the early mississippian frontier in the lower ...anthropology.ua.edu/reprints/14.pdf · the early...

Made in United Slales of AmericaReprintedfromSOUTHEASTERNARCHAEOLOGY

Vol.21,No.2,Winter2002Copyright© 2002bytheSoutheasternArchaeologicalConference

THE EARLY MISSISSIPPIAN FRONTIER IN THE LOWER CHATTAHOOCHEE-APALACHICOLA RIVER VALLEY

John H. Blitz and Karl G. Lorenz

Explanations of Mississippian origins are sometimes present-ed as a choice between in situ cultural developments versuspopulation movement. We examine these contrasting per-spectives in the context of interaction among three regionalpopulations with disparate material culture traditions anddiffering degrees of organizational complexity. We argue thatpopulation movement was the mechanism that established theinitial Mississippian communities in the lower Chattahoochee-Apalachicola region, and subsequent cultural developmentsamong neighboring indigenous populations were influencedby the presence of immigrant settlers. We propose that thedifferent ways in which regional populations responded tointeraction across a geographical; cultural, and sociopoliticalfrontier placed them on divergent paths to Mississippianiza-tion.

logists thought population movement a capnclOushistorical event rather than a cultural process structuredby recurrent principles. Lathrap argued against this per-spective, however, and maintained that understandingpopulation movement as a process is as fundamental tocultural evolution as understanding the processes ofradiation and drift is to biological evolution. Yetthe eco-demographic functionalism of processual archaeology, aspracticed over the decades since Lathrap's essay, merelyreinforced the barrier between culture history andcultural process.

In this article, we examine how these issues haveshaped archaeological debates about Mississippianorigins in the lower Chattahoochee-Apalachicola Riverregion (Figure 1). The lower Chattahoochee Riverextends from the Fall Line to the Flint River confluence,south which the river is known as the Apalachicola. Inthis region, in situ or local cultural development versus

Figure 1. Mound centers in the lower Chattahoochee-ApalachicolaRivervalley,ca.AD 1100-1250.

One problem currently challenging North Americanarchaeologists is to understand the emergence and spreadof the Mississippian cultural phenomenon: the adop-tion of intensive maize agriculture, hierarchical decision-making institutions, and ascriptive ranking. Althoughregionally diverse in population size and material culturecontent, Mississippian societies shared many tech-nological, economic, and organizational patterns, andspecific ceramic vessel styles, architectural forms, andiconographic imagery had wide geographical distri-butions. These pan-regional commonalities cannot beexplained by reference to local adaptive conditionsconsidered in isolation. Instead, complex historicalconnections between populations are implicated.

In the 1950s, the iconoclastic archaeologist DonaldLathrap, among others, proposed that radiation, drift,and other concepts from biological evolution had usefulcultural analogs. Lathrap outlined a process by whichpopulations with successful economic patterns behavedlike radiating species in biology (d. Sahlins and Service1960).As competing populations with different econo-mic patterns expand their range, "this competitionforms the major part of the environment to which asociety must adapt" (Lathrap 1976:528).Lathrap positedthat population movement initiates the competitionamong human populations, which in turn results incultural evolutionary change. At the time of Lathrap'sessay, a deep intellectual schism separated culturehistory from culture process, such that many anthropo-

ROOD o 40km~....,N

t

117

SOUTHEASTERN ARCHAEOLOGY 21(2) WINTER 2002

, population movement as explanations of Mississippianorigins have been at odds for 50 years. Newly availabledata on site distributions and chronology are marshaledto evaluate the two competing interpretations. Weargue that population movement-the migration andresettlement of people from one location to another-was the mechanism that established initial Mississip-pian societies in the region. In the lower ChattahoocheeRiver valley, a regional variant of Mississippian known

, as the Rood phase was the direct product of this'I population movement. Farther south in the adjacent

Apalachicola River valley another regional Missis-sippian variant, Fort Walton, took form among indi-genous Late Woodland peoples through a process ofcontact, interaction and competition with the northernRood Mississippian settlers across a common frontier.Before we consider these specific circumstances, it isuseful to briefly review the intellectual history of theopposition between population movement and localdevelopment as explanations for Mississippian origins.

Changing Perspectives on Mississippian Origins

Into the 1950s,a theory that the Mississippian culturaltradition developed first in the central Mississippi Rivervalley and then spread throughout the EasternWoodlands by migration was the majority interpreta-tion of Mississippian origins (Ford and Willey 1941:350;Eggan 1952:29; Krieger 1951; Willey 1953:371-372;Willey and Phillips 1958:164-165).Given the pre-war"short chronology," the equation of Muskogean migra-tions as related by oral tradition with a Mississippianexpansion seemed plausible. But by the 1950sthe linkingof archaeological remains to specific ethnic groupsbecame less tenable with a growing awareness of thegreat time depth involved (Smith 1984).According toSmith (1984:19),with the publication of the Phillips,Ford, and Griffin (1951) volume, a picture of severaldistinct but interacting areas of core developmentbegan to replace the single area-migration scenario asthe established interpretation of Mississippian origins.While no longer concerned with identifying a singlegeographical area of Mississippian origin, the seeminglyabrupt appearance of the Mississippian culturaltradition in many local sequences, often without anydemonstrable cultural continuity with indigenous LateWoodland traditions, continued to influence archaeolo-gical interpretations. Efforts turned to developingmethods to determine whether cultural elementswithout local developmental precedent were the resultof cultural borrowing through diffusion (i.e., "trait unitintrusion") or the result of population replacementthrough migration (i.e.,"site unit intrusion") (Wauchope1956).

Two studies from this era exemplify the new concernwith "culture contact situations." Rouse (1958)presented criteria for identifying migration in thearchaeological record: there should be evidence ofdiscontinuity with earlier cultures as indicated byartifact styles and settlement patterns; a source cultureor homeland for the immigrants should be identified;there should be independent dating to show that thesource or donor culture is older than the immigrantderivative culture; and there should be evidence of anadaptive advantage that allowed the immigrants tosuccessfully displace or compete with indigenouspopulations in the new territory. Willey (1953)devel-

. oped a model of Mississippian "cultural colonization"in which initial Mississippian migration into theSoutheast established "colonies" at such sites as MaconPlateau in Georgia. At first, indigenous cultural tradi-tions remained unchanged by the presence of theMississippian settlers, but after a period of time, newcultural traditions developed as a product of "diffusion-acculturation" between the immigrant Mississippiansand indigenes.

In the 1970sand 1980s,knowledge of regional Missis-sippian cultures grew rapidly (Smith 1978).Processualarchaeologists rejected earlier migration and diffusionscenarios of Mississippian development as simplistic,and the criteria for identifying population movementsadvanced by Rouse and Willey were dismissed asinadequate (Smith 1984:26-27).Local or in situ Missis-sippian development was attractive to processual.archaeologists because they could discard older con-cepts such as "cultural hearth" and embrace some of theemerging tenets of cultural evolution. Much of the pro-cessual archaeology applied to Mississippian culturalevolution in the 1970s and 1980s de-emphasized ordenied any causal role for population movement (e.g.,Brose and Percy 1978;Peebles 1978;Schroedl et al. 1990;Smith 1984).

Migration, as an external, rapid, and "historical" inputfor change, ran counter to the internal, gradual, and"evolutionary" prime mover favored by processualists:local or in situ adaptation propelled by demographicstress. Populations grew but they did not move. In thisview, external inputs from population movementcarried little explanatory weight, for these "events"merely begged the question of where, when, and howthe original adaptive changes had occurred, whichreturned the focus once more to the local sphere. Butthis line of reasoning betrays an anti-historical biaswhich is comfortable with the implication that thecauses of initial Mississippian development are every-where the same, rooted in local population-resourcestress. In such a view, there is no need to trace historicalrelationships. If one were to accept this position at facevalue, then every regional Mississippian population

118

J


becomes a case of independent or primary chiefdomformation. Certainly this perspective is every bit astotalizing and reductionistic as the simplest KuIturkreisescheme, and harks back to the age-old myth of theprimitive isolate.

Various scholars considered the origins and spread ofMississippian to be a complex problem unsuited to asingle mechanism of change (e.g., Faulkner 1975;Smith1984:21),arid some were willing to consider populationmovement as part of their explanatory framework. Forexample, Morse (1977)applied two well-known modelsof cultural competition and colonization (I.e., Sahlins1961;Willey 1953) to identify what he considered to·bean intrusive Mississippian population in northeasternArkansas. But Morse's interpretation went against theintelleCtual trend of the day because theoretical biasesin processual archaeology both reinforced the historicalevent versus cultural process dichotomy and privilegedlocal population growth as the ultimate explanation. Inthe synthesis volume The Mississippian Emergence, someauthors argued that local eco-demographic stress wasthe primary cause of Mississippian emergence (e.g.,Scarry 1990;Schroedl et al. 1990),while others sugges-ted that both local and external forces shaped specificregional examples of Mississippianism (Smith 1990:2-3).Population movement, as a process of cultural changeor source of cultural variation, was of little concern tomost authors in the volume.

Some influential processual models of Mississippiandevelopment did not neglect population expansion andinter-societal competition as sources of stress, risk, andinformation that fostered new institutional responses(Ford 1977;Peebles and Kus 1977).However, expansionand competition were considered lesser components orconsequences of local subsistence, environmental, anddemographic factors, such as access to agricultural land.Because the analytical focus of these models was thepopulation in a localized ecosystem, culture changewas portrayed as a gradual adaptation through adjust-ments generated within the cultural system. Thisperspective left little room for interest in the large-scale,rapid, and contingent transformations sparked by pop-ulation movement, culture contact, and inter-societalcompetition that an e·arlier generation of archaeologistshad deemed so important.

The isolationist population-resource stress. modelexplained widespread similarities in Mississippian mate-rial culture and organization as analogous responses tosimilar problems by populations with similar subsis-tence needs. But if population-resource stress were theultimate cause of Mississippian, then the rapidappearance of regional Mississippian populationswould mean that local population pressure climaxeduniformly in a short interval across the Southeastamong Late Woodland populations in diverse environ-ments. Why this should necessarily be so remains

unexplained, and there is considerable evidence to thecontrary. Some regions, such as the lower Mississippivalley and central Alabama, have abundant, large Late.Woodland period sites, while in other areas, such asnorthern Georgia, Late Woodland sites are widelydispersed or even absent from seemingly adequateenvironments (Nassaney and Cobb 1991). Proponentsof the local population-resource stress model nevergrappled with the implications of emigration as asolution to local demographic stress or, in the case ofimmigration, as a potential cause of stress.

