HELDIA NA F Anthropology NEW SERIES, NO. 46 Settlement Patterns in the Albarradas Area of Highland Oaxaca, Mexico: Frontiers, Boundaries, and Interaction Gary M. Feinman Linda M. Nicholas November 13, 2017 Publication 1572 PUBLISHED BY FIELD MUSEUM OF NATURAL HISTORY FIELDIANA Mission Fieldiana is a peer-reviewed monographic series published by the Field Museum of Natural History. Fieldiana focuses on mid-length monographs and scientific papers pertaining to collections and research at The Field Museum. Fieldiana is published in two series, Fieldiana Life and Earth Sciences and Fieldiana Anthropology. Eligibility Field Museum curators, research associates, and full-time scientific professional staff may submit papers for consideration. Edited volumes pertaining to Field Museum collections may also be submitted for consideration under a subsidy arrangement. The submission and peer review of these chaptered volumes should be arranged well in advance with the managing scientific editor and the appropriate associate editor. Submission Procedures Submission procedures are detailed in a separate document called “SUBMISSIONS PROCEDURES” available on the Fieldiana web site: (https://www.fieldmuseum.org/science/research/area/fieldiana/fieldiana-submission-procedures) under the Author’s page. All manuscripts should be submitted to the managing scientific editor. Editorial Board: Managing Scientific Editor Gary Feinman, The Field Museum Thorsten Lumbsch, The Field Museum (gfeinman@fieldmuseum.org) (tlumbsch@fieldmuseum.org) Or: Jérg Frébisch, Museum fiir Naturkunde, Humboldt-Universitat, Associate Editor for this Volume Berlin, Germany Ryan Williams, The Field Museum ae P David L. Hawksworth, Universidad Complutense, Madrid, Spain (rwilliams@fieldmuseum.org) A ' a Corrie Moreau, The Field Museum Associate Editors of Fieldiana ; (cmoreau@fieldmuseum.org) Ken Angielezyk, The Field Museum (kangielczyk@fieldmuseum.org) Peter J. Taylor, University of Venda, Thohoyandou, South Africa Luisa Elvira Belaunde, Museu Nacional, Universidade Federal Library Representative do Rio de Janeiro, Brazil Christine Giannoni, The Field Museum Luis Jaime Castillo Butters, Departamento de Humanidades, Editorial Assistant Pontificia Universidad Catélica del Peru, Programa Arqueoldgi- Jonathan Hense, The Field Museum co San Jose de Moro, Peru (fieldiana@fieldmuseum.org) For subscriptions or individual issue purchases, please contact fieldiana@fieldmuseum.org Cover photograph: The western rock face of Llego Yiachi in the Albarradas area of highland Oaxaca (photograph by Linda Nicholas, 1995) PUBLISHED BY FIELD MUSEUM OF NATURAL HISTORY KIELDIANA Anthropology NEW SERIES, NO. 46 Settlement Patterns in the Albarradas Area of Highland Oaxaca, Mexico: Frontiers, Boundaries, and Interaction Gary M. Feinman and Linda M. Nicholas Field Museum of Natural History 1400 South Lake Shore Drive Chicago, Illinois 60605 USA E-mail: gfeinman@ fieldmuseum.org Accepted July 6, 2017 Published November 13, 2017 Publication 1572 PUBLISHED BY FIELD MUSEUM OF NATURAL HISTORY © 2017 Field Museum of Natural History ISSN 0071-4739 PRINTED IN THE UNITED STATES OF AMERICA Table of Contents RE DAC a an es eee eee en eign tk atm ate, Mummers MRK TN Ee TOU ORL EL OS ee Oh bee Ole ae Swe s fmvii Pee reer ROM WCTIONSAN DN DACKGROUND snip iw ees eens a oe eee eee DLE BD MO. PP GEESE elas ] CHAPTER 2: ARCHAEOLOGICAL SURVEY AND THE GUIRUN/ALBARRADAS REGION ........000 00 cee eee eee eee eee ee 7 CHAPTER 3: MIDDLE AND LATE FORMATIVE (ROSARIO PHASE AND MONTE ALBAN I).... 2.0.0.0... 0000 ce eee eee eee eee eee vi eee alien hUNAIMEORMA TIVE: (VIONTE CNLBANI IL) (0-410 oie pc eae wren BI dl I SE be a wb ec eee oS) ee eer ACI ae ASSIGN MIONTP OA LEANGELIA) iy Jean eae, I DS RT OE be oc eect eens 69 CHAPTER E Om oATES GRASSIGNUVIONRERAT BAN LLIEB=aLV) Se aaa cnc chs oe tener eae ew se alo ee de sw eee nee a Se eee 83 eee eee OO Sia NON BAN DEAN ieee Weer eee ett Ae anne een eee SETI Cl oe Si ea oe we was 97 eee a ike OO) re eee ere ee eet ae. WE aes. ee be OL Me De a ee da was. 115 names CRE ae me ae ee ee eg PD P eS od fs Seon Aue ASPs Om ae eet EE Le ee ae oe ee ee He Pea Ree a a ee wwe nee oe ys Pe ee ee ee I AS Ai ie, SN a ew ee 129 List of Figures Ue I ae VEO EO@) a XAC assy CAT COONS rma oucy ysl ane Facade lw ue Sue ee ne teda Soa ee lS Gee eget Od... A A on 3 nt Crm CMCCEALCAS amt O Nand, Oaxaca me cea 9.20.2 essen tp nee ieee cceecabcciskea visit eqsan eve 2a os eM do SD Be 7 Bicureme a Atchacological sites in the Guirun/Albarradas. survey regione. ..:. 5.6. es ce. be ee EM.) Se he 8 Biontes2- ella! paotosraph mosaic of the Guirun/Albarradas survey region’. 1.) on eae ed Ss ee 9 hecoteee mI OU LAinS aueastemmcdterore allevrom Oaxaca men. faml. & am sencblais. oo aleve iG . ots Seals aeealen OL n a aie Pom TOW ECA OngsO lear tOrimica,(OlOTAda | accu ae ut ox SE ERR Pe ey UL DA ml ek oe 9 Po ono tin iane OLOLAG] Sree eeu eo as ess eRe. ie PPS © SE, 2 Ak Oe Aiea AE led eke kk. 10 Cus em Im asic MEslODC SO lLG@ clio CrITONC my ayy eas ae dusts 1c shea Re et Lhe. MA Mh FI Bemarga'h al +. 10 tier. a omlalaval eveiumolLlnwesteriy pall Ol SULVEY TEQION .......,..... . 05 caess Denies Sete ees BLS SE ek. we. 1] Figure 2.9. Northern part of survey region and mountain pass traversed by modern road .................. 000000005 im 1 RORS AVI, IERYS QCAY OTN GARE go SS SIE a oe er a 12 roc MAC LOU lesmlcab sevilla lita Neglo 50S tes aye. cores uiaetedceep ects s beueat ss aetenies 24s secadades silad P Jueapelt Sle ot). 13 Pecatcr | as incipalancientatravel routes, mto the Valley of Oaxaca. <2... #5.0-c Pie TU 1G Ok Some) OU Seed 14 Bolen Moms iCIM A IDATLAG AS GULVe ViTCClONamewan © amr, me Eien: eel. 12 OUI Sib tv gin. ae lt. 2a ee - 15 en ow a a Bel Conc iLOneranicsties UIE UNGSI LON ame wanra geet a bxcd-1ce aya u Caen kets Sp worse, t Saeed Siiieriicss Jon). whe. 16 eu emer clispaniceadobeawallsiat Cruiruni..)- se ce tnohae sd nee fas Laie atell,. dw 2bpaie oat 16 Pence lone AlvecestOne chuciTormll LOmy ct GUITUn weenie 2 oe aes wees SL. bite Poel hoe. ems Sie. a dl be. 16 fou ee Me CCE Ola all seo INCeUCH OLMSLOL) Dewey sce ae eee) eee es ee ES, es CML NE Sok Aasipged ake. I, ee MRM cr cles OpeSqOle Gul Flim wee trey pete arctan yc cea cncetrtacntes dyhs eran SOSOMS GR Metal Dale besuels oF. 17 Rice tO mOulcstoncubicckea la, Cuadradaryuw.1.1 eed Seeeeiee 46 We. OE Kicmel ls alasmwltiog Ji once Suu. ol). lhe Cree ed mis LOKichiuctinccOmeyDIOCKS.at la CUAUTAC Al) .ctet ies secur carey dtpece Sad bs bdo uk & Sl) SOE MAL a, aul de. 18 feu icr = Ie OaIcificdsOLMalons AtsLiIeryvereleA CUA, sien erss ete ayeiel ces sericueees nso ee. 12 MEGS ade ales Wolo dll 18 noice acalcilicdsormaion.and toumst.arca al HiervecleAgua a. ae0.7. 2220.0. 60).0) 2ou4 Geeta Jl aelA vlad... 18 Pacuice. 2 swNMover > Uuntas avahnereastermsedderolitheay alley Ol Oaxacam, 605 10. Sabu et.b ne ce aalh J) aA oe eke. 19 Eto 7 BR OCKEC IL cOnesOULNCASLEI NSIC COU NICVEs UNAS 1604 rep. s Hee sole ees aoe la eked wale. 19 onion omits ispanicwstoncmwalls ON INUCVe PUNIAS 25-0025 4 ccsetaent Ses De SO). GI, San BR k A e 20 et ome Om cclet iat OCKm@accaOl sl CCOmy) 1AC Ml ya ews ep sek. hora een Ree Seto, MOG fai, davelouws oe. 20 fecioe etehlispamcecove walls.om top Or Licgo Y1achiaae) We weeny Same 2 ACE fe Soi PL OO eel we. 20 Pictice 2s es criculiutal land classes, im the GuirumAlbarradas survey region. 207 00. os oe a a ek. 23 Pence omy alle aol Oaxaca subareas aid MelghDOLmNs TESIONS 4. es cee eee ctr UN AIA 25 [vores Ue alleveon ©axacasubarcas ranked by agicultural productivity. 5 ...00.6 004... Be. ORM OG ie bat os. 25 Figure 2.31. Total potential agricultural productivity of Valley of Oaxaca subareas... 2.2... 5.0.0.0 cee ee 25 lnicuncemiene Gpulaiouncensiies CUuLiap the EatlyaPormatve (0... . see © een). 0 SI SMA Oe LO ee 27 iieliices = a OeamOmpaasesse tiements 1 the: Central. Valleys of Oaxaca, o@... 72... eo ee ele Ie 28 Pieurcss-o ss hosano phase settiements in llacolula and the Albarradas area) ..0). oe 29 Petia ero Montes” Ibanepottery athlleSLA=SLA-52..2 3 oe a oe we . PR. Oe ee eee. 30 iene SaLoOpuiaion densities;aurms ther Middle Pormative a. +..0...i eee. Le a ING ats Le. 30 Figure 3.6. Rosario phase archaeological population as a percentage of maximum potential population.................. 3] i et osano phase potential 2oriculturalssurpliss. wae trea yeh ete BLOM Oe IMS Oe Melle OE ed. 31 Prices oy tonic eAlpan bE arlyslssetuementsan the Central) Valleysob Oaxacate, 20 aim SL DE on cee. 32 imeOtems Us topulavion densiticssdurineionteAlbanibarly lyme SS ead dhe a BE ee a a 33 Figure 3.10. Monte Alban Early I settlements in the Guirtn/Albarradas survey region............ 0.000200 0 0 cece eee 34 Pigure3. 1 Monte Alban Early settlements in’ 1 lacolula‘and the Albarradas area. 6/5 ee Oe ee eee ee 35 ee en eat yeoGa Vivores ciel ele a LASS lan eee eee Ta Ene eee ee OTTO RON IT han ol bal AS als 36 Sane me Eee ati ety ates athlon AlVaTIO bitin ek ete Pt See nentee nce omen ARN S BTR BOUL AIIM Bb Ie) eles 36 Prices ionic Albanttatiyiieterrace-sites im the Central Valleys of Oaxaca li, te. IGS We Ws a es 37 Figure 3.15. Monte Alban Early I archaeological population as a percentage of maximum potential population............ 38 Figure 3.16. Figure 3.17. Figure 3.18. Figure 3.19. Figure 3.20. Figure 3.21. Pigure 3.22. Pipure 323: Figure 3.24. Figure 3.25. Figure 3.26. Figure 3.27. Figure 3.28. Figure 3.29. Figure 3.30. Figure 3.31. Figure 3.32. Figure 3.33. Figure 3.34. Figure 3.35: Figure 3.36. Monte Alban Early I potential agricultural surplus... 2. 3...) 2. 6 ee ee ee ee Monte Alban Late I settlements mn the Central Valleys of Oaxacan ae nyse en Population densities during Monte Alban Late I-72 4525 = eee Average site population during Monte Alban; Late Te pyar ence ee ee Monte Alban Late I settlements in the Guirtn/Albarradas survey region y.0) sae ee ee Steep terraced slope of El Calvario... 2... ae ee ee eee ee Monte Alban Late I settlements in Tlacolula and the Albarradas area’. ase) ee ee site map of El Calvario .....s:c6 6 Web jele aed ace oe eo Ss le Oe oe Monte Alban Late I G-12 ceramics from El Calvario .. a... >) eee Terraced, southwestern slope of El Trigo os. 22.5 ose Site map Of El Trig: a sis. eg i b¥eapacese how tole al 5a et yak eee, a ace er Loma de las Tumbas on flat ridgetop above the Hormiga Colorada. 2) a eee Site map of Loma dé:-las,Tumbas << 5.5. 5 sch « ais on Ce nee tee Site map of Sitio de las Tumas. ac s-4.c 2.00 a se ee) ee ew ee View of Cerro del R10. 6s cde hc eg be es esd Di ey ee Site map of Cerro: del Rio. .ce.. cs ccc cece reece bon oe ee wn evas poate 4 nse np Uae open ge View of El Nopal o.cg cc. hiscc ce Hodes thane sue tated cab 0S Que ned cp kee en oe en Site map of El. Nopal cx sacesaciceyaihenncs vaceeneue ecu IT a TO Pte ee Monte Alban Late. I terrace sites in the CentralsValleys:of Oaxaca lua-is1) acre ene ene emcee Monte Alban Late I archaeological population as a percentage of maximum potential population ............ Monte Alban Late I potential agricultural surplus 2.0.2... .. ... -2y ee ee Figure 4.1. Monte Alban II settlements in the Central Valleys of Oaxaca. 00.93 s0ee ee Figure 4.2. Population densities durmg Monte Alban J].7 2h. 2 se ates ee ee Figure 4.3. Average site population during, Monte Alban II. ...-0..,,.., ese eer ie eee) er eee Figure 4.4. Monte Alban II settlements in the Guirun/Albarradas:survey rezionioe) 95 0 ee ee Figure. 4.5.. View of TL-MIT-CC+5 ioc. op scence ectin asd ote sm ohco dws we bnsci Eau ee get sing eg sleciys Oss Skew e-em Figure 4.6, Site map. of TL-MIT-CG>-5. oo gfe c sees ence ce ene os pe ee te ec ee a Figure. 4.7. House wall on Terrace. 38 at TL-MIT-CC-5..... 0. i eter te ciety rete err encore Figure 4.8. Black and White rim and Monte Alban II crema from Loma de las Tumbas .........................2-- Figure.4.9. Shell ornaments from Loma.de las TDumbas 1. 5 12 2 ec ens Figure 4.10. Figure 4.11. Figure 4.12. Figure 4.13. Figure 4.14. Figure 4.15. Figure 4.16. Figure 4.17. Figure 4.18. Figure 4.19. White paste sherds. with fine-line incising from) TL-MIT-CC-53) 7). 2 sere eee ae nee Cafe (imitation crema) and. Black and White sims from El Calvario ayaa eee Lapidary items and worked chert from Loma de las Tumbas? > 02. )..5. 2. iene eee tre en ree Worked chert from El Nopal sci cca en cies ea 4 een eee eee oe coe es ee Monte Alban II settlements in Tlacolula and the Albarradas area <<... .strerenety = on tee eee Black and. White rims from TL-MIT-CC-5 5 «ccc soos tsteen es = OU EE > ae Black and, White rims from El Nopal.i..c. 5255 tees os oe eo Monte Alban IL terrace sites.in the Central Valleys oftQaxacaweeers om frien iene ner ners Monte Alban II archaeological population as a percentage of maximum potential population ............... Monte Alban II potential agricultural surplus:............. . @Rhis seta. hee eee Figure 5.1. Average site population during Monte Alban IIIA... ... ........ See eee ee eee Figure 5.2. Population densities during Monte Alban IITA... 0... 52.2... ee eee Figure 5.3. Monte Alban IITA. settlements in the Central Valleys of Oaxacawy. (igmi.a -.0) ie ee meee een Figure 5.4. Monte Alban IITA settlements in the Guirun/Albarradas survey wesionis an). ee) 2 eee eee Figure 5.5. G=23s from. El. Calvario cove ccccc55 se Rts ots ce te oes se RR Me re Figure. 5.6. Site map. of, Llego. Yiacht 2050... ec oe ee Se ce Figure 5.7. Plan of architecture. on top of Llego Yiachitas Gaaees 16 Fae a eee eee tar Figure 5.8. Stone defensive wall on eastern side of Llego. Yiachi.... .,.,. . egteyeieee, Wenn ke) cent) ene te Figure. 5.9. Terraced. western slopes:of Guirun <2. < 4...¢2.2..< SRR See Se oe te ee Figure 5.10. Figure 5.11: Figure 5.12. Site map. Of GuUsrtm 5.6 ssc vee ns: ece esas oun A A ee TD PAP LP re Survey crew on Terrace 79 at Guirum © ...6<.<.- eae on215 eres sewn cue otek Meo ee ee oe ene Stone foundation wall on Platform 48, at,Guirun....:.2..-0. Pear ee ee ne ee Figure 5.13. Ridgetop with ballcourt:and plaza: withifour-mound ‘eroup*at:Guirini ae eee Figure 5.14. Monte Alban ITA settlements in Tlacolula and the Albarradas area pian eee Figure 5.15. Monte Alban IIIA terrace sites.in the CentraliValleys ofrOaxacamee 9 ars sis sence ren ae ee Figure 5.16. Monte Alban IIIA archaeological population as a percentage of maximum potential population.............. Figure 5.17., Monte Alban IIA. potential agriculturalisurplas Sir irs ene eee ee Figure 6.1. Adjusted population change in Valley of Oaxaca subareas during the Classic and Postclassic periods........... Figure 6.2. Large Late Classic period settlements in the Central, Valleys of Oaxaca Sem sameenct-e et 2 pen er ee Figure 6.3. Population densities during early TIIB-TV oo cic eae oe ee Figure 6.4. Carved stone figure from ballcourt at El Palmulloey-eeece ©) oe ee eee eee ee Figure 6.5. Populationrdensities: during) late LITB-DV 2: Ss, Seer es eee ene en 38 39 40 41 42 43 43 44 45 45 46 46 47 48 49 49 50 50 51 52 53 57 58 58 59 60 61 61 61 62 62 63 63 64 64 65 65 66 67 67 70 70 71 73 74 75 76 76 77 78 78 78 79 79 80 81 81 84 85 86 86 86 ficure Os eanvedstoneaniOormron monkeyuremrriatfomel Ieat-E] PalmillO.ne...c0c fe ie. Se ek OG facie oy Monten’ ban! Vesettlements im the Guirun/Albarradas SUIVEY TESION . a ck ee ce we eens Hoa was era eee mA Ole ke) ls\— 5 LN ee et wa a yd. ce iene MAA. e eel, Ss RAL, 24M OR, OG. ow eee a SRC ADO LtICEVC/Gl ACU See NPY ote One eet kt cunt, hes 1 A ihe Doky ok ey oe ot PVineled Us omollucilUuCturessOlmatcctsleriain Olbiierye.cluAgiia 292 me Weer Ed I RS OE ASI Se Prcuic os otOllesetaining wallsmat base.ol springs atsHierve.elA suago). ee es.20l. Ooi e aN OM Wa. ia ieee me Ossi Zeducana mata Lictivcre ws oUaain Sie eeter ek Neel We: WIR LO SUE ee be OL ea PECL OROMO MN Ae WlOl Tiel Cre’ ol -puenar mee eaneean reas iene A MOLY AN) SPU OS TAM TL AP be eee bb eek es Figure. o.l4) Monte Alban lV settlements im) Tlacolula and the Albarradas area. 20.0. ee es ie oanNontc A ban wl B-LY terrace sites inthe. Central Valleys of Oaxaca (2. 00.6 oe ee ee es Figure 6.16. Early IJIB-IV archaeological population as a percentage of maximum potential population................. Figure 6.17. Late HIB-IV archaeological population as a percentage of maximum potential population.................. Peo V Onge Al baney ssetilements. im the Central Valleystor Oaxaca “= 0. te ee ee et ee ee tt ee mopUl aonidesiiessouricviontesAlbaniVan yee se ee ee ee ee ee ee eee es Poeltier) se wvionic Albanay ssetulements in, the Guirun’Albarradas’survey region ©): >. 2... ee et es eee tear vereerCchsiOne Pendant TOMMmGUIllNaaws.... eee te we cee et ep te eee ee fio Ombre td SoamcmaObenwallatLiccOny IACDing .. uuc8. Stee ee ee a ee bette eee eee ee i emromicy clomaOcciiinc Onuthe western side OL Licgo,Viachi 5. ee ou eae ee et eles ee be wees PeCriReMM nN, eclCLNmiCoc (Ol ICO mY AACN Ie ete mia tiey eee eay) LARP FER ee gar gr rei ees ae aide tie, plgia Rin eos Saupe Chem eel clOumeanceplazaeattOp Olpe It) ti 20.9 webs kareena UPI MT, Fo tan ieee ee ns ede Ba ee Samy ence i wesw Syaile & cure a \lonte Alban, Vosetilements in Ilacolula and the Albarradas area (2... ee i ee i we tee ob Poca me vioniers bans Veterrace sites in the. Central. Valleys of Oaxacas, o2). 1 ae6.i64 - ease «ss ne rib dee sae eS 4 feels Cama SITGAI AT LOMB MC VESTAS ac. (RAK Ware. fete Bn MR eON FUT Ted A sis Ck heal hth ys ols Sh! 05) GEN «Shall ey We ounce Mb mC CSIC CMIAIBTCLEACCS AlANIICVC PUNLAS Zen cacn mani (ged) o> oa eeels Seendes Slave es Putheeeh Kites shes Sys Selo Reale le & Gam eee Tee em al minim DeCrcntinwayalonterkaccst INUGVe PUNLAS J25 ook oo). apa ens dyin deaeet Sew Baye Sealey he ee we Eu Mie ols iciKem moe OCkaCciiisonssouticasterm side. of INNucve Puntass: . «sec en a: PRG Re wae): Senta ch eee awe ergs dine Sp owiars ioe mom eV ROD UCHAVISta TOM TIC PEHIC ee sea. . Gain a id aoc Shonn cate cet cue Sen -yaasd> Resonyed alors = GEE etiomist be 2 Fromme “SG. SVURS TERS OP TRS NUS, Go pep Su US SS ene ee Seen ue et oe ene es ere eee eee eee ea mem emi mma rounlls 1eA lO Aol Lee see ck as Bs, 5 ws Hts ced ops teen SRM, Aye trate lane le bho 4 fain mS Gem eOlM@Lbe SASS Aa) meee dey cue ed gee Fae weet eee nso Gy ahs) soa soe, SARL A OTN. Hak Gil ad reais TNO. Serine yer Gh TBDS SI ETE EC iNed Ie ee eterna a ee re errr ere Figure 7.20. Monte Alban V archaeological population as a percentage of maximum potential population............... inet ce = Ime lotions loamay (potential agricultural SUrplUS sued 2 doce ah ee ks ono ey ee eh oh ce ey so ale ee List of Tables Seyiem bac aronological sequenceior the Valley of Oaxaca and Mesoamerica... gia es. 2). 2s 0S an bak Gila em we Table 2.1. Field numbers and phase numbers for sites recorded during the Guirun/Albarradas survey................... Table 2.2. Distribution of agricultural land classes (in hectares) in the Guirun/Albarradas survey region ................. Paolo ays oOmmative.sitesin Valley of Oaxaca subareas ... . eam ae. Seleeiral. Alig Sata, oar. Saw tees oms ek Oe lee) Bh Osationpiasessites ins Vallcy of Oaxaca subareas, ..:.2. peepee oe Sa ad Mes one vee ti wets oe Yes eo eo ev iiddlcsr ormativerceramics atl U-SLA-SLA-52. 4... ot get dee leas ehetiecbebe er 4s Tow ae. > 36 38 Table 4.6. Estimated archaeological, resource-based, and labor-based potential populations for Valley of Oaxaca subareas in Monte Alban IL .2 .cc5rs 6 cds. L Jae te eS eee eee ee ee ee 65 Table 5.1. Monte/Alban IITA‘ sites inv Valleysof Oaxaca’ subareas 242 > 0.5 ee 69 Table 5.2. Percentage of sourced obsidian at Monte Alban IIIA sites in the Valley of Oaxaca.................0.20000- 73 Table 5.32 Monte: Alban: IITA (sites inithe-Guirun/Albarradas survey regionia saa ee eee ee ee ee 74 Table: 5.4: Monte Alban IITA ceramics ‘at sitesin the Albarradassateay snes cota aetna 75 Table 5.5. Proportion (%) of ceramics by period and obsidian varieties at Guirun by site section..................-.4- dd Table 5.6. Percentage of population residing in terrace sites in Valley of Oaxaca subareas in Monte Alban HIA............ a7 Table 5.7. Estimated archaeological, resource-based, and labor-based potential populations for Valley of Oaxaca subareas in Monte Alban IITA «. so: stews. ule ih PR Wirlakdepaid’? Sey Saeed LAI Leoekeee oy eal cee Sue: ace eee 81 Table 6.1. Late: Classic population. in.Valley,of Oaxaca,subareas 3. re4-n4- Fe See ee ee ee 84 Table 6.2. Percentage: of sourced, obsidian at Late Classic\sitesiim ihighland:Qaxacag 0, se eo eee ne 87 Table 6.3. Monte Alban IJIB-IV ceramics at sites in the Albarradasvarea #4? sa50 + ce ee 88 Table 6.4. Monte Alban IIIB-IV sites in the Guirun/Albarradas survey region... .22...--6 «4-26 Je) eee ee 88 Table 6.5. Percentage of population residing in terrace sites in Valley of Oaxaca subareas in Monte Alban HIB-IV ......... 94 Table 6.6. Estimated archaeological, resource-based, and labor-based potential populations for Valley of Oaxaca subareas in Monte Alban ITEB-IV «cr ae ns on 0s ee Wee eee ek ac a ed aes we oa Ween a 94 Table-7.1. Monte Alban V sitesin Valley of Oaxaca subareas 2. a5 sauce ie ee ee ie 99 Table 7.2. Percentage of sourced obsidian at Early Postclassice sites in Oaxacan... ce © oe eee 100 Table 7.3: Percentage of sourced obsidian at: Late: Postclassic sitesan: Oaxacaye eee ee 102 Table:7.45 Monte Alban V sites in the Guirun/Albarradas survey fegion —— 0. a) 1c ae eee 102 Table 7:5. Monte-Alban V ceramics at sitesin thesAlbarradas area.jerge cs eo es 103 Table 7.6. Percentage of population residing in terrace sites in Valley of Oaxaca subareas in Monte Alban V............. 105 Table 7.7. Estimated archaeological, resource-based, and labor-based potential populations for Valley of Oaxaca subareas in Monte Alban Voie. oosceou- oe sane Aces Seo LE ene ie, ees Sa a ee Se oe 110 List of Appendices Appendix I:,Site Descriptions x«..:...0848 asus He dead Abeta eee © ages yn eee eS 129 Appendix Il: Grid, Square, Maps: a... 44s aed ceoW ae Ae Oo Re ee ee ee a 134 Appendix TI]: Site Survey. Data « -.0 " ) (" 7 @ Te . . a otihuacan | . Tlacolula arm Gulf of Mexico Central Highlands 7 Sets a Southern Highlands Guirun/Albarradas survey region at ot Vas oats ten, ~ El Palmillo ~~ a i a» + N 0 10 a kilometers Fic. 1.1. Valley of Oaxaca survey regions and principal places mentioned in the text. in addition to physiographic features that divide regions, is part of how we have defined marginal or frontier areas. But not all frontiers are zones of transition between core areas; frontiers also separate core regions from more distant sparsely settled areas (Rodseth & Parker, 2005, p. 10). The examination of broad-scale areas provides necessary context for widening our understanding of the nature of past frontiers and how they shifted and changed over time. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 3 In early core—periphery studies, frontiers were seen as the “passive recipients” of innovations that occurred in core areas or as semipermeable cultural barriers that could impede interaction (Lightfoot & Martinez, 1995, pp. 471, 473). More recently, frontiers have come to be seen as porous and fluid zones of transition and interaction that comprise a complicated mosaic of overlapping boundaries and cross-cutting social networks (Lightfoot & Martinez, 1995, p. 472; Elton, 1996, pp. 3-9; Parker, 2006, pp. 80-81; Schon & Galaty, 2006, p. 232). Interactions in frontier zones occur between diverse groups that are not necessarily equivalent in power, status, or degree of autonomy (Trinkaus, 1984, p. 46; Lightfoot & Martinez, 1995, p. 473). Because of the mosaic of overlapping boundaries, the multiplicity of interactions that occur in frontier areas, and differences between cores, outlying areas are not all the same, and changes occur in them over time. Frontiers are shaped by the intensity of the relations and the nature of the networks that link them with neighboring groups (McNeill, 1998, p. 221; Hall, 2009, p. 39; Terrell, 2010; see also Schortman & Urban, 1992). No two frontiers have the same neighbors or intensity of interaction. The processes of contact and interaction also are not unidirectional, with innovations limited to core regions (Lightfoot & Martinez, 1995, p. 487); influence and new cultural conventions can be transmitted in either direction (Lightfoot & Martinez, 1995, p. 487; Schon & Galaty, 2006, p. 232; Hall, 2009, p. 39). This dynamism can make frontiers “socially charged places” (Lightfoot & Martinez, 1995, p. 472) and “unstable zones that exhibit a marked degree of variability through space and time” (Parker, 2006, p. 77). Shifts in boundary relations are a central theme in this work, as our empirical focus is on the eastern physiographic edge of the Valley of Oaxaca. Years back, we and colleagues (Blanton et al., 1982, 1993) defined boundary relations as one of the key dimensions (along with scale, complexity, and integration) that requires attention and consideration when we compare human cooperative networks, political formations, and economic linkages across time and space. Through the focus on these dimensions rather than societal taxonomies or presumed trajectories of change, we have been able to describe, analyze, and compare changes over time while recognizing that human history follows different, alternative pathways and that such time/spatial variation is a key element of what we, as archaeologists, should endeavor to understand (e.g., Feinman, 2012a; M. E. Smith et al., 2012). At the same time, there are important processes and mechanisms that were at work across historical contexts (e.g., Feinman & Nicholas, 2016a). Here, we examine a range of manifestations of scale, such as estimates of population sizes for various spatial units and sites (for a fuller discussion, see Blanton et al., 1993, pp. 13-19). We also consider the scale of monumental architecture at specific localities over time (e.g., Blanton, 1989; Abrams, 1994). Nested spatial scales are, of course, a central element for this investigation and discussion. Complexity, which may be vertical or horizontal in nature, is referenced in regard to the vertical levels or tiers in settlement distributions. We also examine the horizontal components, such as terraces, that make up certain settlements. The competitive or allied political factions, which may or may not share comity and power, are another way that we consider the dimension of complexity. Integration refers to the relations and connections between units, whether individ- uals, households, or larger social formations. It should be clear that we do not wish to imply that any human group or grouping is seamlessly integrated, but the ties and networks that bind and affiliate are a key element of being human, and the strength and nature of these associations are highly variable. In this volume, we are especially concerned with how urban Monte Alban interacted and was interconnected with its hinterland at ever more distant spatial scales as well as how these relations and spatial configurations changed over time. The dynamic of leader—follower relations and how they shift temporally is another focal point of this analysis. By definition, all leadership is relational (Ahlquist & Levi, 2010). The different ways that leaders have to gain and hold power and the nature of the resources available to them to fund their activities/ governance affect the dynamics of ancient polities (e.g., Blanton & Fargher, 2008; Blanton, 2016). Recent scholarship has advanced a continuum of relations, ranging from collective or more representative, where power is shared and divested in different societal segments, to more exclusionary or autocratic, where the principals are more self-aggrandizing and often operate through networks of personal/family ties (Feinman, 1995, 2001a, 2012b; Blanton et al., 1996; Blanton, 1998, 2016; Carballo, 2016). These leadership modes are not static and may change in a given setting over time. They also tend to be tied to differences in a range of other variables, including the accumulation of goods and the nature of public architecture (e.g., Feinman, 200la, p. 160; Blanton & Fargher, 2011; Carballo, 2016, p. 120). An important variable is how the polities are funded (Levi, 1988), with different leadership arrangements underpinned by different sources of revenue (D’Altroy & Earle, 1985; Blanton & Fargher, 2008; Blanton, 2016). Exclusionary polities are more likely to derive their funds from external resources, such as control of spot resources, control of trade routes, or war booty, whereas more collective polities depend on internal resources, such as agricultural surplus from the local populace (Blanton & Fargher, 2008; Feinman & Nicholas, 2016a, pp. 283-284, 2016b, pp. 47-48). In specific political contexts, the way that power is funded has clear implications for how relations are forged between central states and cities and their near and far hinterlands. Our aim here is to examine shifting networks of settlement and interaction between the populations in the Southern Highlands over the pre-Hispanic era. We focus on the relations between the urban capital in the center of the Valley of Oaxaca and the smaller valleys on its borders as the nature of political organization, leadership, and the resources that funded power changed over time. How did the nature and intensity of interactions shift over space when Monte Alban was founded and increased in size and monumentality during the Formative period? What were the differences as Monte Alban began a slow decline during the Late Classic and in the politically decentral- ized but economically commercialized Late Postclassic? How were outlying areas like Albarradas affected (if at all) by these changes in the scale and nature of political organization at the heart of the valley system? At the same time, our other important charge is to report the information that we recovered and mapped at sites in the Guirun/Albarradas area, as, realistically albeit sadly, many of these archaeological localities will never be studied or recorded in greater detail. If the descriptions offered sometimes seem dry or overly detailed in spots, consider that we often provide as much information as we collected on specific settlements, as these written accounts FIELDIANA: ANTHROPOLOGY and the materials collected may be the only enduring record that the past inhabitants of those sites left on the planet. With these aims and new perspectives in mind, we stress that by focusing on a larger spatial landscape than just Guirun/ Albarradas, we have recognized new patterns and relations in these broad-scale data. In a few cases, our opinions have changed from our early interpretations of the Guirtn/Albarra- das area. More frequently, we have reframed or slightly tinkered with or enhanced early views, even those expressed in our own recent monograph (Feinman & Nicholas, 2013). By employing a conceptual lens that examines relationships between political leadership and followers, how governance 1s implemented, and the ways that it is funded (e.g., Levi, 1988; Blanton & Fargher, 2008; Blanton, 2016), we have been able to reconsider prior small enigmas in settlement pattern changes that occurred in pre-Hispanic Oaxaca. Our attention is more focused not just on the links between spatial sectors of this valley system but also on the kinds of flows that may have linked them. We reevaluate why population was so densely settled around Monte Alban in Monte Alban Late I and IIIB- IV but not in Monte Alban II and why settlements east of the mountains that define the physiographic edge of the Valley of Oaxaca in Monte Alban V were situated to exploit agrarian resources and produce a large potential surplus when earlier settlement distributions in the area were not situated to do so. We also add a few new bits of evidence and interpretive wrinkles to our earlier effort to understand the long Classic— Postclassic transition in the Valley of Oaxaca that was embedded in our prior monograph (Feinman & Nicholas, 2013, pp. 141-156). With new vantages and data, we offer newly expanded and, it is hoped, enhanced perspectives relevant to these issues and others, and for that reason we often devote as much or more consideration to larger spatial domains and do not limit our spatial frame to the Guirun/Albarradas region itself. Of course, understanding the Guirtn/Albarradas area would be difficult without that broader context. Organization of This Volume To address the theoretical and empirical goals outlined above, the rest of this volume is organized into an additional chapter of historical and contextual background, five sections that outline the settlement patterns in the Guirun/Albarradas region (and neighboring areas) over time, and a synthetic summation. Each of the five chapters that present the settlement data begins with an opening section that lays out what we think was happening at broader scales in highland Oaxaca and beyond before presenting more detailed information on Albarradas and its relationships to its larger neighbor. These sections of the volume follow Chapter 2, where we discuss the history of systematic regional settlement survey in Mesoamerica and Oaxaca as well as what we knew about this part of the Valley of Oaxaca before we implemented the survey. We focus in Chapter 3 on the Middle and Late Formative (Rosario phase and Monte Alban I), when the earliest settlements were established in the Guirun/Albarradas area. First San José Mogote and then Monte Alban were the principal centers in the Valley of Oaxaca at that time (Table 1.1). We did not find any Early Formative sites in the Albarradas area. The population remained very low until TABLE 1.1. Chronological sequence for the Valley of Oaxaca and Mesoamerica. Oaxaca Mesoamerica 1500 Late Postclassic 1300 Monte Alban V 1100 Early Postclassic 900 Monte Alban IV 700 Late Classic Monte Alban IIIB 500 Monte Alban IIIA Early Classic 300 AD 100 Monte Alban II Terminal Formative 100 BC Monte Alban Late I 300 Late Formative Monte Alban Early I 500 Rosario 700 Middle Formative Guadalupe 900 San José 1100 Early Formative 1300 Tierras Largas 1500 Monte Alban Late I, when a series of small terrace sites were established at the eastern edge of the Albarradas area on the western side of the Hormiga Colorada River. The ceramic artifacts and the nature of the settlement plan and architecture at these sites are very similar to contemporaneous sites in the Valley of Oaxaca. The defensive character of the larger Late I sites in Albarradas is an indication of potential skirmishes and hostilities with neighboring populations, perhaps Mixe peoples to the east, than the presence of a tightly controlled political border. In Chapter 4, we discuss the Terminal Formative (Monte Alban IT) settlements. Populations declined throughout most of the Valley of Oaxaca and neighboring regions between Monte Alban Late I and II as Monte Alban extended its extraregional influence. In contrast, populations rose slightly in eastern Tlacolula and neighboring Albarradas. In Chapter 5, we present our findings for the Early Classic (Monte Alban HIJA), a time of high population growth across highland Mesoamerica, including the Valley of Oaxaca. Teotihuacan was the largest political center in the Mesoamer- ican macroregion and extended degrees of influence and interaction beyond central Mexico during this period. In the Valley of Oaxaca, although Monte Alban grew in size and architectural monumentality, it seems to have lost influence in more distant areas and refocused on the valley. Much of the Albarradas area was depopulated at the same time that the Guirtn site on the western side of the mountains in eastern Tlacolula grew substantially. We suspect that people moving from the Albarradas area contributed to Guirtn’s growth. Most of the remaining population in the Albarradas area was FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 6 defensively concentrated in Llego Yiachi at the eastern edge of the surveyed area. We examine in Chapter 6 the Late Classic (Monte Alban IIIB-IV), when Monte Alban reached its peak dominance and then began its slow decline. Population declined throughout the valley and most of the surrounding regions during this era of political reorganization. At the same time, new powerful centers arose in the Valle Grande and in Tlacolula. The only subregion of the valley that experienced significant population growth was eastern Tlacolula; the population rebound was even greater in Albarradas. Because of long-standing problems with the IIIB-IV ceramic chronology of the Valley of Oaxaca during Monte Alban’s peak and eventual decline (see Feinman & Nicholas, 2013, chapter 9), it has been difficult to elucidate the full suite and contempo- raneity of changes across the Valley of Oaxaca during this significant era of transition. Incorporating findings from subsequent excavations at Classic/Early Postclassic sites in the valley and readings of ethnohistorical records, we have reexamined settlement data from our regional survey projects in and around the Valley of Oaxaca and offered a refined scenario for the population shifts that occurred between the Early Classic and Late Postclassic (Feinman & Nicholas, 201 1a, 2013, 2016b). This scenario largely adheres to the overall findings of the original survey data at the same time that it provides a settlement history more consistent with excavation and ethnohistoric findings for the region. Based on the adjusted demographic profiles, population decline at the end of the Classic period appears to have been more gradual and prolonged than in other interpretations, and significant demographic growth did not occur across the Valley of Oaxaca until the latter half of the Postclassic period. In Chapter 7, we focus on the Postclassic (Monte Alban V), an era of political and demographic decentralization, expanding networks, and increasing commercialization in highland Oaxa- ca and beyond. Following political reorganization at the end of the Classic period, population continued to fall in the northern and western sectors of the Valley of Oaxaca during the early Postclassic. Renewed growth began earlier in the peripheral parts of the valley—Tlacolula and Ejutla—that were the first to pull away from Monte Alban’s hegemony during the later Classic. Although population expanded everywhere across the region by the end of the Postclassic, Tlacolula, especially eastern Tlacolula, continued to be the demographic center of the valley. Settlements extended across the mountain ridges into Albarradas, which had higher population densities than any other edge area. Albarradas also was more densely settled than several subareas of the Valley of Oaxaca. In Chapter 8, we return to themes and questions concerning boundaries and frontiers (with a focus on the Valley of Oaxaca) that we outlined above. The expansion of the systematic archaeological survey into the Albarradas area provides us the opportunity to present new corroborative evidence and interpretive perspectives on the settlement and population changes documented for eastern Tlacolula through prior survey research (Kowalewski et al., 1989). We also gather new strands of evidence that are marshaled to define routes of interaction and exchange that passed through the Albarradas area into the Valley of Oaxaca and that may have been a factor 1n Tlacolula’s demographic and political emergence following the decline of Monte Alban. The specific networks of interaction vary for different edges of the Valley of Oaxaca and offer one potential factor relevant to the distinct demographic histories that we document for the different edges of this region. The Albarradas area was not large nor was it easy to survey systematically due to the rugged landscape. Yet in the context of the large contiguous study area that was surveyed previously, it has been a source both for new empirical findings as well as for provocative new perspectives that have yielded fresh ways to reconsider and interpret elements of the settlement history of the larger landscape of which it is part. FIELDIANA: ANTHROPOLOGY Chapter 2: Archaeological Survey and the Guirtn/Albarradas Region “In multiple ways, since its beginnings some 40 years ago, regional archaeological research based on full- coverage survey has transcended its original goals to find relevance in relation to research problems not anticipated by its founders.” (Blanton, 2005, pp. 295-296) Background to Valley of Oaxaca and Prior Surveys Since the implementation of Gordon Willey’s (1953) regional settlement pattern study of the Vira Valley in Peru in the late 1940s, systematic archaeological surveys have been conducted in many regions across the globe (e.g., Kowalewski, 2008; Fang et al., 2012). Now, after more than half a century, dozens of systematic regional research projects have been completed across the Central and Southern Highlands of Mesoamerica (Nichols, 1996, table 1; C. A. Smith, 2002, table 4-4; Balkansky, 2006; Feinman & Nicholas, 2013, pp. 25-26, figure 3.1). Since their inception, these full-coverage, regional surveys have been among the most important sources of new archaeological knowledge, not only about pre-Hispanic Mesoamerica (Blan- ton, 2005, p. 295) but also for areas across the globe (Sabloff & Ashmore, 2001, p. 14; Cherry, 2003; Renfrew, 2003, p. 313). The first systematic regional surveys in Mesoamerica were conducted in the Basin of Mexico in the 1960s (Sanders, 1965; J. R. Parsons, 1971; Blanton, 1972; Sanders et al., 1979; J. R. e\Cantona o. Cholula ee Puebla TehuacancValley Guerrero Central Mixteca Alta . Tilantongo ” Penoles /; : ~. Monte Alban §-.. Huamelulpan i @Tututepec 0 100 a kilometers Fic. 2.1. ~_2t Albarradas * . Southern - eo? tl la Einua Highlands Miahuatlan Parsons et al., 1982, 1983). The following decade saw the implementation of systematic archaeological survey in the Valley of Oaxaca, beginning with Monte Alban in 197] (Blanton, 1978) and then in all arms of this large Y-shaped valley (Varner, 1974; Kowalewski, 1976; Blanton et al., 1982; Kowalewski et al., 1989). At the same time, a partial survey of the Miahuatlan Valley (Markman, 1981) south of Ejutla was completed, but without using the same full-coverage method- ology. We extended the regional survey from the southern edge of the Valley of Oaxaca into the Ejutla Valley in the mid-1980s (Feinman & Nicholas, 1990, 2013). A decade later, we surveyed the Guirun/Albarradas area at the eastern edge of the Valley of Oaxaca (Feinman & Nicholas, 1996, 1999). At the same time, other investigators carried out regional surveys in areas that abut the Valley of Oaxaca: the Sierra Norte to the north (Drennan, 1989), the Penoles area of the Mixtec Sierra to the west (Kowalewski, 1991; Finsten, 1996), and the Sola Valley to the southwest (Balkansky, 1997, 2002) (Fig. 2.1). Regional surveys in many of the valleys of the Mixteca Alta northwest of the Valley of Oaxaca (Spores, 1972; Byland, 1980; Plunket, 1983; Byland & Pohl, 1990; Balkansky, 1998a; Balkansky et al., 2000; Stiver, 2001; Kowalewski et al., 2009, 2010) have expanded the total area surveyed in highland Oaxaca to more than 9,000 km”, the largest contiguous area surveyed anywhere in the world. The regional survey of the Guirun/Albarradas area was completed during six weeks in July and August 1995, covering an area of 110.8 km*. The area walked over extends the Gulf of Mexico << Veracruz TaReee Sierra Norte Valley of Oaxaca Isthmus of Tehuantepec ; Laguna Zope ' Chiapas Los Horcones 6 Pacific Ocean Surveyed areas in highland Oaxaca and places mentioned in text. FIELDIANA: ANTHROPOLOGY N:S., NO. 46, NOVEMBER 13, 2017, PP. 7-26 a a "709s s Surveyed ‘1980 Archaeological site @ Modern town Fic. 2.2. coverage of the Valley of Oaxaca Settlement Pattern Project (Kowalewski et al., 1989) and is contiguous with the eastern limits of the 1980 survey boundary in the Tlacolula arm of the valley (Fig. 2.2). This area includes the foothills and lower slopes of the high mountain ridges that define the eastern physiographic edge of the large, highland Valley of Oaxaca (Fig. 2.3). The Guirun/Albarradas survey region crosses the mountain ridge and includes terrain belonging primarily to San Pablo Villa de Mitla (west of the ridge) and San Lorenzo Albarradas (east of the ridge). Small areas of land in the southern part of the study area belong to Santiago Matatlan and San Baltazar Guelavila. The previously reported Guirun site, known largely for a partially exposed carved stone cruciform tomb and standing adobe walls (Saville, 1900, 1909; E. C. Parsons, 1936, plate LI, bottom; Feinman & Nicholas, 2004a), is located at the edge of the 1980 survey boundary on the western slopes of Cerro Guirone. At that time, when the kilometer Guirun/Albarradas survey region, showing topography, all archaeological sites recorded on survey, and modern towns. regional survey was implemented, the crew suspected that the site might extend farther up the slopes of Cerro Guirone, but time did not permit an upslope expansion of the survey zone. We picked up the Guirtn/Albarradas survey at the 1980 boundary, covering the highest mountain peaks and the slopes descending on the east to the Hormiga Colorada and the Salado tributary of the Rio Grande/Tehuantepec River. Between 1996 and 1998, we completed intensive surface surveys of three sites in eastern Tlacolula—the Guirun site, El Palmillo, and the Mitla Fortress (see Fig. 1.1; Feinman & Nicholas, 2004a). These studies provided more detailed information for the three sites than was possible to obtain during the regional survey alone (Feinman & Nicholas, 2004a). During our intensive survey of the Guirtn site, we found it was larger and more populous than reported in the original survey volume (Kowalewski et al., 1989) or in our first reporting of the results of the regional survey of the Guirun area (Feinman & FIELDIANA: ANTHROPOLOGY Fic. 2.5. Narrow canyon of the Hormiga Colorada in the southeastern part of the survey region. Nicholas, 1996); here, we include the updated information from the intensive survey in all our discussions of the Guirun site. The Guirtn/Albarradas Survey Region The Guirtn/Albarradas survey region is a rugged, topo- graphically diverse area that includes the high mountain peaks that define the eastern edge of the Valley of Oaxaca (Fig. 2.4). Rising to heights of more than 2,500 m (including Cerro Guirone, El] Tigre, and Nueve Puntas), the mountains descend oe ; in a short distance to the east (approximately 10 km) to 1,200 m Fic. 2.3. Aerial photograph mosaic of the Guirtn/Albarradas above sea level along the banks of the geeply incised Hormiga survey region (reduced from original scale of 1:75,000). Colorada, which largely defines the eastern limit of the surveyed area (Figs. 2.5 and 2.6). There are no large tracks of flat Fins Px ee oa z <. : ey ee 2 oe ae : err Es +e oe : ii ee Spe: Fic. 2.4. The mountains at the eastern edge of the Valley of Oaxaca, viewed from the northwestern part of the Guirtn/Albarradas survey region; peaks from left to right are Cerro Guirone, El Tigre, and Nueve Puntas. