Geographic Information System Modeling of De Soto's Route from …my.ilstu.edu/~jthayn/Publications_files/Sampeck copy.pdf · 2015-10-20 · travel illuminate social, political, eco

Archaeologists have shown that routes oftravel illuminate social, political, eco-nomic, and cosmological worlds through

their orientation, length, and relation to settlements(Bauer 1992; Doyle et al. 2012; Hassig 1991;Hutson et al. 2012; Hyslop 1984; Kantner 1997;Purser 1989; Snead et al. 2009; Sofaer et al. 1989;Trombold 1991). While paths of communicationand exchange have been factors in archaeological

interpretations of political, social, and economicinteraction in many regions of the world, archae-ological examination of the routes themselves hasgarnered relatively less explicit analysis. The studyof extensive, “nonsite” features falls within thepurview of landscape archaeology, an approachthat considers travel, paths, and movement (Ash-more and Knapp 1999; Snead et al. 2009). Geo-graphic information system (GIS) analysis pro-

GEOGRAPHIC INFORMATION SYSTEM MODELING OF DE SOTO’SROUTE FROM JOARA TO CHIAHA: ARCHAEOLOGY AND

ANTHROPOLOGY OF SOUTHEASTERN ROAD NETWORKS IN THE SIXTEENTH CENTURY

Kathryn Sampeck, Jonathan Thayn, and Howard H. Earnest, Jr.

This research revisits the question of the most likely paths traveled during the 1540 entrada of Hernando de Soto andcolonizing efforts of Juan Pardo about 20 years later by utilizing the spatial modeling method of geographic informationsystem (GIS) analysis to evaluate the favorability of different paths and place them within the context of recent archaeologicaland ethnohistoric research. Analysis results make the larger anthropological point that GIS route modeling should explicitlytake into account the size of the party traveling. Routes for small parties are not the same as optimal routes for large armiessuch as de Soto’s, which included hundreds of people, pieces of equipment, and livestock. The GIS-modeled routes correlatewith the distribution of contact-period archaeological sites and attested eighteenth-century routes. More accurate estimationof Spanish routes allows us to better model the Native American social, economic, and political nexus of this period, showingthat the residents in far eastern Tennessee were probably part of a dynamic borderlands between the chiefdom of Coosa tothe west and the ancestral Cherokee heartland to the east. This anthropological refinement in GIS modeling will be usefulin investigating ancient paths of interaction in many parts of the world.

Esta investigación evalúa por el método de análisis espacial del sistema de información geográfico (SIG) la favoribilidad delos caminos viajados durante la entrada de Hernando de Soto en 1540 y la colonización de Juan Pardo aproximadamente 20años más tarde. Esta información sobre caminos favorables considermos en relación a investigaciones recientes de arqueologíay etnohistória. El análisis indica que el modelo de las las rutas debería tomar en cuenta explícitamente el tamaño del grupoviajando. Las rutas para uno o dos viajeros no son igual a las para ejércitos grandes como lo de Soto, que incluyó a cientosde personas, equipamiento, y ganado. Las rutas de SIG modelados son apoyadas por la distribución de los sitios arqueológicosdel período de contacto y los caminos del siglo XVIII tomado por ejércitos durante la guerra de independencia. Estimaciónmás exacta de las rutas españolas nos permite un mejor modelo del nexo social, económico, y político indígena de este período,mostrando que los residentes en el lejano oriente Tennessee eran probablemente parte de una zona fronteriza dinámica entrelas jefaturas de la Coosa por el poniente y la tierra ancestrales de los Cherokee al oriente. Este perspectiva antropológica enel uso de SIG será útil para investigar caminos antiguos en muchas partes del mundo.

Kathryn Sampeck ! Illinois State University, Department of Sociology and Anthropology, Campus Box 4660, Normal, IL61790 ([email protected])Jonathan Thayn ! Illinois State University, Department of Geology and Geography, Campus Box 4400, Normal, IL61790 ([email protected])Howard H. Earnest, Jr. ! Independent Scholar, 507 S. State St., Bloomington, IL 61701 ([email protected])

American Antiquity 80(1), 2015, pp. 46–66Copyright © 2015 by the Society for American Archaeology

DOI: 10.7183/0002-7316.79.4.46

46

Sampeck et al.] GIS MODELING OF DE SOTO’S ROUTE FROM JOARA TO CHIAHA 47

vides a way to model paths based on factors suchas slope and distance. GIS modeling of routes cantest the viability of paths suggested by archaeo-logical evidence, clarifying the cultural processesthat generated them (Doyle et al. 2012; Kantner1997). Even though much archaeological and eth-nohistorical analysis has been devoted to recon-structing Spanish entradas into the U.S. Southeast,no segments of the proposed routes have beenevaluated with GIS in terms of least-cost paths.Recent archaeological survey and excavations inthe Nolichucky Valley of Washington and Greenecounties in eastern Tennessee have recoveredabundant contact-period remains of settlementsthat permit a reevaluation of one area traveledthrough and recorded in the sixteenth century, thetrans-Appalachian zone between Joara, near to-day’s Morganton, North Carolina, and Chiaha, anisland settlement near the modern town of Dan-

dridge, Tennessee (Figure 1). Documentary, lin-guistic, spatial, and archaeological informationdiscussed according to locale provides a view ofshifts in the complex political geographies of theSoutheast in just over 20 years. Though the Span-ish encounters were brief, they may have alteredthe dynamics of Southeastern political consolida-tion, hierarchy, and heterarchy (Beck 2013). Theresults of this study show that roads were a dis-cursive space in the sixteenth-century Southeast-ern political economy: while roads linked alliesand were a means of economic and social ex-change, they also were the place to define culturaldifference. In Purser’s words, “In the end, roadsdo not act, people do. But where roads go, andhow they do or do not link place to place doesshape the lives of the individuals and communitiesthat use them” (1989:134). Optimal routes forsmall vs. large parties in this region are dramati-

Figure 1. The southeastern United States, showing the archaeological study region.

cally different. This variation suggests differencesin the scale of interaction, with contact-period set-tlement heavily favoring large-scale interaction.The analysis of trans-Appalachian routes demon-strates that GIS analysis can be explicitly anthro-pological by taking into account the size of thetraveling party.

Colonial Encounters of the First Kind

Several sixteenth-century accounts relate thetravel of Hernando de Soto and his army in theSoutheast in spring 1540 and the forays of JuanPardo and his soldiers from the fort they estab-lished in 1567. This research builds upon previoussubstantial work on Spanish entradas in theSoutheast (Galloway 1993; Hudson 1990; Mi-lanich 1993; Milanich and Hudson 1993; Worth2007). Digitized images of the Spanish manu-scripts or early imprints examined include the1605 Crasbeek imprint of Garcilaso’s La Floridadel Ynca and Fidalgo de Elvas’s 1557 Relaçamverdadeira, accessed via the digital resources ofthe John Carter Brown Library. The 1851 imprintof Oviedo’s Historia general was the first com-plete edition of all the Oviedo books, some ofwhich had existed previously only in manuscriptform. Digitized manuscripts of complaints, dec-larations of merit, and other legal documents andreports for the de Soto and Pardo expeditionswere available through the Portal de ArchivosEspañoles database of materials curated in severalarchives in Spain. Sampeck translated the pas-sages of these primary source texts included inthis analysis.

Both de Soto and Pardo were interested in be-coming wealthy, being seen as loyal servants tothe king of Spain, and gaining political advantageover the people they encountered in the Southeast.To this end, these accounts record in varying de-grees of detail the natural resources of differentregions; the names and political affiliations ofthe people they met; the size, location, and namesof settlements; and how they traveled from placeto place. Understanding political affiliation hadbeen a key to success in other colonial ventures,so the Iberian authors were careful to notewhether one settlement was part of a larger polityor not. Variations among accounts give clues asto whether the author is over- or underreporting

something, while variations within an accountindicate nuances of resources, production, andemphasis. In accounts of the same time period inother parts of Spanish America that are knownarchaeologically and abundantly documented, thenotation of place-names varies depending on thematter at hand (place of a battle, center of pro-duction of a valuable commodity, and so forth).Even well-known, large settlements may not beincluded in an account because of the writer’sagenda (Sampeck 2007).

Political, Social, and Economic Dynamics,1540 to 1567

After being given the official title of Governorof La Florida, Hernando de Soto began his re-connaissance in what is today the state of Florida,passing through South Carolina and eventuallyturning farther inland in North Carolina. His goalwas to explore the full extent of his realm andideally learn that his protectorate had great wealthin natural resources and people. He was the right-hand man for Francisco Pizarro in the conquestof Peru and had already amassed a great deal ofwealth, and the Florida venture was to give himthe social recognition that came with the officialtitle and the potential for even more riches. JuanPardo was sent on two expeditions to the interiorin the 1560s by the then–Governor of La Florida,Pedro Menéndez de Avilés. His two key directivesfor Pardo were to establish a road from SantaElena, on the coast of present-day South Carolina,to the silver mines of Zacatecas, Mexico, and tosubdue the Indians.