Over the last dozen years, accelerated research insome regions, especially the lower Southeast, has failedto uncover the in situ developmental prototypes forinitial Mississippian material culture predicted by thelocal population-resource stress model (Blitz and Lorenz1999; Lorenz 1992; Payne and Scarry 1998; Williams1994). Instead, radical discontinuities in ceramic style,community plan, settlement patterns, architecture, andsubsistence indicate rapid replacement of antecedentcultural patterns. In some locales, these· new culturalelements are found as additions to the antecedent indi-genous pattern, and in other cases appear as spatially.distinct, yet apparently contemporaneous archaeolo-gical assemblages (Cobb and Garrow 1996;Jackson 2000;Little 1999; Little and Holstein 2001; Seckinger andJenkins 2000;Sheldon et al. 2001).

These researchers interpret this diversity in regionalmaterial cuiture and settlement patterns as evidence ofco-existing, interacting, Late Woodland and initialMississippian populations. The emerging picture of theLate Woodland-initial Mississippian transition, at leastin some regions, is more akin to a cultural mosaic ofcompeting, expanding populations with disparatetechnological, organizational, and economic patterns,as presented in Lathrap's essay, than to the gradual insitu adaptation of the population-resource stress model.If this is indeed the case, then archaeologists investi-.gating .the emergence and spread of the Mississippiancultural phenomenon must construct models .ofculturalchange and variation that include cultural pluralismand uneven development (e.g., Lightfoot and Martinez1995; Rice 1998).

There are -reasons to suspect that the movement ofpeople, as well as products or ideas, is implicated in theevidence for uneven Mississippian. development(Anderson 1997:266).For example, research in Georgiahas revealed a complex process of occupation, use,abandonment, and reoccupation at Mississippian moundcenters (Hally 1996; King 1996; Williams and Shapiro1996). These discoveries imply that aggregation anddispersal of populations was a common pattern sparkedby both sociopolitical and environmental factors.Political instability, created by disputes over successionto chiefly office or similar schisms between factions(Anderson 1994),was one possible cause of community

119

SOUTHEASTERN ARCHAEOLOGY 21(2) WINTER 2002 .

fission and population movement (Blitz1999).Mississip-pian agricultural practices may also have necessitatedfrequent site abandonment and population movement(Baden 1987).

Nevertheless, population movement and its conse-quences of rapid cultural change through replacement,contact, competition, and interaction are not yetadequately revived and integrated into models ofMississippian emergence and development. Interpreta-tions of Mississippian origins throughout the Southeastare still presented as a choice between local adaptation(e.g., Schroedl 1994) and population movement (e.g.,Williams 1994).The problem of "Mississippianization,"defined as "the process whereby regions were incorpo-rating general Mississippian traits," requires explan-atory frameworks that consider the local, regional, andinterregional movements of people, products, andinformation (Cobb and Garrow 1996:21-22).Potentiallyliseful concepts for achieving such a goal may be foundin anthropologically informed studies of abandonrrient,population movement, resettlement, culture contact,and ethnogenesis (e.g., Burmeister 2000;Hegmon et aL1998;Kopytoff 1987;Shennan 2000).

Mississippian Origins in the Lower Chattahoochee-Apalachicola River Region

. Explanations of Mississippian origins in the lowerChattahoochee-Apalachicola River region have long beenexpressed as a dichotomy between population move-ment and local development. In these debates, theemphasis is on spatially distinct distributions of sitesassociated with different ceramic styles. As other re-searchers have done, we must rely on ceramics andsettlement patterns as the primary evidence because noother categories of the archaeological record are suffi-ciently abundant for comparative purposes. As acaveat, we stress that ceramic style may not reflect thefull range of contact and interaction between groupsnor map neatly onto all social and political boundaries.But on the basis of considerable ethnographic evidence,we insist that ceramic style can reveal shared ordivergent social histories and material culture tradi-tions (Conkey and Hastorf 1990;Hodder 1985;Weissner1983; Wobst 1977):

Radiocarbon and stratigraphic evidence (Scarry 1990;Schnell et aL 1981) confirms that two arcl1aeologicalphases, Rood in the lower Chattahoochee River valley·and Cayson (also.known as Early Fort Walton) in thecontiguous Apalachicola River valley, are the oldestMississippian components in the region (Figure 1).

II' Rood and Cayson are the focus of a debate over thecultural dynamics responsible for initial appearance of

,I Mississippian chiefdoms in the region and the originsof the Fort Walton archaeological "culture." Radiocar-

bon dates from the lower Chattahoochee-Apalachicolaregion for Rood and Cayson phases are roughlycontemporaneous, from AD 1100 to 1250 (see Lorenzand Blitz 2002; Scarry 1990, and Schnell et aL 1981).Both Rood and Cayson populations established moundcenters at the beginnings of their respective'develop-mental sequences. Occupation spans and moundconstruction episodes are known for some centers, butothers remain undated. On the upper Apalachicola, theearliest Fort Walton phase, Cayson, is found at thesingle-mound centerS of Curlee (8JA7);Cayson (8CA8),and Yon (8LI2) (Scarry 1984, 1990; White 1982, 1996).Another Fort Walton phase is centered on the LakeJackson site (8LE1)in the Tallahassee Hills area (Griffin1950;Payne 1994).Rood has sometimes been includedwith Cayson as a regional phase of the Fort Waltonarchaeological culture (Scarry 1990; Scarry and Payne1986;White 1982).However, as we demonstrate below,these two archaeological constructs represent contem-poraneous populations with origins in different culturaltraditions.

Both culture history and processual perspectives haveshaped definitions of the Fort Walton concept. FortWalton was initially defined as a ceramic complex,period, and culture in northern Florida (Willey 1949).Similarities in material culture and the presence of largesettlements with platform mounds persuaded Willeythat Fort Walton was derived from Middle Mississip-pian. The term "Mississippian" had come into use as anexpedient label for late prehistoric agricultural societieswith similarities in settlement pattern, subsistence,material culture, and social organization, and distinc-tive regional expressions were identified (Griffin 1967).The primary criteria Griffin and others used to defineMississippian variants were variations in materialculture (especially ceramic styles) with distinct spatialdistributions. Midcontinental Middle Mississippianwas one such variant, often proposed as the develop-mental hearth in age-area explanations for pan-regionalcultural similarities (Smith 1984). Fort Walton culturebecame a Gulf Coastal Plain Mississippian variant.When later definitions of Mississippian placed greateremphasis on social organization (Peebles and Kus 1977)and subsistence practices (Smith 1978), definitions ofFort Walton culture were reformulated. Fort Waltonbecame a "Mississippian cultural adaptation," orga-nized as agricultural chiefdoms, and manifest archae-ologically in several distinct phases (Brose and Percy1978;Scarry 1981,1984;White 1982).

In two thorough syntheses of Fort Walton, Scarry(1984) and White (1982) proposed that Fort Waltonevolved in situ from indigenous societies thrqugh localadaptation. Population growth was identified as theprime mover for a cause-and-effect chain: resourcestress, maize intensification, societal reorganization,and managerial elites. They identified ceramic style

120

continuities that bridged these developments. Whiteapplied the term Early Fort Walton to components withthe earliest Fort Walton ceramic assemblages. Scarrypreferred to designate the same ceramic assemblages theCayson phase.! These ceramic assemblage·s are domi-nated by high frequencies of sand/grit temper, checkstamping, and the initial appearance of. the ceramictype Fort Walton Incised· (zone-punctated decoration

.usually applied to cazuela or shouldered vessel forms).Both Scarry and White argued that these earliest FortWalton ceramic types and modes were derived from theindigenous Late Woodland period Weeden Islandtradition antecedent to Fort Walton. In their studies,Scarry and White focused primarily on the Apala-chicola Rivet valley, but they extended the Fort Waltonconstruct and developmental scheme north to the lowerChattahoochee River valley to include a regionalconcentration of Mississippian sites known as the Roodphase.

Familiar .with the Rood phase from their extensiveexcavation of platform mounds at the Cemochechobeesite, Schnell et al. (1981)held a different view of Roodorigins. Unlike Scarry and White, Schnell and his asso-ciates did not interpret Rood as a regional phase of FortWalton nor did they suggest that Rood was derivedfrom Wakulla. Instead, these investigators consideredRood material culture to be derived from MiddleMississippian. Rood has in abundance what is rare inFort Walton contexts: shell-tempered jars fitted withhandles and incised with arcade or arch motifs. Becausethe nearest contemporary or antecedent MiddleMississippian phases are found to the north and west,they thought it prudent "to consider models involvingpopulation replacement in research bearing on the

. early Rood phase, even though such models are cur-rently out of vogue" (Schnell et al. 1981:244).In otherwords, Schnell and his associates considered RoodMississippian to be the product of an intrusive popula-tion movement with origins in the non-indigenousMiddle Mississippian tradition, and thus distinct fromFort Walton. Scarry and White, on the other hand, bothargued for local adaptation and cultural continuity withindigenous Late Woodland Weeden Island populationsas the primary explanation for Mississippian origins inthe region, and referred to these Mississippian compo-nents, including the Rood phase, as part of the FortWalton concept.

Scarry presented the problem of Mississippianorigins in the·lower Chattahoochee-Apalachicola Riverregion as two opposing hypotheses:

Hypothesis 1: The initial appearance of Mississippiansocietiesin the FortWaltonareaoccurredas a resultofaninvasion of the area by Middle Mississippianpopula-tions.Hypothesis 2: The initial appearance of Mississippiansocietiesin theFortWaltonareaoccurredasa resultofthe

.internalreorganizationofresidentWeeden·Islandgroups(Scarry1984:107).

~ ,

THE EARLY MISsisSIPPIAN FRONTIER IN THE LOWER CHATTAHOOCHEE-APALACHICOLA RIVER VALLEY

In posing the problem as a choice between intrusiveand indigenous origins, Scarry (and White 1982:26-30)was codifying three decades of debate.

Prior to the 1980s, the strongest proponents ofHypothesis 1 were Caldwell (1955), Sears (1967),,andJenkins (1978a, 1978b). In these arguments, sharedceramic styles (and,. to a lesser extent, other materialculture traits) were considered the primary evidence ofhistorical relatedness between archaeological phases.At Rood's Landing site on 'the lower Chattahoochee,Caldwell (1955) identified a stratified sequence withshell-tempered Moundville-like pottery in the lowestlevels, which he interpreted as evidence of a populationmovement from the west. In his influential synthesis,Caldwell (1958:65,77) identified Rood as "early Missis-·sippian," separate from Fort Walton, which he classifiedas "mixed Gulf and Mississippian." Rood was theproduct of a "Mississippian radiation" that represented"the arrival of a new people, bag and baggage, intovarious areas of the east" (Caldwell 1958:65). LikeCaldwell, Sears viewed Mississippian origins in thelower Southeast as a clear case of population movementby peoples whose history was distinct from local indi-genous inhabitants. Sears (1967:57-58)was impressedwith the abundance of Moundville-like pottery atseveral lower Chattahoochee sites, in particular, thehandled Mississippian jars with incised arcade motifs:shell-tempered Moundville Incised and its grit-tempered equivalent, Cool Branch Incised. Sears'sdefinition of the Cool Branch Incised type wasstimulated by his knowledge of Huscher's (1963)exca-vations at a lower Chattahoochee River site, CoolBranch, where Moundville Incised and Cool BranchIncised were prevalent in the deposits. Sears identifiedthe Moundville archaeological culture in west-centralAlabama as the most likely source of an intrusivepopulation that replaced the indigenous Late Wood-land Weeden Island peoples and initiated Fort Walton.Likewise, Jenkins (1978b:83) called attention to thesimilarity of the early Rood ceramic complex andBessemer phase ceramics in north-central Alabama, andproposed that Bessemer and Rood were part of a "waveof Mississippian colonists" ca. AD 1000.