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO D Fic. 2.6. The Hormiga Colorada. agricultural land within the surveyed region. The area to the west, in eastern Tlacolula, is in a rain shadow cast by the aforementioned mountains and is the driest part of the Valley of Oaxaca. In contrast, the eastern slopes of the mountains, the Albarradas area, receive much more rain and are heavily vegetated where not cleared for farming (Fig. 2.7). On more than one occasion during the survey, we returned to clear, dry weather in Mitla (where we lived) after getting drenched by an afternoon shower on the other side of the mountain. The air is decidedly much more humid in the low, narrow valley of the Hormiga Colorada, which is one of the few areas near the Valley of Oaxaca where more tropical plant varieties grow (see Cook, 1983). This area is much lower than the valley floor in eastern Tlacolula, which ranges between 1,600 and 1,800 m above sea level. The Albarradas survey region also includes the Far Fic. 2.7. 10 Eastern slopes of Cerro Guirone (viewed from TL-SLA-SLA-20). small, dry, high valley northeast of Mitla, where the modern highway exits the Valley of Oaxaca (Figs. 2.8 and 2.9). The Valley of Oaxaca is home, past and present, to Zapotec speakers. The present eastern limit of the area inhabited by Zapotec-speaking peoples is close to the eastern boundary of the Guirin/Albarradas survey region. The rugged mountains of the Sierra Mixe lie farther east (Schmieder, 1930, p. 60; Beals, 1973, p. 5), beginning approximately 10 km beyond the eastern edge of our study area (Fig. 2.10). Today, this strip of mountain slopes between the easternmost Zapotec communities and the ethnically and linguistically distinct Mixe is very sparsely settled. There have been historical studies of the Mixe area in the 20th century (e.g., Schmieder, 1930; Beals, 1973; Kuroda, 1984; Lipp, 1991) and limited salvage archaeology of tombs in the Mixe region (e.g., Markens & Winter, 2014) and Albarradas area (Higelin et al., 2013), but the region for the most part remains unstudied archaeologically. A recent report of ceramics from several locations in the Mixe area (Hutson, 2014) has revealed strong affiliations with both the Valley of Oaxaca and the Isthmus of Tehuantepec. Although most of the collected materials in Hutson’s reporting lack adequate provenience, the findings conform to the view that the Mixe region was interconnected with neighboring areas during the pre-Hispanic era. Historically, Mixe settlements were small and dispersed along slopes and accessible gorges, and their residents frequently skirmished with each other (Schmieder, 1930, p. 62; Beals, 1973, pp. 10-11). According to Mixe oral histories, they also had long-standing rivalries with Zapotecs (S. M. King, 2012, p. 236). At the time of the Spanish Conquest, the relations between Zapotecs and Mixes reportedly were hostile: “The extreme aggressiveness of the Mije had so intimidated the Tzapotec that the latter appreciated the proximity of the Spaniards and turned to them for protection” (Schmieder, 1930, Sc ce FIELDIANA: ANTHROPOLOGY y ra a gee i Fic. 2.8. Small valley in the northwestern part of the survey region; the tall peak in the distance is Cerro Guirone. p. 60; see also Burgoa [1670-1674] in Beals, 1973, p. 11). The English translation of a/barradas is “walls,” and there are at least five modern towns in the mountainous areas east and north of the Valley of Oaxaca that include a/barradas in their names, a reference to the defensive features found at sites in the region (Brito de Marti, 1982, p. 623). Accounts from the 1500s include mention of at least four walled settlements or fortresses PP Fe +e < : < z § nats Sa tee a a OTE as in the vicinity of Mitla (Horcasitas & George, 1955, p. 23). Nevertheless, whatever the extent of hostilities, these acrimo- nious relations did not interfere extensively with interaction and exchange, as indicated by Hutson’s (2014) analysis of Mixe ceramics. Well into the 20th century, Mitla was the economic center of the eastern end of the Valley of Oaxaca, located at the juncture see 4 ‘3 : ‘ ‘ 3 Y . ral poe Fic. 2.9. Northern part of the survey region and the mountain pass traversed by the modern road; Cerro Guirone is the tallest peak to the right; several small defendable sites are located on the two ridgetops on either side of the pass, which is in the center of the ridgeline. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 1] Fic. 2.10. Mixe area to the east, near Tamazulapan Mixes. of three natural routes that connected the valley to mountain- ous areas to the east (Fig. 2.11; e.g., Schmieder, 1930, p. 34; E. C. Parsons, 1936, map III; Beals 1979, p. 165). Prior to the construction of paved highways in the middle of the 20th century, more than half of the able-bodied men in Mitla were traders or porters. They traveled by foot not only within the valley but also all the way to the Isthmus of Tehuantepec in pursuit of goods to buy and sell (E. C. Parsons, 1936, pp. 10-14; Beals, 1979, p. 166). The most direct and important route traversed a mountain pass between Mitla and San Lorenzo Albarradas (Beals 1973, p. 5; Cook, 1983, p. 41), passing south of Cerro Guirone. This was the principal route—the Camino Real—through the Mixe area. It served as a passage to bring products from as far away as the Isthmus of Tehuantepec to the valley via Mitla (Beals, 1979, p. 177). The late Fausto Olivera, who grew up in the area and was our principal local field assistant until his untimely death, pointed out parts of this route during our survey, also telling us that this was the principal route to Tehuantepec prior to the construction of the International (Pan American) Highway. This long-used trade route follows the Tehuantepec River from Tehuantepec west through Jalapa to Nejapa (Ball & Brockington, 1978, p. 111), which was an important stop on the route (Fig. 2.12; S. M. King, 2012, pp. 231, 245). From Nejapa, the route follows the river north and passes through the Mixe area and the heart of the Guirtn/Albarradas survey region, although there were other minor routes as well (Ball & Brockington, 1978, p. 111; see also Fahmel Beyer, 2005, 2010). Other routes that wind from the Gulf Coast to the north pass through the Mixe area and intersect the Camino Real before the mountain pass by San Lorenzo Albarradas (E. C. Parsons, 1936, map III). Thus the small Guirtin/Albarradas area was a nexus between larger regions, especially before the construction of the Pan American Highway. Nevertheless, the new superhighway from Mitla to the Pacific Coast, built after this study, does pass through the region. What We Knew before the Survey In the mid-1990s before we began our survey, we were aware of three sites of historic and mythic importance in the area at the valley’s physiographic eastern edge, both in the Tlacolula arm on the western slopes of the Nueve Puntas—Cerro Guirone 12 mountain ridge and in the Albarradas area on the eastern side, that would be within the bounds of the area we planned to investigate (Fig. 2.13). Although some site-specific information was available for these sites, we had basic unanswered questions about all of them. As we carried out the survey, we learned of two additional sites that were not in the literature but that were well known to local people; we are uncertain whether archaeologists previously visited either site. The best known of these sites is Guirin (Feinman & Nicholas, 2004a). The site spreads over high piedmont ridges and mountain slopes in the easternmost part of the Tlacolula Valley, approximately 5 km east of Mitla (Fig. 2.14). Guirun takes its name from Cerro Guirone, one of the defining physiographic features of eastern Tlacolula. The top of this high cone-shaped peak is visible far from the west and from part of the Mixe area to the east. The Camino Real traverses the mountain saddle south of this peak. Sites with standing pre-Hispanic adobe walls are very rare in the Valley of Oaxaca, so reports of standing “Indian” walls drew the first explorers to Guirun in the 1800s. In searching for these walls, or paredones, Dupaix (1834) discovered a subter- ranean cruciform tomb below a ruined adobe structure on a lower slope of the site (Fig. 2.15). A second, more elaborate cruciform tomb on a higher slope and decorated with mosaic stone grecas was not reported until 50 years later (Ober, 1887, p. 542) (Figs. 2.16 and 2.17). Marshall Saville (1900, 1909) visited the site around the turn of the 20th century as part of his study of ancient cruciform tombs in and around Mitla and excavated both tombs. Based on architectural features and the fact that the tomb constructed with carved stones was thought to have never been finished, Saville (1900) concluded that both tombs were Late Postclassic (AD 950-1520). Saville also spent time clearing vegetation from other structures, including a two-room temple that fronts a large enclosed plaza and associated ballcourt (Feinman & Nicholas, 2013, figure 4.6). He recovered debris from stone working and noted that several ancient stone quarries were nearby. Saville, like all the other early visitors to the site, was unaware of (or ignored) the terraced slopes surrounding the tombs and other architecture (Fig. 2.18): “It is not a desirable location for a large settlement,” and “it would have been impossible for the steep sides of the hills to be used for habitation” (Saville, 1900, p. 210). A few scholar/explorers (Tweedie, 1911; Schmieder, 1930; E. C. Parsons, 1936) visited Guirtin following Saville’s sojourn there, but it was not until 1980 that the site was preliminarily mapped and both tombs were placed in fuller temporal and spatial context during the Valley of Oaxaca Settlement Pattern Project (Kowalewski et al., 1989, figure A.IX.53). Findings from the regional survey also led to a conundrum. Saville had earlier concluded that the Guirun tombs were Late Postclassic, yet Classic period ceramics were more abundant on the site than were Postclassic ceramics. Few Late Postclassic sherds were recovered on the high ridge where the carved cruciform tomb was constructed, either during the regional survey or during Saville’s excavations of the tomb. The Guirtin site was surveyed near the end of the 1980 field season and was covered with dense, often thorny vegetation. Time was short, and the regional survey boundary was drawn just upslope of the recorded architectural complexes. Given the elaborate architecture at the site, its size estimate seemed low, around 600 people during its maximum extent in the Late FIELDIANA: ANTHROPOLOGY S/Melchd Betaza ee Domingo =e V. Hidalgo cn ge agacia” ’ \ Yalalag PO ah. a ( ; 4 an : BAN . 8. ee : ‘EE Tiah yfottepe nergy i. . pe 2 " oan a ‘ae ar . oe ; "Yam@eulapan io % % Mixes‘, &, be, a bu . dy fi 9 r . % S. Ana x fe ve ee \ bat Ss, ~ af del Valle V. Di A. 7 Ord 8 Tladbluta Fo S. Marcos 5. Bartolo ee Quiajana FT 7 OS, oy : ieineo te ae 7 j S. S Dioniss i Oc 4 \ . Chichicapan, i i .. 5 Ot on ae: 2 z ao Hebgonal erent SE gO AS 2 surveyed area trade route Fic. 2.11. Classic (Kowalewski et al., 1989, p. 296). We suspected that the site might extend farther up the slopes of Cerro Guirone outside the area surveyed in 1980. Our questions about the Guirtn site were one factor that led us to propose and carry out the regional survey of the Guirun/Albarradas area, not only to possibly complete the survey of the site but also to place it in its fuller spatial context. We did indeed find that Guirun was larger <= S. MeAsuncion Mazatlan Cacalotepse iz: XG migdey = s * etzaltgpec “Asuncion AcdXang) : 5 Vf) ‘ e € 2g arta uu? Chayyan fp J \ Guiavicuzas Santiago™, Lachiguiri S.Maria Totolapilta, Tequisistlan 500 oF Se Sug wh ‘gf Asuncion ee > om cmeonee Tacoltilita Aci. S. Miguel, of ie 10 15 cee | kilometers Travel routes near Mitla in the 1930s (adapted from E. C. Parsons, 1936, map II. than mapped during the 1980 survey, that it extended farther up the slopes of Cerro Guirone and incorporated a large ridge system to the north that includes the La Cuadrada stone quarry (E. C. Parsons, 1936, plate LI, top). Through analysis of materials we collected during the 1995 regional survey, the 1996 intensive site survey of Guirun, and ceramics from Saville’s excavations that are curated at the American Museum of FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 13 Tlacolula arm """**@wcc@, Mitla Gui Pacific Ocean Fige2212) Natural History in New York, we determined that the Guirun tombs were constructed during the Early Postclassic (Feinman & Nicholas, 2013, pp. 146-148, cf. Feinman & Nicholas, 2004a, p. 71). In the Valley of Oaxaca, Early Postclassic pottery is not easily distinguished from earlier Late Classic pottery, so this reassessment appears to resolve the earlier conundrum (Fein- man & Nicholas, 2013, pp. 146-152). The second known site is a large stone quarry on a high piedmont ridge east of Mitla that is locally called La Cuadrada (E. C. Parsons 1936, plate LI, top) due to the large blocks of ignimbrite that were cut from the underlying bedrock but never moved from the site (Figs. 2.19 and 2.20). The quarry had drawn early attention as visitors to Mitla, who were impressed by the large stone blocks used in the construction of Mitla’s elaborate buildings and tombs, sought to find the source of the building stones. Although Saville had noted La Cuadrada and other quarries in the Mitla area, William Henry Holmes (1897) was the first to report extensively on them. Howel Williams and Robert Heizer (1965) later visited the quarries during a geological survey of the Mitla area. The La Cuadrada quarry is just beyond the 1980 survey boundary and was not investigated in greater detail until Nelly Robles Garcia (1994) 14 mae un ATT oP ten, fe, oe saewe, at gt t Sna>* *, a re, . ey, . Pum nse” ‘ 2 "s, 4 . 2 a a ' 3 1 ‘ t \ Mixe region >, *, *,, Tehuantepec Ne a 7 ae - * eeenaee N. = \, ‘. \ Principal ancient travel routes into the Valley of Oaxaca from the north, east, and south. completed her study of known quarries in the Mitla area. She hypothesized that people from Mitla and elsewhere in the area exploited La Cuadrada in pre-Hispanic times and that the quarry could have been the source for cut stone blocks used in Postclassic constructions at Mitla. Yet with few ceramics recovered during her investigation of La Cuadrada and no off-quarry survey of the surrounding area, it was not possible to test the presumption that the quarry was relatively isolated and distant from any concentrated settlements. During our 1995 survey, we found a continuous scatter of Postclassic ceramics that connected La Cuadrada with the impressive ruins at Guirtin. Earlier during the Classic period, concentrated settlement at Guirun was as close as 500 m from the quarry. The quarry was not isolated and open to exploitation by all comers at the time when the Mitla palaces were constructed. During subsequent excavations that we conducted at the Mitla Fortress (2 km west of Mitla) from 2009 to 2011 (Feinman & Nicholas, 2011b), we ascertained that stone-working skills did not develop late in the Mitla area but rather were part of the occupational repertoire in this part of the valley as early as the middle of the Classic period. FIELDIANA: ANTHROPOLOGY ‘ Corral del CONG a 6 —~--—-- core Eastern Tlacolula a De c an) , Be 3/ Hormiga -. \Colorada VN Sy 0 A Archaeological site @ Modern town Fics 2:13: N kilometers The Guirtn/Albarradas survey region showing principal features, modern towns, location of five known sites, the dividing line between eastern Tlacolula and the Albarradas area, and the Camino Real. The third site is Hierve el Agua on the eastern side of the mountains, south of San Lorenzo Albarradas near the modern village of Roaguia. Today, the site is a tourist attraction known for a series of natural, freshwater springs oversaturated with calcium carbonate and two spectacular rock formations like waterfalls that formed over thousands of years (Figs. 2.21 and 2.22). The site also has pre-Hispanic occupations and a series of terraces and fossilized canals that date back as far as 300 BC, although investigators disagree on whether these features were used in agriculture or for salt production (Flannery et al., 1967; Hewitt et al., 1987; Peterson et al., 1989; Neely et al., 1990). With little or no information on other archaeological sites in this area beyond the physiographic bounds of the Valley of Oaxaca, these investigators considered Hierve el Agua unique and, at least implicitly, emphasized the significance of the natural springs in drawing people to reside nearby. During the survey, we recorded a number of contemporaneous sites in similar environmental settings as Hierve el Agua—defendable promontories near well-watered slopes on the eastern side of the Nueve Puntas—Cerro Guirone ridge—so while the fossilized canals at Hierve el Agua may be unique, the physical location of the ancient settlement is not entirely out of line with similar sites (especially in regard to its earliest occupation). Nueve Puntas is the name for the high mountain ridge that defines the southeastern limits of the Valley of Oaxaca (Fig. 2.23). We did not expect that it would be necessary to climb this tall, majestic series of nine peaks that are visible from more than 40 km away in the center of the valley. On its southeastern side, away from the valley, the ridge comprises a steep rock cliff with a 500-m drop (Fig. 2.24). There also were no archaeological investigations or published accounts of any sites on Nueve Puntas. But during the survey, we began to hear from local people that there was a series of walls just below the top of several of the peaks that they thought were colonial or historic corrals. Although we had doubts that there could be a pre- Hispanic site on Nueve Puntas, it seemed even less probable that anyone would have tended a herd of large herbivores on or near the top of this steep mountain. So we changed our original research plan and made the climb. The walls were there, but they were not corrals; they were defensive walls associated with a very well preserved set of residential terraces dating to late in the pre-Hispanic era (Fig. 2.25). FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 15 Fic. 2.14. Cerro Guirone and the Guirun site at the eastern edge of the Tlacolula arm of the valley; the Guirun site extends down the lower slopes of Cerro Guirone to the terraced slope in the center of the image. Llego Yiachi is the name for a distinctive, steep-sided rocky butte that resembles a scaled-down Devils Tower (Fig. 2.26). It rises above rough terrain along the Hormiga Colorada at the eastern edge of the survey area and, from the west, appears impossible to climb. But again, we were told of ancient ruins on the top of the butte, including standing adobe walls. We scrambled up and, indeed, did find well-preserved adobe walls (Fig. 2.27) that rival those at the Mitla Fortress (Feinman & Nicholas, 2004a, pp. 41-44). These five sites are discussed in greater detail in the relevant temporal chapters that present the settlement pattern data through time. Although Guirin, La Cuadrada, Hierve el Agua, Nueve Puntas, and Llego Yiachi may be among the most impressive sites in the surveyed area, they alone cannot tell the ee ene ae a 7 ease ae bd 3 . Fic. 2.15. Pre-Hispanic adobe walls on platform above the enclosed cruciform tomb in the El Paderon section of Guirun. 16 story of the changing eastern frontier of the Valley of Oaxaca. The other 60-plus previously undocumented sites in the region provide the necessary fuller temporal and spatial context. Field and Laboratory Analysis For our study of the Guirun/Albarradas area, we endeavored to maintain consistency with the general methodology of the surveys of the Oaxaca and Ejutla Valleys (Blanton et al., 1982; Kowalewski et al., 1989; Feinman & Nicholas, 2013), which in turn was adapted from the highland survey methodology that G Vue OMe Be ONY, 7 as S ow Fic. 2.16. Carved stone cruciform tomb on flattened ridgetop at Guirtn. FIELDIANA: ANTHROPOLOGY ihe hod hee os fous Ld "Ki ; Fic. 2.17. Interior walls of the cruciform tomb showing the preserved carved greca motifs. ie i William Sanders (1965) and his colleagues (Sanders et al., 1979; J. R. Parsons, 2009) developed in the Basin of Mexico (for fuller discussion, see Feinman & Nicholas, 2013, pp. 26-27). In the Oaxaca and Ejutla surveys, we walked every field and accessible slope and hilltop, checking for archaeological remains and mapping them directly on aerial photographs enlarged to a scale of 1:5,000. Aerial photographs at that scale are ideal for survey in relatively flat, semiarid highland valleys without dense stands of vegetation. They are less useful in mountainous areas where heavy vegetation often hides features of the landscape and also hinders the accurate piecing together of the photographic mosaics. For the Guirin/Albarradas survey, we employed a combina- tion of aerial photographs and topographic maps. Based on our experience using 1:5,000 aerial photographs in the steeper zones Px et Fic. 2.19. ‘ = 3 thE a Sade 6 ES He af: pemon Unfinished cut stone block in place at La Cuadrada. at the edges of the Oaxaca and Ejutla survey regions, we decided instead to enlarge the aerial mosaics to 1:10,000. Aerial photographs at this scale worked well for planning and implementing coverage in most of the survey area. They provided sufficient detail for mapping artifact distributions at the same time that the larger area covered on each photograph made it easier to decide how to position survey transects and to cover ridgetops in often difficult and remote terrain. In the more mountainous, tree-covered parts of the survey area, we relied more heavily on 1:50,000 INEGI topographic maps, as dense distributions of trees on the photographs obscured the terrain. In a few areas where the aerial photographic mosaics were put together poorly (e.g., a bad seam ran through the Guirtn site down to Roaguia, with a large strip of terrain missing), we depended entirely on the topographic maps. No Fic. 2.18. The terraced western slopes of Guirun. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 17 et as POSE A Py ore ; ; ; ES Fic. 2.20. Broken cut stone blocks at La Cuadrada. aerial photographs of appropriate scale were available for the northernmost part of the survey area. In flatter terrain, we traversed the area using the aerial photographs. We walked back and forth in teams of three to four people, checking every field, hill, and unpaved street in villages for surface indications of archaeological remains. For the steeper, more mountainous terrain, we slightly modified those field techniques and utilized procedures similar to those that were employed in the Penoles region (Kowalewski, 1991; Finsten, 1996), the Sierra Norte (Drennan, 1989), and the Sola Valley (Balkansky, 2002). In these rougher areas, we walked and checked the top of every hill and ridgeline and also checked the bases of ridges where they begin to flatten out without necessarily covering all the precipitous and heavily vegetated slopes in between. Contemporary habitations in the survey region and surrounding mountains are almost always con- Fic. 2.22. Calcified formation and tourist area at Hierve el Agua. structed on flat ridgetops or at the base of ridges and almost never on steep slopes. Because of the remoteness of much of the area (far even from unpaved roads); the steep, rough terrain; and the large number of poisonous snakes we encountered, we also often walked in pairs, an unnecessary practice during the valley surveys. When we observed archaeological remains, we mapped their extent on both the topographic map and the aerial photograph when possible. The outlines and the most significant features of larger sites with terraces and mounded architecture were placed on the map and photograph. Terrace and other architectural preservation tended to be good in the Guirtn/Albarradas area, and we were able to measure most residential terraces, house foundations, and platform mounds that we observed. We drew those features to scale and relative to each other on graph paper or in field notes and then every evening melded these maps with Fic. 2.21. Calcified formations at Hierve el Agua, viewed from the north. FIELDIANA: ANTHROPOLOGY Fic. 2.23. the drawings on the photos and topographic maps to produce a final map of the site. In the field, we used the broadly applied 100-m rule (Blanton et al., 1982, p. 40; Kowalewski et al., 1989, pp. 29-30; Balkansky, 2002, p. 30; Feinman & Nicholas, 2013, pp. 30— 31), defining a new site when no pre-Hispanic materials were observed within 100 m of the last occurrence. This practice provides maximum flexibility to record and organize field notes Fic. 2.24. The 500-m rock cliff on the southeastern side of Nueve Puntas. A 8 The distinctive mountain ridge Nueve Puntas defines the eastern edge of the Valley of Oaxaca. during survey and does not preclude combining sites later during analysis and interpretation. We followed the tripartite labeling system used in the Oaxaca and Ejutla surveys to assign field numbers to all sites. Each field name consists of three sets of letters followed by a number (e.g., TL-MIT-CC-1); the first letters refer to the administrative district (Tlacolula), the second set to the municipio (head town), and the third set to the agencia (the smallest administrative unit). Sites are numbered consec- FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 19 Fic. 2.25. Series of pre-Hispanic stone walls on Nueve Puntas. utively within these units as they are recorded in the field. More than 80% of the recorded sites are situated in lands that belong to San Lorenzo Albarradas (SLA), one site belongs to Santiago Matatlan (MAT), and the rest belong to San Pablo Villa de Mitla (MIT). San Lorenzo has one agencia, Roaguia (RO), and Mitla has two, Corral del Cerro (CC, now called San Juan del Paso) and Xaaga (XA). For each archaeological site, we recorded a consistent set of information on the nature of the archaeological materials that were present, the condition of the site, its environmental context, and modern use. Each site was dated and its size estimated by the spatial extent of the pottery visible on the surface, following the established ceramic sequence for the Valley of Oaxaca (see Table 1.1; Caso et al., 1967; Kowalewski et al., 1978, 1989: Feinman, 1980). Collections of ceramics and other surface materials were taken from one or more sectors of all the largest sites and in any other circumstance where we wanted to document more permanently the pre-Hispanic materials and to study them more fully. In total, we made 51 artifact collections from the 69 sites recorded during the survey (Table 2.1). One site lacked visible ceramics and could not be dated. Three sites on the western edge of the Guirun survey area were previously recorded in 1980; our findings at these three sites, including the Guirun site, expand the data collected in 1980. The 68 sites that could be dated include 123 separate jes Fic. 2.26. Western rock face of Llego Yiachi in the southeastern part of the survey area. 20 Fic. 2.27. Standing pre-Hispanic adobe walls on the top of Llego Yiachi. temporal components. More than half of the sites (63%) had ceramics from only one phase, mostly Monte Alban V. To process all the information and materials collected during this research, we followed the basic laboratory and analytical procedures that were employed during the Oaxaca and Ejutla survey projects (for a detailed discussion of these procedures, see Feinman & Nicholas, 2013, pp. 30-32). In the field laboratory, we washed and tabulated all the ceramic collections and made final temporal assessments that were checked against the information recorded on the maps and aerial photographs and in our field notes. We used this information to draw the boundary of each temporal component at every site and then used a compensating polar planimeter to measure the area (in hectares) of each component. We then estimated the population of every site component using procedures initially developed by William Sanders (1965) and Jeffrey Parsons (1971) for the Basin of Mexico surveys that were subsequently adapted for the Oaxaca and Ejutla surveys (Blanton et al., 1982, pp. 10-11; Feinman et al., 1985, p. 336; Feinman & Nicholas, 2013, p. 31). At most sites in the Guirin/ Albarradas area, ceramic densities were similar to those encountered in Oaxaca and Ejutla, categorized as very light in comparison with the heavier densities that had been encoun- tered in the Basin of Mexico (Sanders, 1965). We used the corresponding density figures of 10-25 people per hectare to estimate the population of most sites. A range of 5-10 people per hectare (or fraction thereof for all sites smaller than 1 ha) was used for temporal components for which we had observed only a trace of ceramics. More than 30% of the sites in the surveyed region had residential terraces or other visible domestic architecture that allowed us to estimate the number of houses. Most of the terraces were small and were counted as housing a single residence. We used a figure of approximately 300 m? per house lot (Winter, 1974, p. 982; Blanton, 1978, p. 30) to estimate the number of houses on larger terraces. We also included estimates of house foundations in unterraced areas. Once we had the total number of houses at a site, we used a range of 5-10 people per house. When terraces were too destroyed to count or measure, we used a terrace density of 25— 50 people per hectare for that sector of the site. For sites that included both terraced and unterraced areas, we calculated the population separately for each sector of the site and then summed those figures to obtain the estimated total population of the site. FIELDIANA: ANTHROPOLOGY TABLE 2.1. Field numbers and phase numbers for sites recorded during the Guirtin/Albarradas survey. Monte Monte Rosario Alban Alban Monte Monte Monte Monte Subarea Field site no phase Early I Late I Alban II Alban IIIA Alban IIIB-IV Alban V E. Tlacolula TL-MAT-MAT-95 — — — — = — V-57 Albarradas TL-MIT-CC-1 — — — — — — V-45 Albarradas TL-MIT-CC-2 — == — _— — — V-46 Albarradas TL-MIT-CC-3 = — — — — — V-47 Albarradas TL-MIT-CC-4 — — — — — == V-48 Albarradas TL-MIT-CC-5 — — — II-10 IWA-11 — V-60 Albarradas TL-MIT-CC-6 — — — I]-2 — — V-49 Albarradas TL-MIT-CC-7 — — LI-4 II-3 — — — Albarradas TL-MIT-CC-8 —_— — LI-9 II-7 IHA-7 —— — E. Tlacolula TL-MIT-XA-34 = — V-4]1 E. Tlacolula TL-MIT-XA-35 — V-42 E. Tlacolula TL-MIT-XA-36 — V-43 Albarradas TL-SLA-RO-1 — — Albarradas TL-SLA-RO-2 IA-1 V-1 E. Tlacolula TL-SLA-SLA-7 — V-44 Albarradas TL-SLA-SLA-8 IIA-10 V-59 Albarradas TL-SLA-SLA-9 — — Albarradas TL-SLA-SLA-10 — V-2 Albarradas TL-SLA-SLA-11 — V-3 Albarradas TL-SLA-SLA-12 — V-51 Albarradas TL-SLA-SLA-13 -— V-4 Albarradas TL-SLA-SLA-14 — V-5 Albarradas TL-SLA-SLA-15 a V-6 Albarradas TL-SLA-SLA-16 a V-7 E. Tlacolula TL-SLA-SLA-17 IA-13 V-62 E. Tlacolula TL-SLA-SLA-18 a V-8 E. Tlacolula TL-SLA-SLA-19 ITA-2 V-9 Albarradas TL-SLA-SLA-20 IHA-5 V-55 Albarradas TL-SLA-SLA-21 — V-10 Albarradas TL-SLA-SLA-22 IIA-6 V-56 Albarradas TL-SLA-SLA-23 — — Albarradas TL-SLA-SLA-24 — V-53 Albarradas TL-SLA-SLA-25 HIA-3 V-11 Albarradas TL-SLA-SLA-26 — V-12 Albarradas TL-SLA-SLA-27 — V-54 Albarradas TL-SLA-SLA-28 — V-13 Albarradas TL-SLA-SLA-29 — V-14 Albarradas TL-SLA-SLA-30 — V-15 Albarradas TL-SLA-SLA-31 a= V-16 Albarradas TL-SLA-SLA-32 — V-17 Albarradas TL-SLA-SLA-33 — V-18 Albarradas TL-SLA-SLA-34 — V-19 Albarradas TL-SLA-SLA-35 — V-20 Albarradas TL-SLA-SLA-36 — V-21 Albarradas TL-SLA-SLA-37 — V-22 Albarradas TL-SLA-SLA-38 = V-23 Albarradas TL-SLA-SLA-39 — V-24 Albarradas TL-SLA-SLA-40 — V-25 Albarradas TL-SLA-SLA-41 INWA-12 V-61 Albarradas TL-SLA-SLA-42 — V-26 Albarradas TL-SLA-SLA-43 — V-50 Albarradas TL-SLA-SLA-44 — V-27 Albarradas TL-SLA-SLA-45 — V-28 Albarradas TL-SLA-SLA-46 — V-29 Albarradas TL-SLA-SLA-47 = — Albarradas TL-SLA-SLA-48 — V-30 Albarradas TL-SLA-SLA-49 — V-31 Albarradas TL-SLA-SLA-50 — V-32 Albarradas TL-SLA-SLA-S1 _ V-33 Albarradas TL-SLA-SLA-52 — V-34 Albarradas TL-SLA-SLA-53 IIA-8, 9 V-58 Albarradas TL-SLA-SLA-54 IA-4 V-35 Albarradas TL-SLA-SLA-55 — V-36 Albarradas TL-SLA-SLA-56 — V-37 Albarradas TL-SLA-SLA-57 — ? Albarradas TL-SLA-SLA-58 — V-38 E. Tlacolula TL-SLA-SLA-59 = V-52 E. Tlacolula TL-SLA-SLA-60 = V-39 E. Tlacolula TL-SLA-SLA-61 a= V-40 FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO Analysis of Sourced Obsidian More recently, we have endeavored to examine shifting networks of obsidian exchange temporally and spatially across Mesoamerica by amassing an archive of more than 150,000 pieces of sourced obsidian from published sources (Pires- Ferreira, 1975, 1976; Braswell, 2003) supplemented by our own analyses using a portable X-ray fluorescence (p-XRF) spec- trometer (Golitko et al., 2012; Feinman et al., 2013, 2015; Golitko & Feinman, 2015). This archive includes more than 10,000 pieces of sourced obsidian from pre-Hispanic contexts in Oaxaca, a state for which there are no known obsidian sources. Obsidian circulated widely in Mesoamerica (e.g., Braswell, 2003), with tools and flakes of the material often found in pre- Hispanic contexts far from the original source from which it was mined. By expanding the sample of sourced obsidian and using new analytical methods, such as social network analysis (Golitko et al., 2012; Feinman et al., 2015; Golitko & Feinman, 2015), we have documented that the exchange of obsidian in Mesoamerica was anything but static across time; rather, through time and across space, the networks of interaction were dynamic in intensity, directionality, and the principal sources that were exploited. These shifts reflect and are part of larger-scale processes and changes across Mesoamerica on which we are able to draw to assess what happened more locally in the Valley of Oaxaca and its possible links to broader spatial fields. The Oaxaca samples that we have sourced come from dozens of sites in the Valley of Oaxaca and nearby regions. Jane Pires- Ferreira (1975, 1976) carried out the first sourcing studies in Oaxaca on small samples of obsidian from Early and Middle Formative sites. Using p-XRF, we have expanded the archive by more than an order of magnitude. We have been able to source the obsidian from some Late/Terminal Formative sites but mostly from Classic and Postclassic contexts that were excavated by a wide array of researchers, including ourselves. The archive for the entire state of Oaxaca includes obsidian from 18 different mines or sources. As with the pan- Mesoamerican archive, there are temporal changes and spatial variation in the obsidian that reached highland Oaxaca during the pre-Hispanic era. For the Late Classic, when we have the largest sample, obsidian assemblages varied not only from one site to another but also between houses at individual sites (Feinman et al., 2013). Although most of the obsidian that we have sourced is from excavated contexts, we also have sourced some obsidian that is associated with surface ceramics, including from the Guirun site. We make use of these archives of sourced obsidian when discussing interaction and exchange during the Classic and Postclassic periods. Pre-Hispanic Land Use Analysis To examine the relations between settlements and _ their surrounding catchments and between population and agrarian resources, we extended the pre-Hispanic land use analyses of the Ejutla and Oaxaca Valleys (Kowalewski, 1980, 1982; Nicholas, 1989) into the Guirin/Albarradas survey region (for a more detailed discussion than is presented here, see Feinman & Nicholas, 2013, pp. 32-37). These analyses build on Anne Kirkby’s (1973) seminal study of contemporary farming Nw N practices in the Valley of Oaxaca. Kirkby (p. 26) recognized that water is the critical resource determining agricultural yields in the valley. She married moisture availability with arable land to map the distribution of agricultural resources across the region. Stephen K owalewski (1980, 1982, p. 151) then simplified her classification by grouping land classes with similar yields. The highest yields are possible on Class I land (canal-irrigable and water-table land). Yields on Class Hf land (marginal water- table and good floodwater-farming land) can be as high but are generally more variable temporally. Crops may fail altogether in dry years on Class HI land (poor floodwater and dry-farming land) but can be highly productive in wet years. Class III land on steep slopes is considered only 10% arable. Another important contribution was Kent Flannery’s (Flannery et al., 1967; Flannery, 1970) findings that many simple water manipulation practices, which were employed in the Valley of Oaxaca well into the 20th century, had great time depth. Kirkby also attempted to extrapolate yields back into the pre- Hispanic past. She examined archaeological maize cobs from Tehuacan and Oaxaca and found increasing cob length through time, which correlates with increasing yields (A. V. Kirkby, 1973, pp. 124-126). Using Kirkby’s modern ranges and her estimates of past yields, we extrapolated high and low yields in metric tons per hectare (basically for high and low annual rainfall, respectively) for each land class throughout the pre- Hispanic sequence (Kowalewski, 1980, p. 154, 1982, p. 156; Nicholas, 1989). These yields are based entirely on maize, as we do not have information on pre-Hispanic crop mixes that would allow a more precise accounting. The maize yields serve as a proxy for total crop yields in temporal and spatial comparisons. The yield figures also take into account the critical limit in annual rainfall in Oaxaca (700 mm) that is required for a successful crop by adjusting yield figures in areas where the average annual rainfall is less than that amount to reflect marginal conditions for agriculture (Kowalewski, 1982, pp. 155-156). We recognize, however, that xerophytic plants may have played a larger role in subsistence in the driest (and steepest) parts of the region than is reflected in the maize yields (Feinman & Nicholas, 2005; Feinman et al., 2007). There are no large expanses of flat, valley floor land in the mountainous area at the eastern edge of the Valley of Oaxaca, and less than 1% of the Guirun/Albarradas survey tract is classified as Class I or II land; 40% is Class III land, and 60% is mountain slope that is uncultivable or only 10% arable (Fig. 2.28). These mountains and the area to the east, however, do receive more than 700 mm of annual rainfall and are classed as high-rainfall areas. With sufficient rainfall, Class III land can be as productive as Class II land. As an analytical tool to facilitate spatial comparisons, we extended the 4 x 4-km grid square system used in the Oaxaca and Ejutla Valleys and calculated the total amount of each land class per grid square (Table 2.2). Applying the respective yield ranges to each land class (for each time period) provides an estimate of agricultural resource potentials (i.e., carrying capacity or primary productivity) that can be compared across space and time. For approximations of the amount of food the pre-Hispanic populations that resided in the region during each time period could have produced, we factored in the available labor supply and the land area that a pre-Hispanic farmer could have cultivated. Those calculations are based on A. V. Kirkby’s (1973, p. 73) figure of 2 ha per farmer each year using a coa, the FIELDIANA: ANTHROPOLOGY Guirtin/Albarradas Survey Region E17 E18 E19 E20 WM Class | F234 Class II [EJ Class Ill, 100% Arable EZ” [_] Class III, 10% Arable raemitae oa cae a [1] Uncultivable Fic. 2.28. Distribution of agricultural land classes in the Guirtn/ Albarradas survey region. traditional Mesoamerican hoe (see also Tolstoy, 1988), and also Kowalewsk1’s (1980, p. 156) interpretation of Wayne Kappel’s (1977) past and present population profiles for four communi- ties in the eastern arm of the Valley of Oaxaca, in which he approximated the pre-Hispanic labor force as half the estimated archaeological population. As a proxy for the distance a farmer might travel to his fields (e.g., Chisholm, 1968), we confined all farmers to the grid square in which they resided. But regardless of why they chose to settle where they did, we assumed that they would farm the best land available to them. Within each grid square, we deployed all farmers to Class I and II land before adding in Class HII land. With these figures, we have approximations of producible local food supplies (in metric tons) and how they varied across time and space. An additional analytical step is to convert the potential agricultural yields into levels of population that could have been sustained. Kowalewski (1982, p. 152) reviewed consumption figures in ethnographic studies of communities in Oaxaca and elsewhere in Mesoamerica and arrived at a range of 0.16—0.29 metric tons of maize per person per year. We divided the potential agricultural yields by those consumption figures to calculate ranges of potential population that can be compared to archaeological populations across time and space to identify areas where the production of surpluses was possible and other areas where food imports (or greater reliance on wild, noncultivated resources) would have been necessary. All of FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO TABLE 2.2. Distribution of agricultural land classes (in hectares) in the Guirun/Albarradas survey region. Class Class Class Total Grid Area I Il III arable Uncultivated Rainfall N6E17 202 0 0 18 18 184 low N6E18 12 0 0 ] | 1] high N7E17 483 0 0 48 48 434 low N7E18 642 0 0 63 63 579 low N7E19 1,087 10 4 870 883 204 high N7E20 520 0 2 324 B27) 193 high N8E17 10 0 0 l ] 9 low N8E18 913 0 0 37] 371 541 low N8E19 1,599 0 4 656 659 239 high N8E20 872 0 0 872 872 0 high N9E17 110 0 0 1] 1] 99 low N9EI8 1,202 0 45 539 584 618 high N9E19 1,363 0 0 499 499 864 high N9E20 563 0 0 563 563 0 high N10E17 612 0 0 58 58 554 high NI0E18 910 0 4 17] 175 735 high NI10E19 47 0 0 5 5 42 high Total 11,144 10 58 5,069 5,136 6,007 these calculations involve ranges of yields and population levels. For ease of presentation, we generally refer to mean values. Because of high rainfall in the Albarradas area, farmers past and present can make a living in spite of the lack of large tracts of flat agricultural land and few opportunities for watering their crops beyond rainfall. But there was still risk given the unpredictable timing and nature of rainfall in highland Oaxaca. In average-rainfall years throughout the sequence, farmers in the Albarradas area could have grown more than enough food to supply all residents of the area. Only during the Early Classic, when settlement in the Albarradas area was heavily concentrated in one large site, would farmers have had possible difficulty growing enough food. With reduced rainfall in dry years, local produce during much of the pre-Hispanic period may have been insufficient in many parts of the area. With agricultural risk in addition to the personal risks of living in a cultural transition zone, the residents of Albarradas must have been reliant on exchange networks that followed the principal travel routes through the area on their way to the Valley of Oaxaca. Today, tourist visits to the hot springs at Hierve el Agua help sustain the local economies of communities in this region. Eastern Tlacolula and the Albarradas Area We referred to the area that we surveyed in 1995 as the Guirtn survey region because it centered on the Guirun site and the mountain slopes of Cerro Guirone at the eastern edge of the Valley of Oaxaca. Guirun historically was the most well known settlement in this region. The surveyed area includes not only the high mountains at the eastern end of the Valley of Oaxaca and the eastern slopes of those mountains in the Albarradas area but also area and sites (most notably Guirun) on the western side of the mountains that are part of eastern Tlacolula. As a whole, the surveyed area does not form one coherent physiographic region or subarea that can be compared to the Valley of Oaxaca or any of its constituent subregions. Such cross-regional comparisons facilitated the reporting on the Nw Wo regional survey in the Ejutla Valley at the southern end of the Valley of Oaxaca (Feinman & Nicholas, 2013). Here, we include all sites found during the survey, but for analytical, spatial comparisons we divide the Guirtn survey region along the mountain divide (see Fig. 2.13). We refer to the eastern side of the divide as the Albarradas area and the western side as eastern Tlacolula. For all subregional comparisons, we have added the sites on the western side of the divide to the data that were previously aggregated for the eastern Tlacolula subregion. Thus the figures that we present are an update from what has been reported earlier (e.g., Kowalewski et al., 1989; Feinman & Nicholas, 2013). The changes in both the total number of sites and the population estimates are especially marked for Monte Alban V. Throughout this study, we focus on the shifting relationship (e.g., Stoddart, 2010) between the Albarradas region and adjacent eastern Tlacolula. Eastern Tlacolula is the driest subregion of the Valley of Oaxaca, with the lowest yields and least reliable maize production (Taylor, 1972, p. 102; A. V. Kirkby, 1973; Nicholas, 1989; Druijven & Kruithof, 1992). During the Formative period (ca. 1600 BC-AD 200), popula- tions tended to be sparser in eastern Tlacolula than elsewhere in the valley (Kowalewski et al., 1989, pp. 63-64; Blanton et al., 1993, pp. 55-59). In contrast, for the Classic period, population densities in eastern Tlacolula were as high as or higher than in other sectors of the valley (Nicholas, 1989, p. 497; Feinman & Nicholas, 2013, pp. 96, 127). During the Postclassic period, Tlacolula, especially eastern Tlacolula, was the most densely settled sector (Kowalewski et al., 1989, p. 312; Feinman & Nicholas, 2013, p. 160). The high populations in eastern Tlacolula do not conform to the distribution of land and water resources in the Valley of Oaxaca. We have no evidence of major climatic changes that would have tilted greater availabil- ity of these resources toward eastern Tlacolula (Schoenwetter, 1974). So why did eastern Tlacolula become the most densely settled sector, and how did the population of that sector support themselves in their dry environment? Can the networks that pass through the Albarradas region yield any clues toward understanding these patterns? Craft production is one activity that likely helped sustain Tlacolula’s population (Finsten, 1983, 1995; Kowalewski et al., 1989; Blanton et al., 1993; Feinman et al., 2002). Local stone sources and fibrous xerophytic plants were especially important economic resources in eastern Tlacolula (Feinman et al., 2002; Feinman & Nicholas, 2005). The mountains in eastern Tlacolula are underlain by a layered volcanic tuff (M. J. Kirkby et al., 1986) where numerous veins of silicified tuff, or chert, have formed over time (Marcus & Flannery, 1996, p. 45). Chert outcrops of varying quality were heavily exploited in eastern Tlacolula, and stone nodules were the most common natural resource we observed during the survey of Albarradas. Many sites in the Albarradas area were situated on or near a chert source, and the most common craft specialization that we recorded was chert working. Magueys (Agave spp.) are highly versatile and long have been recognized as economically important in highland Mexico throughout the pre-Hispanic era (e.g., Gongalves de Lima, 1956; MacNeish, 1967; C. E. Smith & Kerr, 1968; M. E. King, 1986: Evans, 1990; J. R. Parsons, 2010; Correa-Ascencio et al., 2014); they are especially abundant in the dry, eastern arm of the valley (Sanchez Lopez, 1989), including on archaeological 24 sites (Martinez y Ojeda, 1996; Middleton et al., 2001). During excavations at several sites in eastern Tlacolula, we found high numbers of stone and bone tools associated with the processing and weaving of fiber into a range of products for immediate use and for exchange (Feinman et al., 2002; Feinman & Nicholas, 2004b). Palmillo, or yucca (Yucca spp.), also was exploited for fiber during the pre-Hispanic era (e.g., Sheldon, 1980; Anawalt, 1981, p. 27; Berdan, 1987, p. 245). Although large groves of palmillo are present on several archaeological sites in eastern Tlacolula (Martinez y Ojeda, 1996; Middleton et al., 2001; Feinman et al., 2002), the natural habitat for palmillo is on the eastern side of the mountains, in the Albarradas area (Schmieder, 1930, pp. 23-24; Cook, 1983, p. 41). Into the 20th century, towns in the Albarradas area continued to produce woven palm goods that were brought to Mitla and sold in the markets (E. C. Parsons, 1936, p. 60; Taylor, 1972, 103; Beals, 1979). Palmillos, however, are rare In nonsite contexts in the Valley of Oaxaca today; the plants most likely were brought into the valley in pre-Hispanic times from areas to the east (Albarradas) and fostered as an economic resource that also was turned into products for exchange and local use. The economic circulation of goods was an important facet of life in the Valley of Oaxaca long before the Classic period (e.g., Flannery, 1976; Kowalewski et al., 1983). Yet there is no doubt that exchanges and economic interdependence intensified across the region with the higher populations of the Classic period and greater valleywide interconnections (Kowalewski et al., 1989; Kowalewski, 2003a, 2012; Feinman & Nicholas, 2012). We have argued that the option to heavily exploit xerophytic plants for both food and economic products could have succeeded and endured in eastern Tlacolula only in the context of an expanding network of circulation and exchange (e.g., Feinman & Nicholas, 2005, 2010, 2012). Comparative Contexts The Valley of Oaxaca is a large, flat, mountain-ringed basin surrounded by many smaller valleys and extensive mountainous areas. Many of the areas that border the Valley of Oaxaca have now been surveyed, including Ejutla (516 km?; Feinman & Nicholas, 2013) and Sola de Vega (370 km?; Balkansky, 2002) to the south, the Pefioles area of the Mixtec Sierra to the northwest (850 km7?; Kowalewski, 1991; Finsten, 1996), the Sierra Norte to the north (650 km*; Drennan, 1989), and the Albarradas area to the east (88 km’). The Miahuatlan Valley (Markman, 1981) south of Ejutla has been partially surveyed, but because of the nature of the survey there (transects and reconnaissance of an area of 300 km’, mostly along the Miahuatlan River), we can consider only certain relevant patterns/findings. Although variable in the level of detail in recording and presenting settlement data, these studies serve to situate the large highland Valley of Oaxaca into a larger spatial context that extends beyond its mountainous confines (see Fig. Day: The physiographic boundary between the southern edge of the Valley of Oaxaca and the Ejutla Valley is not well defined, and we largely consider Ejutla a southern extension, or subarea, of the Valley of Oaxaca (Feinman & Nicholas, 2013). In contrast, the boundaries between the Valley of Oaxaca and FIELDIANA: ANTHROPOLOGY Sierra Norte Western Tlacolula Northern Valle Grande Eastern Tlacolula Albarradas Fic. 2.29. Valley of Oaxaca subareas and neighboring regions that are the principal focus of this volume. Albarradas, Pefioles, Sola, and the Sierra Norte are more sharply defined physiographically. These areas conform more closely to our perception of marginal or frontier areas in that they are separated from the Valley of Oaxaca by mountain ridges. A key element of our analysis here is to compare the Albarradas area with other areas at the edge of the Valley of Oaxaca. The Sola de Vega region has the most comparable detailed reporting (Balkansky, 2002), so we draw the most explicit comparisons with that area. More limited parallels and contrasts are possible with Penoles (Kowalewski, 1991; Finsten, 1996) and the Sierra Norte (Drennan, 1989), which have been less fully reported. We present many of our comparisons in schematic maps that include all four exterior areas, the Ejutla Valley, and the Valley of Oaxaca divided into six subareas (Fig. 2.29). In these comparisons, a clear pattern emerges, namely, that the areas exterior to the valley do not all follow a similar temporal course. The relationship between the valley and these outlying areas does not just reflect core-periphery dynamics but also a series of other factors, including larger networks of interaction that spread across the Oaxaca highlands and beyond. Andrew Balkansky (2002) completed his land use analyses of the Sola Valley following the procedures that we have used, so we have more than one edge area to compare with the Valley of Oaxaca. Sola comprises a very narrow river valley surrounded by steep forested slopes and in many respects resembles the Albarradas area in terms of overall agricultural productivity. We can compare the six subareas of the Valley of Oaxaca, the Ejutla Valley, Albarradas, and the Sola Valley along two dimensions: agricultural productivity per square kilometer and total potential productivity. In these comparisons, Sola and Albarradas are always near the bottom when it comes to agrarian production. Albarradas receives more rain than Rank order of subregions by productivity per km2 Fic. 2.30. Valley of Oaxaca subareas ranked by agricultural productivity per square kilometer. eastern Tlacolula, which is in a rain shadow, so productivity per unit area is higher than in eastern Tlacolula but lower than in the rest of valley, including Ejutla (Figs. 2.30 and 2.31). Although our calculations of agricultural production and potential maize surpluses and deficits are constructed on a series of assumptions, we believe that their consideration is essential to archaeological analysis, as it draws focus to the intersection between political and economic institutions, house- holds, and broader patterns of networks and relations that cross-cut time and space (e.g., De Lucia & Morehart, 2015, pp. 73-74). Albarradas is far from the center of the Valley of Oaxaca, but population densities were not always lower there than in the rest of the valley system. Population densities in Albarradas were lowest relative to the other subregions during the Early Classic, Total productivity of each subarea relative to subarea with highest productivity (1.0) Fic. 2.31. Total potential agricultural productivity of Valley of Oaxaca subareas relative to those with the highest productivity (1.0). FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 25 when the powerful at Monte Alban expanded their sphere of political influence farther into the Valley of Oaxaca proper (possibly encouraging people to move from the marginal Albarradas area to terrace sites at the eastern edge of the valley). In contrast, population densities in Albarradas were relatively larger later, when adjacent Tlacolula became the core of the densest settlement in the valley and demographic densities were reduced elsewhere in the valley system. In the rest of this work, we address a series of historical issues relevant to pre-Hispanic Oaxaca. What took place at the edges of the Valley of Oaxaca and in neighboring areas as Monte Alban was established, reached its greatest extent and power, and then declined? We should be clear that we do not expect that changes in the center of the valley impacted each frontier to the same degree or that the transitions always emanated from the valley’s center; what happened in each area abutting the valley also depends on other factors and exterior contacts. We also do not expect that the “boundaries” between the valley and these outlying areas were static (e.g., Stoddart, 2010): “Spatial scales are never fixed, but are perpetually redefined, contested, and restructured in terms of their extent, content, relative importance, and interrelations” (Swyngedouw, 1997, p. 141). That dynamism across time is what we aim to report and, to the degree possible, understand in the pages that follow. FIELDIANA: ANTHROPOLOGY Chapter 3: Middle and Late Formative (Rosario Phase and Monte Alban I) “Throughout the ancient world, urbanization in- volved the social integration of greater scales of community membership as well as the ratcheting up of social divisions along the continua of wealth, power, and occupational specialization that define urban landscapes.” (Carballo, 2016, p. 2) During the Early Formative period (ca. 1500-850 BC), the era when the first regional settlement hierarchies began to develop across highland Mesoamerica, community sizes and population densities were highly variable (Balkansky, 1998b; C. A. Smith, 2002). At that time, the Valley of Oaxaca, in the Southern Highlands, was one of the most densely settled highland areas. Most of the valley’s population was concentrated in the northern, Etla arm, around San José Mogote (Kowalewski et al., 1989, pp. 63-66; Flannery & Marcus, 1994, 2005, 2015), the largest community in the valley until Monte Alban was established circa 500 BC (see Fig. 1.1). When it was initially founded circa 1400 BC, San Jose Mogote was the only settlement in the valley with nonresidential buildings (Marcus & Flannery, 1996, pp. 87-88). In the Mixteca Alta, to the northwest, there were several small valleys where the population was clustered around one large center, like San Jose Mogote (see Fig. 2.1; Kowalewski et al., 2009). In contrast with those areas, settlements in the rest of highland Oaxaca were smaller Population densities per km? (exclude SJM) Fic. 3.1. Population densities in the Valley of Oaxaca and neighboring areas during the Early Formative. and widely dispersed. In the Valley of Oaxaca, population densities dropped quickly with distance from San José Mogote (Fig. 3.1; Table 3.1); they were especially low in eastern Tlacolula and southern Valle Grande. With the exception of the Ejutla Valley at the southern end of the valley, where there were three small dispersed sites (Feinman & Nicholas, 2013, figure 4.4), no Early Formative settlements are recorded in any of the other areas (including Albarradas) that ring the Valley of Oaxaca (although we cannot rule out the presence of mobile groups who would have left an ephemeral record). Albarradas and other frontier areas at the physiographic limits of the Valley of Oaxaca were sparsely settled, if at all, prior to the founding of Monte Alban. Not only were the areas most distant from San José Mogote (Sola, Albarradas, and Miahuatlan) empty of sedentary settlements, but even the two edge areas that abut the Etla arm of the valley (Penoles and Sierra Norte) were devoid of such occupations. Unlike Monte Alban, the later, much larger capital, San José Mogote did not influence or interact intensively with the residents of adjacent hinterlands much beyond the Etla (northern) arm where it was located. In fact, based on previous research (Plog, 1976), even San José Mogote’s interaction with the community of Huitzo, at the northern edge of the Etla arm, was limited (see Fig. 1.1). Middle Formative (Rosario Phase) The population of the Valley of Oaxaca did not grow appreciably throughout the Middle Formative period (ca. 850— 500 BC). San José Mogote decreased in size but still remained the region’s largest settlement with the most elaborate monumental construction. Yet the region’s inhabitants dis- persed across the valley away from the northern, Etla arm (Table 3.2; Kowalewski et al., 1989, p. 72). The initial dominance of San Jose Mogote lessened as several smaller centers with public architecture were established in the other arms of the valley—Yegiih in the Tlacolula subvalley, Tilcajete in northern Valle Grande, and Tlapacoyan in southern Valle Grande (Fig. 3.2). Clusters of settlements were established TABLE 3.1. Early Formative sites in Valley of Oaxaca subareas. Average Average No. Total site size site Subarea of sites population (ha) population Etla 18 1,573 4.9 87 Central 5 160 1.6 32 N. Valle Grande 9 106 0.4 12 S. Valle Grande 5 43 0.8 14 W. Tlacolula 5 50 0.4 10 E. Tlacolula 1 10 0.3 10 Ejutla 3 24 0.2 8 Albarradas 0 0 0.0 0 FIELDIANA: ANTHROPOLOGY N.S., NO. 46, NOVEMBER 13, 2017, PP. 27-53 ou ie NS ee Tioaiets i & c S\ PG Y ~ eo m., _ ay Ata wi : me uy A , tube? woos L . a cS a gone Wh SAD govt pr sewey SEs ane 3 EN . ‘. # ‘. rs ae me N, . N. y q 2 i 1 M \ C ( me 2 : ‘ : ate \ am So 2 ay & @ - \ Q VA eyo i 7 Neen A w eS Noes cre WwW ne ee ” a TW Qva re Er > S/n S$ NOI me > Za, ee Al 7 = Pa SY a. wr c 3 fs J a Y : Se > ; S Sites smaller than 2.0 ha are marked by triangles. All others are drawn to size and shape. Contour interval: 500 m. N 0 10 7 a kilometers Fic. 3.2. Rosario phase settlements in the Central Valleys of Oaxaca, showing principal places mentioned in the text. around these new centers, which may have been the head towns of small, spatially discrete polities that had variable degrees of autonomy from San Jose Mogote (Blanton et al., 1999, pp. 43- 44). Although there was some competition and conflict between these small polities, they engaged each other sufficiently 28 through exchange ties and other networks to have pottery assemblages that clearly belong squarely within a shared Valley of Oaxaca ceramic tradition. This basic settlement pattern—a dominant center surrounded by a cluster of smaller settlhements and lower densities of FIELDIANA: ANTHROPOLOGY TABLE 3.2. Rosario phase sites in Valley of Oaxaca subareas. Average Average No. of Total site size site Subarea sites* population (ha) population Etla 25 882 2.0 35 Central 9 189 1.0 21 N. Valle Grande 15 199 0.5 13 S. Valle Grande 9 151 0.8 ey W. Tlacolula 13 298 Le 23 E. Tlacolula 6 109 0.6 18 Ejutla 4 40 0.4 10 Albarradas ] 33 1.9 33 * Combined sites. population at greater distances from the center—has been reported in other highland Mesoamerican areas during the Middle Formative (Feinman, 1991), including the Tehuacan Valley (MacNeish et al., 1972), Puebla (Plunket & Urunuela, 2012), eastern Morelos (Hirth, 1987), and the Mixteca Alta (Kowalewski et al., 2009). The largest centers attracted the area’s populace to settle nearby by providing economic opportunities, including access to certain exotic goods, and possibly a greater degree of security. The sharing of certain decorative motifs on pottery across broad swaths of Meso- america at that time (e.g., R. N. Zeitlin, 1979, pp. 53-54) reveals widespread exchange and interaction, but these early settlement systems remained small. The dispersal of settlement across the Valley of Oaxaca extended south into the northern part of the Ejutla Valley, where four small settlements were established during the Rosario phase, circa 600-500 BC (Feinman & Nicholas, 1990, p. 222, 2013, figure 4.6). In the Tlacolula arm, settlement distribution was patchier, more concentrated in western Tlacolula, west of Yegitih. Four small Rosario phase sites near the eastern edge of the Tlacolula Valley were approximately 15 km distant from any other contemporaneous site in Tlacolula (Fig. 3.3; Kowalewski et al., 1989, map 1). The earliest recorded settlement in the Albarradas area (TL- SLA-SLA-52; see Fig. 3.3) was established at the same time in a low field at the eastern edge of the surveyed area, at a very low elevation (1,200 m) near the convergence of the Hormiga Colorada and a principal tributary of the Tehuantepec River. The principal travel route into the Albarradas area from the east follows this tributary. Although the settlement is east of the mountain ridge, it is only slightly more than 10 km from the nearest communities in Tlacolula, the four small Rosario sites in easternmost Tlacolula, which are located near the same route as it descends to Mitla on the western side of the mountain pass. The only other recorded Middle Formative site in the region is a very small settlement at San Dionisio (Fahmel Beyer, 2007, 2010), just south of the physiographic edge of the eastern arm of the Valley of Oaxaca (see Fig. 3.3). This site is approximately 15 km south of the nearest Rosario settlement in eastern Tlacolula Rosario Phase : _Colorada;* San Dionisio 4 — Sites smaller than 2.0 ha are marked by triangles. All others are drawn to shape and size. Contour interval: 500 m. Fic. 3.3. Rosario phase settlements in Tlacolula and the Albarradas area, showing principal places mentioned in the text. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 29 TABLE 3.3. Middle Formative ceramics at TL-SLA-SLA-52. Ceramic description Quantity Medium gris outleaned bowl with flared rim, milky gray 4 surface Medium gris outcurved bowl with flared rim ] Thin gris tecomate rim 2 Thin gris tecomate rim with fluted, squash-shaped body 1 Thin gris outcurved bowl with flared rim, milky gray ] surface Gris bowl base with specular red wash over milky gray i surface Medium café jar rim with red paint on rim Thick café jar rim with curved back neck and red paint on ] rim Café body with red wash on exterior Café tecomate rim (yellow-brown paste) Thick café tecomate with rim flare Thin café tecomate rim Medium tecomate with slight rim flare (micaceous inclusions) See eb but also is near a southern trade route into the eastern arm of the valley (Fahmel Beyer, 2010, pp. 11-12). Pottery at the Albarradas site fits well within the Valley of Oaxaca ceramic tradition. More than a dozen café (brownware) sherds at the site are characteristic of pre-Monte Alban pottery in the Valley of Oaxaca in terms of paste, form, and surface decoration (Table 3.3). Vessel forms include jars with red- painted rims and many fecomates, including one that is similar in form to a tecomate at San José Mogote (Fig. 3.4; Flannery & Marcus, 1994, figure 12.151). The rest of the collection consists of typical Oaxaca graywares, including several with very flared/ everted rims, milky gray surface, or specular red paint characteristic of Valley of Oaxaca pottery in the Middle Formative (Blanton et al., 1982, appendix VII). The Albarradas Rosario phase site was small (1.9 ha) and did not have public architecture. But with an estimated population of around 30, it was still larger than any contemporaneous settlement in eastern Tlacolula. The only other larger sites in Tlacolula were farther west at Yegiiih and Hacienda Alferez in the central part of the Tlacolula arm. TL-SLA-SLA-52, in Albarradas, was among the 10 largest sites in all of the Central Valleys of Oaxaca. A possible factor in the size of this community is its location along a natural route into the Valley of Oaxaca from the east that was heavily traveled throughout the pre-Hispanic era and into the present. Given the site’s isolation and documentary evidence for hostile relations ma cu Fic. 3.4. Pre-Monte Alban pottery at TL-SLA-SLA-S52 includes red-painted grayware vessel rim, tecomate rim, and sherd from a fluted vessel. 30 Rosario phase Population densities per km? (exlcude SJM) Fic. 3.5. Population densities in the Valley of Oaxaca and neighboring areas during the Middle Formative (Rosario). between Zapotecs and Mixes when the Spanish arrived (Schmieder, 1930; S. M. King, 2012), larger size also may have been in part for defense. The low density of population in the Albarradas area was comparable to other areas at the margins of the Valley of Oaxaca at that time (Fig. 3.5), especially Ejutla and the neighboring eastern part of Tlacolula (see Kowalewski et al., 1989, map 2). The southern expansion of population into the Ejutla Valley did not continue into the Sola Valley (Balkansky, 2002) or Miahuatlan (Markman, 1981). There continued to be no sites of this time period in either the Sierra Norte (Drennan, 1989) or the Penoles area (Kowalewski, 1991). Based on land use analysis, regional populations throughout the study area were far below levels that could have been sustained (Fig. 3.6). With such low populations, the residents of all occupied valley sectors could have grown enough food for themselves, with some to spare, on nearby arable land (Table 3.4). Sufficient labor and not arable land was the limiting factor. By far, the greatest surpluses (enough to feed 1000 people) were possible in Etla, the most densely settled sector (Fig. 3.7; Nicholas, 1989, p. 480), where people may have been attracted to settle near San José Mogote. Potential agricultural surpluses were very low in the sectors with the lowest populations— Albarradas and Ejutla. Exchange of foodstuffs beyond small local areas does not appear to have been a significant factor in interaction networks prior to the establishment of Monte Alban. Monte Alban Early I Monte Alban—the Southern Highland’s first urban center— was established on a hilltop at the center of the Valley of FIELDIANA: ANTHROPOLOGY Rosario phase ey Archaeological population as % of subarea maximum potential population Fic. 3.6. Rosario phase archaeological population as a percentage of the maximum potential population in each Valley of Oaxaca subarea. Oaxaca’s three arms circa 500 BC (Blanton, 1978; Blanton et al., 1993, p. 73) during an era of dramatic change across highland Mesoamerica that was characterized by rapid demo- graphic growth, the elaboration of regional settlement hierar- chies, population nucleation in larger settlements than had ever existed before, heightened episodes of public architectural construction, and the production and exchange of higher volumes of utilitarian goods (Blanton et al., 2005; Carballo, 2016). The overall growth rate in the Valley of Oaxaca between the Rosario phase and Monte Alban I (1.4% per annum) was greater than what could have been achieved by natural population increase alone; even a high annual growth rate of 0.6%—0.7% could have been maintained for only brief and localized episodes (Feinman et al., 1985, p. 345; see also Cowgill, 1975; Hassan, 1981, pp. 234, 253). Given low population throughout the region during the Rosario phase, significant in-migration—including the movement and settling TABLE 3.4. Estimated archaeological, resource-based, and labor- based potential populations for Valley of Oaxaca subareas in the Rosario phase. Estimated Resource-based Labor-based archaeological potential potential Subarea population population* population** Etla 882 40,000 1,900 Central 189 19,100 350 N. Valle Grande 199 40,000 350 S. Valle Grande 15] 34,900 300 W. Tlacolula 298 33,100 450 E. Tlacolula 109 15,900 175 Ejutla 40 36,600 70 Albarradas 33 5,900 40 * Based on full exploitation of all agricultural land resources in each subarea. ** Based on available labor supply and equal use of Class I and Class II land near recorded settlements. Rosario phase Surplus rounded to nearest 25 Fic. 3.7. Rosario phase potential agricultural surplus for each Valley of Oaxaca subarea. down of previously mobile groups who have left few easily recoverable archaeological surface remains—likely accounts for a good part of this growth. In-migration from mountainous areas that have not yet been studied by archaeologists also is possible. The location of Monte Alban in what had been a sparsely settled area south of San Jose Mogote markedly shifted the demographic center of the Valley of Oaxaca (Table 3.5; Kowalewski et al., 1989, pp. 90-91; Blanton et al., 1999, pp. 49-53). Monte Alban’s population quickly grew to 5,000 (Blanton, 1978, pp. 33-35). As people were attracted to the new capital, the rural population in the center of the valley expanded rapidly near Monte Alban, and a series of new larger settlements were founded elsewhere in the region (Fig. 3.8; Feinman et al., 1985; Kowalewski et al., 1989, pp. 85-111). Monte Alban I sites were larger, tiers of settlement arranged by size were more hierarchical, and population densities were greater in the Central area around Monte Alban and in the three subregions closest to the center (Etla, northern Valle TABLE 3.5. Monte Alban Early I sites in Valley of Oaxaca subareas. Average Average No. of Total site size site Subarea sites* population (ha) population Etla 60 3,175 2.8 53 Centra! 39 6,793 10.4 174 Central without 38 1,544 all 4] Monte Alban N. Valle Grande 57 1,998 1.8 35 S. Valle Grande 29 526 1.0 18 W. Tlacolula 46 1,814 De, 39 E. Tlacolula 22 34] 0.7 16 Ejutla 21 259 0.6 12 Albarradas 2 59 levi e Sola ] 12 0.3 12 * Combined sites. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 3] Ney, eae x ome Alban _ Lu De ¢ nt eo c anne, Satin gant FO An 7 meen i Grande, and western Tlacolula) than elsewhere in the region (Figs3.o): The location of a new regional capital in the valley’s center between a series of small, competing polities appears to have been an intentional strategy or social contract to establish a new political order or power-sharing arrangement between what had Wo tO Fe cocoa : ‘ A Sites smaller than 2.0 ha are marked by triangles. All others are drawn to size and shape. Contour interval: 500 m. N 0 10 | kilometers been autonomous (at least semiautonomous) groups (Blanton et al., 1993, pp. 69-71, 1999, p. 63; Feinman, 2016)., The formation of large aggregations involves the redirection of people’s time and energy, and they are more likely to be enduring when participants commit to cooperation and collective action and not when ordered by decree (Kowalewski, FIELDIANA: ANTHROPOLOGY Monte Alban Early | Population densities per km? (exclude MA) Fic. 3.9. Population densities in the Valley of Oaxaca and neighboring areas during Monte Alban Early I. 2013, p. 213). At Monte Alban, governance of the early polity exhibits many characteristics of collective political organization, whereas indicators of exclusionary rule, including depictions of individualized rulers and concentrations of wealth in select graves and houses, are rare, if not absent, in the city’s early history. The most impressive public space at the new center was the large Main Plaza, where a large proportion of the site’s population could have gathered at least periodically. At the same time, supernatural symbols, such as Cocijo, provided a broadly shared view of the Zapotec cosmos (Feinman, 2016; Joyce & Barber, 2016, pp. 46-48) that contrasts with the fire- serpent and were-jaguar supernatural opposition of pre-Monte Alban times (Marcus, 1989). The rapid demographic growth of the new urban center, exceeding 5% annually, as well as the demographic boom in immediately surrounding areas, attests to the attractiveness of this reorganization for both those with power and those without (Feinman et al., 1985, p. 345). The large aggregation in a hilltop locality also would have defensive advantages. Concern with defense is one factor that may spur or foment cooperative arrangements at village and larger scales (Kowalewski, 2013; Roscoe, 2013; DeMarrais, 2016). At the same time, in this more secure setting, small-scale labor investments were made in farming strategies (small canals and check dams) that likely increased agrarian production (O’Brien et al., 1982). Markets (Feinman et al., 1984) and reciprocal networks between householders, residing on terraces that shared retaining walls (Feinman, 2006; Kowalewski et al., 2006), helped buffer spatial and annual rainfall periodicities. For those at the top of the social pyramid, the increasing population and potential to produce agrarian surpluses likely helped secure larger funds of power, some of which was returned to the populace through the distribution of public goods and services (Fargher, 2016; Feinman, 2016; Joyce & Barber, 2016). At a domestic scale, the foundation of Monte Alban was timed with significant shifts in residential architecture, household storage, and even food preparation (Blanton et al., 1999). At greater distances, 30-40 km away from Monte Alban, population dropped off markedly (Kowalewski et al., 1989, p. 109). Population densities in eastern Tlacolula and southern Valle Grande were less than a third of the density in the sectors of the region closer to Monte Alban (northern Valle Grande and western Tlacolula). There also was minimal site hierarchy in the eastern and southern extremes of the valley. The 21 settlements dispersed across the Ejutla Valley to the south were all small hamlets or isolated residences (Feinman & Nicholas, 2013, pp. 54-55, 57). Balkansky (2002, pp. 35-37) recorded only one Early I site in the more distant Sola Valley. The site was small (less than 0.5 ha) and located in low-lying terrain along the Sola River, similar to the environmental setting of the earliest site in the Albarradas area. Although the site in Sola is isolated, ceramics on the surface most resemble those from the Valley of Oaxaca (Balkansky, 2002, p. 35). South of Ejutla, there also is one possible Early I site on the banks of the Miahuatlan River near the southern end of the Miahuatlan Valley (Markman, 1981, p. 114, table 2-2). This large site near the modern district head town was occupied throughout the sequence, and there is no indication of how large it might have been in Monte Alban Early I. Both local varieties and Valley of Oaxaca graywares are present at that site (Brockington 1973; Markman, 1981, p. 90). The parts of the Central Valleys of Oaxaca that are most distant from Monte Alban continued to be sparsely settled and not as heavily impacted by the foundation of Monte Alban as were areas closer to the center. The establishment of Monte Alban did not markedly or immediately affect the Albarradas area, which is more than 50- 60 km to the east. Settlement patterns in this outlying region did not change appreciably from the Rosario phase. In contrast, the foundation of Monte Alban appears to have had a greater effect on the growth of population in the mountainous hinterlands that were located closer to the urban center. The earliest sites recorded in the Sierra Norte and the Penoles area date to Monte Alban Early I. All sites with possible Early I occupations in Penoles are in the southern part of the survey tract that is closest to the Valley of Oaxaca; there were no sites yet in the northwestern part of Pefioles that is closer to the Mixteca Alta (Kowalewski, 1991). The pattern is similar in the Sierra Norte, where most Early J sites are in the center and east, with no sites in the most distant, northwestern part of that survey tract (Drennan, 1989, figure 11.4). As the pottery collected at sites in both areas conform to the Valley of Oaxaca ceramic tradition (Drennan, 1989, p. 374; Kowalewski, 1991, pp. 14, 20), we suspect that populations in these mountainous regions inter- acted more closely with valley populations than with Mixtec groups to the west and north. Graywares in the Valley of Oaxaca Monte Alban Early I tradition are extremely rare, if not absent, in the Mixteca Alta (Pérez Rodriguez et al., 2017, p. 589). In addition, the density of settlement in the Sierra Norte was higher than in southern Valle Grande and eastern Tlacolula and all other outlying areas. The site hierarchy also was more developed in the Sierra Norte (Drennan, 1989, p. 376), likely because of greater interaction with the center than was possible in the more distant areas both within the valley and beyond its physiographic boundaries. Although good provenience is FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 53 Monte Alban Early | Settlements in the Guirtn/Albarradas Survey Region E18 N10 <= E19 Spas “V/ Hormiga ‘}-, \Colorada ~~ 9 Sites smaller than 1.0 ha are marked by triangles. All others are drawn to size and shape. Contour interval: 500 m. ‘< ‘4 kilometers Fic. 3.10. Monte Alban Early I settlements in the Guirtn/Albarradas survey region. lacking, the earliest known ceramics in the Mixe area (east of the Valley of Oaxaca and Albarradas) date to Monte Alban I; the ceramics have characteristics that show influences from both the Valley of Oaxaca and the Isthmus of Tehuantepec (Hutson, 2014, pp. 267-268). The distribution of pottery with characteristics of the Valley of Oaxaca tradition beyond the physiographic limits of the valley illustrates that social and economic interaction permeat- ed physical boundaries, but there is no clear indication of political control. The spread of similar ceramics across wide areas is the result of expanding communication networks. The growing urban capital was beginning to have more widespread influence than San José Mogote had in the prior Rosario phase. The political hegemony of the Monte Alban polity in Early I was much closer to the center of the valley, likely including Etla, the northern part of the Valle Grande, and the western part of Tlacolula (e.g., Feinman, 1998, pp. 126-129), but even those relations were not necessarily marked by firm or finite territorial limits. Long-distance relations were fluid, likely affected by 34 changing networks of interaction and the friction of distance. Monte Alban’s greatest political impact and influence likely was limited to areas where the friction of distance did not erode the labor/agrarian taxes that could be exacted. Early I Settlements in Albarradas Two sites in the Albarradas area date to Monte Alban Early I (500-350 BC) (Table 3.6). The Rosario settlement at TL-SLA- SLA-52 almost doubled in size (2.8 ha, estimated population of 50), and one other very small site (TL-SLA-SLA-8) was established on the top of a long piedmont ridge descending the eastern side of Cerro Guirone (Fig. 3.10). Although the settlement at TL-SLA-SLA-8 (El Calvario) was larger in later phases, the Early I occupation covered only 0.5 ha, with an estimated population of less than 10. Neither Early I community in the Albarradas area was definitely associated with public architecture. Still, the site above the Hormiga FIELDIANA: ANTHROPOLOGY Monte Alban Early | | ! cat ! l | | | l ' ! | «a ! ; Hormiga! 