Mid-sixteenth-century Spanish colonists werealready creolized by previous experience in otherparts of Latin America. Several from the de Sotoand Pardo efforts maintained strong ties and heldroyal offices in Guatemala, demonstrated by thevocabulary they used to name items that South-eastern communities gave to the armies: tameme(human porters) and petaca (a woven wickerhamper), both Nahuat (a Mesoamerican Nahualanguage), not Southeastern terms (Fernández deOviedo y Valdés 1851:562; Fidalgo delvas1557:62v; Karttunen 1992:192). Rather than pre-cious metals, agricultural products and indigenousslaves were the key to wealth in Guatemala, pre-occupations evident in the Spanish accounts

48 AMERICAN ANTIQUITY [Vol. 80, No. 1, 2015

(MacLeod 2008). De Soto himself owned andoperated a slave ship (MacLeod 2008:52). DeSoto and his army ate corn and dogs, followedindigenous lines of political schism and alliance,and enslaved people (Fernández de Oviedo yValdés 1851:562).

In the 1540s an acrid debate about the humanityand civil rights of indigenous peoples eventuallyresulted in the abolishment in 1542 of the en-slavement of Indians and recognition of their rightsthrough new policies and procedures, a matter notyet settled when de Soto was in the Southeast in1540. His party’s interactions with Native Amer-icans contrasted markedly with the post-1542 ven-ture of Juan Pardo, who exerted much more effortto secure alliances with local leaders.

The Spanish arrived when Southeastern NativeAmericans were substantially reconfiguring localand regional social, political, and economic rela-tionships (Beck 2013). Archaeological investiga-tions of Late Mississippian settlements show thatthe fifteenth to the sixteenth century was a periodof widespread, profound demographic shifts, withregions such as the Asheville Basin and the Sa-vannah and Etowah river valleys largely aban-doned during the 1400s and 1500s (Anderson1994; Beck 2013; King 2003; Milner and Chaplin2010; Rodning 2008:2–3; Smith 1987). Large rit-ual and administrative centers grew in places dis-tinct from the locations of Middle Mississippianantecedents. Communities and social groups cre-ated new ways to position themselves within ahighly dynamic environment affected in part byfalling population levels in the 1500s (Milner andChaplin 2010). Indigenous residents paid tribute,helped build forts, and placed European objectsin sacred spaces as part of political and social re-alignment in the sixteenth century.

Because roads connected places, they facili-tated management, but at the same time theycrossed boundaries that people created and/ormaintained through actions such as overt hostility(warfare), polity naming, and stylistic variationin material culture. Identifying the location ofroutes helps better characterize the nature of po-litical shifts by identifying who was connectedto whom and the nature of their linkages. Thevarious accounts indicate that de Soto’s army re-lied upon indigenous guides who took them tocenters of population along the main route from

one region to the next, what Garcilaso called thecamino real (Vega 1605:176r).

Spaniards measured route distance in leagues.Several different kinds of leagues were in useduring the sixteenth century; to make our esti-mates comparable to earlier work, we calculatedistances in the legua común of 5.57 km (3.47mi). A daily marching speed of 19 to 32 km perday is a standard estimate for army foot travel(Hassig 1988:67), so a legua común falls in themiddle of the range of 5 or 6 leagues of travelper day that Garcilaso indicated (Beck 1997;Chardon 1980; Hudson et al. 1984:66; Vega1605:1767v). It is possible that some discrepan-cies among accounts may exist because somechroniclers used legua to refer to distance traveledin an hour (Swanton 1939:Appendix E, in Gal-loway 1993:311–354). None of the accounts in-dicate that the distances were measured by in-struments such as chains, in contrast tovoluminous metes-and-bounds surveys conductedin the colonies during this and later periods(Hunter and Sluyter 2011; Sampeck 2014).

Spanish Entradas and Native American Roads

In the 1930s the U.S. Congress appointed JohnSwanton to head a commission to propose theprobable route for Hernando de Soto as part ofan effort in historic conservation. Swanton (1985)compiled extensive historical documentation andarchaeological knowledge to propose a probableroute, which in general skirted northern Georgiaand passed into Alabama. Advances in archaeo-logical, linguistic, and historical research duringthe 1970s to the 1990s fostered a thorough reeval-uation of the Swanton Commission results.Charles Hudson, Chester DePratter, MarvinSmith, and other colleagues proposed a morenortherly route for de Soto, passing through NorthCarolina and east Tennessee. Subsequent workby Robin Beck (1997) further refined the partic-ular segment that is the focus of this study, push-ing the location of Joara a little farther north, tothe Berry site (31BK22), and suggesting thatPardo’s soldiers traveled in some instances alongroutes not traveled by de Soto (Beck et al. 2006).Our recent work in Greene and Washington coun-ties in east Tennessee further supports some of


the identifications suggested by Beck and col-leagues and indicates refinements to the probableroute of Spanish expeditions and our understand-ing of cultural interactions in this area.

Who lived in these places? Place-names andother linguistic evidence in Iberian chronicles sug-gest the locations of a complex array of South-eastern Native American language groups in thesixteenth century (Booker et al. 1992:410). Iro-quoian Cherokee appears to have been centeredin the mountainous regions of eastern Tennessee,western North Carolina, and adjacent areas;Catawban languages and dialects, in the CarolinaPiedmont; and Yuchi, in an uncertain location nearthe Appalachians (Booker et al. 1992:410–411;Goddard 2005; Rudes 2004). Muskogean lan-guages of Choctaw, Chicasaw, and Hitichi appearto have been used for naming in present-day Mis-sissippi, western Alabama, Georgia, and Tennessee(Booker et al. 1992:410–411; Goddard 2005). Af-ter presenting the methodology and results of theGIS least-cost analysis between Joara and Chiaha,specifics of routes are considered in the contextof previous interpretations, archaeological data,and linguistic and ethnohistoric evidence.

GIS Least-Cost Route Analysis

This analysis concentrates on an area anchored bytwo fairly securely identified points. The eastern-most point is a province that the Spanish identifiedas Joara, correlated with the Berry site (31BK22)near Morganton, North Carolina, by Robin Beck,David Moore, and Christopher Rodning (2006),while the western end of the route is the provincethe Spanish recorded as Chiaha (Olamico), identi-fied as Zimmerman’s Island near Dandridge, Ten-nessee (DePratter et al. 1983). The area betweenthese two places is much less well known.

The digital elevation model (DEM) used in thisanalysis uses the Shuttle Radar Topography Mis-sion data collected by the U.S. space shuttle En-deavor in February 2000 (Rabus et al. 2003) andcorrected and cleaned per standard methodologies(Jarvis et al. 2008; Reuter et al. 2007). The spatialresolution of these data is 90 m, a resolution ap-propriate for evaluating route favorability. MikeMoore and archaeologists of the Tennessee Divi-sion of Archaeology provided registered informa-tion about sites used in this spatial analysis.

Least-Cost Analysis and Degrading the Spatial Resolution

Following standard methodology, the shortestthree-dimensional path between two points on theDEM was identified by the least-cost path proce-dure (DeMers 2003:276–284). Least-cost analysispathways based on slope tend to move alongcanyon bottoms, where changes in elevation areminimized. The result of a least-cost path analysisdepends on the cell width of the DEM over whichthe path is being constructed. As a general rule,features on the earth’s surface should be at leasttwice as wide, at the smallest dimension, as thewidth of the DEM’s cells (Jensen 2005:14–18).In essence, valleys narrower than 180 m are toofine to be represented in the Shuttle Radar Topog-raphy Mission DEM data set. Degrading the spa-tial resolution of the data reduced the number ofelevation values recorded in the degraded modelto one-quarter of the number used in the originalmodel. Degrading the spatial resolution does notaffect the accuracy of the DEM data; it merelyrestricts the realm of route choices to paths largeenough for hundreds of people. Narrow canyonswere thus removed from the model as longer vi-able routes for the least-cost pathway, which re-sulted in two significantly different pathways.

Route 1: Southern, Narrow PathThe least-cost path found using the original 90-m DEM passes westward along Highway 70,around the south side of Asheville, and then northalong the French Broad River toward DouglasLake and the island on which Chiaha was located(Figure 2). This southern, narrower route is 272.5km long, including changes in elevation (Figure3). Its maximum elevation is 879.0 m, and themean slope of the route is .83 degrees (1.5 percentgrade). The steepest slope is 17.3 degrees (31.1percent grade), but only 7.6 percent of the routehas a percent grade greater than 5.0 percent.

Route 2: Northern, Wide PathThe least-cost path found using the degraded 180-m DEM moves northward more or less alongHighway 181 toward Elizabethton, Tennessee. Itnext passes southwestward through Johnson City,Tennessee, along the Nolichucky River, generallyfollowing Highway 34, and then turns to the north


side of Greeneville, Tennessee. It stays south ofInterstate 81 and remains on the north side of theNolichucky River just before it drains into Dou-glas Lake (Figure 2). This route is 294.7 km long,including changes in elevation (Figure 3). Itsmean elevation is 496.5 m, and its maximum el-evation is 1,197.04 m. The mean slope along thisroute is .69 degrees (1.2 percent grade). The steep-est slope is 9.6 degrees (16.9 percent grade), al-though only 4.9 percent of the route has a percentgrade greater than 5.0 percent.

