changes occurred in the ceramic assemblage, the basic pattern oflife at Cahokia seemed unchanged. The site, according to Fowler (1972: 89), became more "urbanized," and by the end of the Moorehead phase, Cahokia is seen by some as a city of approximately six square miles in area (Brandt 1972: 65). Pauketat and Lopinot (1997), however, document a dramatic drop in Cahokia Moorehead populations to 3,000-4,500 people. In addition, some architectural changes occurred : circular compounds were replaced by rectangular forms; wall-trench houses were enlarged, perhaps indicating larger families or different household units; and single-post houses and rectangular pits disappeared. Structures tended to be nearly square; special functional forms tended to be present. Lithic assemblages continued unchanged, and exotic materials continued to be present. Limestone tempering, common in earlier phases, was now completely gone.This may suggest a change in the relationship ofCahokia with areas to the south where limestone tempering was prevalent, it may reflect some technological changes, or it simply may reflect the giving up of old habits. Porter (1964) has pointed out that, technologically, the switch from one carbonate (limestone) to another (shell) would represent no difficulty. Shell tempering dominated the scene. Cahokia Cordmarked, with its cordmarked surface and shell tempering, replaced the earlier grit- and grog-tempered, cordmarked types.The shell-tempered bean pot with handles, calledTippets Bean Pot, became more common and more plentiful than a previous form with limestone tempering. Juice presses and crucibles with grog temper continued. Wells Broad-Trailed plates, with their highly polished black surfaces , narrow rims, and broad-trailed decorations, were present. The Cahokia version of Mounds Place Incised was common; however, it appeared as early as A.D. 1125 on Tract 15B (Hall 1972a). A new form, the narrownecked , medium-high water bottle made its first appearance. Both low and rolled rims declined in popularity at the expense of the higher and everted 52 Cahokia and the Archaeology