7 _Colorada,” SS i Sites smaller than 2.0 ha are marked by triangles. All others are drawn to shape and size. Contour interval: 500 m. Fic. 3.11. Colorada (TL-SLA-SLA-52) continued to be larger than any site in eastern Tlacolula, although its size was no longer anomalous for the Valley of Oaxaca as a whole (Fig. 3.11). The new settlement at El Calvario also was positioned within 1 km of the most direct travel route into eastern Tlacolula from the east. The lack of settlements in the rest of the Albarradas area highlights the significance of this route. Pottery diagnostics at both sites continued to conform to the Valley of Oaxaca ceramic tradition, even if the pottery was likely locally made. The most diagnostic Early I ceramics at the two sites in Albarradas are gris (grayware) bowls that have very flaring or everted rims with fine incising on the exterior rim (Figs. 3.12 and 3.13; Table 3.7; types G-15, G-16, G-17; Caso et al., 1967, pp. 23-35; Kowalewski et al., 1978, p. 182; Blanton et al., 1982, pp. 377-378). These decorated graywares (Feinman & Nicholas, 2013, p. 241) are found at sites across the Valley of Oaxaca but are most abundant near Monte Alban and the three closest valley subregions (Etla, northern Valle Grande, and western Monte Alban Early I settlements in Tlacolula and the Albarradas area, showing principal places mentioned in the text. Tlacolula). The decorated (“fancy”) gris comprise 50%—60% of the graywares that can be attributed to Early I in all valley subregions, but in Ejutla and Albarradas they comprise at least 70%. So, although the abundance of fancy gris falls off with distance from the center of the valley, where they likely were made (Kowalewski et al., 1989, p. 94; see also Minc et al., 2016, p. 36), they appear to have been desired by residents of Ejutla and Albarradas, possibly as a sign of their affiliation with the increasingly important city in the center of the Valley of Oaxaca. Given the variety of fine gris sherds in the Albarradas collections, it is likely that at least some were not made in the Albarradas area. Other distinctive Early I pottery varieties, however, were not found at either site; these include distinctive crema (creamware) pottery that was produced near Monte Alban (Feinman, 1980; Joyce et al., 2006; Sherman et al., 2011; Mince et al., 2016, p. 33). Cremas have a spottier distribution across the Valley of Oaxaca, often including locally made copies, what we call “false cremas.” We did recover some false TABLE 3.6. Monte Alban Early I sites in the Guirtin/Albarradas survey region. Grid No. Site name Subarea* square Population ] TL-SLA-SLA-52 AL N7E20 50 2 TL-SLA-SLA-8 (El Calvario) AL N8E19 9 * AL = Albarradas. Mounds Mounds Mound Craft Exotics <1lm >1m volume (m*) Terraces work present 0 0 0 0 chert obsidian ? 0 ? 1 chert obsidian FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 35 mz cu Fic. 3.12. Early graywares at TL-SLA-SLA-52; two G-16s with everted rims are on the right. cremas in Ejutla, mostly in northern Ejutla, but even those false cremas were absent in the Monte Alban I Albarradas collections. Terrace Sites in Early I As populations grew throughout the region during Monte Alban Early I, a few new settlements were located on hilltops where houses were constructed on artificially flattened terraces. Monte Alban, of course, is the first and largest hilltop-terrace site in the Valley of Oaxaca (Blanton, 1978), but three much TABLE 3.7. Monte Alban Early I ceramics at TL-SLA-SLA-8 and TL-SLA-SLA-52. Quantity at SLA-52 Quantity Ceramic description at SLA-8 1 (T111) rim with fine-line carving 17 (T1338) rim from fish plate 15/17 (T1509) bowl rim with notched rim 15/16 (T1511) tecomate body G-15 (T1513) bowl with pinched in rim G-16 (T1515) bowl] with thick everted rim G-15 (T1517) bowl with flared rim and fine- line incising G-15/16 (T1518) bowl with flared/everted l — rim and incising G-12/15 (T1519) bowl with flared rim and fine-line incising Gris bowl/shallow plate with crenulated exterior rim Gris plate rim with curvilinear wide incised | a lines Medium gris outleaned bow! with flared — 4 rim, milky gray surface Medium gris outcurved bowl with flared rim — 1 Gris bowl with heavily flared rim — 1 Thin gris outcurved bow! with flared rim, _— 1 milky gray surface Broad gris everted rim with fine-line incising 1 — G- G- G- G- —- | eeu | | smaller ones were dispersed across the valley: one in Etla, one in northern Valle Grande, and one in Tlacolula (Fig. 3.14). Except for Monte Alban in the Central subarea of the valley, however, the other terrace sites housed only a small percentage of the region’s population (Table 3.8). Although not quantified, terraced hilltop settlements are reported in mountainous areas west and north of the valley—in Penoles and the Sierra Norte a Eicu Early graywares at El Calvario (TL-SLA-SLA-8) include a range of fancy incised vessel rims. Fic. 3.13. 36 FIELDIANA: ANTHROPOLOGY “4 (%, | MonteAlban = - 94 o ONL Fe 5 C 4 Terrace sites de 5 ’ O Defensive walls only N 0 10 eC kilometers Fic. 3.14. Monte Alban Early I terrace sites in the Central Valleys of Oaxaca. (Drennan, 1989, p. 383; Kowalewski, 1991, pp. 14-15)—but not Settlement and Land Use in Early I in mountainous terrain to the east—including the eastern edge of the valley and Albarradas—or to the south—including Population in the Valley of Oaxaca remained far below southern Valle Grande, Ejutla, and Sola. These patterns most regional capacities based on maize farming. Potential food likely reflect the attraction of residing in areas near Monte supplies (contingent on available labor) easily surpassed the Alban that also provided some measure of defense, possibly estimated regional population (Table 3.9; Fig. 3.15). All from other outside groups. subregions were self-sufficient except for the Central sector, FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 37 TABLE 3.8. Percentage of population residing in terrace sites in Valley of Oaxaca subareas in Monte Alban Early I. No. of Terrace % of terrace site Total population in Subarea sites population population terrace sites Etla 1 113 S175 4 Central 1 5,250 6,793 77 N. Valle Grande l 256 1,998 13 S. Valle Grande 0 0 $26 0 W. Tlacolula 0 0 1,814 0 E. Tlacolula i 45 341 13 Ejutla 0 0 29 0 Albarradas 0 0 59 0 Sola 0 0 12 0 where Monte Alban was located (Fig. 3.16). Based on available labor and arable land, the large rural (non—Monte Alban) labor force in the Central subarea would have been able to grow approximately double the food they needed for their own families; the rest of their crops would have fed about one-third of Monte Alban’s population of 5,000. Agricultural surpluses were even higher in Etla and northern Valle Grande; potential surpluses from those adjacent subareas would have been more than sufficient to feed the rest of Monte Alban’s inhabitants. The necessity of importing foodstuffs from nearby areas was one key in the greater integration of the central part of the valley into the polity centered at Monte Alban. Because governance and the population of Monte Alban were depen- dent on this agrarian production, the city likely had to supply certain public goods and services (e.g., defense and ceremonial activities) to the surrounding populations. At greater distances from Monte Alban, where populations were smaller, potential agricultural surpluses were much lower, especially in southern Valle Grande, eastern Tlacolula, and Ejutla. They continued to be negligible in Albarradas and Sola. These all were areas that were not tightly connected into the political system centered on Monte Alban Early | Archaeological population as % of subarea maximum potential population Fic. 3.15. Monte Alban Early I archaeological population as a percentage of the maximum potential population in each Valley of Oaxaca subarea. 38 TABLE 3.9. Estimated archaeological, resource-based, and labor- based potential populations for Valley of Oaxaca subareas in Monte Alban Early I. Estimated Resource-based Labor-based archaeological potential potential Subarea population population* population** Etla Bais 47,600 7,300 Central 6,793 21,700 3,400 N. Valle Grande 1,998 47,300 4,100 S. Valle Grande 526 41,800 1,200 W. Tlacolula 1,814 36,900 2,700 E. Tlacolula 34] 16,700 600 Ejutla 259 39,800 DS Albarradas 59 7,100 90 * Based on full exploitation of all agricultural land resources in each subarea. ** Based on available labor supply and equal use of Class I and Class II land near recorded settlements. Monte Alban and where settlement patterns were much less changed compared to earlier periods. Monte Alban Late I During Monte Alban Late I, Monte Alban tripled in size to approximately 17,000 people, monumental construction on the Main Plaza accelerated, and the urban capital expanded its political influence (Blanton, 1978, pp. 44, 46; Blanton et al., 1993, pp. 77-79). The city became one of the largest, if not the largest, urban center in highland Mesoamerica (C. A. Smith, 2002, p. 46). The greatest population increases and expansion of site hierarchies in the Valley of Oaxaca occurred in the same areas that had experienced the most significant growth in Early I (the Central and adjacent subareas—Etla, northern Valle Grande, and western Tlacolula), which formed the core area of the Monte Alban political system (Fig. 3.17). An inner ring of Monte Alban Early | Surplus rounded to nearest 25 Fic. 3.16. Monte Alban Early I potential agricultural surplus for each Valley of Oaxaca subarea. FIELDIANA: ANTHROPOLOGY 2 2500, us mo Pieaicte) “ cane ASU rv), aie gk oe is > wy > ; > J pod > Kan iee > MARA Mitt a Ce hy Maz ae Fic. 3.17. mounded sites lacked formal architectural groupings, while an outer ring had multiple mounded structures usually associated with plazas (Kowalewski et al., 1989, p. 146). The populations at the inner-ring settlements that lacked formal architecture likely were interdependent with Monte Alban. They may have had access to civic-ceremonial activities at the capital in return for labor or tax compliance. These sectors of the valley likely é ta Sites smaller than 2.0 ha are marked by triangles. All others are drawn to size and shape. Contour interval: 500 m. kilometers Monte Alban Late I settlements in the Central Valleys of Oaxaca, showing principal places mentioned in the text. had the most active communication and interaction with the regional capital. The outer ring included Tilcajete in the Valle Grande and Yegiih in Tlacolula, both of which expanded in size between Early I and Late I. Evidence of burning and building destruction at Tilcajete points to antagonism between Tilcajete (in northern Valle Grande) and Monte Alban (Spencer, 2003; FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 39 Monte Alban Late | Population densities per km? (exclude MA) Fic. 3.18. Population densities in the Valley of Oaxaca and neighboring areas during Monte Alban Late I. Spencer & Redmond, 2003; Redmond & Spencer, 2006; Spencer et al., 2008, pp. 335-336) that supports the interpretation that the entire valley was not peacefully united under Monte Alban and that some local elites had variable and unstable relations with the capital (Elson & Covey, 2006; Feinman, 2007; Feinman & Nicholas, 2013, p. 64). Less work has been carried out at Yegiith, but based on later changes at that site (Feinman et al., 2016), we also suspect that it at least occasionally had adversarial relations with Monte Alban. Based on ceramic distributions and architectural canons, however, both sites were culturally part of the larger Monte Alban system, and interactions between these centers may not have been solely acrimonious. Population densities in outlying valley areas beyond Tilcajete in the Valle Grande and Yegiih in Tlacolula continued to be much lower and settlement hierarchies flatter relative to areas closer to the core of the Monte Alban polity (Fig. 3.18). Nevertheless, these valley edges distant from Monte Alban did experience significant growth between Early I and Late I. In fact, population growth rates between Early I and Late I were higher near the edges of the valley, in southern Valle Grande and eastern Tlacolula, than they were closer to the center of the valley (Feinman et al., 1985, p. 346). Still, the estimated populations at these edges remained lower than closer to the capital. The demographic growth at the edges may reflect households that could make a living through exchanges or other interactions with areas closer to the region’s center but that were insulated by distance from more predatory relations with the rapidly growing capital. Even if Monte Alban did not politically dominate or unify the entire Valley of Oaxaca (Feinman & Nicholas, 2013, p. 62), the effects of Monte Alban’s increasing power and influence were felt across the region, drawing people into the valley from areas beyond its physiographic boundaries. But settlements beyond the outer 40 TABLE 3.10. Monte Alban Late I sites in Valley of Oaxaca subareas. Average Average No. of Total site size site Subarea sites* population (ha) population Etla 150 10,618 Bes 71 Central 119 22,844 7.0 192 Central without Monte 118 5,602 2.4 48 Alban N. Valle Grande 145 8,581 3.3 59 S. Valle Grande 75 2,633 1.7 35 W. Tlacolula 100 4,344 De) 43 E. Tlacolula 45 2,317 1.9 52 Ejutla 59 3,455 3.6 39 Albarradas 12 1,740 3.4 145 Sola 19 1,539 3.3 81 * Combined sites. ring of mounded sites were less connected into the central polity and likely had a different kind of relationship with Monte Alban (Feinman & Nicholas, 2013, p. 63). Beyond the Valley of Oaxaca, population and settlement trends were more variable. In all adjacent areas, there are ceramic affinities with the Valley of Oaxaca, which would seem to indicate cultural affiliations and some degree of interaction with the denser Valley of Oaxaca populace. Although no two exterior regions followed precisely the same pattern of demographic change, growth occurred across most of these areas. The number of sites and population densities increased dramatically in both the Albarradas area and the Sola Valley (Table 3.10). Because populations were so low in Early I, rates of population growth were very high in both areas, yet they also were high in Ejutla, Sola’s neighbor, and the rest of the Valley of Oaxaca, so at least part of the growth must have resulted from in-migration from outside the valley system (Feinman et al., 1985, p. 346; Balkansky, 2002, p. 37; Feinman & Nicholas, 2013, p. 65), even beyond Miahuatlan, which experienced growth in sites and populations similar to Sola (Markman, 1981, p. 63, table 2-1). People who previously lived in surrounding mountains and smaller valleys may have opted to migrate to the central valleys because of emergent regional exchange networks that lessened agricultural risk in a region with elevated seasonal and annual rainfall fluctuations (e.g., A. V. Kirkby, 1973; Flannery, 1983a; Nicholas, 1989). At the same time, the increasing agglomeration of population in larger centers facilitated and fostered the specialized production of utilitarian ceramic pots (Feinman, 1986b) and other utilitarian goods for exchange. The availability of such goods in regional exchange networks also may have encouraged domestic participation even in areas distant from the largest centers. These networks included the more distant Mixteca Alta, where grayware ceramics that were produced in the Valley of Oaxaca have been found (Joyce et al., 2006, p. 586; Perez Rodriguez et al., 2017, p. 589). This era was the beginning of what has been termed a “regional goods transformation” (Blanton et al., 2005, p. 274), when “the intertwined processes of political and domestic restructuring . . . incorporated a vast number of ordinary households as both surplus producers and benefactors of growing political and economic systems expressed as state formation and regional market evolution” (e.g., Blanton et al., 1999, chapter 4). FIELDIANA: ANTHROPOLOGY Monte Alban Late | Average site population (exclude MA) Fic. 3.19. Average site population in the Valley of Oaxaca and neighboring areas during Monte Alban Late I. Despite a high rate of growth, population density in the Sola Valley remained lower than anywhere else in the Valley of Oaxaca, while population densities in the Albarradas area matched those in the central parts of the valley (excluding Monte Alban) and were much higher than in adjacent eastern Tlacolula or in the Ejutla Valley at the southern end of the valley. There were dozens of new sites in Penoles, but all were very small, and population densities continued to be very low there (Kowalewski, 1991). In contrast, the Sierra Norte, one of the two exterior regions closest to Monte Alban, is the only area that lost population during the transition from Early I to Late I (Drennan, 1989, p. 376). Nevertheless, populations doubled in the more distant Mixteca Alta (Kowalewski et al., 2009, pp. 299-301). Monte Alban capped a multitiered settlement hierarchy in the Valley of Oaxaca (Kowalewski et al., 1989, pp. 129-131). This hierarchy of settlements, based on population or the scale of monumental architectural construction, was present even if the spatial focus was limited to those parts of the valley closest and likely most connected to Monte Alban—the Central area, Etla, and northern Valle Grande (Kowalewski et al., 1989, pp. 127-— 138). Settlement hierarchies were flatter in all the edge areas, including Albarradas. Half the sites in Albarradas were in the range of several hundred people (180-440), with no one clearly dominant center. All other sites were very small. In all the other edge areas, there were many more small sites and few settlements with estimated populations of more than 100 people. With so few very small sites in Albarradas, the average site size in Albarradas was two to three times larger than for any other edge area or subarea (excluding Monte Alban) of the Valley of Oaxaca (Fig. 3.19). Sola was very similar to its neighbor, Ejutla, while Albarradas was very different from its neighbor, eastern Tlacolula. Why is Albarradas so different in Late I from the rest of the region? A possible explanation is that aggregation provided a degree of safety in numbers for people living near an important trade route in a frontier area close to the boundary between two linguistic groups. Given the more recent history of hostilities between the Valley Zapotec and the Mixe to the east (Schmieder, 1930; S. M. King, 2012), the settlements in Albarradas may have been prone to village-on- village raiding, a practice that continued in the Central Valleys of Oaxaca into the 20th century (e.g., Dennis, 1987). The continuation of ceramics in the Mixe area with characteristics typical of Late I pottery in the Valley of Oaxaca (Hutson, 2014, p. 268) seems to indicate that interaction across this boundary did not cease. In fact, people may have been drawn to the eastern slopes of Cerro Guirone, possibly to be closer to an important travel route. The flow of lowland coastal goods to the Valley of Oaxaca likely followed riverine transport routes from distant large communities, such as Laguna Zope on the Isthmus of Tehuantepec (R. N. Zeitlin, 1993, p. 87). Late I Settlements in Albarradas The Albarradas area underwent a major settlement transition during Monte Alban Late I (350-200 BC). Not only did this area, like Ejutla at the southern end of the Valley of Oaxaca (Feinman & Nicholas, 1990, 2013), experience significant demographic growth, expanding from only two settlements to 12 (Table 3.11), but the physical nature of the sites and their locations shifted dramatically as well (Fig. 3.20). TL-SLA-SLA- 52, the earlier settlement in the low, humid fields adjacent to the Hormiga Colorada, was abandoned. At the same time, the small site at El Calvario (TL-SLA-SLA-8), located on top of a low but steep-sided hill (Fig. 3.21), grew significantly and became the largest settlement not only in the Albarradas area but also in the larger region including eastern Tlacolula (Fig. 3.22). The Late I occupation at the site (LI-11) expanded to 7.8 ha, with more than 400 people residing on 51 small residential terraces (Fig. 3.23). The terraces enclose an architectural complex that contains one large open plaza, with one small structure on either end, adjacent to a much smaller plaza enclosed on three sides by small structures. Eight additional, new communities were settled in the eastern part of the area, all within several kilometers of the Hormiga Colorada. The most common Late I ceramics on all of the sites are typical Valley of Oaxaca graywares, especially G-12 bowl rims incised with one or two parallel lines (Caso et al., 1967, pp. 25-26, figure 4; Kowalewski et al., 1978, p. 179) and bowl bases with combed bottoms (Caso et al., 1967, pp. 25-26, figure 5; Kowalewski et al., 1978, p. 181) (Fig. 3.24; Table 3.12). A few sherds of C-2 and C-4 crema paste with red paint (Kowalewski etal eel9O7S pope 719 174)y were *present™ but Vrare: * other nondiagnostic crema sherds could pertain to either Late I or II. Café jars and bowls with red paint, especially K-3 (Kowalewski et al., 1978, pp. 189-190), were present in low quantities at several sites. Although all show Monte Alban influence, most of the ceramics likely were made locally. Pottery in Chichicapan, just south of the Tlacolula Valley, also shows Monte Alban influence at this time; the presence of kilnwasters of typical Monte Alban I and II graywares at one site indicates that at least some of the graywares were locally made (Fahmel Beyer, 2012). FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 4] Monte Alban Late | Settlements in the Guirun/Albarradas Survey Region E17 E18 Sites smaller than 1.0 ha are marked by triangles. All others are drawn to size and shape. Contour interval: 500 m. Fic. 3.20. Monte Alban Late I settlements in the Guirtn/Albarradas survey region, showing sites mentioned in the text. TABLE 3.11. Monte Alban Late I sites in the Guirun/Albarradas survey region. No. Site name Subarea* 1 TL-SLA-SLA-25 AL 2 TL-SLA-SLA-40 AL 3. TL-SLA-SLA-47 AL 4 TL-MIT-CC-7 AL 5 TL-SLA-SLA-24 (Hierve el Agua) AL 6 TL-SLA-SLA-27 (Cerro del Rio) AL 7 TL-SLA-SLA-20 (Loma de las Tumbas) AL 8 TL-SLA-SLA-22 (Sitio de las Tunas) AL 9 TL-MIT-CC-8 (El Nopal) AL 10 TL-SLA-SLA-S3 (Llego Yiachi) AL 11 TL-SLA-SLA-8 (El Calvario) AL 12. TL-SLA-SLA-41 (El Trigo) AL 13. TL-SLA-SLA-17 (Guirtn) ET * AL = Albarradas; ET = eastern Tlacolula. ** ch = chert; gs = ground stone; lap = lapidary. 42 Grid square N7E19 N9E20 N9E18 NI0E18 N7E19 N7E19 N8E20 N8E19 NI10E17 N7E20 N8E19 N9E20 N8E18 Population 8 8 12 8 230 180 353 189 225 44 443 40 15 E19 Mounds Mounds >i1m <1lm — BNNeE COCCCSo on fk CO — SF WOWWOrRK OCC SO kilometers Mound volume (m°) Terraces 0 0 0 0 162 101 73 633 780 4,352 803 5 Craft work** ch ch, gs ch, lap ch ch ch, gs ch Exotics present obsidian obsidian? shell, obsidian shell, obsidian nonlocal chert obsidian shell, obsidian shell, obsidian FIELDIANA: ANTHROPOLOGY ye ERE 3 a Msg ha sobs in ei Fic. 3.21. Steep terraced slope of El Calvario (TL-SLA-SLA-8), viewed from the north. Most of the new communities in Albarradas were situated on top of defendable promontories overlooking large barrancas that carry seasonal water draining from mountain ridges. Today, many of these barrancas are tapped to supply water to fields near these ancient sites. Like El Calvario, four of the new sites had compact settlement arrangements composed of 10-30 residential terraces enclosing small architectural complexes on the top of a hill. These complexes consisted of either one small plaza with small structures on one side (TL-SLA-SLA-41; Figs. 3.25 and 3.26) or three sides (TL-SLA-SLA-20, TL-SLA-SLA- 22; Figs. 3.27—3.29) or one large open plaza with one small structure on either end (TL-SLA-SLA-27; Figs. 3.30 and 3.31). The terraces on these sites were arranged on eastern or southern slopes (facing away from the Valley of Oaxaca toward the Hormiga Colorada). Terraces on the eastern side of these settlements often were constructed right to the edge of the steep downward slopes. At Cerro del Rio (TL-SLA-SLA-27), located just above a sharp bend in the Hormiga Colorada, the stone retaining wall of the lowest terrace was particularly high and substantial, serving as a kind of fortification. In contrast, the parts of these sites that faced back to the Valley of Oaxaca were generally left unprotected. Access from the valley side was always easier (less steep) and more direct. Two of the new settlements were established on high promontories in the southern part of the survey area, one on each side of the Hormiga Colorada. We mapped several residential terraces at these two sites that appear to have been constructed in Late I, but in each case modern use or steep, heavily vegetated slopes prevented us from mapping and estimating the total number of terraces, and their populations could be higher than our estimates. One of these sites is Hierve el Agua (TL-SLA-SLA-24; see Fig. 2.22), where the Late I (LI- 5) occupation was centered on the steepest, most defendable ridge above the natural springs. There is one preserved mound on the ridgetop that may have been constructed as early as Late I]. The other site is Llego Yiachi (TL-SLA-SLA-S53), where the Late I occupation (LI-10) was concentrated on the top and upper slopes of the rock feature (see Fig. 2.26). The architectural complex on top of this large outcrop comprises Monte Alban Late | Chichicapan Valley Colorada,” can : \ SS Sites smaller than 2.0 ha are marked by triangles. All others are drawn to shape and size. Contour interval: 500 m. Fic. 3.22. Monte Alban Late I settlements in Tlacolula and the Albarradas area, showing principal places mentioned in the text. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 43 Fic. 3.23. four small plaza-mound groups. Although one building has preserved adobe walls like the Monte Alban V palaces at Mitla (Feinman & Nicholas, 2004a, figure 3.4), one or more of the other small complexes may have been built as early as Late I. TABLE 3.12. Ceramic description (10032) bow] with red paint Psd ee) bow] with red paint T1111) shallow plate with applique (T1207) bowl, 2 incised parallel lines on rim (11227) bowl, 2 incised parallel lines on rim (T1241) bowl, 1 incised line on rim (T1258) bowl, 3 incised parallel lines on rim (T1297) bowl with combed bottom /17 (T1509) bow] with notched rim /1 ( Kw cwbw we He toi tii 7 (T1510) bowl with wide incised lines T1519) bowl, 2-3 incised parallel lines on rim T2010) bowl with dark red paint 19 (T2011) comal /19 (12013A) jar with curved back neck /19 (12013B) jar with thin straight neck /19 (12013C) jar with thick straight neck (12064) bowl with red paint on rim 3 (12065) body sherd with red paint 3 ( 3 ( ( EAP Dies T2072) bowl with beveled rim and red paint T2076) jar with red paint T2085) ARAKRAKRAANKRAQOOADAAAAQAONOD ben * See Table 3.11 for full site name. 44 TL-SLA-SLA-8 Alban II. Site map of El Calvario (TL-SLA-SLA-8). Both sites had unobstructed views of the surrounding area and of each other. Two other new Late I settlements in the east were small isolated residences that did not continue into Monte Monte Alban Late I ceramics at sites* in the Albarradas area. LI-1 11-3 Li-4\_ LI-S* (LE6" 9L1-7 11-8 1-9 ee LE 10m 1 i as es Ze 1 a a e— pe = 1 se = = = 1 — | 5 = 3 | 3 4 - 19 nt 3 | - fs 6 = 2 _ ke 2 2 1 _ = ba = a 1 a 3 7 2 2 a 14 2 = 2 a = os | 2 1 iz = a Ea | ~ “Zs = a= 1 = p) = 1 a aif mas ae oe | = 3 1 bs 1 | a a 2 3 10 _ 1 oe 2 1 | = = x | 2 * 4s = bs = 3 ae 28 Pa BE at | = +s = 1 Ae 4 i e— = fe = ] a ae pit ae | | | | | FIELDIANA: ANTHROPOLOGY Fic. 3.24. Monte Alban Late I G-12 ceramics from El Calvario (TL-SLA-SLA-8). Three defensibly situated settlements also were established in the northern part of the survey area, in the rough hills surrounding the small valley of San Juan del Paso (known as Corral del Cerro when we did the survey), where the modern highway passes after leaving the Valley of Oaxaca near Mitla. Two of these sites were very small (TL-SLA-SLA-47 [LI-3] and TL-MIT-CC-7 [LI-4]); the third one (TL-MIT-CC-8) was one kd ms Fic. 3.25. View of terraced southwest ern slope of El Trigo (TL-SLA-SLA-41). of the four largest Late I sites (LI-9) in the survey area (Fig. 3.32). This very isolated settlement was located on the top of a steep-sided, heavily vegetated hill in the high piedmont. Today, it is known as El Nopal for the large number of nopal cacti that grow with other xerophytic plants on site. The archaeological settlement consists of a complex of small structures and platforms around a small plaza on the very top of the hill, FOSS ‘Eee ee : FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 45 TL-SLA-SLA-41 1820 m Plate I-44 IA-12 IIB-IV-12 V-61 Fic. 3.26. Site map of El Trigo (TL-SLA-SLA-41). surrounded on three sides by a stone defensive wall; two stone wall remnants on lower slopes to the north and west appear to be the remains of a lower defensive wall that encircled at least the northwestern part of the site (Fig. 3.33). There are as many as 30 terraces on the slopes between the lower and upper Se Af dd le id % defensive walls, but we were unable to map most of them because of the heavy xerophytic vegetation. The final Late I site (LI-13) in the surveyed area is the first occupation of the Guirtn site (TL-SLA-SLA-17) in eastern Tlacolula. This initial settlement on an upper ridgeline on the ie ae Fic. 3.27. Loma de las Tumbas (TL-SLA-SLA-20) located on flat ridgetop (in the center of the image) above the Hormiga Colorada. 46 FIELDIANA: ANTHROPOLOGY iN Fic. 3.28. western slope of Cerro Guirone (see Fig. 2.18), north of the mountain pass, was very small. It is located at the juncture of the 1980 and 1995 survey regions and was not detected until the intensive site survey in 1996 (Feinman & Nicholas, 2013), when we observed a Late I K-3 (T2042) bowl fragment and a G-12 (T1241) bowl rim on two adjacent terraces (Kowalewski et al., 19785 ppsld Il 89): Eight of the 12 sites in the Albarradas area are fairly comparable in size, ranging from several hectares to slightly more than 13 ha. These sites are on defendable hilltops and have one or more small architectural complexes on _ their summits, with terraced slopes below. Most have estimated populations of several hundred people. There were only four sites without public architecture and residential terraces; those four were very small isolated residences or hamlets with populations of 12 or less. The four small sites were widely dispersed, while all but one of the larger, defendable villages were roughly evenly spaced (roughly 1.5—3 km apart), forming a band outside the eastern limits of the Valley of Oaxaca. Collectively, these sites could have monitored all of the natural trails and topographic breaks that pass through the high mountains at the eastern edge of the Valley of Oaxaca. TL-SLA-SLA-20 Site map of Loma de las Tumbas (TL-SLA-SLA-20). The discovery of so many relatively large sites dating to Monte Alban Late I in the Albarradas area certainly was not expected. Few Late I sites of any size were recorded at the eastern edge of the Tlacolula arm of the Valley of Oaxaca during the 1980 survey (Kowalewski et al., 1989, maps 2 and 3). The pre-Hispanic settlement at spectacular Hierve el Agua had long been considered an anomalous Late I outlier, positioned to take advantage of the natural spring (see Fig. 2.21). The Albarradas survey provided the larger context to see that Hierve el Agua (TL-SLA-SLA-24) was not so anomalous; it was part a line of terrace sites positioned on defendable promon- tories above the Hormiga Colorada during Monte Alban Late I. This line of defensively located sites would appear to define a defended frontier just beyond the physiographic edge of the Valley of Oaxaca. But given the distance of these sites to Monte Alban and the low population densities and lack of monumen- tal architecture or a clear hierarchy of settlement in the eastern part of Tlacolula, we do not believe that this line of sites forms an eastern political boundary of Monte Alban’s political system, as we postulated in earlier publications (Feinman & Nicholas, 1996, 1999). FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 47 Fic. 3.29. Many of these sites could have served as gateways or stopover points on exchange routes as well as having defensive functions. Marine shell was not a common good on most sites in the Albarradas area and was recorded at only four sites with Late I pottery (El Calvario, TL-SLA-SLA-20, TL-SLA-SLA-22, and El Trigo). None are single component (most of the shell also could pertain to Monte Alban IJ); the association with Late I is least clear at El Trigo, which was occupied continuously from Monte Alban Late I through V; the other three sites are close to the most direct travel route into the valley. Nearness to the travel route could have increased access to goods both from afar and from the valley. Residents of several of these sites also may have made basic goods for exchange. We observed evidence of chert working at a high percentage of Late I sites (7 out of 12) in the Albarradas area. Two of these sites are in the small northern valley (TL-MIT-CC-7 and TL-MIT-CC-8), and five are terrace sites in the east near the travel route (El Calvario, TL-SLA-SLA-41, TL-SLA-SLA-20, TL-SLA-SLA- 22, and TL-SLA-SLA-27). In six cases, though, the sites are multicomponent, and the craftwork also could pertain to Monte Alban II. Only at TL-SLA-SLA-27 is the working of chert (and also ground stone) more definitively associated with 48 TL-SLA-SLA-22 Site map of Sitio de las Tunas (TL-SLA-SLA-22). the Late I occupation. Residents of El Calvario and TL-SLA- SLA-20, respectively, also may have made ground stone tools and lapidary items based on the surface remains observed. The ceramics collected at these sites share close stylistic affiliations with Monte Alban and the Valley of Oaxaca, especially G-12 graywares (see Fig. 3.24), so clearly populations in the Albarradas area engaged in some level of interaction and exchange with communities in the valley, yet the crema ceramics that are identified with Monte Alban were extremely rare in Albarradas. These settlements may have been partially aligned with the Monte Alban polity in the same way that Ejutla was at this time (Feinman & Nicholas, 2013, p. 76). Neither area was directly controlled or politically well integrated into the Monte Alban polity but may have had access to and used ceramic goods that signaled affiliations with the Valley of Oaxaca and the Monte Alban polity (Feinman, 2018). The Late I sites in Albarradas seemingly were part of Monte Alban’s outer circle of influence, but they had a different relationship with the center than did sites closer to the urban capital. In terms of access to services at (and obligations to) the capital, the Albarradas sites were much less interconnected to Monte Alban. FIELDIANA: ANTHROPOLOGY -: P a Fic. 3.30. View of Cerro del Rio (TL-SLA-SLA-27) on ridgetop in bend of the Hormiga Colorada. TL-SLA-SLA-27 LI-6 V-54—w.T12_ Fic. 3.31. Site map of Cerro del Rio (TL-SLA-SLA-27). FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 49 Fic. 3.32. View of terraced site (TL-MIT-CC-8) on El Nopal in the center top of image. The settlement changes in the Albarradas area seem to have been the result of expanding exchange connections and population movements rather than the consequence of persis- tent political interventions. The travel route along the Tehuantepec River to the Isthmus, from where a wide array of coastal products were available (R. N. Zeitlin, 1993, p. 87), passes through the Albarradas area and likely made it an attractive, albeit somewhat risky, place to settle. Since many of the larger Albarradas settlements continued to be occupied in TL-MIT-CC-8 50 Monte Alban II, it is possible that these communities were settled very late in Monte Alban I as the leaders at the capital began to explore new revenue sources by establishing extra- regional links to tropical products and pursuing military forays into more distant realms (e.g., Marcus, 1976). Late I Terrace and Defensive Sites Monte Alban grew much larger, and a number of other terrace sites were established across the region. The formation of these hilltop sites likely was motivated primarily by defensive concerns, as the inhabitants tried to protect themselves from potential raids or attacks from neighbors from both inside and outside the valley (e.g., Roscoe, 1996). Cooperation and a commitment to collective action to provide for community defense is the most effective way to deal with these security concerns (Roscoe, 2013; DeMarrais, 2016). The construction of the artificially flattened terraces behind shared stone retaining walls at these hilltop sites also required significant labor and multihousehold cooperation and organization (Feinman et al., 2002, pp. 271-272; Kowalewski et al., 2006, p. 207). Although terrace sites were widely dispersed in the Valley of Oaxaca, the densest concentration in Monte Alban Late I was in the northern, Etla arm. Except for Monte Alban, they were not significant features of the valleywide landscape. There were FIELDIANA: ANTHROPOLOGY een, 4 Terrace sites O Defensive walls only N 0 10 a kilometers Fic. 3.34. Monte Alban Late I terrace sites in the Central Valleys of Oaxaca. no terrace sites yet in the Ejutla Valley. In contrast, fortified hill towns were established in many neighboring areas, including at some distance in the Mixteca Alta and Cuicatlan Canada, likely as a response to military concerns as Monte Alban grew and attempted to expand its reach (Kowalewski et al., 2006). The first terrace sites in the Albarradas and Sola regions were established at this time (Fig. 3.34; Table 3.13). A marked change in Albarradas in Late I was the establishment of many small terrace sites (8 of the 12 Late I sites have ancient residential terraces) in defendable locations, including isolated promontories. The terrace sites in Albarradas were clustered in a band on the eastern, lower slopes of Cerro Guirone above the FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 5] T wi ) ? PABLE 3.13. Percentage of population residing in terrace sites in Valley of Oaxaca subareas in Monte Alban Late I. No. of Terrace % of terrace site Total population in Subarea sites population population terrace sites Etla 5 3.361 10,618 32 Central 3 17.656 22,844 Teh N. Valle Grande 2 872 8.581 10 S. Valle Grande 3 431 2,633 16 W. Tlacolula ] 473 4,344 11 E. Tlacolula 4 1,070 2317 46 Ejutla 0 0 3,455 0 Albarradas 8 1,704 1,740 98 Sola 8 1,291 1,539 84 Hormiga Colorada and adjacent waterways. Most of them were in line of sight of each other; their proximity to each other, or safety in numbers, may have been an important factor. These compact settlements accounted for 98% of the population in the Albarradas area. In the Sola Valley, 8 of 19 Late I settlements are terrace sites, accounting for 84% of the population (Balkansky, 2002), but the terrace sites there were more dispersed across a larger area than in Albarradas. These two exterior areas alone had almost as many terrace sites as in all of the Valley of Oaxaca (where there were as many as 18). Even including Monte Alban, the percentage of the population residing in terrace sites in the Central subarea was lower (77%). In eastern Tlacolula, 46% of the population lived in terrace sites. In mountainous Penoles and Sierra Norte, there are no large valleys and very little flat land (except on ridgetops), and many if not most sites throughout the sequence consisted of artificially flattened terraces constructed on hillslopes. In the Penoles area, many Late Formative sites were defensively located, including a string of single-phase Late Formative sites (with Monte Alban— style ceramics) situated on a ridge overlooking the adjacent Nochixtlan Valley to the north (Kowalewski, 1991), apparently defining a settlement boundary similar to that in the Albarradas area. Yet site layouts and the total number of terraces occupied in Penoles in Late I are unknown due to larger, later TABLE 3.14. Estimated archaeological, resource-based, and labor- based potential populations for Valley of Oaxaca subareas in Monte Alban Late I. Estimated Resource-based Labor-based archaeological potential potential Subarea population population* population** Etla 10,618 57,600 21,300 Central 22,844 27,200 12,600 N. Valle Grande 8,581 57,300 17,600 S. Valle Grande 2,633 49.700 6,000 W. Tlacolula 4,344 46,700 8,100 E. Tlacolula Qe) 21,900 3.800 Ejutla 3,455 51,000 7,000 Albarradas 1,740 8.300 2.700 Sola 1,539 17,600 2,800 * Based on full exploitation of all agricultural land resources in each subarea. ** Based on available labor supply and equal use of Class I and Class II land near recorded settlements. 