Comparative Cost DistanceThe cumulative cost distance (distance [km] xpercent rise of slope) is an estimate of the effortrequired to travel each route. The cumulative costof the northern route is 667.9, and that of thesouthern route is 764.5. While the northern routeis 8.16 percent longer than the southern route, itscost distance is 12.6 percent less, indicating thatit requires considerably less effort to use the

northern route. The slope of the terrain traversedalong the northern route is more gradual than thatof the southern route. The mean and maximumslopes are less dramatic, and except for the climbon the east side of the mountains, the overallroute is more relaxed.

The relative gentleness and broader pathwaysof the northern route make traveling with a largegroup easier. Large caravans traveling along thesouthern route would have to squeeze their herdsthrough narrower passages and push them upsteeper slopes. Tactical reasons for large armiesto travel along a wider path include protectionfrom attack and less time to get the entire armyon the road (Hassig 1988:68). If the company oftravelers were few, the shorter but steeper south-ern route might be faster. How do these resultsreconcile with other archaeological, linguistic,and historical data? We address this question bydiscussing each place between Joara and Chiahanamed in the de Soto and Pardo accounts.


Figure 2. Least-cost path found using the original 90-m digital elevation model (southern route) vs. the least-cost pathfound using the degraded 180-m digital elevation model (northern route). Locations of contact-period archaeologicalsites and points along the 1780 route of the Overmountain Men are also noted.

Archaeological and Ethnohistoric Evidencefor Native American Places and Roads

Joara: Resources, Polities, and Route PlanningThe relatively long stay of de Soto’s army inJoara depended upon its ample resources (Fer-nández de Oviedo y Valdés 1851:562). The well-preserved archaeological remains of the sixteenth-century Native American village and Spanish fortand town make up one of only two mound sitesin the upper Catawba Valley and one of the largestBurke phase (A.D. 1400–1600) sites in the region(Beck 1997; Beck et al. 2006; Moore 2002). JuanPardo built his principal base of operations, thecity (ciudad) of Cuenca and Fort San Juan, atJoara in early January 1567, citing the large res-ident population and the fertile, mile-wide alluvialfloodplain that forms an ecotone between thePiedmont and the highly dissected uplands of theBlue Ridge (Bandera 1990:f5 [1569]; Beck1997:163; Fernández de Oviedo y Valdés1851:562; Vega 1605:178r). Although Ranjel rec-ognized Joara’s potential for prospecting and goldmines (Fernández de Oviedo y Valdés 1851:562),Pardo’s men were the first to try prospecting. “Inthese mountains is gold and silver” labels the Ap-palachians on maps that postdate 1540, eventhough Joara itself appears inconsistently on six-teenth-century and later maps. Places farther west,

such as Guasili, appear as often as Joara (e.g., deLaet 1625).

Joara’s roads defined ethnic and politicalboundaries. Even though “Xuala” was a provinceseparate from “Cofachiqui,” Joara was under thecacica (female ruler) of Cofitachequi’s rule (Fi-dalgo delvas 1557:61v; Vega 1605:179r). Cofi-tachequi was an extensive polity based in thevicinity of present-day Camden, South Carolina(Hudson 1990:83). The female leader of Cofi-tachequi was important enough that, dependingon the account, she either was kidnapped by orwillingly accompanied the Spanish army on itsjourney to Joara. Garcilaso (Vega 1605) describedJoara as part of “Cofachiqui’s” señorio, a systemof rule in which nobles control access to landand all subjects are vassals. The residents of Joaraserved the Spanish because they treated them asthough they were from the cacica’s señorio; honorshown to them was honor shown to her (Vega1605:179r). Mooney recorded that the Cherokeename of the Swannanoa Gap was “Suwali-Nuinahi, or ‘Suwali trail,’ that being the passthrough which ran the trail from the Cherokee tothe Suwali, or Ani-Suwali, living east of themountains” (1894:57). Even earlier, seventeenth-century maps likewise name the mountain rangenorth of Xuala (Joara) the “Suwala Range,” sug-gesting that Joara perhaps acted as gatekeeper of


Distance

Ele

vatio

n (m

eter

s)

0 50 100 150 200 250 kilometers

400

600

800

1000

Northern route

Southern route

0 10 20 30 40 50 leagues

(Joara) (Chiaha)

Figure 3. Comparison of profiles of the relative costs of paths found using the original 90-m digital elevation model vs.the degraded 180-m digital elevation model.

trans-Appalachian routes. The path described byMooney is the narrow but efficient southern GISpath (Figure 2).

A linguistic break seems to have occurred eastof Joara in the 1540s. Hudson (1990:83) proposesthat the people of Cofitachequi probably spokeMuskogean and Catawban as well as other South-eastern languages. Booker et al. (1992:423, 425)suggest that the people of Joara spoke a linguisticisolate such as Yuchi, but Goddard (2005) makesa strong case that “Joara” can be equated withthe name of the Catawba dialect of Saraw. An-other linguistic boundary probably existed westof Joara: the first settlements west of the BlueRidge may have been of Cherokee speakers, asother interpreters were used by the Pardo armyin addition to the Muskogean-speaking Rufín(Booker et al. 1992:425). By most accounts, deSoto’s army left Joara on Tuesday, May 25. Theyheaded for “Guaxule” because the cacica of Cofi-tachequi’s lands reached that far (Fidalgo delvas1557:61v). In the 1540s Guasili was the next no-table place west and a border region.

Crossing the MountainsThe first day of travel west of Joara passedthrough cultivated fields, and the accounts agreethat the next segment of the journey was over“rough and high” mountains (agras y altas serras)that were difficult to cross (Fidalgo delvas1557:61v). The de Soto army spent five daysafter the first one leaving Xuala traveling through“uninhabited” mountains, with a stated distanceof about 20 leagues (69.4 mi) between Xuala andGuaxule (Vega 1605:180r). The GIS northernroute agrees with this time frame for travel start-ing at the Berry site and entering the mountainsquickly, in contrast to the more southerly routeproposed by Hudson et al. (1984). Beck(1997:164) proposes that the route passed alongState Highway 181, past Table Rock and Hawks-bill Mountain, so that the army entered the up-lands only 15 mi from the Berry site. The GISroute is even more direct, following Warrior Fork,eventually uniting with the route of Highway 181.The key difference between this route and Beck’s(1997) is that part of the GIS path is farther northand east to take advantage of an efficient path.

According to Ranjel, the army crossed a veryhigh ridge and camped “a un montecillo” (Fer-

nández de Oviedo y Valdés 1851:562). Montecillocan be translated as either “small forest” or“hillock” or “hummock” (Clayton et al. 1995:281n. 48 [by John Worth]). The second meaning iscorrect, as the preposition in Spanish is to sleepin (en) a forest but on (á) a hill or plain, the usagefound in other places in the Ranjel account (Fer-nández de Oviedo y Valdés 1851:562). A smallhill just west of the steep easternmost slopes ofthe mountains is the likely location for the firstcampsite during the mountain crossing; one ofthe few relatively low hills between Parks Moun-tain and Cold Mountain lies just west of the firststopping place estimated by the GIS analysis.

By Wednesday, May 26, the de Soto armyreached the summit of the range, where theycamped on a plain (á una savana) and sufferedgreat cold even though it was the end of May(Fernández de Oviedo y Valdés 1851:562). Beck(1997:164) suggests that they camped near JonasRidge. The GIS-estimated spot for the camp is aflat between Elk Park and Laurel Fork, the rela-tively broad fields in the area of Taylor Chapel.In fact, part of the Appalachian Trail is in thisvicinity.

Next, the army “crossed, in water up to theirshins, the river by which they afterward left inthe brigantines that they made and that emptiedinto the sea, shown by the navigation chart that itis the Spiritu Sancto” (Fernández de Oviedo yValdés 1851:562). Ranjel was referring to thecosmographic chart of Alonso de Chaves, madeper the royal order of 1536 by Charles V (SantaCruz 1892:18 [1542]). Beck (1997:164) arguesthat the army crossed over near the present townof Ingalls, North Carolina, to a trail that ran along-side the North Toe River, where U.S. Highway19E is now (Cumming 1966:23–27). We suggestinstead that the army had crossed more of themountain range, so that the notable crossing wasof the Watauga River, which also empties intothe Mississippi (“Spiritu Sancto”).

After the de Soto army crossed this river, prob-ably in the late evening, the lady of Cofitachequiescaped along with two slaves (Fernández deOviedo y Valdés 1851:562). The area was woodedheavily enough that the small detachment couldnot find them (Fidalgo delvas 1557:62v). Thecacica and her companions used the steep terrainto their advantage; the detachment had to return


through the challenges they had largely left be-hind them. It may be that the Nolichucky drainagerepresented a political as well as a linguistic bor-der or that the lady had reached the limit of hertolerance of the Iberians.

By Friday, May 28, the de Soto expedition de-scended enough in altitude to spend the night inan oak wood (robredal) (Fernández de Oviedo yValdés 1851:563). The GIS route places the armyin the vicinity of Elizabethton, Tennessee. Beck(1997:164) argues that the army turned northagain, following a path incorporated into theWashington County–Burke County Road of 1777(today’s U.S. Highway 226) and camping nearWebb, North Carolina, where the Toe and Canerivers join to form the Nolichucky. In our model,the party did not have to turn north; they had al-ready made a turn to the south, down the valley.