In contrast, other investigators found evidence ofindigenous Fort Walton development. Fort Walton wassomehow "related ·to the older cultures of northwestFlorida," but also had "Middle Mississippian connec-tions" (Willey and Woodbury 1942:238).In the upperApalachicola valley, Wakulla Check Stamped, thecommon pottery type of Late Woodland Weeden Island,was found in the same deposits with early Fort Waltonassemblages (Bullen 1958). Continuity of ceramicmodes of decoration and form were also said to charac-terize the Fort Walton sequence at Lake Jackson in theTallahassee Hills (Griffin 1950).Such observations weredifficult to reconcile with population movement andreplacement. Reluctant to ignore a perceived Middle

121


Mississippian connection, a more indirect "influence"was proposed, "a pattern of diffusion-acculturation"(Willey 1953).It is important to note that archaeologistsfamiliar with the Rood phase in the lower Chattahoo-chee valley were advocates of intrusive Mississippianpopulation movement and cultural discontinuity, whilethose investigators working farther south in Floridasupported som.e form of indigenous Fort Waltondevelopment and cultural continuity with Late Wood-land Weeden Island popuiations.

This north-south geographical split in interpretationcontinued in the 1970s. Brose and Percy' (1978)formulated a cultural evolutionary model of autoch-thonous Fort Walton development that drew heavily onsystems theory and eco-demographic functionalism. Intheir view, local population growth was the primemover that stimulated indigenous maize intensifi-cation, concentrated settlements near productive soils,and created a need for hierarchical organization.Although they rejected migration as an explanation forthe origins of Fort Walton (Scarry's Hypothesis 1),Broseand Percy held open the possibility that under con-ditions of population stress, indigenous populationsmight adopt "Mississippian models" for subsistencepractices and social organization, acquired through diffu-sion from distant populations ..

Brose's students, Scarry and White, rejected Hypo-thesis 1 for several reasons:

(1) EarlyMississippianpopulationswere not numerousenoughto supplyimmigrantsfora rapid spreadintonewlands (Scarry1984:114).(2) TheEarlyFortWaltonceramiccomplexis sufficientlydistinctfromthepottery foundin other earlyMississippiperiod populations to precludedirectmigrationor iden-tify the parent Mississippianpopulation (Scarry 1984:147).(3) WakullaCheckStampedin EarlyFortWaltonassem-blagesdemonstratesregionalcontinuityfrom LateWee-den IslandintoMississippian(Scarry1984:147-148;White1982:223-224).(4) Potentialparentpopulationshad tobe olderthanRoodand Cayson,il'was claimed,in order to identifythem asthe source of colonists, but the available radiocarbondates suggestedRood and Cayson were of comparableage (Scarry1984:124-139).Many of these objections against support of Hypo-

thesis. 1 can be refuted and reconciled. First, Scarrycompared Early Fort Walton ceramics to pottery foundin other early Mississippi period phases: Moundville Iin Alabama, Macon Plateau in Georgia, Banks in centralTennessee, and Hiwassee Island in eastern Tennessee.In each phase, he found vessel shapes, rim forms, orcombinations of temper not presenfin Fort Walton, andconcluded that initial Mississippian populations in thelower Chattahoochee, Apalachicola, and Tallahasseeareas did not originate in these distant locales. In orderto support the population movement hypothesis,Scarry insisted that an intrusive ceramic complex,

should duplicate the ceramics of the parent culture.However, this expectation of precise replication of theparent culture in the realm of material goods is.umealistic for pre-industrial pioneer groups (Adams1968;Shennan 2000).Stylistic similarity should be close

. at the time of initial settlement, but divergence fromprototypes will occur under conditions of reducedcontact with the donor culture, different environments,contact with new peoples, and cultural drift (e.g.,Clarke 1968:299-327)..

Certainly, some of the ceramic complexes Scarryexamined do not resemble Early Fort Walton pottery(i.e., Hiwassee Island, Macon Plateau), but his rejectionof Hypothesis 1 was premature because he was un-aware that early Rood sites such as Cool Branch haveceramic assemblages very similar to Middle Missi-ssippian pottery elsewhere and dissimilar to the FortWalton ceramic assemblages on which he based hisdefinition of Fort Walton archaeological culture andphases. Scarry relied primarily on Cayson and otherEarly Fort Walton ceramic assemblages in the Apala-chicola River valley for his ceramic comparisons andassumed that early Rood materials were equivalent.They are not. This failure to recognize that Rood andCayson were rooted in separate material culturetraditions renders the four reasons for rejection ofHypothesis 1 persuasive only when restricted to theEarly Fort Walton Cayson phase of the Apalachicolavalley of Florida. The Rood phase in the lower Chatta-hoochee River valley to the. north is quite a differentmatter.

Knight (1980)noted the lack of antecedent indigenousstyles in Rood pottery at Cemochechobee and identified"specific ceramic type equivalences" that connectedRood to the Middle Mississippian tradition. From theselinks and other similarities of material culture, heconchided that Rood' was most likely founded by apopulation movement from the west. Knight (1980:2)did not identify a specific donor source for the Roodceramic complex, nor did he think it necessary in orderto establish Rood's non-local origins. Proponents ofHypothesis 2 countered that Rood ceramic modes were"eclectic," not homogeneous, and so unlikely to havebeen introduced by immigrants from a single source(white 1982:236-237;Scarry 1984:152).Whether or notthere were single or multiple external sources, what isimportant to the debate, it seems to us, is that Roodceramic style (and additional material culture, such aswall trench architecture and bastioned palisades) is theproduct of a non-local tradition.

Scarry (1984:155)did not deny Knight's evidence ofMiddle Mississippian input altogether, but attributedstylistiCsimilarities to diffusion, wl1ichhe dismissed asunimportant. The external links in material culturecarried no causal significance because "the derivationof a specific ceramic type (e.g., Cool Branch Incised) is

122


Rood and Its Predecessors

Figure 2. Averett (top), Rood (middle), and Wakulla (bottom)ceramicvessels.

What evidence is there for cultural and demographiccontinuity between the Rood population and the non-Mississippian indigenous populations that precededthem? In the lower Chattahoochee River valley, twoindigenous, non~Mississippian archaeological culturespreceded Rood: Averett and Wakulla Weeden Island.These archaeological constructs are identified asspatially distinct distributions of sites associated withdifferent ceramic complexes. In other words, Averett,Wakulla, and Rood ceramic assemblages have differentcharacteristics of form, decoration, and· temper (Figure2).

most probable mechanism for Cayson local develop-ment. Instead, another model (also untested) is equallyplausible: the arrival of Mississippian pioneers in thelower Chattahoochee River valley created a securitythreat for neighboring indigenous groups in the adja-cent Apalachicola River valley, whose· collective responsewas to emulate newcomer technology and organizationin a proces·s of secondary chiefdom formation.

not necessarily the same as the derivation of a culturaladaptation featuring hierarchical organization andsubsistence economy based on intensive agriculture;certainly the ceramic type is not equivalent to the adap-tation" (Scarry 1984:110). We agree, but must point outthat if continuity in ceramic style is used as the primaryevidence for linking Late Woodland Weeden Island toEarly Fort Walton, as Scarry (1984:147-148) and White(1982:29) attest, then a similar continuity between Roodand external sources (and discontinuity between Roodand Weeden Island) must be considered in any expla-nation for the origins of the Rood phase populations.

Finally, two additional objections by· Scarry andWhite to intrusive Mississippian origins can be rejectedor reconciled with Hypothesis 1. It is contradictory toargue that population growth was insufficient toproduce Mississippian immigrants for populationmovements, but necessary and sufficient to transformindigenous Late Woodland peoples into Fort Walton.Mississippians abandoned settlements and moved forreasons having little apparent connection to internalpopulation growth or resource stress, such as attemptsto resolve factional disputes or efforts to form inter-group alliances (Blitz 1999). We agree with Scarry thatone supposed criterion of population movementidentification that cannot be demonstrated with pre-cision at this time is independent dating to show thatthe potential Mississippian source cultures are olderthan Rood or Cayson (Scarry 1984:124-139). The. av<i.il-'able absolute dates indicate broad temporal overlaps,however, so direct cultural ties through populationmovements were possible. If we restrict our focus tothose aspects of material culture often identified withMiddle Mississippian-specifically, shell-temperedpottery and wall-trench architecture-then these traitsare known to occur at an earlier date outside the lowerChattahoochee-Apalachicola' valley (Lafferty and Price1996; McNutt 1996). The remaining factor cited byScarry and White in rejecting Hypothesis .1, the pres-ence of Wakulla Check Stamped in early Fort Waltonassemblages, is applicable only to the Apalachicola region.

In rejecting Hypothesis 1, Scarry and White failed to.realize just how different Rood is from Early FortWalton Cayson. They dismissed the presence of MiddleMississippian cultural elements in Fort Walton com-ponents as having little or no direct bearing on ques-tions of culture process. Consequently, they defined Roodas a local phase of Fort Walton culture and rejectedprotests to the contrary. However, the Rood and Caysonphases originated through· different mechanisms ofchange: the origin of Rood is best explained bypopulation movement (Scarry's Hypothesis 1) and theorigin of Early Fort Walton Cayson fits the evidence forin situ or local development (Hypothesis 2). UnlikeScarry and White, we do not think that the untestedmodel of population-resource stress is necessarily the

123

The northern distribution is known as Averett,radiocarbon dated AD 900-1300 (Chase 1959, 1963; Led-better 1997). Averett settlement is limited to the upperportion of the lower Chattahoochee River valley, wheresites include large, midden-rich artifact and featureconcentrations that suggest intense or long-termhabitation. Small triangular projectile points and maizeremains are present. Apparently, Averett populationsdid not create fortified mound centers in the·Mississippian pattern. Averett mound construction hasbeen found at one site (lRU61), a low mound ofuncertain function capped by the mound stages of alater period (Schnell and Wright 1993:29). On currentevidence, it appears that the Averett settlement patternis differentiated in terms of site activities, but definitiveevidence of political centralization or chiefdom forma-tion is lacking.