52 Archaeological population as % of subarea maximum potential population Fic. 3.35. Monte Alban Late I archaeological population as a percentage of the maximum potential population in each Valley of Oaxaca subarea. occupations (Kowalewski, 1991). Similar patterns are reported in the Canada de Cuitcatlan to the north (Spencer, 1982, 2007; Redmond, 1983; Redmond & Spencer, 2006), although the defendable sites are not as clustered as in the Albarradas area. There were no similar lines of defendable sites in Ejutla (Feinman & Nicholas, 1990, 2013) or at the southern physiographic edge of the Valley of Oaxaca. If the Miahuatlan Valley were to be more fully studied, such sites should be searched for along its southern extent. The populations of the different edge areas of the Valley of Oaxaca responded differently to potential outside threats, influences, and oppor- tunities. Settlement and Land Use in Late I In spite of growing populations, regional capacities based on maize farming continued to far surpass the population in the Valley of Oaxaca, and potential food supplies (based on available labor) easily surpassed the regional population. All subareas were self-sufficient except for the Central sector, where Monte Alban was located (Table 3.14; Fig. 3.35). Based on available labor and arable land, the large rural labor force in the Central area would have been able to grow more than twice the amount of food they needed for their own families; the rest of their crops would have fed about 40% of Monte Alban’s population of 17,242. Surpluses were even higher in Etla and northern Valle Grande; potential agricultural surpluses from those adjacent subregions would have been more than sufficient in most years to feed the rest of Monte Alban’s inhabitants (Fig. 3.36). Monte Alban likely appropriated agricultural surplus and labor in these areas to support governance that in return provided public goods, such as ceremonial spaces, temples, and defensive features (Feinman & Nicholas, 2017, p. 270); these public goods helped draw people to settle near the capital. Most of the other valley subregions could have produced about double what they needed. The lowest surpluses were in eastern FIELDIANA: ANTHROPOLOGY Monte Alban Late | Surplus rounded to nearest 100 Fic. 3.36. Monte Alban Late I potential agricultural surplus for each Valley of Oaxaca subarea. Tlacolula, Sola, and Albarradas. These subareas still were not strongly connected into the political system centered on Monte Alban, and with sufficient crops closer to home, Monte Alban’s rulers likely did not engage these areas for staple products. Any interest in Albarradas more likely focused on trade routes that passed through that area, bringing coastal goods up to the highlands. Middle and Late Formative Summary The Albarradas area, like other edge areas, was settled late compared to the Valley of Oaxaca and was only sparsely occupied until the Late Formative period (Monte Alban Late I). The earliest settlements were situated near the principal eastern trade route (the Camino Real) into the Valley of Oaxaca. Although these small frontier settlements were outside the physiographic limits of the Valley of Oaxaca and near the modern Mixe/Zapotec boundary, the ceramic assemblages at the Albarradas sites mostly comprise typical Valley of Oaxaca pottery and indicate high degrees of connectivity with valley populations. During Monte Alban Late I, the Albarradas area experienced rapid growth, especially in defendable locations where agricul- tural needs also could be met. The Albarradas area stood out from not only the other edge areas but also most Valley of Oaxaca subregions in having high population density, the largest average site size, the greatest nucleation of population in the fewest sites, and the highest percentage of the population in defendable terrace sites. These all represent major changes from Early I. Although our knowledge of the pre-Hispanic history of the Mixe area is limited, we suspect that these changes reflect a defensive posture by people who were weakly integrated with the complex political system in the Valley of Oaxaca. The residents of the Albarradas area participated in the Valley of Oaxaca economic and interactive sphere—they clearly followed the same ceramic tradition and architectural canons—but they do not seem to have been strongly tied into the Monte Alban polity (compared with areas closer to the capital) any more than eastern Tlacolula and Ejutla were at that time. With Monte Alban becoming the largest known political center in the Mesoamerican highlands and the expansion of the Monte Alban polity in Late I, there may have been more traffic/ tension/interaction along the trade route through Albarradas. The people of Albarradas may have been protecting themselves from peoples to the east but also from the valley. Albarradas shares some similarities with the Sola Valley, which also had relatively high average site size and a high percentage of the population in defendable terrace sites, but population densities were much lower in Sola than in Albarradas. The most important trade routes into the Valley of Oaxaca from the south pass through the Ejutla Valley (White & Barber, 2012) and Miahuatlan (Ball & Brockington, 1978, p. 114), not Sola, yet we do not see similar changes in Ejutla. Sola also was part of the same ceramic tradition, but that distant region was probably even less tied in than Ejutla and Albarradas were (cf. Balkansky, 2002, p. 45). Settlement in Sola also had a much more defensive character than in Ejutla and Miahuatlan, where Markman (1981, p. 63) recorded only one Late I hilltop site. Many edge areas may have been dangerous places to live during Late I, when Monte Alban was attempting to expand its influence and hegemony. The spread of ceramics of the Valley of Oaxaca tradition beyond the physiographic bounds of the valley in Monte Alban I is not an indication of the extent of Monte Alban’s political control; rather, it is evidence of the urban center’s influence/ participation in interaction networks that varied across space (e.g., Emberling, 1997, p. 297). At the same time, lack of or lower amounts of specific ceramic types between neighbors should not be seen as a simple reflection of firm political boundaries (cf. Spencer et al., 2008, pp. 335-336). Even secondary centers in the Valley of Oaxaca, especially those at some remove from the urban capital, may have at times actively resisted incorporation into the Monte Alban state at the same time that they interacted with the urban capital and other neighboring communities closer to Monte Alban. Their relationships with the capital also may have been fluid (at times antagonistic), and they may not have been fully integrated into the Monte Alban polity until the end of Late I (Feinman & Nicholas, 2013, p. 64). As noted, even into the 20th century, raiding occurred between communities within the bounds of the Valley of Oaxaca (e.g., Dennis, 1987). Across highland Oaxaca, there were networks of interaction and influence that decreased with distance from first San José Mogote and then Monte Alban. At that point in the historical sequence (and through Monte Alban II), these connectivities mostly were directed south from the Valley of Oaxaca (as Bernal [1965] noted years ago based on stylistic criteria). More recently, studies of sourced obsidian illustrate widespread exchange and interaction across western Mesoamerica at this time (Feinman et al., 2015; Golitko & Feinman, 2015), and from the Central Valley of Oaxaca, the strongest ties to other regions to the south and east likely ran along the Pacific Coast. Such passages likely included the travel route between the Valley of Oaxaca and the Isthmus of Tehuantepec that passes through Albarradas. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 53 o+ ogi vane fae Gla ©“ rahe) p beg > WA Kildina! + eect TS ee i - i * ~ zoel & ofa be 15 a\e-A eae eee i ivge= .' oy 7; Ww i , we! sj aint ile anaes 9) { ‘A bow alel ieee viene , Hotere i y i hat ult a ’ hd Te ¥ 7 J > 7 iA) / TAO PRT \ ties 1m_ volume (m°) Terraces work** present 1 TL-SLA-SLA-56 AL N7E20 8 0 0 0 0 — shell 2 TL-MIT-CC-6 AL NI0EI8 8 0 0 0 0 — obsidian 3. TL-MIT-CC-7 AL NI0OEI18 8 0 0 0 0 ch — 4 TL-SLA-SLA-24 (Hierve el Agua) AL N7E19 49 0 ] 162 2 — obsidian? 5 TL-SLA-SLA-20 (Loma de las Tumbas) AL N8E20 353 I) 0 73 28 ch, lap — shell, obsidian 6 TL-SLA-SLA-22 (Sitio de las Tunas) AL N8E19 189 12 3 633 10 ch shell, obsidian 7 TL-MIT-CC-8 (El Nopal) AL NIOEI17 225 4 3 780 30 ch, cer nonlocal chert 8 TL-SLA-SLA-S3 (Llego Yiach1) AL N7E20 97 1] 10 4,352 >9 obsidian 9 TL-SLA-SLA-8 (El Calvario) AL N8E19 443 14 3 803 51 ch, gs shell, obsidian 10 TL-MIT-CC-5 AL NIOE18 595 24 0 265 62 ch, cer shell, obsidian 11 TL-SLA-SLA-41 (EI Trigo) AL N9E20 40 6 ] 55 ? ch shell, obsidian 12. TL-SLA-SLA-47 AL N9E18 13 0 0 0 0 — obsidian 13. TL-SLA-SLA-17 (Guirtn) BL N9E18 8 0 0 0 ] — — 14 TL-SLA-SLA-17 (Guirtn) ET N8E18 8 0 0 0 ] ad aa 15 TL-SLA-SLA-17 (Guirtn) ET N8E18 8 0 0 0 ] —— —_— 16 TL-SLA-SLA-17 (Guirtn) ET N8E18 8 0 0 0 ] a = 17 TL-SLA-SLA-17 (Guirtn) ET N8E18 8 0. 0 0 ] ed = 18 TL-SLA-SLA-17 (Guirin) ET N8E18 15 0 0 0 2 = aa * AL = Albarradas; ET = eastern Tlacolula. ** ch = chert; gs = ground stone; cer = ceramics; lap = lapidary. abundant at the three large terrace sites that are closest to the Camino Real (TL-SLA-SLA-8, TL-SLA-SLA-20, and TL- SLA-SLA-22) (Fig. 4.8, right) and the newly settled terrace site in the small mountain valley to the northwest (TL-MIT-CC-5) that quickly grew to become the largest settlement in the Albarradas area. These are the same four sites where most of the marine shell was present (Fig. 4.9). One of the sites (TL- SLA-SLA-20) accounted for almost half of the shell and was one of two sites with the most cremas. Although many of the crema vessels likely were locally made copies, some of the crema pottery may have been made in the Valley of Oaxaca (Fig. 4.10 Fic. 4.5. 60 View of terraced slope of TL-MIT-CC-5 in foreground of image. and Fig. 4.11, left). These distributional patterns likely are indicative of more direct networks between the elites of Albarradas and the center of the valley, which would be expected with a shift toward more exclusionary strategies by Monte Alban’s leaders. The large Albarradas terrace sites may have become more direct conduits to the capital for exotic goods moving along the travel route from coastal areas, with fewer effects on other settlement in the area. As in Late I, they also may have produced some goods for exchange. At more than half of the Albarradas sites (7 out of 12), we found evidence of chert working, and at four of those sites there was ® FIELDIANA: ANTHROPOLOGY Fic. 4.6. Site map of TL-MIT-CC-S. evidence of an additional craft (lapidary [TL-SLA-SLA-20], cream-colored surfaces and is usually painted in streaky red or ground stone [El] Calvario], or ceramics [TL-MIT-CC-5 and orange. These imitation cremas are most frequent at Monte TL-MIT-CC-8]) (Figs. 4.12 and 4.13). Alban IJ sites in eastern Tlacolula (Kowalewski et al., 1989, p. An unexpected finding was the dominance at many sites of a 165) and were noted by John Paddock in his excavations in Monte Alban IT ware described as Black and White (Kowa- Monte Alban II contexts at Caballito Blanco, Loma Larga, and lewski et al., 1989, p. 837); it has a darker core with white or Mitla, all in eastern Tlacolula (Fig. 4.14; Paddock, 1983b). We observed Black and White sherds on most of the sites in the a cu ae oe Fic. 4.8. Black and White rim with red paint (left) and Monte Fic. 4.7. Stone remains of house wall on Terrace 38 at TL-MIT- Alban II crema with red paint and postfire scratching (right) from CC-5. Loma de las Tumbas (TL-SLA-SLA-20). Pi a FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 6] TABLE 4.4. Monte Alban II ceramics at sites* in the Albarradas area. Ceramic description I]-1 I]-2 II-3 1-5 11-6 II-7 II-8 II-9 II-10 I-11 IJ-12 C-7 (T0001) bowl with red and orange paint — — — 2 3 — — D -- — — C-11 (10002) specular red paint, fine-line scratching — — — 4 ] — ~ ) — — — G-11 (T0004) specular red paint, fine-line scratching 1 oo — G-7 (T0057) bowl with red paint = — — l — ~- — — — — a Crema (T0379) comal = — = 1 Be as Crema (T610) bowl with hollow feet ae — — — — = ae ee _ oo G-1 (T111) shallow plate with applique — — — — — — == 1 _- _ — G-21 (T1194) bow] — — l — 2 3 _ 4 3 — — G-12 (T1207) bowl, 2 incised parallel lines on rim — — ] ] 3 4 — 19 8 -- 5 G-12 (T1227) bowl, 2 incised parallel lines on rim — — — 3 1 _ — 6 — — — G-12 (T1241) bowl, 1 line incised on rim — — — 2 ] — — —_ — — — G-12 (T1258) bowl, 3 incised parallel lines on rim — a ae l = is at Gris (T1507) bowl with cream slip — — — — _- — — — 1 1 — K-19 (T2011) comal — — 1 =. 3 — 3 -- — K-1/19 (12013A) jar with curved back neck — — — 10 _- 1 p — 1 a K-1/19 (12013B) jar with thin straight neck — — — 2) — — — K-1/19 (12013C) jar with thick straight neck — — — 3 — — aad = —= = = K-8 (T2052) bowl — ] — — ] oo K-8 (T2073) jar 1 = — K-8 (T2085) bowl — — l — a = K-19 (12417) sahumador l — — 3 2 — Dainzu (T3500) bowl 1 5 ] 6 = Black and White (T5100) various forms — , — 36 4 S2 3 61 115 ] > K-3 miniature bowl with red paint l — — _ K-3 thin cylinder with red paint — a -- — _ — 1 — _— — White paste bow] with thin-line incising — — — — — — —_ — 2 —_— —_ Café (imitation crema) with specular red paint a -S — — = 2 — — ] -= — * See Table 4.3 for full site name. Albarradas area (they are represented in 85% of ceramic collections that have Monte Alban II pottery). They are most abundant at two large hilltop terrace sites in the high mountain valley to the northwest (TL-MIT-CC-5 and TL-MIT-CC-8), where we suspect they were made (at both sites, almost 60% of the pottery in our collections are Black and Whites) (Figs. 4.15 and 4.16). This variety also is common (25%-—40%) at two large terrace sites on the eastern slope of Cerro Guirone, both near the trade route that passes through the Albarradas area (TL- SLA-SLA-8 and TL-SLA-SLA-20) (see Fig. 4.8, left, and Fig. 4.11, right). Similar local copies of red-painted cremas were common in the Mixe collections analyzed by Hutson (2014, pp. 269-270). It appears that few true cremas—made near Monte Alban and the Etla arm of the valley—reached even eastern Tlacolula and that residents of eastern Tlacolula and/or the Albarradas area made local copies. Based on these distribu- tions, the residents of eastern Tlacolula may have had as much, if not more, interaction with the Albarradas area than more central parts of the Valley of Oaxaca. The physiographic divide did not stifle interregional interaction. The Albarradas settlement pattern changed little between Monte Alban Late I and IJ, although there were slight shifts. As discussed in more detail in the previous chapter, the architec- tural complexes of the Late I terrace sites that continued to be occupied in Monte Alban II were centered on small plazas situated on the top of promontories that were surrounded by compact arrangements of terraces, whereas the Late I terrace site that was abandoned (TL-SLA-SLA-27) was the only terrace site with a large open plaza (see Fig. 3.31). The abandonment of this site could reflect a change in ritual practices from more open public spaces to more closed, restricted locales as political strategies shifted somewhat from more collective to more exclusionary (e.g., Feinman & Nicholas, 2016c). In addition, the three abandoned sites (two isolated residences and one terrace Fic. 4.9. Marine shell ornaments from Loma de las Tumbas (TL- SLA-SLA-20). 62 Fic. 4.10. MIT-CC-5. White paste sherds with fine-line incising from TL- FIELDIANA: ANTHROPOLOGY me cu ma cu Fic. 4.11. Café (imitation crema) rims with red paint (left) and Black and White rims with red paint (right) from El Calvario (TL- SLA-SLA-8). site) were all near the eastern, potentially more vulnerable edge of the Albarradas area. In some respects, the defensive location of TL-SLA-SLA-27 on the tip of a long narrow piedmont ridge that extends into a sharp bend in the Hormiga Colorada may have made that site especially vulnerable to sieges. In contrast, two of the new sites, including the largest Monte Alban II terrace site (TL-MIT-CC-5), were located in the northwestern part of the Albarradas area, closer to the Valley of Oaxaca. TL- MIT-CC-5 was less defensively located on two flat ridgetops in the small mountain valley just beyond the pass into the northeastern part of the Tlacolula arm. There were at least 62 residential terraces at this site (some were too destroyed to be mapped), with a population estimate of almost 600 people. Even though this was the largest site in the Albarradas area, there was no central architectural complex as at the other terrace sites. Instead, several low platforms were dispersed across the site. The other new settlements in the surveyed area during Monte Alban II are small scatters of pottery on six terraces at the Guirun site (TL-SLA-SLA-2/17) on the western side of the mountain ridge in eastern Tlacolula (see Fig. 2.18). The terraces with Monte Alban II pottery are widely dispersed across the steep ridge systems at the juncture of the 1980 and 1995 survey regions and were not detected until the intensive site survey in 1996 (Feinman & Nicholas, 2004a, p. 69). The best Monte Alban II ceramic diagnostics were red-painted A-9 bowls and one Black and White body sherd (Kowalewski et al., 1978, p. 192, 1989, p. 837). As in Monte Alban Late I, the Albarradas area continued to have a relatively flat site hierarchy in Monte Alban II. Five sites had populations of more than 180, with no one dominant center; most of the other sites were isolated residences or small hamlets of fewer than 50 people. There also was little change in settlement patterns in eastern Tlacolula between Monte Alban Late I and II (see Fig 4.14), where site hierarchies were less developed than elsewhere in the Valley of Oaxaca (Feinman & Nicholas, 2013, p. 63). In contrast, changes in settlement patterns were much more significant at the southern end of the Valley of Oaxaca between Late I and II. Site hierarchies in the Ejutla and Sola Valleys also had been relatively flat in Late I, Fic. 4.12. Lapidary items (left) and worked chert (right) from Loma de las Tumbas (TL-SLA-SLA-20). like in eastern Tlacolula, which we interpreted as evidence that those areas were not yet strongly integrated into the Monte Alban polity (Feinman & Nicholas, 2013, p. 76). By Monte Alban II, however, stronger site hierarchies with one dominant center had developed in the Ejutla and Sola Valleys (Balkansky, 2002, p. 45; Feinman & Nicholas, 2013, pp. 82-83); in Ejutla, there also was a clear tier of secondary centers. We think that Ejutla and Sola were beginning to be more politically incorporated into the Monte Alban polity in Monte Alban I, whereas eastern Tlacolula and the Albarradas area continued to be weakly integrated with the complex political system in the Valley of Oaxaca. The majority of the Albarradas area population continued to reside in terrace sites, forming a band outside the eastern limits of the Valley of Oaxaca, and collectively monitor all of the natural trails and topographic breaks that lead through the high mountains at the eastern edge of the Valley of Oaxaca. With the establishment of the new terrace site (TL-MIT-CC-5) in the northwestern part of the survey area, there was additional monitoring of the northeastern trade route into the Valley of Oaxaca. With a similar settlement pattern and undeveloped site hierarchy as in Monte Alban Late I, defensive/boundary matters continued to be significant concerns for the residents of the Albarradas area in Monte Alban II. Monte Alban II Terrace and Defensive Sites Although there were other significant changes in the Valley of Oaxaca between Monte Alban Late I and IJ, the number of terrace sites in the Valley of Oaxaca did not change significantly (Table 4.5; Fig. 4.17). Except for Monte Alban, these hilltop communities were not significant features on the landscape. Monte Alban was smaller than in Late I but continued to be, by far, the largest terrace site, with more than 14,000 inhabitants (Blanton, 1978, p. 44; Kowalewski et al., 1989, p. 184). Only in the Etla subvalley and adjacent area north of Monte Alban were there several terrace sites with populations between 1,000 and 2,000 people (Feinman & Nicholas, 2013, table 7.4). Elsewhere in the valley, these hilltop settlements generally had populations in the lower hundreds. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 63 Fic. 4.13. There was a change, however, in the distribution of terrace sites away from the broad Valle Grande south of Monte Alban to the Tlacolula arm of the valley. The densest cluster of terrace sites was in eastern Tlacolula and the Albarradas area (which together account for half of the Monte Alban II terrace sites in the combined Valley of Oaxaca/Albarradas area). Most of the Albarradas population (98%) continued to reside in hilltop terrace sites (eight sites), while the proportion residing in these hilltop settlements increased to almost 60% in eastern Tlacolula Es ee . Reet mee n A \\~¢ ya's 2 » in ri A . ia j SAP % a 4 A A ~< western Tlacolula *.*# a A a A Monte Alban II Worked chert from El Nopal (TL-MIT-CC-8). (six sites, counting the six dispersed terraces at the Guirun site as one settlement). Most of these sites are on the slopes, both east and west, of the high mountain ridge that defines the eastern edge of the Valley of Oaxaca. The hilltop centers are especially clustered on the eastern side and appear to collectively monitor travel routes into the eastern end of the Valley of Oaxaca. The eastern end of the Valley of Oaxaca still does not appear to have been as tightly integrated into the Valley of Oaxaca polity as the southern area (Ejutla) was by N 0 4 8 kk) kilometers Hormiga! 4 Colorada,” Sites smaller than 2.0 ha are marked by triangles. All others are drawn to shape and size. Contour interval: 500 m. Fic. 4.14. 64 Monte Alban II settlements in Tlacolula and the Albarradas area, showing principal places mentioned in the text. FIELDIANA: ANTHROPOLOGY me cu Black and White rims with red paint from TL-MIT- Fic. 4.15. CC-5. Monte Alban II and in fact may have been a contested frontier. The close proximity of the easternmost sites to each other indicates that safety in numbers continued to be an important factor in their location. At the same time, the predominance of Valley of Oaxaca ceramics in the Albarradas area indicates that there was close interaction across this boundary and that the Monte Alban polity had influence through elite networks but not control of the Albarradas area. What was happening at the same time in the southern end of the Valley of Oaxaca was considerably different. There were still no defensively located terrace sites in the Ejutla Valley. Although most of the Sola population (91%) continued to reside in hilltop terrace sites, there was little site continuity there, as all but two of seven Monte Alban Late I terrace sites were abandoned and three terrace sites were established in new locations (Balkansky, 2002). The two most isolated Late I terrace sites were abandoned, while the new ones were more clustered together. In contrast to the stability of terraces sites in the Albarradas area, this shift in terrace site location in the Sola Valley and the lack of a defensive posture in the Ejutla Valley indicate greater influence (or possibly some control) by the TABLE 4.5. Percentage of population residing in terrace sites in Valley of Oaxaca subareas in Monte Alban II. No. of Terrace % of terrace site Total population in Subarea sites population population terrace sites Etla 3 4,030 8,237 49 Central 4 16,000 18,694 86 N. Valle Grande 0 0 5,039 0 S. Valle Grande 4 394 2,072 19 W. Tlacolula 3 604 5,404 11 E. Tlacolula 6 1,443 2,475 58 Ejutla 0 0 2,184 0 Albarradas 8 1,99] 2,028 98 Sola 5 756 833 9] Fic. 4.16. Black and White rims with red paint from El Nopal (TL-MIT-CC-8). Monte Alban polity. The data for the other edge areas are not reported in a manner that allows us to examine changes in hilltop settlements during this transition. For example, we know only that there were more than 2,000 residential terraces at 548 sites in the Penoles area (Kowalewski, 1991), so ostensibly hilltop terraces sites were always a significant feature of the landscape. Settlement and Land Use in Monte Alban IT With lower populations and increasing crop yields, regional capacities based on maize farming continued to far exceed the population in the Valley of Oaxaca, and potential food supplies (based on available labor) easily surpassed the regional population. The cushion between actual populations and regional capacities was greater than in Late I except for Tlacolula and the Albarradas area (Table 4.6; Fig. 4.18), where the proportions remained the same. The Sola Valley followed the same pattern as neighboring Ejutla. Overall patterns in potential agricultural surplus did not change. Even though Monte Alban’s population dropped, so did the number of people living in the surrounding area. With lower total yields, the need for importing food did not change at Monte Alban, and the adjacent subareas of the valley could have made up the deficit (Fig. 4.19). Potential surpluses were still much lower in TABLE 4.6. Estimated archaeological, resource-based, and labor- based potential populations for Valley of Oaxaca subareas in Monte Alban II. Estimated Resource-based — Labor-based archaeological potential potential Subarea population population* population** Etla 8,237 66,400 20,200 Central 18,694 30,700 9,600 N. Vaile Grande 5,039 66,100 11,900 S. Valle Grande 2,072 57,400 6,500 W. Tlacolula 5,404 52,200 10,500 E. Tlacolula 2,475 23,200 4,600 Ejutla 2,184 55,500 4,000 Albarradas 2,028 9,500 2,800 Sola 833 20,500 2,800 * Based on full exploitation of all agricultural land resources in each subarea. ** Based on available labor supply and equal use of Class I and Class II land near recorded settlements. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 65 {4 Fic. 4.17. the east (eastern Tlacolula and especially Albarradas) and the south (the Ejutla and Sola Valleys). With sufficient crops closer to home, Monte Alban’s rulers still likely did not engage these areas for staple products. Any interest in these outlying areas more likely continued to focus on trade routes that passed through them. 66 4 Terrace sites O Defensive walls only N 0 10 a kilometers Monte Alban II terrace sites in the Central Valleys of Oaxaca. Monte Alban IT Summary Throughout the early history of the region, changes in the Albarradas area largely mirrored what was happening in eastern Tlacolula. That pattern continued in Monte Alban II. These two areas did not experience the significant settlement FIELDIANA: ANTHROPOLOGY Monte Alban II Archaeological population as % of subarea maximum potential population Fic. 4.18. Monte Alban II archaeological population as a percentage of the maximum potential population in each Valley of Oaxaca subarea. pattern changes and population declines that occurred else- where in the Central Valleys of Oaxaca. These differences may reflect that both eastern Tlacolula and Albarradas were not tightly incorporated into the Monte Alban polity in either Monte Alban Late I or II. In other words, the relationship between these two eastern subregions and Monte Alban remained relatively similar throughout this period. However, that was not the case for other subareas of the valley system, situated farther to the west. Governance at Monte Alban shifted from more internal to external funds of power and relatively collective to more exclusionary political formations. With that change, the settlement pattern and demographic distinctions between the areas close to Monte Alban in Late I that supplied the capital with resources and areas farther away that were less interdependent with the capital diminished to some degree in Monte Alban II. The break between Monte Alban’s immediate heartland and other areas under lesser degrees of hegemony 1s easier to define for Late I (e.g., Feinman & Nicholas, 2013, p. 64, figure 5.12) than for Monte Alban II (e.g., Feinman & Nicholas, 2013, p. 85, figure 6.12). With greater dependence on trade and external networks and perhaps competition between local elites and those at Monte Alban, populations in most subregions declined but became more nucleated, and distance from the capital was a less significant factor in determining patterns of demographic growth, settle- ment, and volume of monumental construction. Excluding Monte Alban, terrace sites were more significant features of the landscape in eastern Tlacolula and Albarradas than in other parts of the valley, probably for defensive concerns. These hilltop communities also experienced greater stability than did terrace sites in other edge areas. The eastern physiographic edge of the valley was not an important interactive boundary at this time, as there may have been more significant interaction between the Albarradas area and eastern Tlacolula than between eastern Tlacolula and the rest of the Valley of Oaxaca. For example, the Black and White ceramic variety was present in significant quantities only at sites in Monte Alban II 800 Surplus rounded to nearest 100 Fic. 4.19. Monte Alban II potential agricultural surplus for each Valley of Oaxaca subarea. eastern Tlacolula and the Albarradas area. This decorated ceramic variety was largely absent elsewhere in the Valley of Oaxaca. This contrasts with settlement and ceramic findings for the Ejutla and Sola Valleys, which appear to have been more fully incorporated into the Monte Alban polity. A hypothesis for why Monte Alban had different relation- ships with its southern and eastern boundaries may lie in the nature of interaction along the trade routes and the neighboring groups who controlled them. The principal trade route into the Valley of Oaxaca from the south passes through the Ejutla Valley (White & Barber, 2012). Although we know little about the areas east of the Ejutla Valley, there were no large or defensive Monte Alban II sites near the entry of this route into eastern Ejutla, which would have provided access to a range of coastal and lowland products. There also was an expansion of sites along or near the Atoyac and Miahuatlan Rivers in the southern part of Ejutla that may have participated in exchange with sites along the southern trade route through Miahuatlan. Instead, the first small defensively situated sites in Ejutla were established in the northern part of the area above the most direct route between the largest Ejutla center and Monte Alban (Feinman & Nicholas, 2013, p. 80). At this time, the population of Ejutla was much more concentrated in one large site than it had been in Monte Alban Late I. Monte Alban may have incorporated the Ejutla area into its realm at least to some degree, in part to exert influence/control over this trade route. In contrast, eastern Tlacolula and Albarradas are the only two subareas that did not lose population during the transition to Monte Alban II. Monte Alban’s sphere of influence always appears to have been stronger in the western part of Tlacolula, out to the Macuilxochitl-Dainzt area, whose principal geo- graphic feature is Cerro Danush, which is visible from Monte Alban. The rest of Tlacolula is not, and eastern Tlacolula and Albarradas are the two subareas least tied into the Monte Alban polity during the Terminal Formative. The population of those two subareas continued to reside in defensively situated hilltop settlements that monitored the principal trade routes into the eastern arm of the valley. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 67 roughout its history, the Valley of Oaxaca was a key intermediary nexus between eastern and western segments of Mesoamerica. As Ignacio Bernal (1965) perceptively noted years ago, its ties tended to extend east through Terminal Formative times. Yet it was during that period (Monte Alban Il) that the rulers and populace of the valley also shifted some of their connections and interests to the north, with apparent conquests in Cuicatlan (Spencer, 1982; Redmond, 1983; 68 Spencer & Redmond, 2006). Although this north-focused imperialism did not endure, the region’s linkages to the north dominated in the subsequent phase with the continued expansion of Teotihuacan in central Mexico. This more northward face for interaction basically continued throughout the rest of the pre-Hispanic era and even into more contemporary times. FIELDIANA: ANTHROPOLOGY Chapter 5: Early Classic (Monte Alban IITA) “[C]ities and towns must be viewed within their wider social and cultural context. The forms and functions of urban settlements are to a large extent determined by the nature of the wider society. To understand ancient cities we need to know the answers to questions such as: was this an empire or a city-state society? was the king considered a god? how was commerce organized? was warfare a problem? did farmers live out among their fields or clustered together in towns? Since urban settlements performed specific functions for their hinterlands, our view of ancient cities needs to take into account their specific geographical regions and settings.” (M. E. Smith, 2002, p. 18) The Early Classic period (ca. AD 200-500) was a time of sustained demographic growth, increasing community size, and greater economic integration between emergent centers and their immediate hinterlands across highland Mesoamerica (C. A. Smith, 2002; Kowalewski, 2004, pp. 92-93). Settlement patterns shifted as polities competed with each other, and highland populations became ever more concentrated in large centers. Nowhere was this more evident than at Teotihuacan in the Basin of Mexico, where the population grew to more than 100,000 people (Hirth, 1980, pp. 57-58; Blanton et al., 1993, p. 134; Cowgill, 2008, p. 969, 2015, p. 144). Teotihuacan was not only the center of political and economic power in the Basin of Mexico but likely held varying degrees of economic and political influence and control over nearby valleys in the Central Highlands (Hirth, 1980, pp. 98-99; Cowgill, 2000, p. 285; Carballo & Pluckhahn, 2007, pp. 621-622), especially adjacent to natural trade and communication corridors (Hirth, 1980, p. 98). TABLE 5.1. Subarea No. of sites* Etla 54 Etla without Suchilquitongo 53 Central 46 Central without Monte Alban 45 N. Valle Grande 154 N. Valle Grande without Jalieza 153 S. Valle Grande 222 S. Valle Grande without Mixtepec 22) W. Tlacolula 304 W. Tlacolula without Macuilxochitl 303 E. Tlacolula 150 E. Tlacolula without E] Palmillo 149 Ejutla 115 Ejutla without San Joaquin 114 Albarradas 11 Albarradas without Llego Yiachi 10 Sola 50 Sola without Los Chilillos 49 * Combined sites. More distant areas, like the Valley of Oaxaca, were not controlled directly by Teotihuacan but were influenced through trade with the urban center and its attempts to monitor key trade routes (Hirth, 1980, p. 99; Cowgill, 1997, p. 134). The nature of Monte Alban’s relationship with the rest of highland Mesoamerica was affected by Teotihuacan’s macroregional dominance, and the Oaxaca capital lost territory and influence beyond the Central Valleys of Oaxaca. In regions outside the valley that previously had Monte Alban II-style pottery, especially to the north, ceramic assemblages began to reflect Teotihuacan influence (Blanton et al., 1993, p. 88). In Cuicatlan, for example, there is no evidence of continued Monte Alban presence in the Early Classic period (Redmond, 1983; Spencer & Redmond, 1997, p. 216). Although the Monte Alban polity retained political autonomy from Teotihuacan (Blanton, 1978, p. 57; Joyce, 2010, p. 205; cf. Winter, 1998), its extraregional activities were curtailed and constrained (Cowgill, 2008, pp. 969-971). With their loss of influence in outlying areas as Teotihuacan expanded its power/influence, Monte Alban’s rulers appear to have refocused their political and economic attentions within the Central Valleys of Oaxaca. Populations increased signifi- cantly across the Valley of Oaxaca and neighboring regions in the Early Classic (Table 5.1). Settlements were larger (Fig. 5.1) and more evenly dispersed across the region than ever before (Fig. 5.2). But the greatest growth occurred in centers. Monte Alban regained most of the population it had lost during the Terminal Formative, and many new large centers were established across the region (Fig. 5.3). Some of these new secondary centers even rivaled Monte Alban in population size—Jalieza in the Valle Grande, Dainzu—Macuilxochitl/ Tlacochahuaya in western Tlacolula, and Cerro Jazmin (Yanhuitlan) in the western Nochixtlan Valley (see Fig. 2.1; Monte Alban IIIA sites in Valley of Oaxaca subareas. Average Average site Total population site size (ha) population 5,492 4.5 102 3,703 3.6 70 18,322 12.4 398 1,822 Pe) 4] 28,118 ea 183 15,284 5.0 100 23,995 3.4 108 20,776 3,2 94 29,171 3.2 96 24,024 2.7 79 16,623 Das) 111 10,466 1.9 70 14,656 8.7 Wy 7,946 3.9 70 1,127 4.3 103 245 12 DS 7,678 5.9 154 5,763 4.6 118 FIELDIANA: ANTHROPOLOGY N.:S., NO. 46, NOVEMBER 13, 2017, PP. 69-82 69 Monte Alban IIIA Average site population (exclude largest site in each subregion) Fic. 5.1. Average site population in the Valley of Oaxaca and neighboring areas during Monte Alban IIIA. Kowalewski et al., 1989, pp. 226-229; Blanton et al., 1993, p. 88; Kowalewski et al., 2009, p. 35). But Monte Alban continued to maintain its monumental and political dominance within the region as new and expanded construction of the city’s two largest structures, the North and South Platforms, greatly enhanced the monumentality of the Main Plaza (Blanton, 1978, p. 46, 1983, p. 128; Flannery, 1983b, p. 132). There was greater formality to the architecture (Blanton, 1989, p. 446), and a number of elite residences were constructed on the edges of the Main Plaza (Fargher, 2016, p. 324) and to its north, but there was minimal glorification or representation of highly individualized leaders (Feinman & Nicholas, 2013, p. 152). Monte Alban’s rulers appear to have shifted back to somewhat more collective rulership strategies as they lost some of their external sources of funding (Fargher, 2016, pp. 322-323). Some named, personalized depictions of rulers in stone are known for the Early Classic period (e.g., Joyce, 2010, p. 213), but they are few in number. Monte Alban likely focused more than in the prior period on the appropri- ation of agricultural surplus and labor from its surrounding hinterland to support governance that in return provided such public goods as ceremonial spaces, temples, defensive features, and roads (Feinman & Nicholas, 2017, p. 270); these public goods helped draw people to settle near the capital. Neverthe- less, the focus on internal resources may not have reached the extent that it had in Monte Alban Late I or during the first centuries of Monte Alban IIIB-IV, as the population in the central part of the Valley of Oaxaca did not recover entirely. Monte Alban’s governance practices, which during the Early Classic period more fully integrated the valley’s population and consolidated power, seem to correspond with what has been referred to as “pseudo decentralization” (Fesler, 1968, p. 373). In addition to Jalieza and Dainzu—Macuilxochitl, several other 70 Monte Alban IIIA Population densities per km? (exclude largest site in each subregion) Fic. 5.2. Population densities in the Valley of Oaxaca and neighboring areas during Monte Alban IIA. valley floor sites in the Valle Grande (Cuilapan, Tejas de Morelos, Zaachila, and Santa Inez Yatzeche) had large complexes of monumental architecture containing many large mounds and multiple plazas (Kowalewski et al., 1989, pp. 226— 231). Public architecture at these sites tended to conform to regional building orientations, and plazas tended to be open and accessible (Blanton, 1989, pp. 444446). For the first time in the history of the Valley of Oaxaca, carved stones in the same style as monuments at the capital were present outside Monte Alban (Blanton et al., 1993, pp. 89-90). In encouraging the buildup of these secondary centers, Monte Alban’s rulers may have been decentralizing many functions to these new large centers. But if architectural scale can be used as a proxy, most governing power and authority continued to be centralized in Monte Alban and not in the large secondary centers. These centers are located in the areas of the valley (northern and southern Valle Grande and western Tlacolula) where the greatest population increases occurred, where the settlement was densest, and where strong site hierarchies developed around them as many smaller centers were established or increased in size. These areas were part of the core of the Monte Alban polity, connected to the capital largely through civic-ceremonial networks that interconnected secondary centers and Monte Alban. Monte Alban’s relationship with outlying areas, notably eastern Tlacolula, remained different. In western Tlacolula, the Macuilxochitl area is marked by a distinctive mountain (Cerro Danush) visible from Monte Alban. Areas farther to the east in Tlacolula are not in the line of sight from the capital. The area around Cerro Danush stretching south to Dainzu and Tlacochahuaya is one of the more agriculturally productive parts of Tlacolula and was more continuously occupied than were areas to the east. During the regional survey, we found FIELDIANA: ANTHROPOLOGY ) ‘ 3 es a Sa nee fa: “Tejas de ; Morelos = FS “ ae Fic. 5.3. that the agricultural fields as well as the ridgetops that surrounded the modern communities of Dainzu, Macuilxochitl, and Tlacochahuaya were covered with sherds but not always of the same chronological period. Artifact densities (and the abundance of sites) dropped off dramatically when we moved to the east. The area between Cerro Danush, Cerro Dainzu, and a Dainzur” Macuilxochit heat es 5 \ ‘©: 1g. spp i - A ss : : yr a . st it 2 ee ye: ron il is ' oe. : ~ Guirtn é El Palmil io" aS Vv) Sites smaller than 2.0 ha are marked by triangles. All others are drawn to size 3 and shape. Contour interval: 500 m. kilometers Monte Alban IIIA settlements in the Central Valleys of Oaxaca, showing places mentioned in the text. the ridge above the modern town of Tlacochahuaya was clearly a dense zone of settlement, although how the