As the group continued to move westward onSaturday, May 29, they spent the night by a “largestream” that they had crossed often during theirjourney (Fernández de Oviedo y Valdés1851:563). The GIS route in fact follows the No-lichucky closely, and the camp was probably inthe area of Johnson City. Beck (1997:164) sug-gests that they were along the Nolichucky, pitchingcamp near Erwin, Tennessee. The Erwin route hasa significant bottleneck where the Nolichuckypasses westward into the valley. This bottleneckis not encountered in the more northerly GIS route.

GuasiliThe de Soto army reached Guasili early in themorning on Sunday, May 30, 1540 (Fernándezde Oviedo y Valdés 1851:563). No mention ofan alliance of Guasili and Joara, even thoughGuasili was perhaps within the sphere of Cofi-tachequi, and the flight of the ruler of Cofi-tachequi well before she arrived at Guasili suggestthat any political affiliation was tenuous at best;Guasili was a place apart. The phonetic structureof the place-name is consistent with Cherokee.Labiovelar clusters, such as /gw/, are common inCherokee but rare to nonexistent in Muskogeanand other languages of the Southeast (Booker etal. 1992). The suffix may be the Cherokee loca-tive /-yi/. Guasili does not appear in the Pardoaccounts, though another name in the Pardo ac-counts may be its 1560s equivalent, just as Chiahais also called Olamico in the Pardo accounts. It

could also be that Guasili was abandoned by the1560s.

The northern GIS route locates Guasili at theJackson Farm/Plum Grove site (40WG17), a cor-relation first suggested by Beck (1997:Figure 1).The combination of late Pisgah shell-temperedand Burke-like sand-tempered ceramics (Boyd1986; Dickens 1980; Keel 1976) suggests thatPlum Grove was occupied during the de Soto era(Beck 1997:164). Sixteenth- and seventeenth-cen-tury European glass beads, an iron ax, and a brassgorget from 40WG17 confirm this dating (Smith1987:49). The middle French Broad, where Hud-son et al. (1984:74) locate Guasili, has no knownlate prehistoric–protohistoric sites (Beck 1997).

On Monday, May 31, the company left Guasiliand camped in an oak wood by a river (Fernándezde Oviedo y Valdés 1851:563). Beck (1997:164)suggests that the army camped along the No-lichucky near Philadelphia, Tennessee. The GISroute places the de Soto party along the No-lichucky until they reached the next settlement,Canosahaqui. This point in the journey is the oneexception to daily travel distances. The only wayfor the rest of the trip to concur with physio-graphically distinct features is for the army tohave lingered for a day along this section of theNolichucky. A reason for this pause in the tripmay have been to interact with the dense settle-ment in the region. The oak forest camp wasprobably along the GIS-defined route, at a slightremove from the many settlements adjacent tothe Nolichucky.

Canosahaqui (Canasoga)Ranjel stated that they passed “Canasoga” onTuesday, June 1, and camped in the open country(al campo) “beyond Canasoga” that evening. El-vas and Beidma concur that the army spent twodays in the vicinity of Guasili and Canosahaqui(Fernández de Oviedo y Valdés 1851:563; Fidalgodelvas 1557:62v). The implication is that Canosa-haqui (the name used in the Pardo accounts;“Canasagua” in Elvas) was reached sometimeduring June 1, and they passed farther west fromthere (Fidalgo delvas 1557:63v).

Hudson (1990:98) suggests that Pardo’sCanosahaqui is equivalent to de Soto’s Canasogaand located in the vicinity of Hot Springs, NorthCarolina. Hot Springs is located along the south-


ern GIS route, but well in the mountains. The deSoto accounts do not mention difficult terrain inthis part of the trip. Lowery (1901) and Mooney(1900:26) identified Canasoga as a frontier townof the Cherokee. Hudson (1990:98) suggests thatthe best etymology of Canosoga is kanosaka,Muskogean for “those of the mound.” Mooneyreconstructed Canasoga as “Gansâgi (or Gan-ságiyi)” (1900:518–519), a name for several for-mer settlements— on the Tuckasegee River; onthe lower part of Canasauga Creek in McMinnCounty, Tennessee; at the junction of Conasaugaand Coosawatee rivers in Gordon County, Geor-gia; and de Soto’s Canasoga. It is probably mis-leading to think of these towns as absolutely fixedin space. Cherokee towns moved during the his-toric period, yet the identity of the town persisted(Booker et al. 1992:415).

We suggest that Canasoga, Canasahaqui, andNolichucky are variants of the same name basedon the morphological similarity of these place-names and archaeological remains dating to thetime span in question. Ramsey (1853:110–111)stated that the original name of the river was“Nonachunheh,” which appears on Jacob Brown’s1775 deed of conveyance and plat for the No-lichucky Valley lands. The orthography in thesixteenth-century versions of the name indicateda saltillo (glottal stop), making the word mor-phologically similar to Nonachunheh, what laterbecame Nolichucky.

The border of this polity was defined by awelcoming party of 20 Indians, each carrying abasket filled with mulberries (cesto damoras) (Fi-dalgo delvas 1557:63v). Mooney suggested thatthe berries were “more probably the deliciousservice-berry of the southern mountains, whichripens in early summer, while the mulberry ma-tures later” (1900:26). The significance of mul-berries is less as a food source and more how itrelated to commercial production. The de Sototo Pardo time span (1540s to 1560s) was duringthe economic boom of silk in Mexico (Borah1943). Garcilaso commented that “There werelikewise among the foliage many trees with di-verse fruits, and large mulberry trees and moreabundant, than those up until that point one hadseen. We always take particular note of this treebecause of its nobility and for the utility of silk,and its universally high value” [auia assi mismo

entre las ramadas muchos arboles con diuersasfrutas, y grandes morales mayores y mas viciosos,q los q hasta alli se auia visto. Damos siempreparticular noticia deste arbol por la noble a del, ypor la vtilidad de la seda, q dò quiera se deue es-timar en mucho] (Vega 1605:163r). The publishedtranslation, “which ought to make it valuableeverywhere,” misses the point (in Clayton et al.1993:289). Silk and mulberry fodder for it werealready a means to great wealth (Borah 1943).Naturally occurring mulberry trees must haveseemed an unusual bounty to these Spanish ven-turers, as they consistently noted them.

Beck (1997:164) suggests that Canasahaquicould be 40GN9 or 40GN11. Archaeological sur-vey and excavation confirm that 40GN9 was alarge settlement, while 40GN11 was a hamlet.Several kinds of artifacts confirm 40GN9’s six-teenth-century occupation, including altajía(enameled brass) buttons and an iron bar, recov-ered in previous excavations (Smith 1987:49).Our recent intensive survey and excavations at40GN9 have recovered more Spanish artifactsthat further support this dating.

Carefully defined borders between Guasili andCanasahaqui may have been important becausesixteenth- to seventeenth-century towns and poli-ties were tightly packed in the Nolichucky Valley.

40GN9 Spatial Organization and Material CultureOur work in the Nolichucky Valley has employedthe methodologies of intensive survey of artifactdistributions and the study of “nonsite” landscapefeatures such as roads, viewsheds, and access toresources, as well as excavation of structures andfeatures. The survey to define site limits was sys-tematic, comprehensive, and intensive. The over-all size of the 40GN9 town, 2 ha, is comparableto large towns of about the same time period,such as the 4.2-ha Coweeta Creek in North Car-olina (Rodning 2008) and the 2.2-ha King site inGeorgia (Hally 2008). The core of the town hadstructures that surrounded a wide plaza (Figure4). Six 2-x-2-m excavation units tested the north-ern and southern edges of the central cluster ofarchaeological remains, recovering a wealth ofwell-preserved artifacts and features. All of theexcavated soil was water-screened to recoversmall historic artifacts and faunal and botanical


remains. The preservation of materials at the sitewas excellent, and items included bone tools anda wide array of wild food remains. Even thoughthis site has been plowed for agriculture, remainsfrom sub-plow-zone levels were undisturbed, in-cluding an extensive sheet midden and post moldsof varying size and construction methods (Figure5). The northern excavation units recovered evi-dence of a wall constructed of large posts, whichcould have been part of a palisade (Figure 6).

Analysis of the material culture is still underway, but enough has been completed to suggesta substantial Middle Qualla (A.D. 1500–1700)

ceramic assemblage accompanied by Spanish ar-tifacts (Rodning 2008). Forms and decorationscorrelate with three distinct ware types, a blendof several traditions much like Franklin et al.(2010) observed farther north and east at HollistonMills (Figures 7, 8, and 9; Supplemental Table1). Neither Overhill nor Little Egypt (Coosa) dur-ing the Barnett phase has the nearly equal pro-portion of grit-tempered, shell-tempered, andsand-tempered vessels as do the Nolichucky Val-ley assemblages (Hudson 1990:101). Tin-enam-eled earthenware (majolica) was also presentamong the ceramics (Figure 10). These Spanish


Figure 4. Map of 40GN9 showing the locations of excavation units.

Figure 5. Site 40GN9 excavation profile showing preserved features and the midden deposit.

wares included Columbia Plain and Yayal Blue-on-white vessels (Deagan 1987; Goggin 1968).As at Joara and Fort San Juan, the Spanish arti-facts at 40GN9 include quotidian Spanish ceram-ics, not just trade goods. Common vessels maybe a good marker of Spanish presence.

