Society for American Archaeology Exchange Implications of Obsidian Source Analysis from the Lower Rio Verde Valley, Oaxaca, Mexico Author(s): Arthur A. Joyce, J. Michael Elam, Michael D. Glascock, Hector Neff and Marcus Winter Source: Latin American Antiquity, Vol. 6, No. 1 (Mar., 1995), pp. 3-15 Published by: Society for American Archaeology Stable URL: http://www.jstor.org/stable/971597 . Accessed: 20/05/2014 23:42 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Society for American Archaeology is collaborating with JSTOR to digitize, preserve and extend access to Latin American Antiquity. http://www.jstor.org This content downloaded from 128.138.73.68 on Tue, 20 May 2014 23:42:30 PM All use subject to JSTOR Terms and Conditions l_ - EXCHANGE IMPLICATIONS OF OBSIDIAN SOURCE ANALYSIS FROM THE LOWER RIO VERDE VAT,T,1ZY,OAXACA, MEXICO Arthur A. Joyce, J. Michael Elam, Michael D. Glascock, Hector NeS, and Marcus Winter This article considersthe resultsof instrumentalneutron-activation analyses of 61 obsidianartifacts recoveredfrom excavations at four archaeologicalsites in the lower Rio Verdevalley on the Pacifc coast of Oaxaca, Mexico. Determinationsof sourcelocations of theseartifacts permit the examinationof changesin obsidianexchange patterns spanning the late Middle Formativeto the Classic period. The resultsshow that throughmost of this period the importationof obsidianinto the lower Verderegion was dominated by sourcesin the Basin of Mexico and Michoacan. The data allow us to evaluateprevious models of interregionalrelations during the Formativeand Classicperiods, includinginteraction with the highlandcenters of Monte Alban and TeotEhuacan. Este articulo considera los resultadosde analisises instrumentalesde activacion neutronicade 61 artefactosde obsidiana recuperadosde excavacionesen cuatro sitios arqueologicosdel valle inferiordel Rio Verdeen la costa Pacifea de Oaxaca, Mexico. Determinacionesdefuente de estos artefactospermiten la examinacionde cambiosen los patronesde intercambiode obsidianadesde el Formativoal perfodoClasico. Los resultadosmuestran que durante muchode este periodola importacionde obsidianaal valle inferiordel Rio Verdefue dominadapor yacimientosen la cuenca de Mexico y Michoacan. Los datos nos permiten evaluarprevios modelosde relacionesinterregionales durantelos periodosFormativo y Clasico, inclusa interaccioncon los centrosde Monte Albany TeotEhuacan. Recent archaeologicalresearch in the low- today coastal and highland populations con- er Rio Verde valley on the Pacific coast verge at regionalmarkets such as Putla, Tlax- of Oaxaca (Figure 1) has stressed the impor- iaco, and Miahuatlan to exchange goods. tance of interregionalinteraction in the social However, many of the items exchanged be- development of the region (Joyce l991a, tween the coast and highlandsare rarelypre- l 99 lb, 1993). A key stimulus of interaction served in the archaeologicalrecord and, even has been the ecological complementarityof if nonlocal materialsare preserved,it is often the coast relative to highland regions. The difficult to identify them and establish their ethnohistoricalrecord shows that at the time point of origin. of the Spanish conquest a vibrant economy Despite these methodological problems, existed between the coast and highlandswith archaeologicalresearch in the lower Rio Ver- coastal productssuch as shells, cacao, cotton, de valley has begun to outline Prehispanic feathers, salt, fish, and purpura dye trans- exchange routes and patterns of interaction ported to the interior in exchange for items (Joyce 199 1a, 199 1b, 1993). The data have such as cochineal, obsidian, and pulque (del confirmedthe importanceof high1andregions, Paso y Troncoso 1981; Spores 1993). Even especially the Valley of Oaxacaand the Basin ArthurA. Joyce * Departmentof Anthropology,Vanderbilt University, Box 6050-B, Nashville, TN 37235 J. Michael Elam, Michael D. Glascock,and Hector Neff * ResearchReactor Facility, University of Missouri- Columbia,Research Park, Columbia,MO 6521 1 Marcus Winter* Centro Regional de Oaxaca, InstitutoNacional de Antropologiae Historia, Pino Suarez715, 68000 Oaxaca,Oaxaca, Mexico Latin AmericanAntiquity, 6(1), 1995, pp. 3-15. CopyrightC3 1995 by the Society for AmericanArchaeology 3 This content downloaded from 128.138.73.68 on Tue, 20 May 2014 23:42:30 PM All use subject to JSTOR Terms and Conditions 4 LATIN AMERICANANTIQUITY [Vol. 6, No. 1, 1995 AS cEj- QSlin C^n rrS 1 Tub a Orizaba 2 T otihuacan b Zaragoza 3 Cuicullco c Guadalup Victorb 4 Xochicalco d Paredon S