area was occupied and built on shifted greatly during the pre-Hispanic era (Faulseit et al., 2016). Up through Monte Alban IIIA, population densities were always higher in western Tlacolula than in eastern Tlacolula, FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 71 during the Early Classic period an earlier settlement at \inzii-Macuilxochitl/Tlacochahuaya expanded by an order of magnitude to become the largest settlement in Tlacolula and one of the largest secondary centers in the entire valley. At this time, Monte Alban appears to have more fully consolidated its reach in western Tlacolula, which had always had a stronger relationship with the capital than did communities in eastern Tlacolula. Lambityeco/Yegtiih, a smaller secondary center approximately 8 km southeast of Cerro Danush, was less affected and may have even lost population. Yegiiih, the earlier part of the site, does not appear to have been as closely tied into the Monte Alban’s polity and may have at times had an adversarial relationship with the Monte Alban state even as its inhabitants interacted with the capital and other neighboring communities to the west. Communities beyond the polity’s core area, like Yegiith, may have had fluid (at times antagonistic) relationships with the capital and still were not fully integrated into the Monte Alban polity. Spatially, there are parallels with these relations that have been documented textually for the Late Postclassic period (Horcasitas & George, 1955; Oudijk, 2008). At that later time, the large center at Macuilxochitl had close ties with Cuilapan, situated in the valley center basically at the foot of Monte Alban. Yet Mitla, farther to the east, was a longtime adversary of Macuilxochitl. By the end of the Early Classic period (1.e., Middle Classic, AD 500-600), Monte Alban engaged in another effort to strengthen and expand its hegemony farther east in the Tlacolula Valley and south in the Ejutla Valley. In both areas, existing large centers were repositioned to be closer to Monte Alban, and in both cases the new centers had much more monumental architecture than earlier centers. The Early Classic center at Yegitth was moved approximately 1 km west to Lambityeco, where the new civic-ceremonial core was rebuilt with much larger and taller mounds, including a large ballcourt that closely resembled the principal ballcourt at Monte Alban in layout and size (Feinman et al., 2016, pp. 49, 52). The new location was closer to and had a better view of Cerro Danush, where monumental architecture was newly constructed on the mountain’s summit (Faulseit, 2013, p. 15). The layout and the construction materials used in the building of Lambityeco signal access both to knowledge and to goods that were likely not entirely local (Feinman & Nicholas, 2016c; Feinman et al., 2016). In the Ejutla Valley, the largest settlement with the most monumental architecture at the end of Monte Alban II was situated under the contemporary district head town (Ejutla de Crespo). Although that site grew as well during the Early Classic period, by the end of the Early Classic it had been completely eclipsed by a new center at San Joaquin, which was established on a series of low hills 5 km to the north. The new center consisted of three large complexes of monumental architecture and plazas and became the fourth-largest center in all of the Central Valleys of Oaxaca (Feinman & Nicholas, 2013, pp. 96, 106). The repositioning of the principal center in the Ejutla Valley to a previously underutilized area may signal the growing importance of interaction and exchange with Monte Alban, which increased its focus on Ejutla as a gateway to lowland products entering the valley system from the south. This gateway may have been with areas east of Ejutla (toward Tehuantepec) rather than to the south through Miahuatlan (see G2 Fig. 2.12), which was one of the few areas in the Southern Highlands that lost population during Monte Alban HIA (Markman, 1981, p. 67). Even though the population of Ejutla overall increased sevenfold, southern Ejutla also lost popula- tion. In the Sola Valley, a new large center, Los Chilillos, also dominated its hinterland (see Fig. 5.3; Balkansky, 2002, p. 51). Unlike San Joaquin, which was situated on a series of previously unoccupied low hills in the northern part of the Ejutla Valley, Los Chilillos in the Sola Valley (Balkansky, 2002, p. 57) and the Guirun site at the eastern edge of the Tlacolula arm grew from small, dispersed settlements in Monte Alban II into centers of several thousand people in Monte Alban IIIA. Both were defensibly located on top of high mountain ridges and had significant sectors with monumental public architec- ture. Population growth at the Guirtn site was coincident with depopulation in the Albarradas area beyond the eastern physiographic edge of the valley. According to Hutson (2014, p. 270), Early Classic ceramics in the Mixe area also show weaker affiliations with the Valley of Oaxaca. The only area within the Valley of Oaxaca to lose population was the northern, Etla arm, where no new large centers were established until the end of the Early Classic or during the transition to Monte Alban IIIB-IV. Etla had been one of the most densely settled areas in the valley for a millennium. The Sierra Norte at the northern edge of the Valley of Oaxaca was almost entirely abandoned (see Fig. 5.2); Robert Drennan (1989, p. 378), who surveyed the area, reports finding only one sherd on the surface that could be assigned with any certainty to Monte Alban IIIA. As populations in the Valley of Oaxaca shifted to the east and south and with the growing threat/ competition with Teotihuacan to the north, the Sierra Norte may have become an undesirable area in which to live. There was overall population growth in the Penoles area west of Monte Alban, and many new settlements were established in the eastern part of Penoles that abuts the Valley of Oaxaca (Finsten, 1996, figure 7.3), but population densities still remained much lower there than in the valley. These different patterns of settkement and demographic growth are a result of Monte Alban’s pulling away from the north and focusing more on the southern and eastern parts of its realm. Closer in, administration was carried out through a well-developed hierarchy of centers. In more outlying areas— eastern Tlacolula, Ejutla, and Sola—these activities were more heavily concentrated in one or two large centers. The largest site in Ejutla, San Joaquin, was located near an important southern trade route into the valley. These settlement changes may not have been uncontested, however, as defensively located hilltop terrace sites became much more significant features of the landscape during Monte Alban IIIA, circling eastern Tlacolula and the southern edge of the valley into Ejutla and Sola. In concert with much larger populations throughout the region, the volume of economic circulation and interconnection increased in Monte Alban IIIA (Kowalewski et al., 1989, p. 213). This was especially true in eastern Tlacolula, where increasing populations lived in drier areas where crops were not reliable and many sites specialized in crafts; the greater likelihood of food imbalances made it necessary for communi- ties to engage in exchange for food (Feinman & Nicholas, 2005, p. 246). Our subsequent excavations at two sites in eastern Tlacolula—E] Palmillo and the Mitla Fortress—have provided FIELDIANA: ANTHROPOLOGY TABLE 5.2. Percentage of sourced obsidian at Monte Alban IIA sites in the Valley of Oaxaca. Site No. El Chayal Ei Paraiso’ G. Victoria Orizaba Dainzi S2 — ao =e 9 Ejutla 5 _- a 20 —_ E] Palmillo 258 — — — | Jalieza 49 — — y 18 Macuilxochitl 6 _ a we leans Mitla Fortress 13 _ = a == Monte Alban = 256 4 4 3 4 Yegtiih 32 — = 13 16 evidence of interaction between communities and even between houses at the same communities that were producing different ranges of goods for exchange (Feinman & Nicholas, 2005, 2010, pp. 92-94, 2012, pp. 241-248). The economic options selected in eastern Tlacolula in the Early Classic could not have succeeded or endured outside the context of burgeoning networks of Otumba Pachuca Paredon Ucareo Zacualtipan Zaragoza oes 22 = a _ 69 = 80) oe =e: =. a 3 6 | 2 | 87 a 10 = 2 - 67 = = = = 17 83 = 8 = bes ie 92 8 63 3 | we 21 6 44 19 eo _ 3 circulation and market exchange (Feinman & Nicholas, 2010, 2017): Our recent sourcing studies of excavated collections of obsidian from sites across the valley are providing information on these networks of exchange. Earlier during the Late/ Terminal Formative, the Guadalupe Victoria source accounted Monte Alban IIIA Settlements in the Guirun/Albarradas Survey Region E17 E18 N10 Sites smaller than 1.0 ha are marked by triangles. All others are drawn to size and shape. Contour interval: 500 m. E19 E20 kilometers Fic. 5.4. Monte Alban IIIA settlements in the Guirtin/Albarradas survey region, showing sites mentioned in the text. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 73 TasLe 5.3. Monte Alban IIIA sites in the Guirtn/Albarradas survey region. Grid Mounds Mounds Mound Craft Exotics No. Site name Subarea* square Population <1 m >1m volume (m*) Terraces work** present 1 TL-SLA-RO-2 AL N7E19 8 0 0 0 0 aan aa 2 TL-SLA-SLA-19 ET N8E18 128 16 0 0 1 — obsidian 3. TL-SLA-SLA-25 AL N7E19 18 0 0 0 0 — — 4 TL-SLA-SLA-54 AL N7E20 8 0 0 0 0 — — 5 TL-SLA-SLA-20 (Loma de las Tumbas) AL N8E20 8 0 0 0 1 — obsidian? 6 TL-SLA-SLA-22 (Sitio de las Tunas) AL N8E19 23 yD 0 0 D — obsidian 7 TL-MIT-CC-8 (El Nopal) AL N10E17 32 4 3 780 ? ch — 8 TL-SLA-SLA-53 (Llego Yiachi) AL N7E20 882 13 10 4,352 3] — obsidian 9 TL-SLA-SLA-S3 AL N7E20 92 7 0 10 0 — — 10 TL-SLA-SLA-8 (El Calvario) AL N8E19 8 6 0 66 ] ch obsidian 11 TL-MIT-CC-5 AL NIOE18 8 0 0 0 1 — obsidian 12. TL-SLA-SLA-41 (EI Trigo) AL N9E20 40 6 ] 55 ? ch obsidian 13. TL-SLA-SLA-17 (Guirtn) ET N8E18 2,475 52 DS 21,108 280 ch, gs obsidian * AL = Albarradas; ET = eastern Tlacolula. ** ch = chert; gs = ground stone. for almost 50% of the obsidian in the Valley of Oaxaca, but during the Early Classic that source was largely replaced by obsidian from nearby Zaragoza, which formed more than 50% of the assemblage (Table 5.2). Obsidian was exchanged through several different routes during the Classic period (see Fig. 2.12,). Pachuca (close to Teotihuacan) obsidian remained more prevalent at Monte Alban than in most other sites in the region and likely arrived through a northern route through Etla (Feinman & Nicholas, 2013, p. 115; Feinman et al., 2015, figure 12). The Zaragoza source in Puebla was the most prominent source for obsidian at most other sites in the valley, especially at sites in Tlacolula. Although our sourced sample of Early Classic obsidian in the Mixe region is small, it is almost exclusively from Zaragoza. Cantona, a smaller but still important Classic period center in the Central Highlands, heavily exploited the nearby Zaragoza source (see Fig. 2.1; Garcia Cook & Merino Carrion, 1998, p. 213). The easiest route for Zaragoza obsidian to reach the Valley of Oaxaca would have been along the Gulf Coast and up through the Mixe area and into eastern Tlacolula. The proportion of Zaragoza obsidian in Valley of Oaxaca Classic period assemblages generally is lower in sites in western parts of the valley, especially at Monte Alban and the Ejutla site (Feinman & Nicholas, 2013, figure 7.29; Feinman et al., 2015, figure 12). Given this spatial distribution, it would appear that the main entry for the Zaragoza obsidian was the eastern travel route that passes just south of the Guirun site. Social network analysis of obsidian distributions also has been employed, and based on these studies, we see Early Classic period links between eastern Tlacolula and the southern Isthmus (Golitko & Feinman, 2015, figures 7 and 8; Feinman cw G-23s from El Calvario (TL-SLA-SLA-8). Fic. 5.5. 74 et al., 2015, figures 11-13). These connections illustrate that the Central Valleys of Oaxaca were still connected economically with the Pacific Coast (through sites such as Los Horcones in Chiapas), a key node between eastern and western Mesoamerica at that time (see Fig. 2.1). Although the Albarradas area may have been too dangerous for much permanent settlement at this time, exchange still moved through the region, and the Guirun site had a key vantage to track it. Monte Alban IIIA Settlements in Albarradas Settlement patterns shifted dramatically in the Albarradas area with the transition to Monte Alban IIJA. Although the total number of settlements only dropped from 12 to 11, the area lost almost half of its population as most of the Monte Alban II sites were either abandoned entirely or lost much of their earlier population. The eastern slopes of Cerro Guirone and the high mountain valley in the northwestern part of the surveyed area were largely depopulated (Fig. 5.4). When Classic period materials were noted on some of the Late/Terminal Formative terrace sites, the occupation was restricted to a small sector of the earlier site, usually only at the summit of these settlements. Three-quarters of these as well as new settlements were isolated residences or small hamlets of fewer than 50 people (Table 5.3). In contrast to eastern Tlacolula (and Albarradas during the Terminal Formative), craft production in Early Classic Albarradas was minimal, evidenced only at three multicomponent sites that were larger in other periods. Most of the Albarradas population from the Monte Alban II sites that were abandoned easily could have moved to one of two sites that became much larger during Monte Alban HIA. Both are situated in highly defendable localities: Llego Yiachi in Albarradas and Guirtn in eastern Tlacolula. Although we know next to nothing archaeologically about the Mixe region to the east, Zapotecs and Mixes reportedly had hostile relations (Schmieder, 1930, p. 60) at the time of the Spanish Conquest. This depopulation and relocation to very defendable locations may be evidence for the deep history of this animosity. The ceramics in the Albarradas area continued in the Valley of Oaxaca tradition, defined largely by the ubiquitous Classic period G-35 conical bowls. Several varieties that are more FIELDIANA: ANTHROPOLOGY TABLE 5.4. Monte Alban IIIA ceramics at sites* in the Albarradas area. G-12 (T1241) bowl, | line incised on rim — G-12 (T1258) bowl, 3 incised parallel lines on rim a G-35 (T1263) bowl with tripod hollow supports an ao G-23 (T1264) outleaned wall bow] with exterior carving — — G-23 (T1265) hemispherical bow] with exterior carving — 1 G-35 (T1274) bowl with nubbin support oe — Gris (T1507) bowl with cream slip — — Café (T1508) large thick storage jar ] — K-1/19 (T2013A) jar with curved back neck — 1 A-8 (T3410) bowl with exterior carving — - A-3 (T3411) thin orange bowl — a= Dainzua (T3500) bowl -- — Ceramic description IWA-2 TITA-3 G-35 (T1126) conical bowl 10 1 G-21 (T1194) bowl — S G-12 (T1207) bowl, 2 incised parallel lines on rim — 1 1 * See Table 5.3 for full site name. ** Ceramics collected during Guirun site survey in 1996. diagnostic of the Early Classic—G-23, A-8, A-3 (thin orange), and Dainzt bowls—are present in low numbers on most of the sites (Fig. 5.5; Table 5.4). We did not observe a single piece of pottery that appeared exotic, so even as this area was being abandoned, it remained in the Valley Zapotec cultural tradition. Classic period ceramics at the Guirutn site also include the ubiquitous G-35 bowls and low quantities of G-23s and A-3s (thin orange). The population increased 10-fold at Llego Yiachi (TL-SLA- SLA-53), an outpost hilltop terrace site on the eastern side of the Hormiga Colorada at the edge of the surveyed area (Fig. 5.6). The four small plaza-mound groups on the top of the rocky promontory continued in use as new settlement spread down the eastern slopes, where there were a series of terraces on WHA-5 TA-7 WIA-8 TTA-10) WA-11— WA-12— Guiran** be | 3 10 = 1 177 = = _ 4 1 = = ~~ 4 = 19 6 fu = = = is = 4 gas sexs ae 1 = ec ca = = 2 a = se 5 | = a 3 oe se 5 So = = 1 a — as = = a = _ es 4 at = it — | 1 | =e | Ss ~ = = 2 2 | | 2 be | 3 = = _ = | _ 4 = _ = 5 1 6 = the easiest approach to the top of the rock feature and a tall defensive wall near the base that protected this slope (Figs. 5.7 and 5.8). The steep slopes of Llego Yiachi and the surrounding rough terrain provide a highly defendable location that could have monitored travel along the trade route to the east. The site looks down on the recently improved road (and old foot trail) to Santa Ana del Rio along a tributary of the Tehuantepec River in the Mixe region. The residents of Llego Yiachi clearly chose this location for defense (see Fig. 2.26). We did not find any evidence of craft production at the site. Even during Monte Alban II, the residents of Llego Yiachi may have been less under the influence of Monte Alban, as we found only a few of the Black and White pottery sherds that were abundant at other Monte Alban II sites in the Albarradas area. The specific Toy 9 fp (o50 0 0 00¢e eee te OF) FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO aD LI-10 II-8 INA-8 IIB-IV-14 V-58 Fic. 5.7. Plan of architecture on the top of Llego Yiachi (TL- SLA-SLA-53). relations between Llego Yiachi and Monte Alban remain unknown. It may have been a redoubt for those in opposition to or seeking distance and defense from Monte Alban, a fortified outpost that monitored a key trade corridor to the Valley of Oaxaca, or both. An even larger proportion of the Albarradas population appears to have moved to the western side of the mountains, to the Guirtn site in eastern Tlacolula (Fig. 5.9). This hilltop terrace site grew from fewer than 100 inhabitants in Monte Alban II to one of the largest sites (almost 2,500 inhabitants) in eastern Tlacolula by the end of Monte Alban IIIA (Feinman & Nicholas, 2004a, pp. 69-70). The new center was located on a series of high ridgetops and the steep western slopes of Cerro Guirone (Fig 5.10). Complexes of platforms and structures were constructed on every flat ridgetop and saddle, with groups of terraces on the steep slopes below (Figs. 5.11 and 5.12). A four- mound group and adjacent small ballcourt were constructed on the top of a lower slope (Fig. 5.13, in site sector ITA; Feinman & Nicholas, 2004a, appendix 2.c). The Guirtn site is situated on ridges above the major trade route that enters the eastern end of the valley from San Lorenzo Albarradas and would have had a clear view of traffic along the trail. This trail continued to be heavily used until the paved highway was constructed north of the Cerro Guirone ridge system in the 1960s. We have not excavated at the Guirtn site, which was a large settlement from Monte Alban IIIA through the Late Postclas- sic, so most of the obsidian from the site that we have sourced cannot be definitely assigned to a period. But in the sector of the site with the Classic period ballcourt (sector IIA) and where we observed much higher quantities of Classic period pottery on the surface than Monte Alban V pottery, the sourced obsidian is almost entirely from Zaragoza (Table 5.5). On another lower slope (sector IB), where there are standing adobe walls that date 76 Fic. 5.8. Stone defensive wall on the eastern side of Llego Yiachi (TL-SLA-SLA-S3). to the Postclassic period and where the quantity of Monte Alban V pottery was twice that of Classic pottery, there was collectively more obsidian from Orizaba and Pachuca than from Zaragoza. In general, the two former sources are much more abundant in excavated Postclassic contexts in the Valley of Oaxaca. Thus residents of the Guirtn site may have played a role in the importation of Zaragoza obsidian into the valley in Monte Alban IIIA. The population in eastern Tlacolula also grew significantly (from approximately 2,500 to more than 16,000), so it is unlikely that all—or even most of—the growth at Guirun came from within the valley. The populations at other sites in eastern Tlacolula, especially El Palmillo and the Mitla Fortress, also grew tremendously during Monte Alban HIA (Fig. 5.14). As Monte Alban’s rulers refocused their interests inside the valley, they may have made an attempt to attract people from surrounding mountainous areas, including the Albarradas area. The Early Classic period growth at Macuilxochitl may have encouraged the occupants of eastern Tlacolula to produce for that expanding center and the economic networks into which it was linked. Goods, such as maguey fiber, pulque, chert preforms, and other xerophytic plant products, may have been collected and produced for exchange with the more western, agricultural sectors of the valley, where these resources are less abundant. These opportunities may have fostered rapid FIELDIANA: ANTHROPOLOGY vs Fic. 5.9. Terraced hilltops and western demographic growth in the eastern sector of the valley (Feinman & Nicholas, 2005). Monte Alban IIIA Terrace and Defensive Sites Hilltop terrace sites were well established as a form of settlement by Monte Alban IIIA, but they became much more imposing and numerous in the Valley of Oaxaca at that time (Feinman & Nicholas, 2013, p. 101). In general, Classic period terraces sites tended to be much larger than earlier terrace sites, and most of the largest population centers during the Early Classic period were hilltop, terraced localities. A majority of the Monte Alban HIA population in the valley resided in these compact, defendable communities, and a band of these sites ringed the eastern and southern edges of the valley (Fig. 5.15). The establishment of these communities occurred in areas that saw the greatest population growth in Monte Alban HIA. More than 70% of the population in Tlacolula and almost 90% in the Sola Valley resided in hilltop TABLE 5.5. Proportion (%) of ceramics by period and obsidian varieties at Guirun by site section. Site Classic Postclassic section ceramics ceramics Zaragoza Orizaba Pachuca IA 70 30 - ~ ~ IB 34 66 46 37 17 HA 88 12 92 4 4 IIB 73 DY 5 0 5 I Dy 43 83 17 0 IV 27 73 43 57 0 terrace sites (Table 5.6). In contrast to the Late and Terminal Formative, there was only one terrace site of any size in Albarradas, yet that one site, Llego Yiachi, housed almost 80% of the total population in the area. The site is in such a defendable location and has defensive walls, so defense was clearly an important concern. Located on hilltops far above the best farmland in the valley, terrace sites were dense (if not always large) concentrations of people that served as loci of nonagricultural production, especially in the Tlacolula subvalley (Kowalewski et al., 1989, pp. 242-243). The largest terrace sites also tended to have precincts of public monumental architecture. By the very nature of their locations, most terrace sites had some concern with defense. In addition to their defendable locations, many of the hilltop sites had high stone retaining walls that provided additional protection. Some hilltop sites appear to have been military garrisons. Yet the residents of these sites clearly TABLE 5.6. Percentage of population residing in terrace sites in Valley of Oaxaca subareas in Monte Alban IIA. No. of Terrace % of terrace site Total population in Subarea sites population population terrace sites Etla 2 2,504 5,492 46 Central I 16,507 18,322 90 N. Valle Grande 4 13,823 28,118 49 S. Valle Grande 19 14,718 23995 61 W. Tlacolula 19 20,968 29,171 2, E. Tlacolula 8 133750 16,623 83 Ejutla 9 2,199 14,656 is Albarradas 3 954 1,127 85 Sola 21 6,773 7,678 88 FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 77 Guirun “1 @® House or residential terrace = [= Mound or platform N C) Plaza — Wall 0 250 =a X Cruciform tomb Contour interval 20 M ae ; Fic. 5.11. Survey crew on Terrace 79 at Guirun (TL-SLA-SLA Fic. 5.12. Stone foundatio 17). SLA-SLA-17). ¥ ret e § n wall on Platform 48 at Guirtn (TL- 78 FIELDIANA: ANTHROPOLOGY ie, ty Fic. 5.13. Ridgetop with ballcourt and plaza with four-mound group at Guirin (TL-SLA-SLA-17). ! | ' 1% L > BS ~ -oNee p aa) 0 4 8 : , Eee “. Fan : ' ‘ kilometers Mh ge Ak a Sameee a woul "el nag ed yee is sees, a a a si fs aa “4 - 4 Ba I a4 Ps oe. eee tt 4), \ a ‘ SA as 4 é =, ae e ae th Oe, Cie Ae a a i 4 1eastern Tlacolula 4 7 oe . ees 2009 —— oe ee ee oe ! z | El Palmillo ; aA 4 ae a 4 Hormigaf < o => tree / Chichicapan Valley San Dionisio Valley i buns Sites smaller than 2.0 ha are marked by triangles. All others o are drawn to shape and size. Contour interval: 500 m. Fic. 5.14. Monte Alban IIIA settlements in Tlacolula and the Albarradas area, showing principal places mentioned in the text. 2 FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 79 Monte Alban’. 2000 } Fic. 5.15. interacted with their neighbors. Residents of many of them likely exchanged specialized goods for agricultural produce that they could not grow themselves. Architectural canons and artifact assemblages at all of these hilltop sites, even the most outlying, are fully within the Valley of Oaxaca tradition, so clearly their greatest interactions were with other settlements in the valley than beyond. 80 eee seem 4 Terrace sites O Defensive walls only N 0 10 a kilometers Monte Alban IJA terraces sites in the Central Valleys of Oaxaca. The new large hilltop settlement at Guirtin was one of many terrace sites in the Tlacolula arm of the valley. It grew to become the second-largest site in eastern Tlacolula. The amount of public architecture at the site and its location right above the principal eastern travel route into the valley from the east indicates that this site had some links to Monte Alban and the center of the valley. The movement of people from the eastern FIELDIANA: ANTHROPOLOGY Archaeological population as % of subarea maximum potential population Fic. 5.16. Monte Alban IIIA archaeological population as a percentage of the maximum potential population in each Valley of Oaxaca subarea. slopes of Cerro Guirone to a location on the western slopes, closer to the center of the valley, mirrors the establishment of the large terrace site of Los Chilillos in the northeastern part of the Sola Valley (Balkansky, 2002, pp. 52-54). Settlement and Land Use in Monte Alban IIIA Using maize as a proxy, potential crop yields continued to increase in highland Oaxaca but not as fast as the population did. The estimated population, based on the archaeological evidence, continued to be below regional capacities based on maize farming, but the cushion was much less than ever before (Table 5.7; Fig. 5.16), especially in the dry Tlacolula arm of the valley, where several large new settlements were established in TABLE 5.7. Estimated archaeological, resource-based, and labor- based potential populations for Valley of Oaxaca subareas in Monte Alban IIA. Estimated Resource-based Labor-based archaeological potential potential Subarea population population* population** Etla 5,492 80,200 17,700 Central 18,322 37,100 6,900 N. Valle Grande 28,118 80,200 44,800 S. Valle Grande 23,995 70,200 36,400 W. Tlacolula 29,171 63,600 39,100 E. Tlacolula 16,623 28,400 10,200 Ejutla 14,656 67,800 19,000 Albarradas e227 11,800 1,200 Sola 7,678 24,700 9,900 * Based on full exploitation of all agricultural land resources in each subarea. ** Based on available labor supply and equal use of Class I and Class II land near recorded settlements. Monte Alban IIIA 100 6,400) Surplus rounded to nearest 100 Fic. 5.17. Monte Alban IIIA potential agricultural surplus for each Valley of Oaxaca subarea. spite of limited agricultural potential. Factoring in labor supply, farmers would have been able to produce significant surpluses in many parts of the valley, especially in the populous subareas of the valley closest to Monte Alban (Fig. 5.17). Farmers in Etla and the Valle Grande could grow more produce than they needed, supplying some of that surplus to Monte Alban as the capital continued to outstrip the surplus crops of farmers in the Central area. As populations in these areas allocated more labor to farming to provide agricultural surplus for Monte Alban, they would have increased their dependence on markets/other exchange mechanisms to obtain household goods and handi- crafts that they could not make themselves (Blanton & Fargher, 2010, p. 220). Potential surplus production was much lower in the southernmost part of the region (Ejutla and Sola), where it undoubtedly would have been much less practical to transport agricultural surplus to the capital. More signif- icantly, most of the Monte Alban HIA population in eastern Tlacolula resided in hilltop sites far from what little decent farmland there is in the area. This is the driest part of the valley, and farmers there could have grown enough food for about two-thirds of the population. Agricultural produce from other parts of the valley would have been necessary to feed all of the inhabitants of the large terrace sites, who produced a range of household and other craft goods to exchange for foodstuffs (Feinman & Nicholas, 2005, 2010, pp. 92-94). In exchange for these goods, the farmers in western Tlacolula could have grown more than enough agricultural produce to provision themselves and all the large, hilltop terraces sites in both western and eastern Tlacolula. This interdependence—and the exchange and market networks that likely sustained it—underpinned the rapid demographic growth that occurred in eastern Tlaco- lula. The population was so low in the Albarradas area, concentrated mostly in Llego Yiachi in a highly inhospitable area, that little agricultural surplus would have been available FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 8] \s life beyond the physiographic edge of the valley may have become more dangerous in the Early Classic, defensive concerns and not access to good farmland was the principal factor in the changing settlement patterns in Albarradas. At a more local scale, there was a large increase in Monte Alban IIIA in the number of people living in large sites in areas of the Valley of Oaxaca where the production of large agricultural surpluses was not possible, especially not in dry years (Nicholas, 1989, pp. 482-487). The presence of integrated exchange networks, some centered on markets, would have been necessary before such large concentrations of people moved to hilltop settlements where they could not have reliably provisioned themselves. A prerequisite was interdependence between subareas for basic subsistence items. At the same time, defense appears to have been a major concern in the decision to move to hilltops and leave more productive farmland vacant. This move is coincident with the greater levels of specialized production that we see in Monte Alban HIA than ever before (Kowalewski et al., 1989, p. 213; Feinman & Nicholas, 2013, pp. 120-121). Based on distributions of good farmland across the region, one would not predict that Tlacolula, especially eastern Tlacolula, would have population densities as high or higher than other valley areas. Increasing commercialization, both within and beyond the valley, including exploitation of local stone resources and xerophytic plants in craft activities for exchange, may have been a key factor in demographic growth in the easternmost part of the valley, while the more agriculturally productive Valle Grande and Etla arms of the valley continued to be the breadbaskets of the region. Monte Alban IITA Summary For the first time, the settlement trajectory of the Albarradas area did not match what was happening in eastern Tlacolula. The Valley of Oaxaca as a whole, including eastern Tlacolula, experienced rapid population growth. This growth extended south into the Ejutla and Sola Valleys but not north into the Sierra Norte or east into the Albarradas area. As Teotihuacan expanded its influence and Monte Alban withdrew from northern areas to focus more on the valley and exchange routes from the south, areas like the Sierra Norte may have become more dangerous, less desirable places to live. The Albarradas area lost almost half of its population, and what remained resided in defendable locations close to the principal trade route through the area. The move of so many people from the Albarradas area to the Guirun site on the western slopes of Cerro Guirone likely reflects the same processes that brought higher populations into eastern Tlacolula, where they also settled in defendable hilltop locations. Monte Alban’s rulers made an effort to draw outlying populations into the valley as the capital lost access to the extractive resources gained previously from areas beyond the physiographic limits of the valley. At the same time, edge areas like Albarradas became too dangerous for all but highly defendable communities (like Llego Yiachi). The powerful at Monte Alban shifted their focus toward greater reliance on internal resources, but the means of generating their funds to support power was based less on the attraction of labor to regions close to the capital coupled with agrarian intensification and, instead, relied on strengthening relations with secondary centers, such as Jalieza and Dainzu—Macuilxochitl, and expansion at these sites as well as other terraced communities. These centers became nodes in an expansive exchange-market system that fostered demographic growth, agrarian production (especially in the Valle Grande and western Tlacolula), and intensified craft production across much of the valley. The fund of power for the governance of Monte Alban likely came from taxes on labor, agricultural produce, and exchange markets. People in outlying parts of the valley system, like Albarradas, may have been influenced/encouraged to move closer as the outlying areas became more dangerous. With larger popula- tions moving into the valley and more, larger centers, competition between local secondary elites may have intensified. The defensive posture of many of the large hilltop settlements may have been as much for guarding themselves from their neighbors, with whom they exchanged as much as they competed. They also may have tried to resist high degrees of political control by Monte Alban’s leaders. The closer to Monte Alban secondary elite were, the more control and influence likely was ceded to the capital. Beginning during the Terminal Formative and continuing into Monte Alban IIIA, the Tlacolula arm of the valley began to equal and often surpass the population densities in other, more agriculturally productive subareas of the valley. Through- out this period, densities were always higher in western Tlacolula, which was more closely integrated with the Monte Alban polity, than in eastern Tlacolula. Tlacolula continued its demographic dominance after Monte Alban began to decline at the end of the Classic period and throughout the Postclassic period. But after the Early Classic, the demographic histories of eastern and western Tlacolula began to diverge, with eastern Tlacolula, the driest subarea in the valley, overtaking western Tlacolula as the most densely settled part of the valley. This ascendance would not have been possible without well- integrated exchange/market networks. That some of these networks connected to areas outside the valley, following the important trade routes through the Albarradas area to the Pacific and Gulf Coasts, was likely a key factor in the region’s later history. Although Monte Alban eventually fell, later in the Classic period, many of the networks and economic patterns that it seems to have encouraged endured beyond the apogee of the capital. FIELDIANA: ANTHROPOLOGY Chapter 6: Late Classic (Monte Alban IIIB-IV) “[T]here is never a shortage of strong men who enjoy getting a fortune from tax receipts.” (Olson, 1993, p. 573) The Late Classic/Epiclassic (ca. AD 500—-850/900) was a period of regional reorganization across highland Mesoamer- ica. The large expansive states under the hegemony of the great Classic period cities began to break up into smaller polities as the urban centers reached their apogees and then began to decline in size, influence, and control. Both Teotihuacan and Monte Alban retained sizable populations, but they were no longer the dominant political and demographic centers in their respective regions by the end of this period (Blanton et al., 1993, pp. 91-99; C. A. Smith, 2002, pp. 165-168). In Oaxaca, the smaller polities that subsequently formed generally were centered at prior secondary centers where local ruling elites were able to take advantage of the resulting power vacuum to promote themselves into new positions of authority and autonomy. In the Valley of Oaxaca, this change is visible in a new kind of carved stone monument—genealogical registers—that began to be erected at Monte Alban and more than a dozen other sites in the valley toward the end of the Late Classic period (Marcus, 1983b, 1992b, 2006; Urcid et al., 1994; Urcid, 2003; Feinman, 2007). Emerging rulers began to employ more exclusionary leadership strategies and erect stones or other illustrative means that typically show marital pairs and trace the descent of powerful individuals from their forebears in an attempt to legitimize their right to rule. These representations, in contrast to earlier stone monuments, generally were positioned in highly private settings where only a narrow societal segment could view them. Long-standing problems and debates concerning the ceramic chronology for the Classic and Early Postclassic periods in the Valley of Oaxaca have hampered our understanding of Monte Alban’s decline and the contemporaneity of population and settlement shifts across the region. The archaeological surveys of the Valley of Oaxaca (Blanton et al., 1982; Kowalewski et al., 1989) produced the first valleywide picture of settlement patterns for this era and recorded important changes over time. But the spatial distributions for Monte Alban IIIB and IV seemed a bit improbable to many. The near lack of spatial overlap in settlement patterns in the Valley of Oaxaca for these two periods has raised questions for decades (Winter, 1989; Lind, 1991-1992, 1994), with some researchers suggesting that the two phases should be combined into one long period (Lind, 1994; Markens, 2004). A complicating matter is that during surveys in more outlying areas, such as the Ejutla Valley, the Sola Valley, the Albarradas area, Penoles, and the Sierra Norte, there often were few diagnostics beyond the ubiquitous Classic period G-35s to distinguish between Monte Alban IIIB and IV. In our surveys of the Ejutla Valley and the Albarradas area, we found very little pottery that we could definitely attribute to Monte Alban IIIB in comparison to a somewhat greater abundance of Monte Alban IV diagnostics. Although we could not confirm the presence of Monte Alban IIIB at some sites, we also could not rule it out. The lack of clear Monte Alban IIIB diagnostics matched but was even starker than patterns previously observed in southern Valle Grande and eastern Tlacolula, the subregions adjacent to Ejutla and Albarradas. There likely is some temporal overlap between the ceramics traditionally referred to as Monte Alban IIIB and IV. Yet there also are diagnostics of Monte Alban IV that clearly extend centuries later in time (Brockington, 1973; Bernal & Gamio, 1974; Winter, 1997; Feinman & Nicholas, 2013, pp. 141-156), while archetypical Classic forms, notably the iconic ceramic funerary urns, generally do not. A key diagnostic that we used on survey to define Monte Alban IV was imitation Balancan (Z) Fine Orange (Kowalewski et al., 1978, pp. 191-192), which elsewhere has been dated to contexts that postdate the time frame for Monte Alban HIB (Marcus & Flannery, 1990, p. 194). Imitation Fine Orange was most abundant in Tlacolula and has been found in the Ejutla Valley (Feinman & Nicholas, 2013, figure 8.10) and in Miahuatlan (Brockington, 1973; Markman, 1981, pp. 101, 164). Imitation Fine Orange has been recovered in a limited area at Monte Alban (Blanton, 1978, figure 4.37) that does not spatially correspond with the site’s greatest expansion in Monte Alban IIIB. Imitation Fine Orange instead was more abundant in those parts of the valley system that were the first to break away from Monte Alban and that were most accessible to the Gulf Coast, where Fine Orange pottery is abundant and was made (Arnold et al., 1993). Differences between the settlement patterns for Monte Alban IIIB and IV also support the inference that areas more distant from Monte Alban likely were the first to break away from the capital’s hegemony (Kowalewski et al., 1989, maps S—7). Areas in the south and east tended to have more Monte Alban IIA and IV ceramic diagnostics and fewer IIIB ceramic markers. In subsequent excavations at Ejutla in the south and El Palmillo in the east, we also found some ceramics that typically would be classified as Monte Alban HIA in contexts that postdate AD 500 but where we had not found clear Monte Alban IIIB diagnostics on the surface (Feinman & Nicholas, 2013, p. 151). The more typical Monte Alban HIB pottery found at Monte Alban and surrounding areas did not reach very far into the eastern and southern hinterland. As Monte Alban IIIA-—style ceramics declined in use in these areas, they were relatively rapidly replaced with more typical Monte Alban IV diagnostics that first appeared in outlying parts of the region while more typical IIIB pottery continued to be used at Monte Alban (Feinman & Nicholas, 2013, pp. 150-151). In spite of problems with the ceramic chronology, there are spatial and temporal patterns that would be lost if the Monte Alban IIIB and IV populations simply were added together and treated as one long period. Such a scheme obscures potentially significant temporal and spatial variation that was consistently observed in the field. The original settlhement pattern maps present three spatially defined patterns (Kowalewski et al., 1989, maps 5-8). In Etla and around Monte Alban, the population was large in Monte Alban IIIB, declined in IV, and rebounded some in V. Tlacolula had a large population in Monte Alban IIIA, which declined in IIIB, recovered in IV, and FIELDIANA: ANTHROPOLOGY N.S., NO. 46, NOVEMBER 13, 2017, PP. 83-96 83 cE 6.1. Late Classic population in Valley of Oaxaca subareas. Subarea Monte Alban IIIB Late IIIB-IV Etla 24,053 15,000 Central 39,189 18,000 Subarea Early HIB-IV Monte Alban IV N. Valle Grande 23,000 18,678 S. Valle Grande 13,000 9.439 W. Tlacolula 21,000 15,761 E. Tlacolula 18,000 24,132 Ejutla 8,000 3,029 Albarradas 1,000 2,406 Sola 7,000 7,066 expanded even more in V. The Valle Grande had a large population in Monte Alban IIA that was much reduced in HIB and IV and then was large again in V. To address these issues in a way that preserves the broad findings of the original surveys, incorporates new information from recent excavations, and recognizes that Monte Alban IIIB-IV was not a prolonged era of stasis, we reexamined the settlement data from across much of the region. We looked at all multicomponent sites with occupations pertaining to Monte Alban IIIA through V for the Valleys of Oaxaca and Ejutla and the Albarradas area and systematically made adjustments for interim phases that had not been detected or well defined during the surface surveys (Feinman & Nicholas, 2013, pp. 142-152). We use specific language to distinguish adjusted populations from the original survey data. Instead of Monte Alban HIB and IV, we refer to early IIIB-IV and late IIIB-IV for the adjusted figures and interpretations drawn from these new assessments. For the most part, this procedure produced adjusted popula- tions for late IIIB-IV in Etla and the Central area (where we did not detect much Monte Alban IV on the surface) and adjusted populations for early IIIB-IV in the rest of the valley (where Monte Alban IIIB diagnostics were sparse) (Table 6.1). This exercise also resulted in population estimates for the Early 50,000 40,000 Postclassic, which we refer to as early V in the subsequent chapter. The discussion below incorporates the new demo- graphic profiles (Fig. 6.1). However, we also present the original data and maps for Albarradas (dividing this temporal era into two phases: Monte Alban IV and V) that are based directly on our original field investigations. Chronological refinements for the Valley of Oaxaca are important, as Teotihuacan’s decline started early in the Late Classic period, and its fall likely had ramifications for the region and its main center, Monte Alban. The decline of Teotihuacan began circa AD 500-550, when parts of city were abandoned and major structures, especially in the civic-ceremonial core of the city, were burned (Cowgill, 2015, p. 233). The depopulation and eventual fall of the great city is thought to have been provoked by a range of factors, including internal social, political, and economic troubles that weakened the polity so that it was more vulnerable to invasion and/or increasing competi- tion from nearby polities (Cowgill, 2015, p. 239; Nichols, 2016, p. 29). The fall of Teotihuacan and the retraction of its foreign interactions led to changes in macroregional networks. Although we do not see any clear evidence that Teotihuacan conquered or ruled Monte Alban (cf. Winter, 1998), which during this period retained distinctive Oaxacan canons and traditions of pottery, architecture, writing, and burial, there are indications that the elites at the two Classic period centers did have a special relationship, which was recorded toward the end of the Early Classic period on carved stone monuments and painted murals (Marcus, 1983e). In the early part of the Late Classic (early IIIB-IV, ca. AD 500-700), perhaps initially bolstered by its ties with Teotihua- can, Monte Alban reached its largest size and greatest monumentality. A large construction episode across much of the civic-ceremonial core of the city, including the expansions of the great platforms at the northern and southern ends of the Main Plaza, was implemented (Bernal, 1965). At the central city, resource extraction was focused more heavily on internal resources and agricultural production to feed the capital. This emphasis, in turn, promoted demographic growth in and near Western Tlacolula +Eastern Tlacolula 30,000 20,000 iN. Valle Grande we S. Valle Grande & Central 10,000 0 IIIA Early IIIB-IV 84 Late IIIB-IV Albarradas Late V Fic. 6.1. Adjusted population change in Valley of Oaxaca subareas during the Classic and Postclassic periods. Early V FIELDIANA: ANTHROPOLOGY WY rm P 4 ee . ” aroma del fa ches an p rie ae Fy ary ‘72500 ay fared Kamera ney Pic. 6.2. Monte Alban. Both in subdivisions or neighborhoods of Monte Alban (Blanton, 1978, pp. 19-20) and in smaller centers near the capital (Kowalewski et al., 1989, pp. 262-270), architectural complexes with relatively consistently aligned groups of platform and plazas were built (Blanton, 1989; Kowalewski et al., 1989, p. 263). In-migration and the civic-ceremonial integration of an expanding population was fostered through 3 ‘, Ve Py Monte Alban., a4 om A t\ Sa Ce C cee , Macifioen ain § a { A é 2000 ty is me: “vale OD. ~ al? "Lambinfeca a, a é eee ae | es - Mitla Fortress. OES go Late Classic (Monte Alban IIIB-IV) hae Fl Palmillo ,- i “\C-20007 & 4 population >1000 4 population 400-1000 N 0 10 a kilometers Large Late Classic period settlements in the Central Valleys of Oaxaca, showing principal places mentioned in the text. public rituals and events likely centered in these plazas (Feinman & Nicholas, 2013, p. 152, 2017, p. 270). As people were drawn (not forced) to the capital and its immediate surrounds, population and settlement in the Central area around Monte Alban and in the northern, Etla arm reached their highest pre-Hispanic levels. Several large secondary centers in Etla—Reyes Etla, Suchilquitongo, and Loma del FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 85 Early IIIB-IV estimated population densities per km? Fic. 6.3. Population densities in the Valley of Oaxaca and neighboring areas during early HIB-IV. Trapiche—had large complexes of monumental public archi- tecture (Fig. 6.2; Kowalewski et al., 1989, p. 260) that match the scale of Early Classic centers in the Valle Grande. Based on recent excavations and the assessment of earlier investigations at the secondary center of Lambityeco (e.g., Lind & Urcid, 1983, 2010; Paddock, 1983a; Feinman et al., 2016), we see early HIB-IV as a time of relatively collective organization at Monte Alban. Although the governing elite was neither Fic. 6.4. 