Averett ceramics are mostly undecorated sand-tempered vessels: small semi-conical jars with slightlyincurved rims, larger semi-conical and globular jarswith restricted necks, large shoulder nodes, and flaringrims, and hemispherical bowls. Brushing, rectilinearincising, punctation, and red filming are rare decorativetreatments (Ledbetter 1997; Schnell 1998). In sitediversity, subsistence, and technology, Averett does notappear radically different from terminal Late Woodlandphases (e.g., Woodstock, Coker Ford) in adjacentupland Georgia or Alabama. But perhaps because of alate temporal placement, maize production, and jarswith superficial similarities to Mississippian forms(although they lack handles), Averett has beencharac-terized as "quasi-Mississippian" (Knight and Mistovich1984:223) and "Emergent or Early Mississippian"(Garrow and Associates 1995:129).

Several Averett components have yielded smallquantities of Rood pottery or copies of Rood forms(Chase 1963; McMichael and Kellar 1960:212-214;Schnell1998:113), although Averett ceramics are not present inRood phase assemblages. Etowah Complicated Stampedpottery has been recovered in several Averett assem-blages (Chase 1963; Gresham et al. 1985:239-244;Schnell1998:113), but it is not found in Rood assemblages.Averett populations were clearly engaged in some formof interaction with Rood and Etowah phase popula-tions.

South of Averett territory was another population, .Wakulla Weeden Island, radiocarbon dated AD 750-lobo in Florida (Milanich 1994; Scarry 1984,1990; White1982). Better known and more widespread than Averett,Wakulla is the terminal expression of the indigenousWeeden Island cultural continuum in northwesternFlorida and the Apalachicola River valley. In the lowerChattahoochee River valley, Wakulla reaches a northern

·limit of distribution. Like Averett, Wakulla populationshad maize, small triangular points, a non-hierarchicalsettlement pattern, and no indicators of chiefdom orga-niza tion.


Wakulla ceramics are distinct from Averett. The mostcommon vessel forms are sand-tempered, semi-conical orglobular jars with distinctive folded rims. Check-stamped(the pottery type known as Wakulla Check Stamped) orplain surfaces dominate Wakulla ceramic assemblages,but a minority of incised and punctated decorations maypersist from earlier periods. Check-stamped pottery, withthe checks partially obliterated, was identified at severalsites along the lower Chattahoochee River in the 1960s.This pottery was dubbed the "Cat Cave complex" (Kellyet al. 1962) and tentatively assigned to the Late Wood-land period. Now regarded as a local variety of WakullaCheck Stamped (Knight and MistoviCh 1984:222;Schnellet al. 1981:240; Schnell 1998:111-112),these componentsare included here as Wakulla. No site has producedAverett and Wakulla pottery in association. A correctedradiocarbon date of AD 950 ± 140 (Mielke and Long1969:166) from the Lynn's Fish Pond site (lBR21) is theonly dated Wakulla phase component in the lowerChattahoochee valley. The corrected age of this woodcharcoal sample can be calibrated (Stuiver and Reimer1993), with a two-sigma standard deviation, to AD 778-(1037,1143,1148)-1298. Thelarge standard deviation forthis date makes it uncertain if the end of the Wakullaphase overlaps in time with the beginning of the Roodphase.

Early Rood ceramics replicate Middle Mississippianvessel shapes and decoration distributed from theOhio-Mississippi confluence to central Alabama. Themost common vessel form is a wide-mouth globular jarwith a restricted neck, straight-to-flaring rims, and loopor narrow strap handles with nodes. When decorated,these handled jars are incised with simple arcade orarch motifs. Also present are open bowls with effigy rimadomos and lugs (animal heads and tails), shallow bowlsor plates, beakers and bottles. Rood pottery is temperedwith shell, sandi grit, or a combination of both. Theseattributes of temper (especially shell temper), decoration,and form are very distinctive, and render the Rood cera-mic style instantly recognizable as the same tradition thatspawned such Mississippian phases as Jonathan Creek inKentucky (Webb 1952);Obion in Tennessee (Garland 1992);Owl Creek, Mississippi (Rafferty 1995);and Summerville I(Blitz 1993), Bessemer (DeJarnette and Wimberly 194"1;Welch 1994) and Moundville I (Steponaitis 1983; Knightand Steponaitis 1998) in Alabama.

Concurrent with the appearance of the Middle Missis-sippian ceramic tradition in these regional phases iswall-trench architecture. Buildings with wall trenches

. are present in early Rood contexts at the Cool Branch(Lorenz and Blitz 1999), Cemochechobee (Schnell et al.1981), and Singer-Moye sites (Schnell and Wright 1993).No wall-trench buildings have been found in Averett orWakulla contexts, We consider improbable the pan-regional dissemination of wall-trench architecture by themechanisms of long-distance diffusion or independentinvention. Wall trenches are signatures of a structurally.

124

I

I


In the lower Chattahoochee River valley, the spatialdistribution of Averett, Rood, and Wakulla sites hasimportant implications' for the identification of aMississippian population movement. As mentionedpreviously, the beginning dates for Averett and Wakullaare both older than the beginning date of Rood,suggesting that these populations inhabited the regionprior fo Rood populations. However, Averett, Rood,and Wakulla sites are found in different physiographiczones along the lower Chattahoochee River (Figure 3).Distinct site distributions might be related to the poten-tial of zones to sustain different subsistence strategies(Schnell and Wright 1993:4-11), or the presence ofindigenous inhabitants in those zones may have Figure3. Distributionof Averett,Rood,and Wakullacom-restricted the settlement choices of immigrant popula- ponents across five physiographic zones of the lowertions. If the initial appearance of Mississippian sites ChattahoocheeRivervalley:A=LowerPiedmont,B=FallLineoccured. in an unpopulated location with few or no Sand Hills, C=ChattahoocheeRed Hills, D=FallLine Redcontemporary.or immediately antecedent indigenous Hills, E=DoughertyPlain.The PelhamEscarpment(F)andsites, the case for intrusive settlement is strengthened TallahasseeHighlands(G)formthe southernboundary.

125

because there was no pre-existing local population basefor cultural continuity.

A frequency histogram (Figure 4) illustrates the num-ber ofAverett, Rood, and Wakulla components recordedin each physiographic zone along the floodplain, fromOliver Reservoir in the north to the Chattahoochee-Flintriver confluence (Lake Seminole) to the south. Compo-nents were identified from state site files and inven-tories compiled through CRM surveys, including sitesrecorded prior to reservoir construction (Belovich et a1.1982: Tables 10, 11; Chase 1963; Garrow and Associates'1995:45, Figure 11; Knight and Mistovich 1984: Tables3.2, 3.3, 3.4, 8.1; Ledbetter 1997:260; McMichael andKellar 1960: Tables 19,20; White 1982: Tables 11, 12, 14).Interfluvial site locations away from the valley are exclu-ded from the sample due to inadequate survey coverage;thus, any interpretations drawn from these site distri-butions apply only to floodplain locales.

• Averett* Rood• Wakulla

,,./

F .//

" G"

.•...

.....•..

o 40km~

N

" ...

"-\ \\ "-

"\ \\. \

"-\

" " .•.. .•..Mississippian Population Movement into a Frontier

specific architectural form, the details of which werenot likely to be easily transmitted by casual observationor indirect sources. Furthermore, it has long been esta-blished that folk architecture carries a high symbolicload often emblematic of cultural identity and, there-fore, resistant to change (Deetz 1977; Glassie 1975). Forthese reasons, we think wall-trench architecture was oneelement of the cultural blueprint carried by the MiddleMississippianpioneers who initiated the early Roodphase.

Viewed from the lower Chattahoochee River valley,the Mississippian tradition that most resembles Roodextends north and. west, and cultural elements thatcoalesced into this new way of life-fortified politicalcenters with platform mound-plaza arrangements, walltrench architecture, similar shell-tempered vesselforms, inferred chiefdom organization-have contem-porary or antecedent links outside the' region. Averettand Wakulla have none of these cultural elements.Early Rood and the Mississippian phases mentionedabove are pan-regional expressions of shared materialculture; these phases appear as widely spaced settle-ment dusters with greater material culture similarity toeach other than to the local indigenous traditions that'preceded them. Such a situation suggests that thesefar-flung populations were composed of communitieslinked together by shared histories. Although wedisavow the simplistic notion that all Middle Missis-sippian populations or innovations originated at asingle geographical homeland or point of dispersal,long-distance diffusion or local adaptation actingindependent of other populations may not be the bestexplanation for such striking similarities in materialculture~In many cases, such as the Rood phase, popu-lation movements are implicated .


Figure4. Number of recordedcomponentsin physiographiczones,lowerChattahoocheeRivervalley.

Averett components are concentrated in the physio-graphic zones on either side of the Fall Line at the nor-thern end of the valley; Wakulla components dominatethe two southernmost zones. Only the centrally locatedChattahoochee Red Hills has both Averett and Wakullacomponents, including the southernmost Averett site(9SW124, the· only Averett site in this zone) and thenorthernmost Wakulla site (lBR27); the two sites areseparated by 5.6km. There are only six Wakulla compo-nents in the Chattahoochee Red Hills zone (includinglBR27), widely dispersed along the river for a distanceof approximately 50 km. Farther south, the number anddensity of Wakulla components increase dramatically.The Chattahoochee Red Hills is clearly peripheral toboth Averett and Wakulla settleinent distributions, butit contains the majority of Rood components.

The peculiar concentration of Rood sites in thephysiographic zones between the demographic centers

. of Averett and Wakulla is further revealed when thesesame data are plotted on a regional map (Figure 3). Asnoted above, Averett and Wakulla site distributionsexhibit no spatial overlap. Much of the ChattahoocheeRed Hills appears to have been an unirihabited orunderpopulated "buffer zone" (Hickerson 1965)betweenthe hcvopopulations. Rood components are clustered in

. the spatial gap between Averett and Wakulla territories.Given these site distributions, it is unlikely that Rood

"emerged" in place as an autochthonous transformationof indigenous Averett or Wakulla parent populations.If Rood were merely a local expression of Fort Waltonand thus the product of internal demographic growthout of local Late Woodland populations, as the adap"tationist model would have it, then we would expectRood mound centers to develop in the areas most

,I heavily occupied by Averett or Wakulla. That did notII happen. No early Rood mound center that has been'I investigated in any detail has evidence of Averett, occupation. Wakulla Check Stamped is absent from

excavation samples at such Rood mound centers asCool Branch, Mandeville (Mound A cap), Singer-Moye,

and Rood's Landing, and is present only in tracequantities in mound-fill contexts at Cemochechobee(7 of 33,320 sherds, Schnell et al. 1981:240).Figure 3reveals some spatial overlap of Rood components withAverett and Wakulla distributions, but it is doubtfulthat these specificcomponents were contemporary. Roodpopulations probably did not occupy the Fall Line (a'common pattern for Mississippia,n sites elsewhere inGeorgia) because Averett groups were already there. Itis improbable that Averett was the original donor popu-lation for Rood because radiocarbon dating and rareexamples of exchanged or copied Rood pottery in Averettcontexts indicates that Averett and Rood were contem-porary (Ledbetter 1997). If Averett had been the donorpopulation for Rood, then we would have to explainwhy some Averett groups retained their traditionalmaterial culture in their northern home territory whileothers moved a short distance to the south anddeveloped into Rood Mississippian. If Wakulla was thedonor population for Rood, then we must explain whyRood did not retain Wakulla-derived material culture(check-stamped pottery), as did their supposed EarlyFort Walton Cayson phase brethren to the south.