lliatacapan * Pachuca 6 Corro d Ias Mina t Otumba 7 Rio VInJo 9 Ucar o 8 Corro do la Cruz 9 La Consntida 10 Charco iRedondo t >, possibl obsidbn 11 Barra Oumbrada *xchangc routs 12 Thrras Largas 13 Mont AIban U Laguna Zop 1S Acaputahua 16 1s1a Corritos 17 Chichon Itza 18 San Juan Figure 1. Mesoamerica,showing regions, sites, obsidiansource locations, and possibleexchange routes mentioned of Mexico, in the externalrelations of coastal cific source locations of 99 to 100 percent of populations throughout the Prehispanic pe- the artifacts recovered in excavations any- riod. The data suggestthat manne shell may where in Mesoamerica (Cobean et al. 1991; have been transportedto the highlandswhile Glascock et al. 1994; R. Zeitlin 1979). The obsidian and pottery were sent to the coast. present study uses data produced by a rapid Source studies have demonstratedthat some instrumental neutron activation analysis of the pottery imported to the lower Verde (INAA) proceduredeveloped by researchers came from the Valley of Oaxaca(Banker and at the Missouri University ResearchReactor Joyce 1991). These data, coupled with com- (MURR) (Elam et al. 1992; Glascock et al. parativestudies of ceramic styles and the ico- 1988, 1994) to determine the source loca- nography of carved stone monuments, have tions for 61 obsidian artifactsfrom the lower been used to develop an outline of the nature Verde. The sample for INAA source deter- and impact of external relations on lower minations consisted of 44 obsidian artifacts Verde society (Joyce 1993; also see Urcid from the site of Rio Viejo, 11 artifactsfrom 1993; J. Zeitlin 1993). This paper continues Cerro de la Cruz, S artifacts from La Con- analysis of the externalrelations of lower Rio sentida, and 1 from BarraQuebrada (Figure Verde valley populations by presenting the 1). There is no doubt that the obsidian was results of chemical source determinationsfor imported into the lower Verde, because 61 obsidian artifactsfrom that region. sourceshave not been located within the state of Oaxaca. Methods The artifacts for INAA source determi- Obsidianhas proven to be especiallyeffective nations were selected from a sample of sev- in studies of Prehispanic exchange because eral hundred obsidian artifacts excavated chemical characterizationcan identify spe- during 1986 and 1988 field seasons.The 1986 This content downloaded from 128.138.73.68 on Tue, 20 May 2014 23:42:30 PM All use subject to JSTOR Terms and Conditions Joyce et al.] OBSIDIAN EXCHANGE PATTERNS IN MEXICO 5 4.5 4.0- / Pachuca s / AM/ SY o Guadalupe Victona / t 3.s- 4 V+ z Zaragcza g 8/ O * b Figure 2. Scatterplotof Mn Ucareo t Otumba versus Na for obsidianartifacts from the lower Rio Verde valley Paredon 2.s comparedto the 95 percentconfi- o 20al 40al 600 800 1000 1200 14aoo dence ellipses for central-Mexi- Mn (ppm) can sources. projectwas a pilot study designedto examine The rapid INAA procedurewas conducted early settlement in the region and to sketch at the Missouri University ResearchReactor the entire Prehispanicsequence of the lower (MURR) using a five-second irradiation,25- Rio Verde because little research had been minute decay, and a 12-minute measure- conducted in the region prior to that time ment. This procedure yields concentration (Gillespie 1987; Grove 1988;Joyce and Win- values for six elements:barium (Ba), chlorine ter 1989). Fieldwork involved site recon- (C1),dysprosium (Dy), potassium (K), man- naissance and test excavations at five sites, ganese (Mn), and sodium (Na). Earlierwork including Barra Quebrada and Cerro de la found the elements Ba, Mn, and Na to be Cruz. The 1988 project expanded on the re- very useful for source diffierentiation(Glas- sults of the previous work, and was focused cock et al. 1994). Source attribution is primarily on social developments at the end achieved by examining scatter plots of ele- of the Formative (Joyce 199 1a, 199 1b). Re- ment pairs for the artifactsprojected against search included broad horizontal exposures the 9S percent confidence ellipses generated of the Late Formative occupation at Cerro from source specimens in the MURR obsid- de la Cruz, excavation of five deep trenches ian data base. For example, Figure 2 shows at Rio Viejo that sectioned deposits from the the artifactsin this studywhen comparedwith Middle Formative to the Late Classic, and the 95 percentconfidence ellipses for the ma- test excavations at two other sites including jor Mexican sources. Although the Paredon the late Middle Formative site of La Con- and Otumbasources overlap in Figure2,