86 Late IIIB-IV Population densities per km? Fic. 6.5. Population densities in the Valley of Oaxaca and neighboring areas during late IIIB-IV. invisible nor entirely faceless at this time, the broadly distributed genealogical registers that commemorated several generations of rulers tended to have been carved later in IIIB- IV. Likewise, at Lambityeco and El] Palmillo, the elevation and elaboration of palaces to the center of governance took place later in this period (Feinman & Nicholas, 2013, pp. 141-157). In contrast, during the Middle Classic and early in the Late Classic period, processions (e.g., Gonzalez Licon, 2015) and ritual mm mic Carved stone figure from the ballcourt at El Palmillo; face of figure was intentionally knocked off prior to its discard in ballcourt fill. FIELDIANA: ANTHROPOLOGY mm me cv Fic. 6.6. Carved stone in form of monkey from Platform 11 at El Palmillo. events with wide participation, held in plazas, were significant at sites in the political sphere of Monte Alban, including the early occupational episode at Lambityeco (Feinman & Nicholas, 2013, 2016c; Feinman et al., 2016). The significant growth in population and monumental building that took place in Etla was ramified in the southern part of the Penoles area, which is adjacent to Etla, where many small Late Formative hilltop sites grew into important Classic period centers, with lots of terraces and platforms. We do not have fine chronological control for Penoles, but population increases there may have occurred coincidently with the growth in Etla (Fig. 6.3). Given its proximity to the northern arm of the valley, the population growth that occurred in the Sierra Norte during Monte Alban IIIB-IV (Drennan, 1989, p. 378) also may have started early in that era. As Monte Alban’s rulers focused on enhancing connections with areas to the north, populations declined in southern parts of the valley, especially in southern Valle Grande and the Ejutla Valley (see Fig. 6.3). As Monte Alban’s interest in southern networks that had focused on Ejutla waned, the Early Classic center at San Joaquin began to decline. Although population decline was less pronounced in the adjacent Sola Valley, effects of Monte Alban’s directional shift are evident in settlement discontinuities and a lower level of political unification (Balkansky, 2002, p. 57). At the same time, populations grew in Miahuatlan, with many sites located in more defendable locations (Markman, 1981, p. 67). Miahuatlan may never have been as tied in with Monte Alban as Ejutla and Sola were, so the patterns of change differed. Recent excavations have provided a fuller picture of Late Classic period changes in Tlacolula (e.g., Feinman et al., 2016). At the time that populations declined somewhat in western Tlacolula, they increased in eastern Tlacolula during early HIB- IV. Initially, this may have been spurred by efforts of Monte Alban’s governors to extend the core area of their most intense degree of influence and interdependence from Dainzu—Mac- uilxochitl, a major secondary center in Monte Alban IIIA, to Lambityeco in the center of the Tlacolula arm. Yegiuih, situated in the eastern part of the larger Lambityeco/ Yegitiih site, had been an important secondary center in Tlacolula during Monte Alban HIJA, but it was outside Monte Alban’s core heartland, which extended to Dainzt—Macuilxo- chitl. Early in the Late Classic period, the civic-ceremonial precinct at Lambityeco/Yegiiih was moved several kilometers to the western, Lambityeco part of the site (Feinman et al., 2016), which is closer to Macuilxochitl/Cerro Danush. Residential population also moved to the northern and western parts of Lambityeco (Paddock, 1983d). The Yegiiih section of the site was basically abandoned. We believe that this move to a location closer to and visible from Cerro Danush, an area that had always been more closely tied to Monte Alban, reflects a change in the relationship between Lambityeco and Monte Alban. Although mound groupings at both Yegiih and Lambityeco had linear arrangements, the new civic core at Lambityeco was far more monumental than the earlier civic precinct at Yegitth and was constructed following architectural canons common at Monte Alban. The building of Lambityeco’s core was well planned and rapidly implemented. All buildings shared a common orientation (Lind & Urcid, 2010, p. 106; Feinman & Nicholas, 2016c). The site’s palaces were construct- ed on the north side of the civic-ceremonial core (as they were at Monte Alban), and the rooms of the palaces were positioned around a single patio, also like at Monte Alban (Lind & Urcid, TABLE 6.2. Percentage of sourced obsidian at Late Classic sites in highland Oaxaca. El G. Other/ Subarea No. Chayal Victoria Orizaba Otumba Pachuca Paredon Tulancingo Ucareo Zacualtipan Zaragoza unknown Etla 5 aoe a = aa 40 20 — — 40 — Central 74) l 1] 5 3 17 3 wl 8 ] 55 3 N. Valle Grande 102 ae 2 1 2 13 1 ] 3 2 1S aa W. Tlacolula 2191 1 ] 2 6 a 1 Ey) 5 2 75 4 E. Tlacolula 2174 3 aa 2 6 10 a2 = 4 77 3 Ejutla 1859 a 2 ] 5 8 2 oo 28 52 = Mixe 223 aa — 6 ] ‘ D — 5 = 87 ae Mixteca Alta 21 — 14 — 19 24 — — 33 — 10 — FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 87 TABLE 6.3. Ceramic description IV-3 IV-4 3-1 (T1120) jar — es (11123) sahumador — — G-35 (T1126) conical bowl 10 — G-3onuien) — — G-35 (T1363) bowl with tripod hollow supports — — G-35 (T1277) spiked vessel — — Café (T1508) large thick storage jar ] oe Gris (11520) bowl — == K-1/19 (T2013A) jar with curved back neck — 1 Café (T2014) comal _ — K-14/22 (T2086) — — Café (12418) comal with raised shelf — a Thick, poorly finished hemispherical gris bow] — oe Fine Orange (T3030) bowl = = * See Table 6.4 for full site name. ** Ceramics collected during Guirun site survey in 1996. 2010, pp. 115-116). At this time, a new ballcourt that was similar in form, size, and orientation to the ballcourt on Monte Alban’s Main Plaza (Feinman et al., 2016) was constructed at Lambityeco. The Lambityeco ballcourt was positioned so that its long axis ran north to south, which differs from the earlier, smaller Yegith ballcourt, which ran east to west. The iconography at Lambityeco included prominent representations of -Cocijo (Cuind &s Urcidy 20107 ps 112). the principal supernatural associated with Monte Alban, which were arrayed in the patio of an important palace. El] Palmillo is another, larger secondary center in eastern Tlacolula where there also were significant changes between Monte Alban IIIA and IIIB-IV. Based on survey and excavations, the hilltop terrace site reached its greatest extent in Monte Alban IIIA or early in IIIB-IV and then began a slow decline during the Late Classic (Feinman & Nicholas, 2004a, pp. 95-96). On most of the excavated terraces, we mapped intact residential architecture pertaining to Monte Alban IIIA. But the earliest intact levels in the three palaces at the top of the hill, all with one patio, were constructed in early IIIB-IV, more or less contemporaneous with the move from Yegiiih to Lambityeco. Some Monte Alban IIIA and earlier ceramics that were present under the lowest levels are all that remains of Monte Alban IJIB-IV ceramics at sites* in the Albarradas area. IV-7 IV-10 IV-11 IV-12 IV-13 Guiran** = = 2 7 = 26 = 4 _ — 22 6 at 1 3 10 6 142 = Lo = Es 1 =a — at 2 a 1 4 = = oe = = 4 = = = as = 2 = | | 4 = 1 = 2 | | 2 = = ss ae 3 = 3 a a 2 4 = 3 a = = me a 1 1 == aa a — es =< = = = = 1 residential architecture that was leveled during the construction of the new Late Classic palaces (Feinman & Nicholas, 2009, p. 64). Thus major changes occurred during the middle of the Classic period, even at El Palmillo, which is distant from Monte Alban. When this shift occurred, the palaces were built with a plan similar to that at Monte Alban. But later in the Late Classic period at both Lambityeco and El] Palmillo, the size of main palaces was expanded to include multiple patios; that plan deviated from the architectural conventions previously shared with Monte Alban. At the same time, a small ballcourt was constructed between two of the palaces at E] Palmillo (Feinman & Nicholas, 201 1c). During late ITIB-IV, Monte Alban lost power and began a slow decline. Changing macroregional networks provided new opportunities for local elites at secondary centers who grew in importance and began to assert greater autonomy. At as many as a dozen sites, local elites began to commission and erect genealogical registers that emphasized their personal biography and bilateral descent (Marcus, 1992b, 2002, 2006; Urcid, 1992, 2003; Urcid et al., 1994; Masson & Orr, 1998; Feinman, 2007). The funding of power may have shifted to external resources, especially the control of trade routes. One of those sites was Lambityeco, where closer ties to Monte Alban did not last long TABLE 6.4. Monte Alban HIB-IV sites in the Guirtn/Albarradas survey region. Grid No. Site name Subarea* square Population 1 TL-SLA-RO-1 AL 719 8 2 TL-SLA-SLA-9 AL 819 26 3. TL-SLA-SLA-19 ET 818 128 4. TL-SLA-SLA-23 AL 719 17 5 TL-SLA-SLA-30 AL 819 8 6 TL-SLA-SLA-33 AL 819 8 7 TL-SLA-SLA-56 AL 720 8 8 TL-SLA-SLA-61 ET 819 8 9 TL-SLA-SLA-43 AL 920 68 10 TL-SLA-SLA-24 (Hierve el Agua) AL 719 1,294 11 TL-SLA-SLA-S53 (Llego Yiachi) AL 720 882 12. TL-SLA-SLA-41 (El Trigo) AL 920 87 13. TL-SLA-SLA-17 (Guirtn) ET 818 2,918 * AL = Albarradas; ET = eastern Tlacolula. ** ch = chert; gs = ground stone. 88 Mounds Mounds Mound Craft Exotics <1 m >1m_ volume (m*) Terraces work** present 0 0 0 0 — _ 0 0 0 0 —- obsidian 16 0 0 1 — obsidian ] 0 46 ? ch — 0 0 0 0 — — 0 0 0 0 = obsidian 0 0 0 0 — — 0 0 0. 0 — — 8 0 0 1 — — 2) 2 385 >3 — obsidian 13 10 4,352 31 — obsidian 6 1 5D 7) ch obsidian, shell 18 26 21,615 BLS ch, gs obsidian FIELDIANA: ANTHROPOLOGY Monte Alban IV Settlements in the Guirun/Albarrradas Survey Region Ed E18 N10 X : y . & iets & Bey } Sites smaller than 1.0 ha are marked by triangles. All others are drawn to size and shape. Contour interval: 500 m. Figs Oy. once the region’s largest urban center began to decline. At Lambityeco, palaces were reconstructed with a two-patio plan not seen at Monte Alban (Lind & Urcid, 2010, pp. 123-125). The civic-ceremonial core at Lambityeco was modified in ways that changed its use, including the remodeling of the ballcourt to a layout less similar to the one at Monte Alban (Feinman et al., 2016). At the same time, a large plaza associated with the ballcourt was closed off, and ritual space on the plaza was restricted to use by much smaller groups (Feinman & Nicholas, 2016c, pp. 31-33). In later excavation levels at Lambityeco, representations of Cocijo are rare, largely replaced by other imagery; this change would indicate a strong declaration of broadening ideological emphases with less focus on Cocijo (Feinman & Nicholas, 2016c, p. 50). In the Mound 190 palace at Lambityeco, the iconic Cocijo frescoes were covered over during a remodeling episode and never replaced. These changes are related to shifts in political strategies from the more collectively organized power structures present in the Valley of Oaxaca during the Early and Middle Classic to more exclusionary tactics of emerging elites at smaller centers. The FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO E19 E20 ipo ay ‘ Hormiga - Colorada ‘S 0 kilometers Monte Alban IV settlements in the Guirtn/Albarradas survey region, showing sites mentioned in the text. location of the ballcourt at El] Palmillo between two palaces, with direct access by members of those households, instead of in the site’s main public plaza fits the expanding use of exclusionary tactics by emerging elites. The decentralization of political influence away from Monte Alban and the empower- ment of local nobles at secondary sites led to greater instability, and regional populations declined across much of the valley during late IIIB-IV. Lambityeco itself was largely abandoned by the end of the Late Classic period. El Palmillo continued to be occupied until possibly the beginning of the Early Postclassic, but there, too, was upheaval. The ballcourt was short lived. During its abandonment, a carved stone sculpture of a regal-looking figure (a possible ruler), with its head lopped off (defaced), was thrown into the ballcourt during an intentional fill event that ended the use of the ballcourt (Fig. 6.4). Eastern Tlacolula, notably the area east of Lambityeco, was the only subarea of the valley where populations grew. Even though eastern Tlacolula is the driest subarea, it became the new demographic center of the valley (Fig. 6.5). 89 TL-SLA-SLA-43 Fic. 6.8. The adjacent Albarradas area was the only other area in the region to see significant population growth during late IIIB-IV (see Fig. 6.3). Given the focus on craft specialization in the absence of great expanses of good farmland that began in Monte Alban IIIA, eastern Tlacolula became the commercial center of the valley. Analysis of ceramic compositions using instrumental neutron activation (Faulseit et al., in press) has shown that even utilitarian pottery was exchanged across the valley. That El Palmillo, in eastern Tlacolula, had the highest proportion of imported pottery of any site in the study illustrates how important exchange connections were to the dry, eastern Tlacolula arm of the valley. The adjacent Albarradas area benefited from these changes as the amount of exchange and interaction increased in eastern Tlacolula. The higher proportions of Imitation Fine Orange pottery at sites in eastern Tlacolula (Feinman & Nicholas, 2013, figure 8.10) iS One indicator that trade links to the east, toward the Gulf Coast and Cantona (eastern Puebla), were a key factor that allowed Tlacolula elites to break from Monte Alban and thrive as that center fell. Influences from and interactions with Veracruz and the Gulf lowlands likely arrived in the valley mainly by way of the eastern travel route through the Albarradas and Mixe areas or by the southern route into the Ejutla Valley. Fine Orange also is present in the Mixe collection analyzed by Hutson (2014, p. 271). In our excavations at four sites in Oaxaca that were occupied throughout the Classic period, we found limited amounts of Imitation Fine Orange in Ejutla (in the south) compared to higher quantities at the three sites in Tlacolula (El Palmillo, Mitla Fortress, and Lambityeco) (Feinman & Nich- olas, 2015, table 6). Although the abundant samples of Imitation Fine Orange at Lambityeco were almost certainly locally made at Lambityeco itself (Abascal et al., 1973), the relative abundance implies links to the east, likely via the Mixe region. A stone sculpture of an anthropomorphic monkey recovered in a palace at El Palmillo (Fig. 6.6) is similar to carved stone figures in the Mixe collection analyzed by Hutson (2014, figure 31). Network analyses of obsidian distributions also provide support for the increasing importance of the eastern travel 90 IIIB-IV-9 Site map of TL-SLA-SLA-43. route into the Valley of Oaxaca but with a shift from the southern Isthmus to the Gulf Coast (Feinman et al., 2015; Golitko & Feinman, 2015, figures 9 and 10). A principal route from the Gulf Coast, to the north, intersects the eastern travel route along the Tehuantepec River in the Mixe area (see Fig. 2.12). This is the most likely route for obsidian from the Zaragoza obsidian source, which is situated near the Puebla/ Veracruz border, to reach the Valley of Oaxaca. During the Late Classic period, Zaragoza obsidian became a major source in highland Mesoamerica, especially in Oaxaca, where its abun- dance in obsidian assemblages continued to be higher in eastern Tlacolula than in western parts of the valley (Table 6.2; Feinman & Nicholas, 2013, figure 7.29; Feinman et al., 2013, figure 5, 2015, figure 15). Zaragoza obsidian also accounts for more than 85% of our sourced sample of obsidian from several sites in the Mixe region (Feinman et al., 2017). Zaragoza was heavily exploited by Cantona (Garcia Cook, 2003, pp. 339, 341; Garcia Cook & Tenorio, 2010), which became the largest and most important Late Classic center in the Central Highlands after the decline of Teotihuacan (Garcia Cook, 1994, 2003, pp. 311, 341; Garcia Cook & Merino Carrion, 1998; Montero Guzman, 2004, p. 56; Gomez Chavez & Spence, 2012, p. 288; Cowgill, 2015, p. 237). Proximity to the Zaragoza source and its exploitation may have been a significant factor in the rise of Cantona as a large, important city (Garcia Cook & Merino Carrion, 1998, p. 213). Pottery and other goods from faraway regions (including Oaxaca, central Mexico, the Gulf Coast, and Campeche) have been found at Cantona and yield indications of long-distance connections (Garcia Cook & Merino Carrion, 1998, p. 213; Plunket & Urunuela, 2005, p. 105), which increased trade along the Gulf Coast during the Late Classic period. Monte Alban IIIB-IV Settlements in Albarradas During the survey of the Albarradas area, we recovered few definitive diagnostic ceramics of Monte Alban IIIB (only at the Guirtin site). We dated 10 sites with typical Valley of Oaxaca FIELDIANA: ANTHROPOLOGY ee TL-SLA-SLA-24 L HIB-IV V . Roaguia : LI-5 II-4 : IIB-IV-10 \. V-53 Fic. 6.9. Site map of Hierve el Agua (TL-SLA-SLA-24). IIIB-IV pottery (ubiquitous G-35 bowls) to Monte Alban IV, or part of the Guirtn site. In Chichicapan, Fahmel Beyer (2012) what we call here late IIIB-IV (Table 6.3). Although the late found evidence that local potters continued to be influenced by IIIB-IV diagnostics on sites in the Albarradas area did not Monte Alban and molded figurines and whistles that replicated include Imitation Fine Orange, that ware was present in one those found at the capital during the Late Classic. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 9] y - Fic. 6.10. Small structures on flatter terrain in the northern part of Hierve el Agua (TL-SLA-SLA-24). In terms of settlement, the largest change in the Albarradas area in Monte Alban IIIB-IV was the complete abandonment of the small, high mountain valley in the northwest and the establishment of a large site at Hierve el Agua in the southeastern part of the Albarradas area (Fig. 6.7). Llego Yiachi remained a large outpost with significant population, and except for the addition of Hierve el Agua, the overall pattern in the location of smaller sites in defendable locations on the eastern slopes of Cerro Guirone continued. But this similarity masks the abandonment of all small sites in the Albarradas area by the end of Monte Alban IIA. The only two settlements that continued to be occupied in HIB-IV are two of the most defensibly located sites: Llego Yiachi and El Trigo (TL-SLA-SLA-41) (Table 6.4). El Trigo is one of the few places in the Albarradas area that was occupied throughout the pre- Hispanic sequence, beginning in Monte Alban Late I. It is a terrace site situated on the top of a steep and very defendable hill in the middle piedmont, with a great view of areas to the east and north. El Trigo is situated above the contemporary highway to the Mixe region, which largely follows an ancient road. The site consists of two dozen residential terraces, although it is not clear that they were all constructed by this time, as we found Classic period pottery on only about one- third of the terraces. Postclassic ceramics were found on all the terraces, and many of them may not have been constructed until AP Fic. 6.11. Stone retaining walls at the base of springs at Hierve el Agua (TL-SLA-SLA-24). 92 Fossilized canal at Hierve el Agua (TL-SLA-SLA-24). FignG:l> then. Evidence of chert working at the site also may pertain to Monte Alban V. As populations moved back into the Albarradas area in Monte Alban HIB-IV, people settled in different but still very defendable locations, often on ridgetops or promontories with good views of the surrounding area, so it appears that the residents of these settlements still feared raids or had other defensive consider- ations. One of these sites is TL-SLA-SLA-43, where we mapped dispersed house foundations on the top of a long, narrow ridge between two deep barrancas (Fig. 6.8). Hierve el Agua, located below the eastern slopes of Nueve Puntas, became the largest site in the Albarradas area (Fig. 6.9). More than 90% of the population of the Albarradas area resided in Hierve el Agua and Llego Yiachi, both of which had defensive postures. We found no evidence that craft production was a significant activity at either site, although chert working was present at a small settlement very close to Hierve el Agua. Overall emphasis on craft production continued to be minimal in Albarradas. Hierve el Agua, near the modern village of Roaguia, had been occupied earlier during Monte Alban Late I and II. The earlier population settled on top of a narrow ridge above the natural springs. Although we did not recover any definite Monte Alban IIIA pottery on the surface, Classic period pottery covered a wide area of flatter, less defendable terrain, and we cannot rule out that there was some occupation at the site in the Early Classic. Clearly by Monte Alban IIIB-IV, though, the settlement expanded over an area of 86 ha of well-watered terrain at the base of the mountains and above the natural springs and fossilized canals (Figs. 6.10—6.12). The site is defendable from the south and east but not from the north (Fig. 6.13). Today, a modern dirt road from eastern Tlacolula to Roaguia crosses a secondary pass in the mountain ridge behind Hierve el Agua, and the site’s past inhabitants may have used this route to travel into eastern Tlacolula for both defense and trade. The site also lies right above the current path that follows the Hormiga Colorada, leading south to San Baltazar Guelavila and beyond. In spite of defensive concerns, all of the IIIB-IV settlements were in the eastern sector above the Hormiga Colorada and its tributaries, an indication that location near the principal trade route into eastern Tlacolula continued to be a key factor. Guirin continued to be one of the largest sites in the eastern end of the Tlacolula Valley as settlement spread farther up the slopes to the top of the highest ridge on the western side of Cerro Guirone (see Fig. 5.10). Local rulers at Guirtn may have FIELDIANA: ANTHROPOLOGY Fic. 6.13. View of Hierve el Agua (TL-SLA-SLA-24) from the southeast. Monte Alban IV 8 a western Tlacolula .Lambityeco . if a a v= 2 c “as, West et a ner a 44% tae! ve eastern Tlacolula * #4 ya * ( — 3 aa NG a = + eoU eis wv fe El PalmilloXS~ ./* BOS (05 Ce GS 3! ° =_Hormiga) ae Colorada ) v Ss RY A Guelavila Sites smaller than 2.0 ha are marked by triangles. All others are drawn to shape and size. Contour interval: 500 m. Fic. 6.14. Monte Alban IV settlements in Tlacolula and the Albarradas area, showing principal places mentioned in the text. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO fas_e 6.5. Percentage of population residing in terrace sites in Valley of Oaxaca subareas in Monte Alban IIIB-IV. No. of | Terrace % of terrace site Total population in Subarea sites population population terrace sites Etla (IIIB) 6 14,043 24,053 58 Central (IIIB) 3 27,359 39,189 70 N. Valle Grande (IV) 2 17,298 18,678 93 S. Valle Grande (IV) 6 7,032 9,439 74 W. Tlacolula (IV) 6 2,310 15,761 15 E. Tlacolula (IV) 7 11,406 24,132 46 Ejutla (IIIB-IV) 6 1,529 3,029 50 Albarradas (IIIB-IV) 5 2,301 2,406 96 Sola (IIIB-IV) 12 4,067 7,066 58 continued to exploit their control/monitoring of the eastern route into the Valley of Oaxaca and develop their own networks, and this may have been an important factor in the site’s continued growth. Most of the large sites in the Tlacolula arm during the later Classic were located in the northern half of this region, near other possible entry points into the valley from the north (Fig. 6.14). The southern part of the Tlacolula arm was almost devoid of settlement, and El Palmillo may have begun to decline in size and importance by the very end of the Late Classic (Feinman & Nicholas, 2004a, p. 95). Both of these patterns may reflect a waning importance of southern routes into the Valley of Oaxaca, one of which entered the valley west of El Palmillo. This mirrors a decline in population in Ejutla at this time, as the focus of interaction was mainly with areas to the east. Monte Alban IIIB-IV Terrace and Defensive Sites Although populations declined across the Valley of Oaxaca by the end of the Late Classic period and there were fewer terrace sites overall, the hilltop centers continued to house significant proportions of the population (Table 6.5). At their peaks early in IIIB-IV, between 60% and 70% of the population in Etla and the Central area lived in these hilltop communities. Monte Alban’s hilltop presence was reduced by the end of the Classic period, while other terrace sites continued as large population centers (Fig. 6.15). Jalieza, in the northern Valle Grande, expanded to become the largest hilltop center in the entire region and also was the principal population center in the valley, with more than 16,000 residents (Kowalewski et al., 1989, p. 287). Several other large Monte Alban IIIA terrace sites in the Valle Grande continued to be occupied in IHIIB-IV so that more than 75% of the population of the Valle Grande resided in these hilltop centers. The change between Monte Alban IIJA and IIIB-IV was much greater in Ejutla. The proportion of population in terrace sites in Ejutla jumped to 50% from 15% as the large center at San Joaquin lost much of its population by the end of IIIB-IV (Feinman & Nicholas, 2013, p. 130). Only one of the Monte Alban IIA terrace sites in Ejutla continued to be occupied, while all the other terrace sites were new occupations in defensive locations. The great discontinuity in settlement locations in the Ejutla Valley illustrates the significant disruption that occurred even at the valley’s edges coincident with Monte Alban’s decline. With Monte Alban’s lessened 94 TABLE 6.6. Estimated archaeological, resource-based, and labor- based potential populations for Valley of Oaxaca subareas in Monte Alban IIIB-IV. Estimated Resource-based Labor-based archaeological potential potential Subarea population population* —_ population** Etla (IIIB) 24,053 97,800 51,600 Central (JIB) 39,189 45,200 27,900 N. Valle Grande (IV) 18,678 106,000 12,900 S. Valle Grande (IV) 9,439 92,300 12,300 W. Tlacolula (IV) 15,761 85,300 22,400 E. Tlacolula (IV) 24,312 39,000 23,300 Ejutla (IIIB-IV) 3,029 92,000 10,000 Albarradas (IIIB-IV) 2,406 15,400 4,500 Sola (IIIB-IV) 7,066 32,700 16,200 * Based on full exploitation of all agricultural land resources in each subarea. ** Based on available labor supply and equal use of Class I and Class II land near recorded settlements. regional influence and greater interelite factionalism, some of which began to extend beyond physiographic borders, residents of outlying areas became increasingly vulnerable and chose more defensive locations. More than one-third of the Monte Alban IIIB-IV terrace sites in the valley were in the Tlacolula arm, and many of them were large population centers. But not even the largest IIIB-IV settlement in Tlacolula—Macuilxochitl/Tlacochahuaya, where only a very small percentage of the population resided on terraces—was even half as large as Jalieza. The large terrace sites in Tlacolula, including the Guirtn site, were in the eastern half of that arm, where the largest buildup of population occurred in IIIB-IV. Many had been large centers in Monte Alban IIIA. Yet as population growth in eastern Tlacolula also occurred in sites that lacked terraces, the population in terrace sites encompassed a smaller proportion of the region’s overall population than in the Valle Grande (see Table 6.5). In edge areas like Albarradas and Sola, large proportions of the population continued to reside in terrace sites. But there was disruption in site distributions similar to what happened in Ejutla, especially in Albarradas, where most of the Monte Alban IIIA sites were abandoned and IIIB-IV sites were placed in different locations. These changes also likely are part of the reorganization that occurred across the region with and after Monte Alban’s decline. Settlement and Land Use in Monte Alban IIIB-IV With increasing maize yields per hectare and declining populations across most of the Valley of Oaxaca toward the end of the Late Classic period, archaeological populations remained below regional capacities (Table 6.6). The only exceptions are the Central area and eastern Tlacolula in early IIIB-IV and eastern Tlacolula in late IIIB-IV (Figs. 6.16 and 6.17). In spite of higher agricultural yields per hectare, the smaller populations in most subareas were not distributed optimally relative to good land and would not have been able to grow as much agricultural surplus as was possible in Monte Alban IIIA (Nicholas, 1989, p. 480). The residents of the largest settlements in the Central area, northern Valle Grande, and FIELDIANA: ANTHROPOLOGY Monte Alban... Pics 6:13: eastern Tlacolula would have had to engage in exchange in foodstuffs to feed themselves (see Table 6.6). Clearly, concerns with defense often rivaled efficient agrarian placement, leading to the positioning of settlements and population in less-than- optimal locations for agriculture. In northern Valle Grande, population was so concentrated in one center—Jalieza—that its residents would have had to travel considerable distances to farm available agricultural land or Monte Alban IIIB-IV ie Terrace sites with IIIB ceramic diagnostics 4. Terrace sites with IV or IIIB-IV ceramics N 0 10 ol kilometers Monte Alban HIB-IV terrace sites in the Central Valleys of Oaxaca. import foodstuffs grown elsewhere in the valley. Monte Alban would have continued to require agricultural surplus produced elsewhere, all of which could have come from Etla. At the southern end of the Valley of Oaxaca, large surpluses also were possible in the Ejutla and Sola Valleys. Settlements in both areas were more broadly distributed across their regions in line with agricultural resources (Balkansky, 2002, p. 60; Feinman & Nicholas, 2013, p. 138). FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 95 Archaeological population as % of subarea maximum potential population Fic. 6.16. Early HIB-IV archaeological population as a percent- age of the maximum potential population in each Valley of Oaxaca subarea. The large centers in eastern Tlacolula would have continued to depend on exchange for foodstuffs grown elsewhere in the region. Most of that food may have come from the Albarradas area, where farmers could have produced much more agricul- tural surplus than ever before. Although there is limited flat or gently sloping land in the Albarradas area, the eastern slopes of Cerro Guirone receive much higher rainfall than eastern Tlacolula, which is located in a rain shadow, and high crop yields are possible with sufficient labor. The large settlement at Hierve el Agua was located on some of the flattest, well-watered land in the Albarradas area, and its inhabitants could have grown far more food than they required. Although we could not confirm their age, there are agricultural terraces on a southern ridge of the site, above the fossilized canals, that likely were used for farming (Flannery et al., 1967; Neely et al., 1990; cf. Hewitt et al., 1987; Peterson et al., 1989). Monte Alban IIIB-I1V Summary The decline of Monte Alban coincided with an episode of population disruption and reorganization across highland Mesoamerica during the final centuries of the Classic period (e.g., Diehl & Berlo, 1989; Hirth, 1995; C. A. Smith, 2002, p. 313; Kowalski & Kristan-Graham, 2007). Many regions and their principal centers declined in population at the same time that populations grew and new centers were established in other areas. Although demographic declines occurred across most of the Valley of Oaxaca, populations grew in Tlacolula as Monte Alban was losing population. Eastern Tlacolula became the new 96 Monte Alban Late IIIB-IV Archaeological population as % of subarea maximum potential population Fic. 6.17. Late IIIB-IV archaeological population as a percentage of the maximum potential population for each Valley of Oaxaca subarea. demographic and commercial center of the valley. In concert with this shift, populations declined in the southernmost parts of the region: southern Valle Grande and the Ejutla Valley. Those areas may never have been as strongly integrated politically or economically into Monte Alban polity, and local elites in those areas may have initially lost influence as market and exchange connections to the capital weakened. The large Monte Alban HIJA centers declined as local elites gained more autonomy and moved to establish new settlements and networks of interaction. Most of the largest centers in IIIB- IV were in Tlacolula, especially eastern Tlacolula, where the ruling elite at sites that were becoming the new population centers of the valley began to erect genealogical registers to legitimate their right to rule (Marcus, 1983b, 1992b, 2006; Urcid et al., 1994: Masson & Orr, 1998; Urcid, 2003; Feinman, 2007). As with southern Valle Grande and Ejutla, eastern Tlacolula never had been as tightly integrated with the Monte Alban polity as the IIIA centers in western Tlacolula, like Macuilxo- chitl, had been. With regional reorganization and changing networks of interaction, ruling elites in eastern Tlacolula may have been less able to obtain desired goods from their western neighbors and instead strengthened ties to the east, through the Albarradas area, perhaps to Cantona, Cholula, and the Gulf Coast. In addition, sites in eastern Tlacolula from early in the Classic period relied heavily on spot resources, like salt (Lambityeco), and craft production. As governance and power shifted to more autocratic/exclusionary forms dependent on external means of support, the local elites in eastern Tlacolula may have had advantages over other polities where internal resources and maize farming were still more important. FIELDIANA: ANTHROPOLOGY Chapter 7: Postclassic (Monte Alban V) “In commercialized times such as the Mesoamerican Postclassic, power is built by moving value about freely, which is difficult where there are corporate obstacles (Blanton et al., 1996). Commercial power is easily transferred from place to place. It makes boundaries dissolve. In the language of money, this is liquid, financial capital, as opposed to a more slowly moving industrial capital, which would be closer to the landed, corporate form of power that was more in evidence in the Classic period.” (Kowalewski, 2003b, pp. 9-10) The decline of Teotihuacan, Monte Alban, and other large Classic period centers in highland Mesoamerica marked the beginning of a long period of instability, political fragmenta- tion, and population reorganization that lasted well into the Early Postclassic period. Early in the Late Postclassic period, this long episode of upheaval culminated in a much more heavily populated landscape of small, competing states. The fall of the great centers led to power vacuums and disruptions in exchange networks and trade routes that provided new opportunities for local elites to assert their independence and form new networks of interaction. Local elites became the rulers of smaller semiautonomous polities that were economically interdependent as they competed to establish new networks of interaction and exchange and to attract followers. The proliferation of small polities coincided with increasing commercialization, expansion of market exchange, and increas- ing economic interdependence as population growth occurred across the highlands. Regional and extraregional boundaries became more permeable than ever before, and in many regions new networks and relations were established (Kowalewsk’ et al., 1983, p. 52; Kowalewski, 1990, 2003b; M. E. Smith & Berdan, 2000; Berdan et al., 2003; Golitko & Feinman, 2015, p. 233). Small polities had fluid, changing boundaries, and alliances between these polities shifted with regularity. Not until the end of the pre-Hispanic era, marked by the late rise of the Aztec empire with its capital at Tenochtitlan, did a large polity dominate the landscape as Teotihuacan had during the Terminal Formative and Early Classic periods. During the Postclassic (Monte Alban V), the large region that is today the state of Oaxaca was politically fragmented into networks of small, semiautonomous petty states, or cacicazgos (Oudijk, 2002, 2008). These small states generally were part of alliance networks that often extended well beyond physiographic regions, such as the Valley of Oaxaca (Oudijk, 2000; Pohl, 2003a, 2003b, 2003c). In the valley, cacicazgos were periodically linked to the Mixteca Alta to the north, to the Pacific Coast to the south, and the Albarradas area to the east (Spores, 1965; Oudijk, 2000). At times, small states aggregated under powerful lords. During the Postclassic, economic networks shifted as Monte Alban lost power and emerging leaders of small polities employed exclusionary political strategies to build networks of personal connections to attain power (Kowalewski, 2003b, p. 10; Fargher et al., 2011). Across Mesoamerica, the economy became more commercialized, fostered by an increasing reliance on water transport, new technologies (including metallurgy), and more voluminous flows of “bulk luxuries,” such as obsidian, white salt, fine textiles, and highly decorated and hard-fired pottery (Kepecs, 2003; Blanton et al., 2005). The goods that were moved in these networks across the Meso- american highlands also became key sources of revenue for leaders and lords, especially for those who were able to broaden networks and alliances and attract followers. In Oaxaca, the small Late Postclassic states were economically interdependent and at the same time competitive, sometimes forming larger alliances and confederations that were often ephemeral (Spores, 1974; Pohl, 2003a). Ethnohistoric accounts detail competition and shifting alliances (military and marriage) among these small late pre-Hispanic polities in Oaxaca (Spores, 1967; Paddock, 1983c; Pohl, 2003a). The Valley of Oaxaca was never again united (or even mostly unified) under one dominant center. Much of the growth in highland Oaxaca occurred through the establishment of small settlements that proliferated across the region. This dispersed settlement pattern is found across the Southern Highlands, in mountainous zones as well as in broad valleys. The large number of small settlements reflects an active market system that allowed at least a subset of householders in these communities to buffer agrarian imbalances and exchange craft products. At the same time, demographic expansion occurred at many levels of the site hierarchy, with a significant proportion of the population living in larger centers of 1,000 people or more (Kowalewski et al., 1989, table 10.2). By the Late Postclassic, these larger communities were generally the political centers of between 15 to 25 semiautonomous cacicaz- gos that spread across the Oaxaca and Ejutla Valleys (Kowalewski et al., 1989, pp. 344-345; Kowalewski, 1990, p. 42, Feinman, 1998, p. 115, 2001b, p. 365; Feinman & Nicholas, 2013, p. 148). These later polities had populations as high as 25,000 (Feinman & Nicholas, 2013, table 10.1), much smaller than the earlier Classic period Monte Alban polity but comparable to estimated population ranges for small states in other parts of the world (e.g., Pounds, 1969; Renfrew, 1982). A similar trend in polity size occurred across highland Meso- america. There were as many as 30 semiautonomous petty kingdoms in the Mixteca Alta in the Late Postclassic, most of which were in the size range of the Valley of Oaxaca polities (Kowalewski et al., 2009; Feinman & Nicholas, 2013, table 10.2). Many Late Postclassic polities in central Mexico also fell within the size range of the Oaxacan polities (Sanders et al., 1970, pp. 443-462; Bray, 1972; M. E. Smith, 2005), although, of course, a few were much larger. In the Valley of Oaxaca, two of the largest polities (Macuilxochitl and Mitla) were situated in the Tlacolula arm of the valley (Fig. 7.1; Feinman & Nicholas, 2013, table 10.1), which became the most densely settled, commercialized, and market-dependent subarea of the valley during the Late Postclassic (Kowalewski, 2003c, p. 73). Population densities followed a clear falloff with greater distances from Tlacolula (Fig. 7.2); the northern Etla arm, which had been so dominant during the Early and Middle Formative, was only slightly more FIELDIANA: ANTHROPOLOGY N.:S., NO. 46, NOVEMBER 13, 2017, PP. 97-113 OF , ae “4 ee ee hs a “ ae q [UL A a Biv aoe > ke. Zaachilai., oS . } . ne a = \ aaa / Po a4 ST ee enese S. \ a 3 , at 44 we —G uh, rg 8 ems ‘aa a & S ras sed 3 if aE Foto eo nln Nk > ta > : SC ee eal Ste BiGa7et! densely settled than Ejutla, where the population density was only about 30% of the density in eastern Tlacolula. Farther south, the Sola and Miahuatlan Valleys were settled even less densely. Populations more than doubled in Penoles (Kowalew- ski, 1991) and the Sierra Norte (Drennan, 1989), yet population densities in those areas also remained much lower than in any subarea of the Valley of Oaxaca. 