We think the lower Chattahoochee River valley was aMississippian frontier, a geographical space betweensettlements as well as a sociopolitical space where con-tact between distinct cultural groups could be initiated,or in some cases, avoided. Frontiers are visible inmaterial culture distributions and settlement patterns.Frontiers may mark social and political boundaries, butthese are hard to measure archaeologically. In this case,Rood sites appear without local precedent between thedemographic centers of Averett and Wakulla· settle-ment. In the Chattahoochee Red Hills, where Rood com-ponents cluster, there was no antecedent populationsource for local Rood emergence. The parsimoniousexplanation is that Rood populations did not originatelocally, but instead belonged to a tradition that"emerged" elsewhere. It was the movement of Mis-sissippian pioneers bearing the products of this non-local tradition into a frontier that created the initialRood archaeological remains.

The Cool Branch Site:A Mississippian Frontier Settlement

Located in the pn~-existing buffer zone orunirihabited gap between Averett and Wakulla sitedistributions, the previously mentioned Cool Branchsite (9QU5) represents an early Rood phase frontiersettlement (Lorenz and Blitz 1999). An 850-m longpalisade with tower bastions spaced 35 m apart waserected around a single platform mound early in theRood sequence (Figure 5). Remains of wall-trencharchitecture, pre-mound structure patterns, and many

o

~hysiographic Zones

78

Sand lIill~ Chnunhoochce Red Hills fall Line Red lIill~ Dougherty Plain

20

RO

60

~B 40

8-"6u

126


Rood and Fort Walton

The principal evidence marshaled by proponents ofindigenous Fort Walton origins is the presence of'Wakulla Check Stamped in both Wakulla and Early FortWalton assemblages. In the Apalachicola River valley,the appearance of Mississippian-derived ceramic stylesidentical to those of Rood, in assemblages otherwisedominated by Wakulla Check Stamped (and FortWalton Incised), is diagnostic of the Early Fort WaltonCayson phase (Scarry 1984;White 1982).Such is not thecase in the lower Chattahoochee region, where WakullaCheck Stamped pottery is rarely found in Rood assem-blages, and then only in small quantities. A few otherexcavated sites have produced both Rood ceramics andWakulla Check Stamped ceramics, but in every case thecontext is such that it is uncertain if the artifacts belongto the same or different components (Schnell et aL 1981:240). In short, the high frequency. of Wakulla CheckStamped in Early Fort Walton Cayson assemblages isinterpreted as evidence of cultural continuity by inves-tigators in the upper Apalachicola River valley, but inthe lower Chattahoochee River valley there is no evi-dence of ceramic style continuity between Rood, Wakulla,or Averett.

The debate over the place of Rood and Early Fort Wal-ton Cayson in regional Mississippian origins has lacked

. comparison of ceramic attribute frequencies between. the two phases. With the competing origin hypothesesin mind, ceramic attribute frequencies in provenienceunit samples at regional mound centers are comparedto (1) clarify the distinctive composition of early Roodand Cayson ceramic complexes, and (2)identify temporal-spatial patterns in ceramic types or attributes that implycontact and interaction between early Rood and Cayson.

a Feet 100>-----<o MeIers 30

"- Palisade

•.•• 4- --&-~ OJ

MoundEx.cavations -.

" ~,

Q,/~

<l.

~j)""/~~II1/11

1,'11///

{II{J/1/1111

/10/1,1ft"II'

// If1

111

'Bulldozed'Trenches

specifics of material culture are similar or identical tocounterparts found at Cemochechobee, and at MiddleMississippian sites north and west of the lower Chat-tahoochee valley. Indeed, these external similarities aremore pronounced at Cool Branch than at Cemoche-chobee, which is located 27 km to the south. Thepredominance of shell-tempered jar attributes, such asthe incised arcade motif on lobed jars, and identicalnoded narrow-strap handle forms, bears a strikingresemblance to ceramic assemblages from the Bessemerphase (DeJarnette and Wimberley 1941: Figure 62;Welch 1994:22),Moundville I phase (Steponaitis 1983:323c326)and Langston phase (Walthall 1980:200-205).These ceramic attribute similarities reinforce the impres-sion of an early Mississippi period chronologicalplacement (ca. AD 1100-1250)for the Cool Branch site.In fact, two radiocarbon samples' have been dated: onefrom the pre-mound ceremonial precinct (Beta-130241,840 ± 50 BP,wood charcoal, ~\3C = -25.9, cal AD 1040-(1210)-1280 at two-sigma); and one from an earlymound building construction.stage (Beta-130242,740 ±50 BP,wood charcoal, ~\3C = -26.9, cal AD 1210-(1280)-1300at two-sigma). Thus, both dating ranges fall rough-ly within a two-sigma calibrated (Stuiver and Reimer1993)range of AD 1100-1250,supporting an early Roodphase assignment for the site.

The effort devoted to palisade construction at Cool. Branch is impressive. The multiple-mound centers ofRood's Landing and Singer-Moye also have evidencefor fortifications, but these features remain undated.Fortifications have not been found at Averett or Wakullasites. Fortification buildiflg may have played a criticalrole in successful Mississippian colonization. Becausethe Cool Branch settlement was established early in theRood sequence, and because it exhibits a material cul-ture without regional antecedents, perhaps the Cool'Branch site represents a planned effort to establish afortified community of immigrant settlers in a frontierbuffer zone between the indigenous settlement clustersof Averett to the north and Wakulla to the south. Thereis no Averett or Wakulla pottery at Cool Branch. Efforts.to expel the invaders were unlikely to succeed oncesuch a fortress was erected, because fortified defendersheld a tactical advantage over attackers in bpw warfare.(Hassig 1997), and there is no evidence that South"eastern peoples practiced siege tactics (Stout and Lewis1998:175). Territorial acquisition would be the endresult of these conditions and the indigenous people tothe north and south of Rood phase settlements wouldbe forced to adjust to the new situation. Viewed in thisway, fortification building was more than a passivedefense. Instead, we must consider the possibility thatpalisade. construction was employed as a strategictechnology to acquire and hold new territories on theMississippian frontier .

. .Figure 5, Plan view of the Cool Branch site (9QU5).


As other investigators have done, we assume that atraceable continuity in the use of ceramic temper,decoration, and vessel shapes (as expressed in ceramictypes) from earlier times through the advent of Roodand Fort Walton indicates some degree of cultural con-tinuity in the region. The ceramic samples are drawnfrom these sources of early Rood and Cayson assem-blages: Cayson (Scarry 1980), Cool Branch (this study),Curlee (White 1982), Mandeville (Kellar et al. 1961),Omussee Creek (Neuman 1961; this study), Rood'sLanding (Caldwell 1955; Schnell 1977, 1981), Singer-Moye (Knight 1979),and Yon (White 1996).

In Table 1, we compare the four most commondecorated types in contemporaneous early Rood andCayson assemblages. Shell-tempered Moundville Incisedand its grit-tempered equivalent, Cool Branch Incised,are characterized by the incised arcade decorationfound on standard Mississippian jars with handles.These types dominate early Rood, but occur in lowfrequencies in Cayson. Fort Walton Incised (zonepunctation) and Wakulla Check Stamped are absent orrare in early Rood assemblages, but are the decoratedmajority types in Cayson. Clearly, northern Roodcenters and southern Cayson centers have roots inseparate material culture traditions, but. there isevidence of interaction .

.Stratigraphic' superposition and radiocarbon dating(Caldwell 1955;Knight 1979;Schnell et a1.1981)supportthe contention that, as a general temporal trend,assemblages with higher frequencies of shell temper areolder than those assemblages with low f~equencies.Shell-tempered pottery is found in the earliest Caysondeposits. Most likely, shell temper and the standardMississippian handled jar passed from early Rood toCayson,but shell temper was never as popular in Caysonas in early Rood. Although Early Fort Walton Cayson

populations did adopt Mississippian jars as a minorityware, most were made with indigenous temperingagents, not shell temper. The origins of the type FortWalton Incised are unclear, but the definitive zoneC

punctate treatment has deep antecedents in the Gulftradition (Caldwell 1958). The fact that Fort WaltonIncised diverges in temper, form, and distribution fromRood ceramics implies that it was not derived from thatsource. Hybrid combinations of intrusive and indigenousceramic modes, such as handles added to indigenousvessel forms or shell-tempered equivalents of thedecorated types Fort Walton Incised or Wakulla CheckStamped, are not in evidence. Without any such hybridsthat could be interpreted as developmental prototypes,there is no basis to argue that the handled jar form orshell temper originated as localized in situ innovations.For the most part, Mississippian ceramics (primarily thehandled jar form) were introduced into the Wakulla"Fort Walton continuum from elsewhere, with littlealteration except the use of tempering agents other thancrushed shell. In the realm of ceramic style, early Roodcultural boundaries were apparently rather resistant tostylistic inputs from Early Fort Walton, whereas EarlyFort Walton cultural boundaries were more permeable.The relatively low-level, one-sided style transmissionbetween early Rood and Cayson suggests the relation-ship was one of "negative reciprocity" (Hodder 1985).Even if the two groups were mutually antagonistic,transmission of ceramic styles could have occurredthrough marriage alliances or abduction of potters.

In addition to the temporal dimension, there is spatialpatterning in the frequency of shell-tempered pottery inthe lower Chattahoochee-Apalachicola region. Forexample, when ceramic samples from early Rood phasemound centers are compared, a north-south spatial clinein the frequency of shell-tempered pottery is evident:

Table1. Frequency(%) ofthemostabundantdecoratedpotterytypesinearlyRoodand Caysonphasesamples.

Site/Provenience Moundville CoolBranch FortWalton WakullaCheck TotalNIncised Incised Incised StampedEarly RoodRood'sLanding 100 0 0 0 3Pre-MoundD

CoolBranchPre- 83 12 3 0 135Mound& Mound

Mandeville 73 20 7 0 44Mound A Cap

Singer-Moye 67 22 0 0 8. Mound COmusseeCreek 0 100 0 0 2Pre-Mound

CaysonCurlee 0 1 20 79 701Yon 3 13 51 34 104CaysonLower 9 0 76 15 33CaysonUpper 4 2 92 1 133

128


highest frequency in the north at Rood's Landing andCool Branch, lower frequency at Mandeville andSinger-Moye to the south and east (Figure 6). We can beconfident'that these frequency changes over geographicspace are not the result of change over time becausethese lower Chattahoochee mound assemblages con-tain relatively high frequencies of Moundville Incised, akey marker type for the early Rood phase. Becauseearly Rood ceramic assemblages are most like Missis-sippian phases to the north and west, this patternsuggests that the point of initial Rood immigration intothe. iower Chattahoochee would have been in thenorthern valley below the Fall Line. Perhaps these nor-thern Rood sites maintained greater contact with theircommunities of origin to the north or west and had lessdirect interaction with early Fort Walton peoples to thesouth. Compare early Rood assemblages with contem-porary Cayson assemblages farther south in the upperApalachicola valley; shell tempering is present in Cayson,but in lower frequencies.

