This study presents an alternative perception of cultural change. Culture change is seen as the inevitable result of a system, far from equilibrium, adapting to fluctuating conditions. Such a system displays a dissipative structure (Friedman 1982; Nicolis and Prigogine 1977) in that, while maintaining local stability at some material/energy cost, it eventually reaches a threshold where it evolves into an unstable order in response to fluctuations. It is under these conditions that new phases are produced (morphogenesis). One goal of explanatory archaeology should be the identification of such structures. Sufficient non-archaeological data indicate that prehistoric agricultural systems were subject to such fluctuations. In this study, both historical and empirical observations were used to specify an approximate model of a dissipative, Mississippian system which could be compared to the documented late prehistoric development scheme of East Tennessee.
From a regional standpoint, Mississippian morphogenesis has been associated with the question of Overhill Cherokee origins (Schroedl 1986). The level of pre-Mississippian IV social organization has been correlated with elaborate material remains and implicitly characterized as being more complex than the later Cherokee order. This tendency to correlate material elaboration with social complexity necessarily demands an explanation for the obvious material discontinuity between Mississippian III and IV, not only in East Tennessee but throughout the Southeast. Yet, ceramics, mounds, structures, and so on are not equiv-
alent to ethnic groups, social organizations, or linguistic categories. If there has been a single weakness in any explanation, it is this leap from material order to social order ignoring the dynamic nature of culture as an evolving system. In particular, such an approach ignores the fact that society tends to reproduce itself through structured processes influenced by external parameters. Lacking a solid theoretical basis for such a transformation, various logical rationalizations have been presented as "explanations".
Early Mississippian research emphasized bridging the macroscopic discontinuity between the more complex mound building populations and their supposed descendants, the various historic Indian groups. The only available inductive tool was the axiomatic assumption that proximity in archaeological space was equatable with cultural similarity, regardless of material discontinuities (Harrington 1922, Thomas 1894). Two cultural manifestations separated in time, but not in space, must be ethnically equivalent.
Synchronic trait associations (Lewis and Kneberg 1946, Webb 1938, among others) and ethnohistoric interpretations (after Swanton 1928, 1946) were later used to promote a population replacement model (i.e. migrations) as the mechanism explaining material differences dividing the subphases of the Mississippian period. This particularistic model rejected the concept that change was internally induced by adjustments to fluctuating conditions. Instead, they chose a one-to-one equivalence between material assemblages and ethnic units. Whatever could not be explained by ethnic conflicts was resolved by calling on external disruptions like European contact and the spread of disease.
We now recognize the impact of agriculture on settlement systems. Villages should be seen as semi-sedentary in response to the need to be relatively close to their fields. When a land reservoir became exhausted, other river valleys would be needed for colonization. If such unclaimed locales existed, Mississippian "traits" would spread at the expense of extending the geographic limits of influence (perhaps beyond some limit defined by information theory). If not, conflicts between valley inhabitants would develop unless some social mechanism existed that could bind villages together or facilitate the shifting settlement system. The dispersed hamlet system of the historical period would have maximized aerial coverage while minimizing population disruption (by raising fs).
These mechanisms, or social operators, would provide an order to the interactions between villages. If the underlying subsistence system failed, new operators would be developed to perpetuate the society. For Mississippian cultures these transition periods correlate with the occurrence of instabilities in our potential function. Such sudden transformations, predicted by the rate of change in potential functions, closely follow Thom's (1975) Theory of Catastrophes which has been applied to other cultural phenomena (Friedman 1982; Poston 1979; Renfrew and Poston 1979). Here, society lies along a trajectory determined by how it challenges its social and physical environments. It will tend to move further and further from equilibrium (assuming there are no external constraints). Under such conditions, the potential function serves as a measure of the stress placed on the production subsystems.
When these mechanisms undergo rapid change, the society crosses a threshold of instability and rapid transformations can occur.
Turning back to the question of Cherokee (and Creek, Chickasaw, Choctaw, etc.) origins, the social cost of maintaining nucleated, egalitarian villages would be small in comparison to that required by Mississippian II and III systems built around exchange networks. Yet such exchange systems (and their associated material assemblages) were needed to maintain the extensive agricultural system presented here. When that system could no longer be supported, the social order binding it together could devolve into a less complex structure (Mouse Creek?). In areas where extensive agriculture was less feasible (e.g. the Appalachian Summit) and therefore more quickly constrained, we would expect the pinnacle social structure to be less complex (Dickens 1986). Under these circumstances, society's ability to maintain itself beyond some equilibrium point would be more limited. Such a system would produce a totally different potential curve and accompanying social order.
Summarizing, the East Tennessee Mississippian sequence is seen as an ebb and flow of unstable to stable conditions wherein the current phase state is defined by choices made by the previous one. Although the Spanish (and later English and French) presence late in the period served as an external "mutation", opening new trajectories for Mississippian evolution, the primary evolutionary force affecting Southeastern aboriginal populations originates in the adjustments required to maintain some potential function under conditions far from equilibrium. Given a specific set of behavioral rules and a bounded environment, an expanding society's potential function can be calculated and used to
predict periods of instability. This study challenges our ability to recognize contemporary sites, demands a more accurate accounting of the space-time continuum linking archaeological sites with their original cultures, and begins to address Netting's (1974) call for a more systematic examination of agricultural processes.
Although it may be tempting to use the results of this application to "explain" Cherokee morphogenesis, the reader must recognize that maize agriculture served as only one part of the total socio-techno complex we call Mississippian. Other dissipative factors, such as the depletion of fire wood resources and the dependence on inadequate information systems, worked in conjunction with the agricultural limitations to define an overall potential function. Broader examinations are needed that combine all of these subsystems to fully deal with Mississippian morphogenesis. This study serves as a first step toward such a synthesis.