Historic beads include wire-wound blue glassbeads and green seed beads (Figure 11). Wire-wound beads of several solid colors co-occurwith Nueva Cadiz and faceted chevron beads,which securely date from the early to mid–six-teenth century (Mitchem and Leader 1988). Greenseed beads also co-occur with well-dated beadsof the sixteenth and seventeenth centuries (Smithand Good 1982).

40GN11: A Hinterland OccupationAfter leaving Canasahaqui, the de Soto partycrossed the Nolichucky, probably near Ripley Is-land as it is one of the best fords of the river inthis region (Beck 1997:164). Adjacent to the fordon the south side of the river is 40GN11. Beck(1997:164) suggests that crossing the Nolichuckyhere avoided more treacherous fords downriver

and also implies that they knew where they weregoing or were following a well-established route.

Excavations at the Yellow House site(40GN11) recovered evidence of a few structureswithout extensive rebuilding, indicating a rela-tively short-term occupation (Figure 12). Thebulk of the ceramic assemblage is stylisticallymuch like that of 40GN9 and includes a mix ofwares but no Spanish artifacts.

After crossing the river, the de Soto groupcamped in the open near Greeneville, Tennessee(Beck 1997:164). The next place along the route,encountered on Wednesday, June 2, can be posi-tively identified on the basis of physiography.Beck (1997:164) suggests that the army traveledwest along the north side of the Nolichucky (to-day’s State Highway 340, noted in Carey’s 1801map and a postal route in the 1790s), which agreeswith the GIS route (Phillips 1990:30). There, thede Soto army ate a “very great infinity” of mul-berries and eventually camped beside abog/marsh: “á par de una ciénaga” (Fernándezde Oviedo y Valdés 1851:563). Published Englishtranslations use “swamp,” but ciénaga indicates


Figure 6. Site 40GN9 post molds after excavation, indicating a possible palisade wall.

a seasonally inundated place, while pántano is apermanently inundated area. The “swamp” is al-most certainly where Lick Creek meandersthrough a broad, deflated karstic flat near the in-tersection of Highway 11E and Interstate 81 (thelocation of the Lick Creek mound site) (Beck1997:164; cf. Hudson et al. 1984:74), one of fewplaces in the region that floods seasonally at aboutthe time of year of de Soto’s travel.

On Thursday, June 3, 1540, the de Soto armycontinued to follow the Nolichucky along itsnorth side (Fernández de Oviedo y Valdés1851:563). Beck argues that the army continuednorthwest from Warrensburg to the Morristownarea and “then turned south and camped just northof present Witt” (1997:164). The GIS analysisplaces the army in this vicinity, near Susong anda recorded archaeological site.

ChiahaThe people of Chiaha marked their eastern bound-ary with several welcoming parties. The first wasof 20 Indians, each carrying a basket of mulberries.On Friday, June 4, more Chiaha Indians “came inpeace and brought corn” (Fernández de Oviedo yValdés 1851:563), after which the army continuedsouth, pitching camp beside Long Creek near Dan-dridge, Tennessee (Beck 1997:164). That the wel-coming parties greeted the army so far from themain settlement suggests that this was indeed aprovince with a large territory and multiple set-


Figure 7. Shell-tempered ceramics from the NolichuckyValley.

Figure 8. Sand-tempered ceramics from the NolichuckyValley.

Figure 9. Grit-tempered stamped jars from theNolichucky Valley.

tlements. Two leagues before arriving at Chiaha,15 Indians again brought maize, told the armyabout barbacoas full of maize awaiting them, andpledged that everyone was at their service (Fidalgodelvas 1557:63r–63v). The pattern of the firsttown of Chiaha being a zone of encounter/gate-keeper, with the center of political power fartherwest, occurred in the Pardo accounts as well. Thetime spent traveling to Chiaha varies among dif-ferent accounts, but they all agree that the day ofarrival was July 5. Overall, the time frame for ar-

riving at Zimmerman’s Island along the No-lichucky route fits better than Hudson et al.’s(1984) French Broad route (Beck 1997:165; De-Pratter et al. 1983:145–156).

Events in the de Soto and Pardo accounts sup-port that Chiaha was a secondary center of theparamount chiefdom of Coosa and relatively largerthan centers such as Maniatique (Beck 1997:167;Booker et al. 1992:408; DePratter et al. 1983:146;Hudson et al. 1984:75; Hudson et al. 1985). Chiahaplayed a crucial role in the Iberian occupation of


Figure 10. Yayal Blue-on-white majolica from 40GN9 excavations.

Figure 11. Glass beads recovered from 40GN9 excavations.

the Southeast as a place that the Spanish felt com-fortable enough to linger, build a fort, and relyupon during excursions to other areas (Fidalgodelvas 1557:64v). Pardo and his officers also de-pended upon Chiaha as an ally, with Moyano build-ing a small fort in the principal town of the chiefof Chiaha in 1567 (Martinez 1990:320). The townplan of a palisaded village was different fromplaces farther east. Booker et al. (1992) make aconvincing case that the people in the region ofChiaha spoke Koasati, and Koasati place-namesare found west of there. The chief’s florid wel-coming speech indicated that he was sent fromGuaxulle, suggesting some degree of political in-teraction, but the western edge of what may havebeen the northern reaches of Cherokee settlementdid not extend much beyond present-day GreeneCounty (Fidalgo delvas 1557:63v). The overallimpression is that Chiaha was a political, linguistic,and social entity distinct from areas to the east.

Chisca

Chisca was a political player that did not lie alongthe camino real and therefore is unlikely to havebeen along the Nolichucky (Beck 1997:166).Chisca was to the north of Chiaha and had a metalfoundry, piquing Spanish interest (Fidalgo delvas1557:67r, v). A primary reliance on Southeasterninstead of Great Lakes copper sources occurredrelatively late during the Mississippian period(Beck 1997:166; Goad 1978; Treveylan 2004),along with a settlement shift farther west andsouth of the Fort Ancient region on the north-eastern Mississippian periphery, perhaps con-tributing to more interaction with eastern Ten-nessee (Pollack et al. 2002:207; Williams andShapiro 1990). After A.D. 1500, Fort Ancient–region mortuary contexts contain the second-greatest concentration of Citico-style rattlesnakegorgets outside of eastern Tennessee (Drooker


Figure 12. Structural remains from 40GN11.

1997:294; Polhemus 1987; Schroedl 1998).Scouts sent by de Soto returned after 10 days

of travel in mountain ranges so steep that campingor marching was impossible, making the distanceto Chisca about 25 leagues (86.75 mi) (Fernándezde Oviedo y Valdés 1851:564; Fidalgo delvas1557:64v, 66v). Pardo accounts indicate thatChisca included Maniatique and Guapere, townsdestroyed by Moyano in 1567. Guapere couldhave been on the Watauga River in upper eastTennessee (cf. DePratter et al. 1983:131) butcould not be Cherokee, as Cherokee has no /p/(Booker et al. 1992). Luisa Mendez, cacica ofManiatique, reported that her town had saltsprings (Hudson 1990:190, 201; Martinez1990:320), perhaps those of Saltville (Beck1997:166). A cost distance analysis comparingthe route to Saltville vs. a Fort Ancient site,Hardin Village, shows that the cost is very highto head north, while the northeastern route toSaltville better fits distance estimates given inthe Spanish accounts; but the foray returned alongpart of the camino real, something accounts typ-ically would have noted (Figure 13). Those ofChisca never sent reinforcements to Chiaha dur-ing Spanish incursions; the political break presentin 1540 appears to have persisted in Pardo’s time.The copper working, salt sources, and non-Iro-quoian linguistic affiliation all point to some de-gree of Fort Ancient affiliation for Chisca.

Alternate Routes

Pardo and soldiers of Fort San Juan made twoforays from Joara to the west in 1567, and thoseaccounts name places not mentioned in the deSoto chronicles. The Pardo expedition that left onSeptember 24, 1567, was small and trying to reachMoyano quickly to rescue him from a siege at hisfort in Chiaha. Both Beck (1997) and DePratteret al. (1983) suggest that the route they took wasthrough the Swannanoa Gap, along the quicksouthern route of the GIS analysis, but evidencesupports travel along the north route. The newtown names in 1567 may reflect a rise to promi-nence of polities in the Holston and Nolichuckydrainages in the years since de Soto’s expedition.

Tocae'eAfter traveling three days from Joara, Pardoreached Tocae'e (TocaE, with the saltillo indicatedby E; Tocae'e was also spelled Tocae, Tocar[e],and Toca), “a place that is over the top of theridge,” the western side of the Blue Ridge (Ban-dera 1990:266 [1569]). Booker et al. (1992:425)suggest that Tocae'e was Toqua or possibly Dak-wai, along the French Broad River according tonineteenth-century Cherokees, about 6 mi aboveHot Springs in Madison County, North Carolina(Hudson 1990:95). The stated distance of a littlemore than 14 leagues and the potential for six-

Sampeck et al.] GIS MODELING OF DE SOTO’S ROUTE FROM JOARA TO CHIAHA 61E

leva

tion

(met

ers)

200

300

400

500

600

700

0 200100 400 600 700 kilometers500300

0 4020 80 12010060 140 leagues

(Chiaha)

(Saltville)

(HardinVillage)

Distance

Figure 13. Cost distance analysis of routes to possible locations for Chisca: Saltville and the Fort Ancient area of south-eastern Kentucky.

teenth-century settlement convinced DePratter etal. (1983:143) and Beck (1997:167) that Tocae'ewas near the present town of Asheville, NorthCarolina. The GIS routes show that the southernroute spends the first leg of the journey on rela-tively flat ground and would be near the highestelevation after three days’ worth of travel,whereas on the northern route the travelers couldindeed have crossed the range.