It is important to note here how confusion regardingassignment of the Rood phase to the Fort Walton culturecame about in the first place. Researchers recognizedrelatively high frequencies of Fort Walton Incisedpottery in some lower Chattahoochee valley mound site

assemblages, but (before this study) fine chronologicalcontrol was not yet established. Therefore, the Roodphase was linked to Fort Walton culture by virtue ofsimilar ceramic styles (i.e., the Rood phase incised-arcadetypes Moundville Incised and Cool Branch Incised, andthe Early Fort Walton Cayson phase zone-punctatedtype, Fort Walton Incised) found at the same site. Butwhen each lower Chattahoochee valley mound siteceramic assemblage was examined in stratigraphicorder, it became clear that Fort Walton Incised did notbecome a common type in the lower Chattahoocheevalley until the Singer phase (ca. AD 1350-1450) (Lorenzand Blitz 2002; Schnell and Wright 1993), when itoccurred with Lamar Plain, a rim mode of notched orpinched applique strips of clay added to the exterior ofthe vessel. Moundville Incised and Cool Branch IncisedtyPes declined to low frequencies. After ca. AD 1400,the stylistic vestiges of Rood's Middle Mississippianorigins were replaced by the continued popularity ofFort Walton Incised and an infusion of Lamar (SouthAppalachian Mississippian) ceramic styles from theeast. So unlike sites in the Apalachicola River valley,relatively high frequencies of Fort Walton Incised in thelower Chattahoochee River valley are a post-AD 1400phenomenon ..

NON-SHELL SHELL.

Rood's Landing '" .EARL Y Pre-Md D '40% 60%.ROOD N=60

Cool BranchPre-Md & Md 38% 62%

N=4073MandevilleMd A cap 59% 41%

N-IOO6Singer-MoyeMdC 60% 40%

N-452 •Omussee CreekPre-Md 58% 42%·

N-249. II

Cayson Lower .

cAYSON N=298 79% 21%·Cayson Upper. Is

. . ..

I IN=I709 90% 10%Curlee IN=3633 94% 6%Yon

I •..... ...i§i~/o\N=1420••

.. ', p. 3%Ii

Figure6. Frequency(%) of shell-temperedpottery.and non-sheil-temperedpottery in earlyRood and Caysonphase samples.

129

III


Conclusions

The campeting hypatheses af papulatian mavementand lacal develapment have lang been prapased to.explain Mississippian arigins in the lawer Chatta-haachee-Apalachicola River regian. Invaking Rause's(1958) criteria far evaluating evidence af papulatianmavement, the hypathesis that papulatian mavementsinto. the lawer Chattahaachee River valley establishedthe Mississippian Raad phase cauld nat lJe rejected farthe fallawing reasans:

(1) There is regianal discantinuity in material culture;Raad ceramicstyle and architecturehave no.precedentinthe area af Raad settlement.(2) Raad site distributians reveal a demagraphic discan-tinuity with antecedent lacal papulatians because Raadsettlements·were established in an uninhabited area.between indigenaus papu·latians.Raad sites rarely haveindigenausAverettar Wakullacampanents.(3) In a general manner, the extra-regianal saurceterritary far Raad Mississippiansettlers is identified anthebasis af similaritiesin materialcultureand settlementpattern.These culturalelementsappear first in thesaurceterritary.(4) Thearganizatianalcapabilityaf erectingfartificatiansto. defend settlements is evidence af an adaptiveadvantage that allawed the immigrants to.successfullydisplacear campete with indigenaus populatians in thenew territary.In cantrast to. Raad arigins, the hypathesis that the

Early Fart Waltan Caysan variant af Mississippian ori-ginated from intrusive papulatian mavement can berejected for the fallawing reasans:

(1)There is regianal cantinuity in material culture withantecedentWakullapapulatians, primarily in the farm afceramicmades af decoratian,farm, and temper.(2)Fart WaltanCaysan site distributians reveal a dema-graphiccantinuitywith antecedentWakullapapulatians.EarlyFortWaltanCaysan sites cammanlyhave Wakullacampanents.

Additianal evidence was presented which suggests thatcontact with Mississippian pianeers had a significant,lang-term develapmental impact anindigenaus LateWaadland papulatians:

(1) Regianaltemparal-spatialpatterns in ceramicmadesaf decaratian,vesselfarm, and temper indicatea limited,asymmetricaltransmissianbetween intrusive Mississip-pians and indigenaus Wakullagroups.Aseriatian af cer-amic samples fram regianal maund centers is presentedto. illustratethis pattern.(2) The relative lack af shared ceramic. mades isinterpreted as negative recipracity ar a law level afinteractian between intrusive Mississippians andindigenausWakullaand Averettgroups.Weconclude that Raad originated as a series af accupa-

tians by Mississippian immigrants in a vacant ar sparselyinhabited frontier zane between Averett and Wakullasettlement clusters. Aspropased by Scarry, White, andathers, the Fart Waltan archaealagical culture appears

to.represent the rearganizatian af indigenaus, Wakulla-derived papulatians into.maund centers, accampaniedby bath change and cantinuity in the ceramic camplex.Did the intrusian af Raad settlers have lang-term devel-apmental cansequences far the neighboring indigenausinhabitants? Unfartunately, the· data are taa fragmen-·tary for definitive answers. There has been less exca-vatian at Averett sites, so.we knaw relatively little abautthe Averett respanse to.the expanding Raad papulatian,but we can tell that it was quite different from theWakulla-ta-Fart Waltan transfarmatian in the sauth. No.significant changes in Averett material culture ar set-tlement pattern have been identified. Perhaps Averettceramic style was emblematic af cultural identity andthus resistant to. external influence. Patentially, Averettaccupatian af the :FallLine was af strategic impartanceto. intermediate trading partners with Mississippianchiefdams lacated up and dawn the ChattahaacheeRiver. Based an radiacarban determinatians, Averettpapulatians co.-existed with Raad populatians, but thetwo. groups do. nat appear to. have had many inter-actians, as judged by the law frequencies af Raad phaseceramic types within the Averett assemblages and thetatal absence af Averett types in Raad assemblages.Hawever, Averett peaple do. appear to. have beenclasely affiliated with papulatians lacated to. the narththat used complicated stamped pattery, arid perhapsthis narthern cannectian was an alliance farged as adefense against intrusians fram Raad papulatians.Whatever the case, Averett culture disappeared from.the archaealagical recard around AD 1300.

Unlike Averett, the Wakulla-derived Early Fart WaltanCaysan papulatians underwent profaund changes.Because initial Raad settlement and the Wakulla-ta-Caysan phase transitian appear clase in time and space,it seems to. us that Mississippian papulatian mave-ments and the rearganizatian af indigenaus, Wakulla-derived papulatians need nat be viewed as mutuallyexclusive ar unrelateddevelapments. At the verybeginning af the Fart Waltan sequence, when the Wakulla-derived Early Fort Waltan Caysan papulatians esta-blished their maund centers, there was an injectian afMiddle Mississippian material culture into. the Apala-chicola River valley: minar quantities af shell-temperedMississippian jars are faund in the lawest levels afCaysan phase depasits (Figure 6 and Table 1; also. seeWhite 1982:242-243).On present evidence, Caysan centersdid nat farm prior to. contact with Middle Missis-sippians. Raad and. Early Fart Waltan Caysan mayrepresent different develapmental paths to. Mississip-pian, linked tagether in a sequential cause-and-effectrelatianship through cantact and interactian across acornman frontier. The intrusian af campetitive Raadpapulatians may be the primary causal factor in FartWaltan chiefdam formatian. Perhaps Wakulla papula-tians centralized into. maund centers and initiated the

130


Acknowledgments. An earlier version of this paper waspresented at' the 1999 Southeastern ArchaeologicalConference. We thank Adam King and Maureen Meyers for

131

Fort Walton variant of Mississippian as a regional,defensive reaction to the real or perceived threat ofintrusive Rood populations on their northern frontier.

Most investigators who championed indigenous FortWalton origins have acknowledged some degree ofexternal input, which they attributed to long-distancediffusion (Brose and Percy 1978; Scarry 1984:155-156).To them, this vague external input was not so dramaticas to indicate population movement, nor was it soimportant as to take precedence over local. adaptivefactors such as internal population growth. But in pre~industrial societies, long-distance diffusion is anunlikely mechanism for the complete replacement ofutilitarian artifact styles, such as found in the contrastbetween Rood and its predecessors (Adams 1968:200-201). Furthermore, we doubt that long-distance diffu-sion was sufficient to impart the "Mississippianmodels" that may have aided the transformation ofWakulla into Fort Walton (Brose and Percy 1978:105)..Instead, the fact that Rood and Cayson occupied acommon frontier and shared a limited number of cera-mic attributes suggests a more direct form of contactand competition (Schnell et al. 1981:244). _

Much more archaeological data are required todocument the local and external factors that producedthe Early Fort Walton Cayson phase. Specifically, weneed to collect more radiocarbon samples anddecorated ceramic assemblages from stratified moundcontexts in the Apalachicola valley to gain chrono-logical control over changes in the archaeologicalassemblages from each mound center (Marrinan andWhite. 1998). Identifying changes in the patterns ofartifact, feature, and ecofact assemblages from moundcenters should shed light on those factors related to the'evolution of Fort Walton. However, the hypotheticalcause-and-effect chain cited by proponents of the eco-demographic mQdel-resource stress, maize intensifi-cation, societal reorganization, strong leadership,formation of mound centers-might just as plausibly beexplained as a reaction to the external threat of Roodsettlers on an expanding Mississippian frontier (d. Willey1953). Archaeologists attempting to understand theemergence and spread of the Mississippian culturalphenomenon cannot afford to ignore the observationoffered by Lathrap and others years ago: migration, andmore specifically the ensuing contact and competitionbetween intrusive and indigenous populations, is afundamental process of cultural evolution. Chains ofantecedent cause-and-effect relationships are not neces-sarily confined to a single locality or region. Not everyriver valley is a pristine scene of primary chiefdomformation.

Notes

inviting us to participate in the symposium. Thanks are also. due to James Krakker and other members of the Departmentof Anthropology staff at the Smithsonian mstitution Cura-torial Facility in Suitland, Maryland, who assisted us in ourefforts to examine collections from the Cool Branch andOmussee Creek sites. Nancy White's comments led toimprovements in the article's contents, although her generousassistance does not imply agreement with our arguments.1While both Scarry and White reject population movement asan explanation for the origins of Fort Walton in theApalachicola River valley, it is important to note that Whiterejects Scarry's Fort Walton phases as valid constructs becausethey are not based on published stratigraphic excavationS(Marrinan and White 1998).

References Cited

Adams, William.Y.1968 Invasion, Diffusion, Evolution? Antiquity 17:194-215.Anderson, David G.1994 The Savannah River Chiefdoms: Political Change in the Late

Prehistoric Southeast. University of Alabama Press, T.usca-loosa.

1997 The Role of Cahokia in the Evolution of SoutheasternMississippian Society. In Cahokia: Domination and Ideologyin the Mississippian World" edited by Timothy R. Pauketatand Thomas E. Emerson, pp. 248-268. University of Neb-raska Press, Lincoln.

Baden, William W.1987 A Dynamic Model of Stability and Change in Missis-

sippian Agricultural Systems. PhD dissertation, Departmentof Anthropology, University of Tennessee, Knoxville.

Belovich, Stephanie J., David S. Brose, Russell M. Weisman,and Nancy M. White1982 Archaeological Survey at George W. Andrew's Lake and

Chattahoochee River. Cleve/and Museum of Natural HistoryArchaeological Report 7. Cleveland, OH ..

Blitz, John H.1993 Ancient Chiefdoms of the Tombigbee. University of Alabama

Press, Tuscaloosa.1999 Mississippian Chiefdoms and the Fission-Fusion Process.

American Antiquity 64(4):577-592.Blitz, John H. and Karl G. Lorenz1999 . The Early Mississippian Frontier in the Lower Chatta-

hoochee Valley. Paper presented at the 56th SoutheasternArchaeological Conference, Pensacola, FL.

Brose, David S., and George W. Percy'1978 Fort Walton Settlement Patterns. m Mississippian Settle-

ment Patterns, edited by Bruce D. Smith, pp. 81-114. Aca-demic Press, New York.

Bullen, Ripley P.1958 Six Sites near the Chattahoochee River in the Jim

Woodruff Reservoir Area, Florida. Smithsonian Institution,Bureau of American Ethnology, River Basin Surveys Papers 14.Washington, DC.

Burmeister, Stefan2000 Archaeology and Migration. Current Anthropology 41(4):

539-567.Caldwell, Joseph R.1955 Investigations at Rood's Landing, Stewart County,

Georgia. Early Georgia 2(1):22-49.


1958 Trend and Tradition in the Prehistory of the Eastern Griffin, James B..United States. American Anthropological Association Memoir 1967 Eastern North American Archaeology: A Summary.88.Washington, DC Science 156:175-191.

Chase, David W. Griffin, John W.1959 The Averett Culture. Coweta Memorial Association Papers 1950 Test Excavations at the Lake Jackson Site. American

1.Columbus, GA Antiquity 16(2):99-112.1963 A Reappraisal of the Averett Complex. Joumal of Hally, David J.

Alabama Archaeology 9(2):49-61. 1996 Platform-Mound Construction and the Instability ofClarke, David 1. Mississippian Chiefdoms. In Political Structure and Change1968 Analytical Archaeology. Columbia University Press, New in the Prehistoric Southeastem United States, edited by John

York. F. Scarry, pp. 92-127.University Press of Florida, Gaines-Cobb, Charles R., and Patrick H.Garrow ville.1996 Woodstock Culture and the Question of Mississippian Hassig, Ross

Emergence. American Antiquity 61(1):21-37. 1997 Anasazi Violence:A View from Mesoamerica. In Deci-Conkey,Margaret W., and Christine A Hastorf (editors) phering Anasazi Violence, edited by P.Y.Bullock,pp. 53,68.1990 Uses of Style in Archaeology. Cambridge University Press, HRM Books,Santa Fe, NM.

Cambridge, UK. Hegmon, Michelle, Margaret C Nelson, and S. M. Ruth. Deetz, James 1998 Abandonment and Reorganization in the Mimbres Re-

1977 In Small Things Forgotten: The Archaeology of Early gion of the.American Southwest. American AnthropologistAmerican Life. Anchor Books,Garden City,NJ. 100(1):148-162..

DeJarnette, David 1., and Steven B.Wimberly Hickerson, Harold1941 Bessemer Site: Excavation of Three Mounds and 1965 The Virginia Deer and Inter-Tribal Buffer Zones in the

Surrounding Village Areas near Bessemer, Alabama. Upper Mississippi Valley. In Man, Culture, and Animals,Geological Survey of Alabama, Museum Paper 17.Tuscaloosa. edited by A Leeds and Anthony P. Vayda, pp. 43-65.

Eggan, Fred ~merican Association for the Advancement of Science,1952 The Ethnological Cultures and their Archaeological Washington, DC

Backgrounds. In Archeologt) of Eastem United States, edited Hodder, Ianby James B. Griffin, pp. 35-45. University of Chicago 1985 Boundaries as Strategies:An EthnoarchaeologicalStudy,.Press, Chicago, 11. 'In The Archaeology of Frontiers and Boundaries, edited by

Faulkner, Charles H, Stanton W, Green and Stephen M. Perlman, pp, 141-159.1975 The Mississippian-Woodland Transition in the Middle Academic Press, Orlando, F1.

South. Southeastern Archaeological Conference Bulletin 15:38- Huscher, Harold A45. 1963 The Cool Branch Site, 9Qu5, Quitman'County, Georgia:

Ford, James A, and Gordon R. Willey A Fortified Mississippian Town with Tower Bastions.1941 An Interpretation of the Prehistory of the Eastern Paper presented at the Eastern United States Archaeo-

United States. American Anthropologist 43:325-363. logical Federation, Philadelphia, PAFord, Richard 1. ' Jackson,Paul D.1977 Evolutionary Ecology and the Evolution of Human 2000 The Coeval Occupations of West Jefferson and Mound-

Ecosystems:A Case Study from the Midwestern U.s.A In ville I Phase Sites in West-Central Alabama, PaperExplanation of Prehistoric Change, edited by James N. Hill, presented at the 57th Southeastern Archaeological Con-pp. 153-184.University of New MexicoPress,Albuquerque. ference, Macon, GA

Garland, Elizabeth '8. Jenkins, Ned J.1992 The Obion Site,An Early Mississippian Center in West- 1978aTerminal Woodland-Mississippian Interaction in Nor-

em Tennessee.Mississippi State University, Cobb Institute of ' ' ,them Alabama: The West Jefferson Phase. SoutheasternArchaeology, Report of In!lestigations 7. Starkville. Archaeological Conference Special Publication 5:21-27.

Garrow and Associates, Inc. 1978b Prehistoric Chronology of the Lower Chattahoochee1995 Prehistoric Preservation Plan for the Cultural Resources Valley: A Preliminary Statement. Journal of Alabama'

on U.s. Army Installations at Fort Benning Military Archaeology 24(2):73~87.Reservation, Chattahoochee and Muscogee Counties, Kellar,James H., A R Kelly,and Edward V.McMichaelGeorgia, and Russell County, Alabama, volume 2, 1961 Final Report on Archa'eological Explorations at theTechnical Synthesis. Report submitted to the National Mandeville Site, 9Cla1, Clay County, Georgia. SeasonsPark Service,Atlanta, GA 1959,1960,and 1961.University of Georgia, Athens.

Glassie, Henry . Kelly,A R., David W.Chase, arid Frank T.Schnell, Jr.1975 Folk Housing in Middle Virginia. University of Tennessee 1962 Report on the Cat Cave Creek Site, Clay County,

Press, Knoxville. 9Cla28.In Survey of Archaeological Sites in ,ClayCounty,Gresham, Thomas H, Charles O. Braley,W. Dean Wood, and Georgia, Other than Mandeville, 9Cl<i1,by AR. Kelly,Kay G. Wood Richard Nonas, Betty J. Broyles, Clemens de Baillou,1985 The Carmouche Site:Archaeology in Georgia's Western David W. Chase, and Frank T. Schnell, Jr., pp. 19-25,

Fall Line Hills. Southeastern Archaeological Services, Inc., University of Georgia,Laboratoryof Archaeology,Athens,Athens, GA Report submitted to the U.s. Army Corps of King, AdamEngineers, Savannah, GA 1996 Tracing Organizational Change in Mississippian Chief-

132

THE EARLY MISSISSIPPIAN FRONTIER IN THE LOWER CHATTAHOOCHEE-APALAcHICOLA RIVER VALLEY

domsof the Etowah Valley. PhD thesis, Department ofAnthropology, Pennsylvania State University.

Knight, Vernon James, Jr.1979 Ceramic Stratigraphy at the Singer-MoyeSite, 9S~2.

Journal of Alabama Archaeology 25:138-151.1980 Interregional Reiationships and the Study of Fort

Walton Style. Paper presented at the 38th Annual South-eastern Archaeological Conference, New ,Orleans, LA

Knight, Vernon James, Jr., and Tim S. Mistovich1984 Walter F. George Lake, Archaeological Survey of Fee

Owned Lands, Alabama and Georgia. University ofAlabama, Office of Archaeological Research, Report ofInvestigations 42. Tuscaloosa.

Knight, Vernon James, Jr., and Vincas P.Steponaitis (editors).1998 Archaeology of the Moundville Chiefdom. Smithsonian

Institution Press, Washington, DC.Kopytoff, Igor (editor)1987 The African Frontier: .The Reproduction of Traditional

African Societies. Indiana University Press, Bloomington.Krieger, Alex D.1951 A Radiocarbon Date on the Davis Site in East Texas.

American Antiquity 17:144-145.Lafferty, Robert H., III, and James E. Price

"1996 Southeast Missouri. In Prehistory of the CentralMississippi Valley, edited by Charles H. McNutt, pp. 1-45.University of Alabama Press, Tuscaloosa.

Lathrap, Donald W.1976 Radiation: The Application to Cultural Development of

A Model from Biological Evolution. In The Measures ofMan: Methodologies in Biological Anthropology, edited by E.Giles and J. S. Friedlaender, pp. 494-532.Peabody MuseumPress, Harvard University, Cambridge, MA

Ledbetter, R. Jerald1997 The Victory Drive Site, 9ME50, Muscogee Couf\ty,

Georgia. Georgia Department of Transportation, OccasionalPapers in Cultural Resource Management 8. Atlanta. "

Lightfoot, Kent G., and Antoinette Martinez1995 Frontiers and Boundaries in Archaeological Perspec-

tive. Annual Review of Anthropology 24:471-492.Little, Keith J.1999 The Role of Late Woodland Interactions in the Emer-

gence of Etowah. Southeastern Archaeology 18(1):45-56.Little, Keith L and Harry O. Holstein2001 Terminal Woodland Manifestations in the Coosa Valley

of Northeast Alabama. Paper presented at the 58th South-eastern Archaeological Conference, Chattanooga, TN.