Tocae'e was linguistically distinct from Joara,as other interpreters besides the Muskogean-speaking Rufín were needed to communicate withits orata. Pardo met with four Cherokee-speakingoratas at Tocae'e, and Tocae'e is consistently as-sociated with Cherokee towns (Hudson 1990:96).The travel distance and physiographic descriptionfit well with Watauga Old Fields, near Elizabeth-ton, Tennessee.

CauchiOn October 2, 1567, after leaving Tocae'e, Pardonext encountered Cauchi, in an area of very largebottomlands. Suggested locations of Cauchi in-clude the vicinity of present-day Marshall, NorthCarolina (Hudson 1990:96), and the GardenCreek site, on the upper Pigeon River near Can-ton, North Carolina (Beck 1997:167; DePratteret al. 1983:144; Dickens 1976:207). Cauchi fitsbetter with the upper Pigeon than with the middleFrench Broad, where alluvial lands are narrow,and the Swannanoa River may have been aban-doned by the time of the de Soto and Pardo en-tradas (Hudson 1990:100). The upper Pigeon,however, is still much narrower than the No-lichucky Valley. One of the best clues aboutCauchi’s location is a set of distances in the“long” Bandera (1990:f18v [1569]) relación. Inthe list of distances to several other Spanishforts/towns from Joara, Cauchi is northeast fromJoara, which makes it impossible for it to be inthe Asheville Basin to the southeast but doesplace it within the Nolichucky Valley (Table 1).Booker et al. (1992:437) and Hudson (1990:97)tentatively identify Cauchi as Cherokee and asde Soto’s Guasili (Booker et al. 1992:406). IfCauchi was Guasili, then it was likely in the No-lichucky Valley. The dense concentration of con-tact-period settlements was a strategic locale fora Spanish fort to manage tribute collection andfoster political alliances. The upper Pigeon does

not appear to have the same human resources,nor is the valley as broad.

Why did Pardo name some settlements that deSoto did not? Established settlements were not al-ways noted in each account, and some settlementsmay have shifted even closer to the camino realduring the 20 or so intervening years since deSoto’s visit. The route may have acted as a magnet,drawing closer new or consolidating establishedalliances as part of political jockeying within thecontext of Spanish tribute demands. Tocae'e andCauchi reiterated ties between them, as well asCherokee towns whose names persisted into theeighteenth century. Cauchi was probably just eastof a cultural, linguistic, and political divide, beyondwhich lay the polity of Coosa, where no Cherokeeoratas met with Spaniards (Beck 1997:167;Booker et al. 1992:408, 429; Hudson 1990:97).

Overall, the locations of sixteenth- and sev-enteenth-century towns align along the two pro-posed GIS routes. Further, the locations of manyof the towns correspond with 5-league stoppingpoints along both paths. This finding suggeststhat typical army travel time affected settlementlocation in the sixteenth-century Southeast.

A Revolutionary-Period RouteA later historical example supports the GIS north-ern route and offers important refinements. Theclosest approximation of Spanish expeditions interms of the number of people, pieces of equip-ment, and livestock is the military foray of theOvermountain Men from Sycamore Shoals, Ten-nessee, to Quaker Fields on their way to fight theBattle of King’s Mountain in fall 1780 (Alderman1970; Draper 1881; U.S. Congress 1928). Likede Soto’s army, the Overmountain Men numberedabout 400 and brought weapons and livestockwith them. From their start near Elizabethton,Tennessee, they traveled along the Doe River to


Table 1. Distances to Native American and Spanish Townsand Forts from Joara.

Sixteenth-Century Town/ DistanceFort Name Direction (Leagues)Chiaha/San Pedro West 50Cauchi/San Pablo Northwest 28Guatari/Santiago/Salamanca Northeast 40Canos/Santo Tomás Southeast 45Canos to Santa Elena South 55

its headwaters and then passed to the North Toeheadwaters. The route is close to the GIS pathand farther north than Beck’s (1997), but theSycamore Shoals approach offers an even widerpassage to the mountains.

Discussion and Conclusion

This research shows that routes in the Southeastwere vectors of political and social affiliation anddifference, as well as how GIS modeling of pathscan take into consideration the crucial factor ofthe scale of interaction that created these socialdynamics. How many people traveled together ata time is a question relevant to many parts of theworld in different periods. The way NolichuckyValley residents related to others via their roadslends a more nuanced view of their place as cul-tural pioneers and shows that the movement oflarge groups of people might have been a basisfor dramatic changes of the period. Sixteenth- toseventeenth-century settlement was aligned to ac-cess the wide, northern travel route, a main arteryof exchange in the Southeast interior that contin-ued in use from de Soto’s to Pardo’s time despitesubstantial shifts in the political landscape. Theemphasis on the movement of large groups seemsto have fostered power for Nolichucky and Joarapolities compared with Cofitachequi and Coosa;perhaps an emphasis on mobility and flexibilityrather than a large territory made polities moredurable. Social and political power often involvesissues of scale, but in this case it is engagementwith the scale of traveling groups and flows ofpeople more so than the size of territory, factorsthat come to light through the analysis of roads.

Sixteenth-century accounts of the Nolichuckysupport Cherokee linguistic affiliation and statusas a distinct polity, defined along these pathwaysby social displays of welcoming parties. The ma-terial culture of the Nolichucky settlements hasstrong stylistic ties to known Cherokee towns onthe east side of the Blue Ridge, yet also Burkeand Dan River wares as well as possible precur-sors of later Overhill Cherokee ceramics. Thisunusual combination of material culture traditionswas facilitated by the close association of No-lichucky towns with the wide highway and theinteraction and movement of large groups of peo-ple as well as the efficient travel of small parties

to the Cherokee heartland. Archaeological evi-dence indicates that Nolichucky sixteenth- to sev-enteenth-century towns were new and did not lastlong, evacuated before any influx of CharlesTown trade goods. Nevertheless, they were theold towns of note to Cherokees in the 1760s:

The only settlements noticed by the Cherokeeswere near Nolichucky; but having been longsettled there, they could not go farther back toothers . . . prior to the year 1652, or as early asthe period when Cornelius Dogherty came tothe nation, Mr. Hicks says about the year 1697[Haywood 1823:237].

The hundred years or so of this component oc-curred during the waning of Cofitachequi and riseof Joara sovereignty, defined by material and so-cial actions along the northern route (Beck 1997;Booker et al. 1992:406; Ethridge 2010; Merrell1989). Later Qualla pottery at Overhill sites prob-ably represents the movement of some Cherokeehouseholds and towns to Overhill settlements dur-ing the late 1600s and 1700s (Goodwin 1977) asa response to encroachment by Europeans in thesouthern Appalachians, as well as destabilizationdue to the deerskin and slave trade and new kindsof conflict and warfare (Ethridge 2006; Gallay2002; Marcoux 2008; Smith 1987). This researchsuggests that Cherokee reshaping of the politicallandscape to include westward movement beganbefore any European arrived in the region.

This study of routes emphasizes that crucialelements of human activity are not just “sites”—single, discrete, bounded units of human activity(Dunnell 1992). The road is a dynamic force, aplace of encounter for people that also createsphysical constraints. Rather than neutral back-drops for movement, the roads themselves affecthow and where time is spent and with whomtravelers interact. The ethnohistoric and archae-ological data support that sixteenth-century roadnetworks were least-cost paths and that Spanishentradas followed them. Factoring in the size ofthe traveling party in the GIS analysis resulted intwo significantly different pathways, with onlythe northern route conforming to the vivid ac-counts of the de Soto expedition. The surprisingcorrelation of both routes and their 5-league stop-ping points with protohistoric archaeological sitesunderscores that not only the routes themselves


but typical travel times also affect settlement lo-cation. Roads are human routes and need to bemodeled in anthropological terms.

Acknowledgments. Several seasons of archaeological researchwere made possible by institutional support from Illinois StateUniversity, several grants from the Tribal Historic PreservationOffice of the Eastern Band of Cherokee Indians, and a grantfrom the Cherokee Preservation Foundation. This researchwould not have been possible without the very generous co-operation of Mike Moore and Susanne Hoyal at the state ofTennessee Archaeology Division. Susan Danforth, George S.Parker Curator of Maps and Prints, and Ken Ward, Maury A.Bromsen Curator of Latin American Books at the John CarterBrown Library, Brown University, both graciously providedinvaluable assistance for this research. We also wish to thankRuss Townsend, Chris Rodning, Beau Carroll, Johi Griffin,Miranda Panther, Bran Burgess, and Tyler Howe for their in-sightful comments and support of our endeavors. Joey andAngie Webb, Bobby Broyles, and Billy and Rosa Lea Proffittprovided invaluable logistical support and encouragement, forwhich we are thankful. Any errors or omissions are our own.

Data Availability Statement. The artifacts, survey and exca-vation records and drawings, and ecodata are all being curatedat Illinois State University in collaboration with the EasternBand of Cherokee Indians Tribal Historic Preservation Officewhile analysis is ongoing. The digital databases of site ageand location are managed by the state of Tennessee, Divisionof Archaeology.

Supplemental Materials. Supplemental materials are linkedto the online version of the paper, which is accessible via theSAA member login at www.saa.org/members-login. Supplemental Table 1. Characteristics of Sixteenth to Sev-enteenth-Century Indigenous Ceramic Wares in the No-lichucky Valley.

References CitedAlderman, Pat 1970 The Overmountain Men. Overmountain Press, Johnson

City.Anderson, David G. 1994 The Savannah River Chiefdoms: Political Change in

the Late Prehistoric Southeast. University of AlabamaPress, Tuscaloosa.

Ashmore, Wendy, and A. Bernard Knapp (editors) 1999 Archaeologies of Landscape: Contemporary Per-

spectives. Blackwell Publishers, Malden, Massachusetts.Bandera, Juan de la, II 1990 [1569] Proceedings for the Account Which Captain

Juan Pardo Gave of the Entrance Which He Made intothe Land of the Floridas. Translated by Paul Hoffman. InThe Juan Pardo Expeditions: Explorations of the Carolinasand Tennessee, 1566–1568, by Charles Hudson, pp.205–296. Smithsonian Institution Press, Washington, D.C.

Bauer, Brian 1992 Ritual Pathways of the Inca: An Analysis of the Collasuyu

Ceques in Cuzco. Latin American Antiquity 3:183–205.Beck, Robin A. 1997 From Joara to Chiaha: Spanish Exploration of the

Appalachian Summit Area, 1540–1568. Southeastern Ar-chaeology 16:162–169.

2013 Chiefdoms, Collapse, and Coalescence in the EarlyAmerican South. Cambridge University Press, Cambridge.

Beck, Robin A., Jr., David G. Moore, and Christopher B. Rodning 2006 Identifying Fort San Juan: A Sixteenth-Century

Spanish Occupation at the Berry Site, North Carolina.Southeastern Archaeology 25:65–77.

Booker, Karen M., Charles M. Hudson, and Robert L. Rankin 1992 Name Identification and Multilingualism in the Six-

teenth-Century Southeast. Ethnohistory 39:399–451.Borah, Woodrow 1943 Silk Raising in Colonial Mexico. University of Cali-

fornia Press, Berkeley.Boyd, C. Clifford, Jr. 1986 Archaeological Investigations in the Watauga Reservoir,

Carter and Johnson Counties, Tennessee. University ofTennessee Department of Anthropology Report of Inves-tigations No. 44 and Tennessee Valley Authority, Publi-cations in Anthropology, No. 46.

Chardon, R. 1980 The Elusive Spanish League: A Problem of Mea-

surement in Sixteenth-Century New Spain. HispanicAmerican Historical Review 60:294–302.

Clayton, Lawrence, Vernon James Knight, Jr., and Edward C.Moore (editors)

1993 The De Soto Chronicles: The Expedition of Hernandode Soto to North America in 1539–1543. 2 vols. Universityof Alabama Press, Tuscaloosa.

Cumming, William P. 1966 North Carolina in Maps. North Carolina Division of

Archives and History, Raleigh.Deagan, Kathleen 1987 Artifacts of the Spanish Colonies of Florida and the

Caribbean, 1500–1800, Vol. 1: Ceramics, Glassware, andBeads. Smithsonian Institution Press, Washington, D.C.

de Laet, Johannes 1625 Betchinvinghe van West-Indien door Ioannes de Laet

Tweede druck: In ontallÿcke plaetsen ver: Betert, ver-meerdert, met eenige nieuwe Caerten, beelden van ver-scheyden dieren ende panten verciert. Elsevier, Leiden.

DeMers, Michael N. 2003 Fundamentals of Geographic Information Systems.

2nd ed. John Wiley and Sons, Somerset, New Jersey.DePratter, Chester, Charles Hudson, and Marvin Smith 1983 Juan Pardo’s Explorations in the Interior Southeast,

1566–1568. Florida Historical Quarterly 62:125–158.Dickens, Roy S. 1976 Cherokee Prehistory: The Pisgah Phase in the Ap-

palachian Summit. University of Tennessee Press, Knoxville. 1980 Preliminary Report on Archaeological Investigations

at the Plum Grove Site (40W gl7), Washington County,Tennessee. Unpublished MS on file, Department of An-thropology, Georgia State University, Atlanta.

Doyle, James A., Thomas G. Garrison, and Stephen Houston 2012 Watchful Realms: Integrating GIS Analysis and

Political History in the Southern Maya Lowlands. Antiquity86:792–807.

Draper, Lyman C. 1881 King’s Mountain and Its Heroes: History of the

Battle of King’s Mountain, October 7th, 1780, and theEvents Which Led to It. Cincinnati.

Drooker, Penelope B. 1997 The View from Madisonville: Protohistoric Western

Fort Ancient Interaction Patterns. Memoirs No. 31.Museum of Anthropology, University of Michigan, AnnArbor.


http://www.saa.org/members-login

Dunnell, Robert C. 1992 The Notion Site. In Space, Time, and Archaeological

Landscapes, edited by Jacqueline Rossignol and LuAnnWandsnider, pp. 21–41. Plenum, New York.

Ethridge, Robbie 2006 Creek Country: The Creek Indians and Their World.

University of North Carolina Press, Chapel Hill. 2010 From Chicaza to Chickasaw: The European Invasion

and the Transformation of the Mississippian World, 1540–1715. University of North Carolina Press, Chapel Hill.

Fernández de Oviedo y Valdés, Gonzalo 1851 Historia general y natural de las Indias, islas y tierre-

firme del mar océano, / por el capitan Gonzalo Fernandezde Oviedo y Valdés . . . ; Publicala la Real Academia de laHistoria, cotejada con el códice original, enriquecida conlas enmiendas y adiciones del autor, é ilustrada con la viday el juicio de las obras del mismo por d. José Amador delos Rios. Real Academia de la Historia, Madrid.

Fidalgo delvas 1557 Relaçam verdadeira dos trabalhos q[ue] ho gouernador

do[m] Ferna[n]do d[e] Souto [e] certos fidalgos portuguesespassarom no d[e]scobrime[n]to da prouincia da Frolida [sic].[Évora]: Foy impressa . . . em casa de Andree de Burgos . . .acabouse aos dez dias de febreiro do anno de mil [e] quinhentos[e] cincoenta [e] sete annos. na nobre [e] sempre leal cidadede Euora. André de Burgos, Évora. Electronic document,http://josiah.brown.edu/search~S6?/tRela{u00E7}am+verdadeira/trelacam+verdadeira/-3%2C-1%2C0%2CB/frameset&FF=trela-cam+verdadeira&1%2C1%2C.

Franklin, Jay, Elizabeth K. Price, and Lucinda M. Langston 2010 The Mortuary Assemblage from the Holliston Mills

Site, a Mississippian Town in Upper East Tennessee. InMississippian Mortuary Practices: Beyond Hierarchyand the Representationist Perspective, edited by Lynne P.Sullivan and Robert C. Mainfort, Jr., pp. 250–269.University Press of Florida, Gainesville.

Gallay, Alan 2002 The Indian Slave Trade: The Rise of the English

Empire in the American South, 1670–1717. Yale UniversityPress, New Haven.

Galloway, Patricia (editor) 1993 The Hernando de Soto Expedition: History, Histori-

ography, and “Discovery” in the Southeast. Universityof Nebraska Press, Lincoln.

Goad, Sharon I. 1978 Exchange Networks in Prehistoric Southeastern

United States. Ph.D. dissertation, Department of Anthro-pology, University of Georgia, Athens.

Goddard, Ives 2005 The Indigenous Languages of the Southeast. An-

thropological Linguistics 47:1–60.Goggin, John M. 1968 Spanish Majolica in the New World: Types of the

Sixteenth to Eighteenth Centuries. Publications in An-thropology Vol. 72. Yale University Press, New Haven.

Goodwin, Gary C. 1977 Cherokees in Transition: A Study of Changing Culture

and Environment Prior to 1775. Research Paper No. 181.Department of Geography, University of Chicago, Chicago.

Hally, David 2008 King: The Social Archaeology of a Late Mississippian

Town in Northwestern Georgia. University of AlabamaPress, Tuscaloosa.

Hassig, Ross 1988 Aztec Warfare: Imperial Expansion and Political

Control. Civilization of the American Indian Series No.188. University of Oklahoma Press, Norman.

1991 Roads, Routes, and Ties that Bind. In Ancient RoadNetworks and Settlement Hierarchies in the New World,edited by Charles Trombold, pp. 17–27. Cambridge Uni-versity Press, Cambridge.

Haywood, John 1823 The Aboriginal and Natural History of Tennessee,

Up to the First Settlements Therein by the White PeopleThis Year 1768. George Wilson, Nashville.

Hudson, Charles 1990 The Juan Pardo Expeditions: Explorations of the

Carolinas and Tennessee, 1566–1568. Smithsonian Insti-tution Press, Washington, D.C.

Hudson, Charles, Marvin Smith, and Chester DePratter 1984 The Hernando de Soto Expedition: From Apalachee

to Chiaha. Southeastern Archaeology 3:65–77.Hudson, Charles, Marvin Smith, David Hally, Richard Polhemus,

and Chester DePratter 1985 Coosa: A Chiefdom in the Sixteenth-Century South-

eastern United States. American Antiquity 50:723–737.Hunter, Richard, and Andrew Sluyter 2011 How Incipient Colonies Create Territory: The Textual

Surveys of New Spain, 1520–1620s. Journal of HistoricalGeography 37:288–299.