Lorenz, Karl G.1992 Big Men in theBig Black Valley: Small Scale Late Prehistoric

Community Organization in Central Mississippi. PhDdissertation, Department of Anthropology, University ofIllinois, Urbana. University Microfilms, Ann Arbor, ML

Lorenz, Karl G., and John H. Blitz1999 The Cool Branch Fortified Mound Center and the

Moundville-Fort Walton Frontier. Paper presented at the64th Annual Meeting of the Society for American

.• Archaeology, Chicago, IL. .- 2002 The Mississippi Period in the Lower Chattahoochee

Valley: A Statistical Approach to for Establishing aRegional Chronology. Paper presented at the 67th AnnualMeeting of the Society for AmericanArchaeology, Denver,CO.

Marrinani Rochelle A, and Nancy M. White1998 Smoke and Mirrors in"Modeling Fort Walton Culture,

Northwest Florida. Paper presented at the 55th South-eastern Archaeological Conference, Greenville, Sc.

McMichael, Edward V, and James H. Kellar1960 Archaeological Salvage in the Oliver Basin. University of

Georgia, Laboratory of Archaeology Series Report 2. Athens.McNutt, Charles H.1996 The Central Mississippi Valley: A Summary. In The

Prehistory of the Central Mississippi Valley, edited byCharles H. McNutt, pp. 187-257. University of AlabamaPress, Tuscaloosa.

Mielke, James E., and Austin Long1969 Smithsonian Radiocarbon Dates. Radiocarbon 7(1):166.Milanich, Jerald T.1994 Archaeology of Precolumbian Florida. University Press of

,Florida, Gainesville. 'Morse, Dan F.1977 The Penetration of Northeast Arkansas by Missis-

sippian Culture. In For the Director: Research Essays inHonor of James B. Griffin, edited by Charles E. Cleland,pp. 186-211. University of Michigan, Laboratory of Archae-ology Series Report 2, Museum of Anthropology, Anthro-pologicalPapers 61.Ann Arbor .•

Nassaney, Michael S., and Charles R. Cobb1991 "Patterns and Processes of Late Woodland Development

in the Southeastern United States. In Stability, Transfor-mation, and Variation: The Late Woodland Southeast,editedby Michael S. Nassaney and Charles R. Cobb, pp. 285-322.Plenum Press, New York.

Neuman, Robert W.1961 Domesticated Corn from a Fort Walton site in Houston

County, Alabama. Florida Anthropologist 14 (3-4):75-80.Payne, Claudine1994 Mississippian Capitals: An Archaeological Investigation of

Precolumbian Political Structure. PhD dissertation, Univer-sity of Florida, Gainesville. University Microfilms, AnnArbor, ML

Payne, Claudine, and John F.Scarry1998 Town Structure at the Edge of the Mississippian World.

In Mississippian Towns and Sacred Spaces: Searching for anArchitectural Grammar, edited by R. Barry Lewis and C.Stout, pp. 22-48,University "ofAlabama Press,Tuscaloosa.

Peebles, Christopher S.1978 Determinants of Settlement Size and Location in the

Moundville Phase. In Mississippian Settlement Patterns, editedby BruceD. Smith, pp. 369-416.Academic Press, New York.

Peebles, Christopher S., and Susan M. Kus1977 Some Archaeological Correlates of Ranked Societies.

American Antiquity 42(3):421-448.Phillips, Phillip, James A Ford, and James B. Griffin1951 Archaeological Survey In the Lower Mississippi

Alluvial Valley: 1940-1947.Harvard University, Papers of thePeabody Museum 25. Cambridge, MA.

Rafferty, Janet1995 Owl Creek Mounds: Test Excavations at a Vacant Missis-

sippian Mound Center. Mississippi State University, CobbInstitute of Archaeology, Report of Investigations 7. Starkville.

Rice, Prudence M.1998 Contexts of Contact and Change: Peripheries, Frontiers,

and Boundaries. In Studies in Culture Contact: Inter-

133


action, Culture Change, and Archaeology, edited by J. G.Cusick, pp. 44-66. Southern Illinois University, Center forArchaeological Investigations, Occasional Paper 25.Carbondale.

Rouse, Irving1958 The Inference of Migrations from Anthropological Evi-

dence. In Migrations in New World Culture History, editedby R. H. Thompson, pp. 63-68.' University of ArizonaSocial Science Bulletin 27. Tucson.

. Sahlins, Marshall D.1961 The Segmentary Lineage: An Organization of Predatory

Expansion. American Anthropologist 63:322-345.Sahlins, Marshall D., and Elman RService1960 Evolution and Culture. University of Michigan Press, Ann

Arbor.Scarry,John E1980 The Chronology of Fort Walton Development in the

Upper Apalachicola Valley, Florida. Southeastern Archaeo-logical Conference Bulletin 22:38-45.

1981 Fort Walton Culture: A Redefinition. SoutheasternArchaeological Conference Bulletin 24:18-21.

1984 Fort Walton Development: Mississippian Chiefdoms in theLower Southeast. PhD dissertation, Department of Anthro-pology, Case Western Reserve University, Cleveland.University Microfilms, Ann Arbor, MI.

1990 Mississippian Emergence in the Fort Walton Area: TheEvolution of the Cayson and Lake Jackson Phases. In TheMississippian Emergence, edited by Bruce D. Smith, pp.227-250.Smithsonian Institution Press, Washington, DC

Scarry, John E, and Claudine Payne1986 Mississippian Polities in the Fort Walton Area: A Model

Generated from the Renfrew-Level XTENT Algorithm.Southeastern Archaeology 5(2):79-90.

Schnell, Frank T.1998 Ceramics in the Southern Half of the Chattahoochee

Valley.Journal of Alabama Archaeology 44(1-2):99-130.Schnell, Frank T.,Vernon J. Knight, Jr., and Gail S. Schnell1981 Cemochechobee: Archaeology of a Mississippian Ceremonial

Center on the Chattahoochee River. University Press ofFlorida, Gainesville.

Schnell, Frank T., and Newell O. Wright, Jr.1993 Mississippi Period Archaeology of the Georgia Coastal

Plain. University of Georgia, Laboratory of Archaeology Series26. Athens.

Schnell, Gail S.1977 The View from Rood's Landing, Then and Now. Early

Georgia 6:12-21.1981 A Preliminary Political Model for the Rood Phase.

Southeastern Archaeological Conference Bulletin 24:23-24.Schroedl, Gerald E1994 A Comparison of the Origins of Macon Plateau and

Hiwassee Island Cultures. In Ocmulgee Archaeology 1936-1986, edited by David J. Hally, pp. 138-143.University ofGeorgia Press, Athens.

Schroedl,Gerald E, Clifford C Boyd, Jr., and R P. StephenDavis, Jr.1990 Explaining Mississippian Origins in East Tennessee. In

The Mississippian Emergence, edited by Bruce D. Smith, pp.175-196.Smithsonian Institution Press, Washington, DC

Sears, William H.1967 The Tierra Verde Burial Mound. Florida Anthropologist

20(1-2):25-73..

Seckinger, Ernest W., and Ned J. Jenkins2000 A Plural Society in Prehistoric Alabama. Journal of

Alabama Archaeology 46(1):43-57.Sh~ldon, Craig T.,John W. Cottier, and Jason A. Mann2001 A Late Woodland-Mississippian Culture Contact in

Central Alabama. Paper presented at the 58th South-eastern Archaeological Conference, Chattanooga, TN.

Shennan, Stephen2000 Population, Culture History, and the Dynamics of

Culture Change. Current Anthropology 41(5):811-835.Smith, Bruce D. (editor)1978 Mississippian Settlement Patterns. AcademicPress,New York.1984 Mississippian Expansion: Tracing the Historical Develop-

ment of an Explanatory Model. Southeastern Archaeology3:13-22.

1990 Introduction. In The Mississippian Emergence, edited byBruce D. Smith, pp. 1-8. Smithsonian Institution Press,Washington, DC

Steponaitis, v.P.1983 Ceramics, Chronology and Community Patterns: An

Archaeological Study at Moundville. Academic Press, NewYork

Stout, Charles, and R. Barry Lewis1998 Mississippian Towns in Kentucky. In Mississippian

Towns and Sacred Spaces: Searching for an ArchitecturalGrammar, edited by R Barry Lewis and Charles Stout; pp.151-178.University of Alabama Press, Tuscaloosa.

Stuiver, M., and P.J. Reimer1993' Extended 14CData Base and Revised Calib 3.0 14Cage

Calibration Program. Radiocarbon 35:215-230.Walthall, John A.1980 Prehistoric Indians of the Southeast: Archaeology of Alabama

and the Middle South. University of Alabama Press,Tuscaloosa.

Wauchope, Robert (editor)1956 Seminars in Archaeology: 1955. Society for American

Archaeology, Memoir 11.Webb, William S.1952 The Jonathan Creek Village, Site 4, Marshall County,

Kentucky. University of Kentucky, Reports in Anthropology8(1). Lexington.

Welch, Paul D.1994 The Occupational History of the Bessemer Site.

Southeastern Archaeology 13(1):1-26.White, Nancy M.1982 The Curlee Site (8JA7! and Fort Walton Development in the

Upper Apalachicola-Lower Chattahoochee Valley. PhDdissertation, Case Western Reserve University, Cleveland.Uriiversity Microfilms International, Ann Arbor, MI.

1996 Test Excavations at the Yon Mound and Village Site(8Li2), Middle Apalachicola Valley, Northwest Florida.Report Submitted to the Division of Historical Resources,FloridaDepartment of State, Tallahassee.

Weissner, Polly1983 Style and Social Information in Kalahari San Projectile

Points. American Antiquity 48(2):253-276.Willey, Gordon R.1949 Archeology of the Florida Gulf Coast. Smithsonian Mis-

cellaneous Collections 113.Washington, DC1953 A Pattern of Diffusion-Acculturation. Southwestern

Journal of Anthropology 9:369-384.

134


Willey, Gordon R., and Philip Phillips1958 Method and Theory in American Archaeology. University

of Chicago Press, Chicago, IL.Willey, Gordon R, and Richard B. Woodbury1942 A Chronological Outline for the Northwest Florida

Coast. American Antiquity 7:232-254.Williams, Mark1994 The Origins of the Macon Plateau Site. In Ocmulgee

Archaeology 1936-1986,edited by David J. Hally, pp. 130-137.University of Georgia Press, Athens.

Williams, Mark, and Gary Shapiro1996 Mississippian Political Dynamics in the Oconee Valley,

Georgia. In Political Structure and Change in the PrehistoricSoutheastern United States, edited by John F. Scarry, pp.128-149.University Press of Florida, Gainesville.

Wobst, H. Martin1977 Stylistic Behavior and Information Exchange. In Papers

for the Director: Research Essays in Honor of James B.Griffin, edited by Charles Cleland, pp. 317-342. Universityof Michigan, Museum of Anthropology, Anthropological Paper61. Ann Arbor.

135

the early mississippian frontier in the lower ...anthropology.ua.edu/reprints/14.pdf · the early...

Documents