Hutson, Scott R., Aline Magnoni, and Travis Stanton 2012 “All that Is Solid . . .”: Sacbes, Settlement, and Semiotics

at Tzacauil, Yucatan. Ancient Mesoamerica 23:297–311.Hyslop, John 1984 Inka Road System. Academic Press, New York.Jarvis, A., H. I. Reuter, A. Nelson, and E. Guevara 2008 Hole-Filled SRTM for the Globe Version 4. Electronic

document, http://www.cgiar-csi.org.Jensen, John R. 2005 Introductory Digital Image Processing. 3rd ed.

Pearson Prentice Hall, Upper Saddle River, New Jersey.Kantner, John 1997 Ancient Roads, Modern Mapping: Evaluating Pre-

historic Chaco Anasazi Roadways Using GIS Technology.Expedition 39(3):49–60.

Karttunen, Frances 1992 An Analytical Dictionary of Nahuatl. University of

Oklahoma Press, Norman.Keel, Bennie C. 1976 Cherokee Archaeology: A Study of the Appalachian

Summit. University of Tennessee Press, Knoxville.King, Adam 2003 Etowah: The Political History of a Chiefdom Capital.

University of Alabama Press, Tuscaloosa.Lowery, Woodbury 1901 The Spanish Settlements Within the Present Limits of

the United States. G. P. Putnam’s Sons, New York.MacLeod, Murdo 2008 Spanish Central America: A Socioeconomic History,

1520–1720. University of Texas Press, Austin.Marcoux, Jon Bernard 2008 Cherokee Households and Communities in the English

Contact Period, A.D. 1670–1740. Ph.D. dissertation, De-partment of Anthropology, University of North Carolina,Chapel Hill.

Martinez, Juan 1990 Relation. Translated by Paul Hoffman. In The Juan

Pardo Expeditions: Explorations of the Carolinas andTennessee, 1566–1568, by Charles Hudson, pp. 317–372. Smithsonian Institution Press, Washington, D.C.

Merrell, James H. 1989 The Indians’ New World: Catawbas and Their Neigh-

bors from European Contact Through the Era of Removal.University of North Carolina Press, Chapel Hill.


http://www.ingentaconnect.com/content/external-references?article=0002-7316()50L.723%5Baid=9268828%5D


http://www.cgiar-csi.org

Milanich, Jerald T. 1993 The Hernando de Soto Expedition and Spain’s Effort

to Colonize North America. In The Expedition of Hernandode Soto West of the Mississippi, 1541–1543, edited byGloria A. Young and Michael P. Hoffman, pp. 10–28.University of Arkansas Press, Fayetteville.

Milanich, Jerald T., and Charles Hudson 1993 Hernando de Soto and the Indians of Florida. Uni-

versity of Florida Press, Gainesville.Milner, George R., and George Chaplin 2010 Eastern North American Population at A.D. 1500.

American Antiquity 75:707–726.Mitchem, Jeffrey, and Jonathan Leader 1988 Early Sixteenth-Century Beads from the Tatham

Mound, Citrus County, Florida: Data and Interpretations.Florida Anthropologist 41:42–60.

Mooney, James 1894 Siouan Tribes of the East. Bulletin No. 22. Bureau of

American Ethnology, Smithsonian Institution, Washington,D.C.

1900 Myths of the Cherokee. Annual Report No. 19. Bureauof American Ethnology, Smithsonian Institution, Wash-ington, D.C.

Moore, David 2002 Catawba Valley Mississippian: Ceramics, Chronology,

and Catawba Indians. University of Alabama Press, Tuscaloosa.Phillips, P. J. 1990 Never a Safe Road: Postal Communications in the

Southwest Territory. Journal of East Tennessee History62:18–32.

Polhemus, Richard 1987 The Toqua Site, 40MR6: A Late Mississippian, Dallas

Phase Town. Tennessee Valley Authority, Knoxville.Pollack, David, A. Gwynn Henderson, and Christopher T. Begley 2002 Fort Ancient/Mississippian Interaction on the North-

eastern Periphery. Southeastern Archaeology 21:206–220.Purser, Margaret 1989 All Roads Lead to Winnemucca: Local Road Systems

and Community Material Culture in Nineteenth-CenturyNevada. In Perspectives in Vernacular Architecture, editedby Thomas Carter and Bernard Herman, pp. 120–134.University of Missouri Press, Columbia.

Rabus, Bernhard, Michael Eineder, Achim Roth, and RichardBamler

2003 The Shuttle Radar Topography Mission— A NewClass of Digital Elevation Models Acquired by SpaceborneRadar. ISPRS Journal of Photogrammetry and RemoteSensing 57:241–262.

Ramsey, J. G. M. 1853 The Annals of Tennessee to the End of the Eighteenth

Century: Comprising Its Settlement, as the Watauga As-sociation, from 1769 to 1777: A Part of North-Carolina,from 1777 to 1784: The State of Franklin, from 1784 to1788: A Part of North-Carolina, from 1788 to 1790: TheTerritory of the U. States, South of the Ohio, from 1790 to1796: The State of Tennessee, from 1796 to 1800. J.Russell, Charleston, South Carolina.

Reuter, H. I., A. Nelson, and A. Jarvis 2007 An Evaluation of Void-Filling Interpolation Methods

for SRTM Data. International Journal of Geographic In-formation Science 21:983–1008.

Rodning, Christopher 2008 Temporal Variation in Qualla Pottery at Coweeta

Creek. North Carolina Archaeology 57:1–49.Rudes, Blair 2004 Multilingualism in the South: A Carolinas Case

Study. In Linguistic Diversity in the South: Changing

Codes, Practices, and Ideology, edited by Margaret C.Bender, pp. 37–49. Southern Anthropological SocietyProceedings Book 37. University of Georgia Press, Athens.

Sampeck, Kathryn 2007 Late Postclassic to Colonial Landscapes and Political

Economy of the Izalcos Region, El Salvador. Ph.D. dis-sertation, Department of Anthropology, Tulane University,New Orleans.

2014 From Ancient Altepetl to Modern Municipio: Surveyingas Power in Colonial Guatemala. International Journalof Historical Archaeology 18:175–203.

Santa Cruz, Alonso de 1892 [1542] Map of the World. Explanations by E. W.

Dahlgren. Royal Printing Office, P. A. Norstedt Si Sóner,Stockholm.

Schroedl, Gerald F. 1998 Mississippian Towns in the Eastern Tennessee Valley.

In Mississippian Towns and Sacred Spaces, edited byBarry Lewis and Charles Stout, pp. 64–92. University ofAlabama Press, Tuscaloosa.

Smith, Marvin 1987 Archaeology of Aboriginal Culture Change in the

Interior Southeast: Depopulation During the Early HistoricPeriod. University Press of Florida, Gainesville.

Smith, Marvin, and Mary Elizabeth Good 1982 Early Sixteenth-Century Glass Beads in the Spanish

Colonial Trade. Cottonlandia Museum Publications,Greenwood, Mississippi.

Snead, James E., Clark L. Erikson, and J. Andrew Darling (editors) 2009 Landscapes of Movement: Trails, Paths, and Roads

in Anthropological Perspective. University of PennsylvaniaPress, Philadelphia.

Sofaer, Anna, Michael Marshall, and Rolf Sinclair 1989 The Great North Road: A Cosmographic Expression

of the Chaco Culture of New Mexico. In World Ar-chaeoastronomy, edited by Anthony Aveni, pp. 365–376.Cambridge University Press, Cambridge.

Swanton, John R. 1985 Final Report of the United States De Soto Expedition

Commission. Smithsonian Institution Press, Washington, D.C.Trevelyan, Amelia M. 2004 Miskwabik Metal of Ritual: Metallurgy in Precontact

Eastern North America. University of Kentucky Press,Lexington.

Trombold, Charles D. (editor) 1991 Ancient Road Networks and Settlement Hierarchies in

the New World. Cambridge University Press, Cambridge.U.S. Congress 1928 Historical Statements Concerning the Battle of Kings

Mountain and the Battle of the Cowpens, South Carolina.70th Congress, 1st Session, House Document No. 328.U.S. Government Printing Office, Washington, D.C.

Vega, Garcilaso de la 1605 La Florida del Ynca: Historia del adelantado Hernando

de Soto, gouernador y capitan general del reyno de laFlorida, y de otros heroicos caualleros españoles è indios.Impresso por Pedro Crasbeeck, Lisbon.

Williams, Mark, and Gary Shapiro (editors) 1990 Lamar Archaeology: Mississippian Chiefdoms in the

Deep South. University of Alabama Press, Tuscaloosa.Worth, John 2007 The Struggle for the Georgia Coast. University of

Alabama Press, Tuscaloosa.

Submitted February 3, 2014; Revised July 9, 2014;Accepted July 19, 2014.

66 AMERICAN ANTIQUITY [Vol. 79, No. 4, 2015]









Documents

Geographic Information System Modeling of De Soto's Route from …my.ilstu.edu/~jthayn/Publications_files/Sampeck copy.pdf · 2015-10-20 · travel illuminate social, political, eco