98 ha 7S a aS ee Sites smaller than 2.0 ha are marked by triangles. All others are drawn to size and shape. Contour interval: 500 m. kilometers Monte Alban V settlements in the Central Valleys of Oaxaca, showing principal places mentioned in the text. The similarities in size of the Late Postclassic polities highlight the interdependency and importance of economic exchanges, often through marketplaces that were linked into wide-ranging networks of interaction. At least for the Basin of Mexico and the Mixteca Alta, all polity centers are thought to have had at least one marketplace (Lind, 2000, p. 573; M. E. Smith, 2003, p. 60). It was not until the late expansion of the FIELDIANA: ANTHROPOLOGY Monte Alban V Population densities per km? Fic. 7.2. Population densities in the Valley of Oaxaca and neighboring areas during Monte Alban V. Aztec empire that one of the largest Late Postclassic polities was established at Coixtlahuaca in the northern Oaxaca highlands as a commercial center for Aztec intrusion/expansion into Oaxaca (see Fig. 2.1; Kowalewski et al., 2008, 2010). But not all residents of the Valley of Oaxaca benefited from increasing commercialization and expanding market exchange. As emerging elites began to form personal networks as a basis to gain power, wealth differences throughout the region increased. During the Late Postclassic, more people than ever before resided in settlements that had few goods beyond utilitarian ceramics (Kowalewski & Finsten, 1983, table 2). In addition, the average number of different kinds of goods found at each site was lower than for any other phase of occupation (Kowalewski & Finsten, 1983, table 3). Some of these trends began during the Late Classic period but then accelerated during the Postclassic. During the Late Postclassic, the greatest impoverishment was in settlements located in the interstices between small polities (Kowalewski & Finsten, 1983, p. 419). At the same time, Postclassic palaces were larger, more elaborate, and more central to settlement layout than earlier in the pre- Hispanic sequence (e.g., Gallegos Ruiz, 1978; Pohl, 1999; Barber & Joyce, 2006). Fewer resources were invested in the construction of temples and other public spaces as the emphasis shifted to royal palaces where ruling lords exercised more exclusionary power and used ritual to validate their right to rule (Winter et al., 2007, p. 205). For Oaxaca, the trends traditionally associated with the Postclassic period more generally, such as significant demo- graphic growth, more dispersed settlement patterns (see Fig. 7.1), expanded exchange and interpersonal networks beyond physiographic regions, and increased economic commercializa- tion (Kowalewski et al., 1989; Feinman, 1996; Kowalewski, 2003b; Markens et al., 2008), are more characteristic of the final centuries of the pre-Hispanic era (late Monte Alban V), when TABLE 7.1. Monte Alban V sites in Valley of Oaxaca subareas. No. of Total Average Average site Subarea sites* population site size (ha) —_ population Etla 270 15,404 2.9 57 Central 158 20,839 6.6 132 N. Valle Grande 219 24,938 aS) 114 S. Valle Grande 276 23,919 4.1 87 W. Tlacolula 358 41,255 5.6 115 E. Tlacolula 260 40,119 jal 154 Ejutla 322 19,970 3.3 62 Albarradas 5] 5,416 4.9 106 Sola 83 9,168 47 110 * Combined sites. populations reached their pre-Hispanic peaks (Table 7.1; Feinman & Nicholas, 2013, figure 9.16). Yet the processes that provoked these shifts were initiated at the end of the Classic period with political fragmentation, an increasing reliance on external resources as a base of power, and the intensification of long-distance exchange networks. Throughout many parts of the Valley of Oaxaca, populations were lower in the Early Postclassic (early Monte Alban V) than they had been at the end of the Classic period (see Fig. 6.1). The southern and eastern ends of the valley were exceptions. Populations continued to rise in the Tlacolula arm of the valley in early Monte Alban V in spite of more limited agricultural resources, especially in eastern Tlacolula. In a landscape of competitive cacicazgos (Oudijk, 2002, p. 75), Yagul and Mitla became the new centers of power during the Early Postclassic period (Bernal & Gamio, 1974; Robles Garcia & Juarez Osnaya, 2009; Feinman & Nicholas, 2013, p. 155), with larger and more elaborate palaces than ever before. Mitla’s location near the entry point of three natural routes into the valley (e.g., Schmieder, 1930, p. 34; E. C. Parsons, 1936, p. 568; Beals, 1979, p. 165) was an important factor in its rise as an economic and political center in early Monte Alban V. Yagul also sits at the opening to a northward extension of the Valley of Oaxaca (Diaz Ordaz) that is a gateway to Yalalag, the Cajones River, and eventually Veracruz (e.g., Ortiz Diaz, 2002) (see Fig. 2.12). At the outset of Monte Alban V, populations also appear to have rebounded faster in the Ejutla Valley and the Albarradas area, In part because of their locations relative to important trade routes (see Fig. 2.12). Based on site sizes, populations almost doubled in Miahuatlan (Markman, 1981, table 2-1). The southern travel route (White & Barber, 2012) into the Ejutla Valley had been less important in the Late Classic period. But by early Monte Alban V, expanding exchange networks revived interaction and exchange along the southern trade route, including intensified interregional networks between the Mix- teca Alta and the Pacific Coast of Oaxaca (S. M. King, 2008, p. 279; Levine et al., 2011, p. 127). The growth of Tututepec on the Oaxaca coast beginning in the Early Postclassic period likely was a key impetus for the expanded importance of this southern route (see Fig. 2.1; Spores, 1993; Joyce et al., 2004; Levine, 2011). To the east, the travel route through the Mixe and Albarradas areas continued to be a significant interaction corridor, as indicated by the diverse materials in the Mixe region with stylistic links to the Valley of Oaxaca, the Mixteca, and Veracruz (Hutson, 2014, p. 272). By the middle of the Postclassic period, the center of political power began to shift from eastern Tlacolula back to the middle FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 99 TABLE 7.2. El G. Subarea No. Chayal Victoria Orizaba Otumba Pachuca Paredon Tulancingo Sierra Norte 17 — 24 12 _— 65 Etla 8 = — 38 — 62 Central 9 — —- 67 22 — N. Valle Grande 18 — — 11 11 1] W. Tlacolula 5 — — — — 60 E. Tlacolula 861 rl 2 4 3 65 Ejutla 33 — — — 3 24 Mixe 296 — — 6 — 1 S. Oaxaca coast 55 a= _ 13 9 9 Southern Isthmus 54 2 ~- 50 6 28 of the Valley of Oaxaca. Earlier in the Classic period, Macuilxochitl, in western Tlacolula, had been closely tied to the capital at Monte Alban, but then it decreased in importance in early Monte Alban V, when Yagul and Mitla were in ascendance. But then, according to documentary accounts (Oudijk, 2008; Joyce, 2010, p. 265), Macuilxochitl (western Tlacolula) joined a confederation with Zaachila (northern Valle Grande) and Mixtec cacicazgos to oppose the rulers of Mitla, whose palaces show evidence of being dismantled and abandoned by late Monte Alban V (Esparza, 1983; Robles Garcia & Juarez Osnaya, 2009, p. 396). According to early colonial era documents, the peasants of Mitla seldom ate corn because it was not available in sufficient quantities (Horcasitas & George, 1955, p. 21). In the dry Mitla area, agriculture has never been a reliable endeavor. Once hostilities with Macuilxo- chit] increased, Mitla’s exchange networks with Albarradas may have strengthened and at least partly replaced earlier ones with western Tlacolula. Agricultural surplus from either area alone would have been sufficient for Mitla’s inhabitants as long as they had the goods, such as craft products, to exchange for it. A direct Mixtec presence in the valley does not appear to have been significant until the later part of the Postclassic period, when Macuilxochitl and Zaachila formed the confederation with the Mixteca Alta. Rapid demographic growth has been argued for the Mixteca Alta at that time (Kowalewski et al., 2009, p. 317, 2010, p. 81), coincident with explosive growth in central Mexico during the Middle Postclassic (Beekman & Christensen, 2003; Berdan & Smith, 2003). Some areas in western parts of the valley, where there is better agricultural land, may have owed political fealty to Mixtec lords, but there is no evidence of large numbers of Mixtecs in the valley (Byland & Pohl, 1994, pp. 182-184). In the Mixteca Alta, craft specialties, such as stonework and metalwork, were highly developed, but large patches of productive agricultural land are limited. Toward the end of the pre-Hispanic era, Mixtec lords established marriage links to just those parts of the Valley of Oaxaca where the largest, most productive patches of land were located, namely, Etla and northern Valle Grande. In these areas, relatively high maize yields could be produced without large inputs of labor. Most communities in Oaxaca today are very compact, but several modern dispersed communities (La Union and Cuilapan) in the western part of the valley where Mixtec lords moved people to farm their estates are exceptions. In contrast to those areas, the economy in Tlacolula was more craft based, with fewer good agricultural opportunities (Fein- man et al., 2002, pp. 254-255; Feinman & Nicholas, 2005, p. 232; Blanton & Fargher, 2010, pp. 220-221). 100 Percentage of sourced obsidian at Early Postclassic sites in Oaxaca. Other/ Ucareo Zacualtipan Zaragoza unknown La a Zu a | etal | w| be || oe NO ES Ww | During the Late Postclassic period, the Mixtecs who migrated to the Valley of Oaxaca mostly settled as agriculturalists, tied to the estates of their lords, in the area southwest of Monte Alban (near Cuilapan) and in Etla (Whitecotton, 1977, p. 151). Because they were brought into the valley as landless dependents to farm the terrain on their lords’ estates, they were directly subservient to the rulers, somewhat like serfs (Spores, 1974; Chance, 1976, p. 605). In that sense, these estates were external resources used by rulers to fund more autocratic political structures. The lack of a tight collective community organization among the in-migrants may help account for the sparsity of artifacts that we found at so many small Monte Alban V archaeological settlements (Kowalewski & Finsten, 1983). From an archaeological perspective, it is hard to find intact Late Postclassic commoner houses in Oaxaca, and this may in part be the consequence of the more modest techniques used in their construction. Obsidian Exchange in the Postclassic Period The sourcing studies of obsidian that we have conducted for sites in the Valley of Oaxaca and neighboring areas have revealed significant changes in networks of interaction after the fall of the great Classic period cities and during the Postclassic period. These shifts in networks of exchange coincide with observed changes in settlement patterns in Oaxaca and elsewhere in western Mesoamerica. The Classic period prepon- derance of Zaragoza obsidian in Oaxaca fell dramatically in the Early Postclassic, and obsidian from the Pachuca source in central Mexico became the dominant obsidian in the Valley of Oaxaca (Table 7.2; Feinman et al., 2015, figure 18, 2017). The decline in the importance of Zaragoza obsidian in Oaxaca is coincident with the decline and abandonment of Cantona around AD 1000 (Garcia Cook & Merino Carrion, 1998, p. 213). Cantona was located less than 10 km from the Zaragoza source, and people at that site heavily exploited it, receiving in exchange goods from across large parts of Mesoamerica. The social network graphs for all of Mesoamerica illustrate a complete breakdown of prior Classic era networks in the Early Postclassic (Feinman et al., 2015, figures 16 and 19). During the Postclassic period, northern exchange networks from the Valley of Oaxaca to the Mixteca Alta and central Mexico increased in significance. During the Late Postclassic, after Cantona’s fall, Zaragoza was even less important, replaced largely by other Gulf Coast FIELDIANA: ANTHROPOLOGY Monte Alban V Settlements in the Guirin/Albarradas Survey Region E17 N10 Sites smaller than 1.0 ha are marked by triangles. All others are drawn to size and shape. Contour interval: 500 m. Fics): and central Mexican sources (Table 7.3). Pachuca continued to be the dominant source in Oaxaca and the rest of western Mesoamerica, while Orizaba, another Gulf Coast source, surpassed Zaragoza (Feinman et al., 2015, figure 21). The increasing dominance of Pachuca obsidian reflects the growing importance of the Aztec empire and its expanding networks of interaction and exchange. Following the reorganization of the Early Postclassic period, the networks of interaction were reconstituted and had an entirely different structure during the Late Postclassic (Feinman et al., 2015, figures 19 and 22). The changes in the network graphs of obsidian distributions are consistent with increasing commercialization and economic relationships intensified through markets (Golitko & Feinman, 2015, p. 233) and the entanglements between marketplace transactions and politically driven tribute associated with empire (e.g., Berdan, 1977, 1985, 2003b). In Oaxaca, most of the obsidian from Pachuca arrived largely through northern routes, while obsidian from Orizaba seems to have arrived along the same routes from the Gulf Coast that Zaragoza obsidian did, passing through the Albarradas area. kilometers Monte Alban V settlements in the Guirtn/Albarradas survey region, showing sites mentioned in the text. Monte Alban V Settlements in Albarradas Settlement pattern change in the Albarradas area mirrors broader patterns of transition across most of the Valley of Oaxaca during the Late Postclassic period. In the Albarradas area, the number of settlements increased fivefold to 51, while the overall population more than doubled. Although the number of sites expanded at all levels of the site hierarchy, the greatest change was an explosion of small settlements or hamlets that were broadly dispersed across the landscape, in both defendable and less defendable locations (Fig. 7.3). Three- quarters of the Albarradas sites had 30 or fewer inhabitants; only two sites (Llego Yiachi and Hierve el Agua) had more than 1,000 inhabitants (Table 7.4). Most of the sites are dispersed across the lower eastern slopes of Cerro Guirone. A few small settlements were established or resettled in the small mountain valley north of Cerro Guirone, and six small sites were established in very defendable locations on both sides of the high mountain pass north of Cerro Guirone. From these locations, residents of these six sites could have monitored the FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 101] TABLE 7.3. Percentage of sourced obsidian at Late Postclassic sites in Oaxaca. El G. Other/ Subarea No. Chayal Victoria Orizaba Otumba Pachuca Paredon Tulancingo Ucareo Zacualtipan Zaragoza unknown Sierra Norte 47 — — 1] 4 40 — _ 26 6 13 — Etla 42 — — 12 7 — —- 2 = ct ad Central 116 l ] 3 — — — — — 3 ] W. Tlacolula 128 — — 64 2 — — oo — a — E. Tlacolula 541 — 2 5 4 64 —- — 4 DS 2 Ejutla 80 — ] 66 = 3] — -- —— _ ] — Mixteca 31 = = 55 13 — — — 13 6 Saat S. Oaxaca coast 1,138 — 1 54 2 42 eo _- 1 — — Southern Isthmus 35 — — 46 6 49 — _ — — — _— pass, a historic trade artery. In Albarradas, there continued to be little emphasis on craft production, as we noted chert working at only six sites and ceramic production at one. Most Monte Alban V sites in Albarradas were identified by the thin, hard-fired G-3M sherds that are ubiquitous on the surface of Late Postclassic sites in the Valley of Oaxaca (Table 7.5; Kowalewski et al., 1978, p. 177). Café sahumadors with narrow hollow tubes (Kowalewski et al., 1989, pp. 833-834) also were fairly common. At most sites, there were few other ceramic diagnostics of Monte Alban V. The only polychrome (Kowalewski et al., 1978, p. 193) we observed was in the sector (IB) of the Guirun site with the heaviest density of Postclassic pottery and where there are standing adobe walls (see Fig. 5.10). As in the Valley of Oaxaca, where approximately 20% of the TABLE 7.4. Monte Alban V sites in the Guirun/Albarradas survey region. Grid No. Site name Subarea* square Population 2 TL-SLA-SLA-10 AL N8E19 18 3. TL-SLA-SLA-11 AL N9E19 30 4 TL-SLA-SLA-13 AL N9E19 23 5 TL-SLA-SLA-14 AL N9E19 8 6 TL-SLA-SLA-15 AL N9E19 12 11 TL-SLA-SLA-25 AL N7E19 18 14. TL-SLA-SLA-29 AL N7E19 93 1S TL-SLA-SLA-30 AL N8E19 Di 16 TL-SLA-SLA-31 AL N8E20 32 17 TL-SLA-SLA-32 AL N8E20 Me 18 TL-SLA-SLA-33 AL N8E19 15 19 TL-SLA-SLA-34 AL N8E20 277 21 TL-SLA-SLA-36 AL N8E19 159 22 TL-SLA-SLA-37 AL N9E20 D3 23. TL-SLA-SLA-38 AL N9E20 131 25 TL-SLA-SLA-40 AL N9E20 14 27 TL-SLA-SLA-44 AL N9E20 is 32. TL-SLA-SLA-S0 AL N9E19 1S 34. TL-SLA-SLA-52 AL N7E20 i 37 TL-SLA-SLA-56 AL N7E20 8 38 TL-SLA-SLA-58 AL N7E20 8 43 TL-MIT-XA-36 ET N7E17 8 46 TL-MIT-CC-2 AL N9E18 IS 47 TL-MIT-CC-3 AT N9E18 8 49 TL-MIT-CC-6 AL NIOEI18 66 50 TL-SLA-SLA-43 AL N9E20 68 $51 TL-SLA-SLA-12 AL N9E19 83 $2. TL-SLA-SLA-S9 ET N7E18 203 53. TL-SLA-SLA-24 AL N7E19 1,458 54. TL-SLA-SLA-27 AL N7E19 8 55 TL-SLA-SLA-20 AL N8E20 8 56 TL-SLA-SLA-22 AL N8E19 oe 57 TL-MAT-MAT-95 (Nueve Puntas) ET N7E17 863 58 TL-SLA-SLA-S53 (Llego Yiachi) AL N7E20 2,191 59 TL-SLA-SLA-8 (EI Calvario) AL N8E19 8 60 TL-MIT-CC-5 AL NI0E18 60 61 TL-SLA-SLA-41 (El Trigo) AL N9E20 188 62 TL-SLA-SLA-17 (Guirun)*** El N8E18 2,828 * AL = Albarradas; ET = eastern Tlacolula. ** ch = chert; gs = ground stone; cer = ceramics. Mounds Mounds Mound Craft Exotics <1 m >1m_ volume (m*) Terraces work** present 0 0 0 0 — — 3 0. 0. >1 — — 0 0 0 3 — — 0 0 0 0 a obsidian 0 0 0 0. ch obsidian 0 0 0 0 — — 0 0 0 0 ch _ 0. 0 0 0 — — 0 0 0 0 — greenstone 0 0 0 0 — _ 0 0 0 0 — obsidian 0 1 345 0 — obsidian 0 0 0 0 — obsidian 0 0 (0) 0 oe -— 0 0 0 0. — obsidian 0 0 0 0 — — 1 0 0 0 — — D 0 0 0 — — 0 0 0 0 ch obsidian 0 0 0 0 — shell 1 0 0 0 — — 0 0 0 0 — obsidian 0 0 0 0 — obsidian 0 0 0 0 ch — 0 0 0 0. — obsidian 8 0 0 >1 — — 2 2 373 1] — — 13 0 93 23 — — 2 D 385 3 — obsidian 0 0 0 1 = — 0 0 0 1 — obsidian? 10 2 593 fi ch shell, obsidian 12 0 57 115 — — 58 11 4,759 sy emi obsidian 0 0 0 0 — obsidian? 0 0 0 8 — obsidian 6 1 5D 28 ch shell, obsidian 79 27 23,075 278 ch, gs_ obsidian, greenstone *** Craft work and exotic items recovered during regional and intensive site surveys. 102 FIELDIANA: ANTHROPOLOGY * * Bl atrataane- tna st = ao wo 5 OQ a ‘e nm an = AN > —= ° ouontn NNO CN > S Ae “ ee > nN SC eet at | ti S se CM ni eon | nam == = Fic. 7.4. Small, carved greenstone pendant from the highest hilltop of section HI at Guirin (TL-SLA-SLA-17). population had few goods beyond utilitarian ceramics (Kowa- = _ lewski & Finsten, 1983, table 2), many sites in the Albarradas area also had low artifactual diversity during the Postclassic, ae | | | {| especially the Late Postclassic. This seeming material cultural | “impoverishment” was more pronounced than ever before. Approximately 33% of the rural population (not including the two large centers of Hierve el Agua and Llego Yiachi) and 60% of all Late Postclassic sites in Albarradas had no artifacts other than chert from nearby local sources and a small assortment of ceramic varieties. Only at the Guirun site itself did we find such Sadao ONAI MU jereltc: Mie aealll | | tf artifacts as a small, carved stone pendant (Fig. 7.4). Patterns of settlement change at the two largest sites in the _ _ Albarradas area, Llego Yiachi and Hierve el Agua, were markedly different. Hierve el Agua expanded on the lower, less steep slopes of the Nueve Puntas—Cerro Guirone ridge system, where the site’s inhabitants had access to a large spring and the largest expanses of good, well-watered land on either side of the mountain ridge. The layout of the Late Postclassic occupation was relatively open and not particularly defensive (see Fig. 6.9). Sar uit bat! | Llego Yiachi, in contrast, is in one of the most defendable locations of any site in the region (see Fig 5.6). o Llego Yiachi was always an important site in the Albarradas area, first as part of a string of Late/Terminal Formative terrace sites above the Hormiga Colorada, not far from the modern Zapotec/Mixe boundary, and then as the only large site in the area during Monte Alban HIJA, when most of the Albarradas Monte Alban V ceramics at sites* in the Albarradas area. ea —— V-16 V-27 V-34 V-37 V-43 V-46 V-49 V-53 V-54 V-55 V-56 V-57 TABLE 7.5. V-2 V-5 V-I1 3 | ** Ceramics collected during Guirun site survey in 1996. s 5 nam o ama » ‘c= 8 3S Sa eee aS Oo — S =e ate neo See e325 2 ilove as OE Seer 2 en caeee 2 ma It ase = — — _ ar BS) es82eSs 8 5 BEF a Dn Ta aces iS LO f-s oy 29) BRS og Seer ° = 6-220 # — S28 Ss) Oc A |lee ees oes So esee sc its 2) SSESBSES4 SOS SFFAR OS IT E SCH NNO Dag, Ons Se Bt Oa RI ASTOSCLSRS AOR Aee [2 cooocoococoxTroHn — BHR RBBB BRANNAN RE S Bee ye ee tea Se Sa inn) Eee Ou . fs get: 05) 67) On om On OO oSSS3 232 % Fic. 7.5. Pre-Hispanic adobe wall at the top of Llego Yiachi (TL- ee Se ce Set es, OSTORO) Ora ag G G G G G G G SLA-SLA-S3). FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 103 Fic. 7.6. Lower ridgeline with archaeological remains descending into the clouds on the western side of Llego Yiachi (TL-SLA-SLA-S3). area was depopulated. The natural rock cliff is the most ceramics from Monte Alban Late I through V on the top of the prominent topographic feature in the area and is highly rock cliff in association with the architectural complexes, but defendable (see Fig. 2.26). But given the architectural complexes the standing adobe walls in one building (Fig. 7.5) resemble on the summit that include three small plazas and associated those at the Mitla Fortress (Feinman & Nicholas, 2004a, pp. buildings, it was always more than a military outpost. We found 44-45), and that building may have been constructed near the By Fic. 7.7. The western ridge of Llego Yiachi (TL-SLA-SLA-53), viewed from near the top of the site; the ball between the rock cliff and the western slope. x Geri: court is located in the saddle 104 FIELDIANA: ANTHROPOLOGY ie; Fic. 7.8. Remains of platform and plaza at the top of El Trigo (TL-SLA-SLA-41). end of the occupational history of the site, when it more than doubled in size. In Monte Alban V, Llego Yiachi spread beyond the top and steep eastern slope of the natural rock cliff to the long ridge on the west that slopes more gradually to the Hormiga Colorada (Fig. 7.6). Along this ridgetop, we mapped dozens of residential terraces, and dozens more were too badly eroded to measure. Some areas on the northern side of this ridge were steep and heavily vegetated, and we could not cover and check all areas (Fig. 7.7). Our estimate of 55 residential terraces 1s a minimum number; the actual number could be two or three times higher. Several small architectural complexes 20002 esiern acolul Semeee “=, Ty etme ~ — ‘ vege iy Ca ee a As, Lo } : a en TABLE 7.6. Percentage of population residing in terrace sites in Valley of Oaxaca subareas in Monte Alban V. No. of Terrace % of terrace site Total population in Subarea sites population population _ terrace sites Etla 3 460 15,404 3 Central 5 16,953 20,839 81 N. Valle Grande 4 7,558 24,938 30 S. Valle Grande 16 4,719 23,919 20 W. Tlacolula 15 10,542 41,255 26 E. Tlacolula 12 13,490 40,119 34 Ejutla 13 5,091 19,970 26 Albarradas 9 4,260 5,416 719 Sola 25 5,914 9,168 64 were dispersed along the ridgetop. On top of this ridge at the western base of the natural rock cliff is a small I-shaped ballcourt associated only with Monte Alban V pottery. The ballcourt location, outside the main civic-ceremonial complex at the site, corresponds to the location of two Late Postclassic ballcourts at the large hilltop terrace site of Coatecas Altas at the eastern edge of the Ejutla Valley (Feinman & Nicholas, 2013, figure [X.la). The ballcourts at these two sites likely were built in areas where “foreigners” could be brought to the site to engage in ballgame rituals. The placement of the ballcourts at these two sites in unusual settings at the edge of their settlements may have been a practice where members of the Monte Alban V ~ SAS a Pt v dP RNG C A caps ee % X 0 aN , ° SAN £ i ) 2 oe? Hormiga! <6 ¢g 7 “A Colorada,” tc (\ Sites smaller than 2.0 ha are marked by triangles. All others are drawn to shape and size. Contour interval: 500 m. Fic. 7.9. Monte Alban V settlements in Tlacolula and the Albarradas area, showing principal places mentioned in the text. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 105 Toy <> Pica 710: “visiting team” were not trusted to enter the community’s civic- ceremonial core. The next two largest Postclassic sites in the Albarradas area are TL-SLA-SLA-34 and El Trigo (TL-SLA-SLA-41). These two sites also are diametrically opposed in situation and character. TL-SLA-SLA-34 was a new community located on top of a small rock outcrop and surrounding flat terrain on the 106 roe re mee we AS nS ADO YS Be a o fu VS) A 0 PLAT, ge eae Nuéve Puntas 6 wT / . 4. Terrace sites O Defensive walls only N 0 10 a kilometers Monte Alban V terrace sites in the Central Valleys of Oaxaca. eastern slopes of the mountain ridge that is some of the best agricultural and well-watered land in the Albarradas area. On top of the outcrop was a large mound with a stone base that had been looted in the past. Like Hierve el Agua, the principal focus of this site may have been agricultural. There was no artifactual evidence of any craft activities. In contrast, El Trigo, like Llego Yiachi, had been occupied since Monte Alban Late I FIELDIANA: ANTHROPOLOGY TL-MAT;MAT-95 Fic. 7.11. but did not reach its greatest size until Monte Alban V, when all of the two dozen terraces at the site were occupied. At the center of the site is a small platform and plaza (Fig. 7.8). Like Llego Yiachi, it also sits in a more defendable location (see Fig. 3.26). Growth in the Albarradas area and on the high mountain ridgetop that defines the eastern end of the Valley of Oaxaca coincided with significant growth at Mitla in eastern Tlacolula (Fig. 7.9). Even the surveyed ridgetop between eastern Tlacolula and the Albarradas area had 11 sites, eight of which were very small isolated residences. The Guirtn site continued to be one of the largest communities in eastern Tlacolula, surpassed only by Mitla. In Monte Alban V, it extended farther up the slope closer to the base of Cerro Guirone and also to the north to incorporate the La Cuadrada stone quarry (see Fig. 7.3). Stone from this quarry clearly was used in the cruciform tomb at Guirtn that Saville (1900, 1909) first investigated a L4171 Aas Tg Di BO od © Residential terrace f& Mound or platform 7 Wall Z Ditch 0 50 hh =m Site map of Nueve Puntas (TL-MAT-MAT-95). century ago; the quarry also was one of several in the area from which stone was obtained for palace construction at Mitla (Robles Garcia, 1994). Although the spatial extent of the site expanded in Monte Alban V, much of the occupation appears to have been dispersed (low density) so that the estimated population did not increase. At Guirun, low stone walls blocked many of the more accessible approaches to the site, including the north side of the stone quarry. The sectors of the site that were newly occupied in Monte Alban V tended to be more defendable than the parts that were settled earlier. The site was always situated to monitor/control the principal trade route into eastern Tlacolula (which passes just south of Cerro Guirone); with its northern expansion to the quarry in Monte Alban V, it also had view of another route that passes through the small mountain valley on the north side of Cerro Guirone. Guirun clearly had strong ties FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 107 Fic. 7.12. walls at Nueve Puntas (TL-MAT-MAT-95). to Mitla and was a node of interaction with the Albarradas area. Monte Alban V Terrace and Defensive Sites Terrace sites continued as densely settled features of the Late Postclassic landscape in the Central Valleys of Oaxaca. Throughout highland Oaxaca, Monte Alban V sites often were defensively situated, especially in marginal or peripheral areas such as Ejutla (Feinman & Nicholas, 1990), Pefoles (Kowa- lewski, 1991), the Sierra Norte (Drennan, 1989), and Albarra- das; at the same time, multitudes of new communities were distributed across the valley floor and in other less defendable locations so that lower proportions of the total population resided in the hilltop settlements (Table 7.6). Hilltop sites continued to be most prevalent in the eastern and southern arms of the valley, extending into the Albarradas area to the east and into the Ejutla and Sola Valleys to the south (Fig. 7.10). These sites also may have served as periodic redoubts for select residents of more open and less defendable settlements. Defensive posture and practice differed from earlier eras when the defense of large, compact communities may have been a more broadly shared civic responsibility of community resi- dents. nse Aint se Fe ; 2G = Fic. 7.13. Stone-lined ramped entryway (covered with leaves) to terrace at Nueve Puntas (TL-MAT-MAT-95). 108 Fic. 7.14. Rock cliff on the southeastern side of Nueve Puntas (TL-MAT-MAT-95). Nevertheless, the persistence of hilltop terrace settlements, in defendable locations, indicates a continued concern with defense. Many of the Monte Alban V terrace sites are situated at the edges of their respective valleys. The residents of these sites may have been defending themselves from neighbors with whom they often traded but also from groups living in mountainous areas outside the valley. In Monte Alban V, settlements spilled over from the Valley of Oaxaca into neighboring areas. The physiographic limits of the Valley of Oaxaca were in many ways more permeable and open at the same time that intraregional relations continued to be factionalized, contested, and antagonistic (Feinman & Nich- olas, 1999, p. 185). Many of the larger, defendable settlements faced inward into the valley, an indication of regional political decentralization and contestation (Elam, 1989). Early explorers reported at least four fortresses, or al/barradas, in the vicinity of Mitla (Horcasitas & George, 1955, p. 23). The possible sites that are referred to are found on both sides of the mountain ridge, in eastern Tlacolula and Albarradas. Two new sites—Nueve Puntas and Buenavista—were estab- lished on the ridgetop between eastern Tlacolula and the Albarradas area. Both are remotely situated terrace sites on the highest spots along the ridge, with excellent views to the west (inside the valley) and east (outside the valley). Nueve Puntas consists of 115 residential terraces and several long defensive walls constructed on the top and western face of three of the nine points that gives the mountain ridge its name (Fig 7.11). Parts of the walls are still 2-3 m tall today. Several deep ditches also limited access to the terraced areas. Because the site has never been plowed, the terraces are very well preserved, with house foundations and entry ramps visible on many of them (Figs. 7.12 and 7.13). The opposite, eastern side of Nueve Puntas is a stone cliff that drops more than 500 m (Fig. 7.14). Buenavista is located on top of a steep-sided peak on the long, narrow ridge between Nueve Puntas and El Tigre (Fig. 7.15), about halfway between Guirun and Nueve Puntas. The site consists of two small platforms, each associated with one small structure. One is on the highest point of the ridge and the other about 100 m down the ridge to the south. We mapped 21 small, dispersed terraces on both sides of the peak, and there were visible house foundations on and near some of the terraces (Fig. 7.16). Given the small size, remote location, and excellent visual vantage of this site, it likely was a military outpost or redoubt. FIELDIANA: ANTHROPOLOGY Fic. 7.15. Although fewer than a quarter of Monte Alban V settlements in the Albarradas area were terrace sites, more than 75% of population resided in these hilltop communities, the largest of which was Llego Yiachi. Many other smaller sites also were in defendable hilltop locations. Unlike the earlier defendable, often fortified sites in the Albarradas area in Monte Alban I and II, the defensively located sites in Monte Alban V were not part of multisettlement linear arrangements. There were no clear gaps in settlement or a single, aligned frontier. The inhabitants of many of the terraced and other defendable sites on the eastern slopes of Cerro Guirone in the Albarradas area, as in the valley, likely were protecting themselves as much from neighbors as from people from outlying areas. But others, especially at the northern edge of the Albarradas survey, were small and defensively located, often with several small structures or small numbers of terraces (like Buenavista). Most of these Monte Alban V sites were on ridgetops where they could monitor travel through the area, especially the northern route into the eastern arm of the Valley of Oaxaca, similar to what we had observed for the Ejutla Valley at the southern edge of the Central Valleys of Oaxaca. In the Ejutla Valley, many of the small terrace sites at the valley’s edge appear to have been positioned to monitor/guard the southern and eastern borders of the Ejutla Valley. Two sets of small defendable sites—two on the south side of the pass and two on the north—were in positions to monitor the principal northern route into the eastern arm of the Valley of Oaxaca. Two small sites are located on El Portazuelo, a narrow mountain ridge located just south of the saddle where the modern highway to Mitla from the Mixe region passes. We mapped one terrace and three small cut stone structures at TL- SLA-SLA-11 (Fig. 7.17). There were several short walls on one slope that appeared to be defensive. The site had an excellent view to both the east and the west and appears to have been View of Buenavista (TL-SLA-SLA-59), the terraced hilltop on the ridgeline at center top of image. positioned to monitor this pass. The other site (TL-SLA-SLA- 12), on the highest point of El Portazuelo, consisted of 11 terraces and four small platforms and structures of cut stone blocks, all in linear alignment with the slope (Fig. 7.18). Downslope from this site is another very small site with only three terraces above a low defensive wall (TL-SLA-SLA-13; Pig) 19): On the mountain ridge just north of the pass and El Portazuelo are two small defendable sites. TL-SLA-SLA-50 consists of two small stone house foundations. At TL-SLA- SLA-51, cut stone foundation walls of one structure are visible on the surface. Both sites have great views of the surrounding area, including into the eastern end of the Tlacolula Valley. These four sites were new occupations established during the Postclassic, but to the east is another very defendable site above the pass that had been occupied since the Late Classic. TL-SLA-SLA-43 stretches along the top of a long narrow ridge between two deep barrancas near the northern edge of the surveyed area (see Fig. 6.8). The remains of eight dispersed house foundations and one terrace are visible on the surface. Settlement and Land Use in Monte Alban V In spite of tremendous demographic growth as populations in the Central Valleys of Oaxaca reached their pre-Hispanic maxima during the Late Postclassic period, archaeological populations remained far below regional capacities. Maize yields continued to increase, and populations were more evenly dispersed across the region. The inhabitants of most valley subareas could have grown more than twice the amount of food they needed to sustain themselves (Table 7.7). Even populations FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 109 TL-SLA-SLA-59 Fic. 7.16. Site map of Buenavista (TL-SLA-SLA-S59). in the Central sector were self-sufficient following the decline of Monte Alban and the shift to a more dispersed settlement pattern there. The only exception in the Valley of Oaxaca was eastern Tlacolula. Beginning in the Early Classic period, the population of eastern Tlacolula was at risk of food shortages. During the Late Classic and Postclassic, the highest population densities were in Tlacolula, especially in the easternmost part of that valley arm, and the risk of food imbalances continued (Fig. 7.20). Yet these risks were not as pronounced as they had been during the Early Classic, when the first large hilltop terrace sites were established in eastern Tlacolula. Many of the Early Classic sites were located in areas where a range of natural resources were available that could be modified into goods for exchange. The varied commercial networks that connected Tlacolula to different parts of the region were integral parts of the economy that allowed these communities to be situated in less agriculturally productive places where they produced goods to exchange for food grown elsewhere. At the edges of the valley system, settlement patterns changed in different ways that affected how much produce farmers could grow above their own needs. During the Postclassic, large settlements in Sola were situated along the principal transport corridor and not near the largest patches of good farmland (Balkansky, 2002, p. 71). In contrast to most of the Central Valleys of Oaxaca, excluding eastern Tlacolula, population in 110 the Sola Valley was more concentrated in fewer large sites in areas of less productive farmland. As a consequence, even though the population was higher and potential per-hectare ylelds were higher than in the Late Classic, fewer people could have been sustained by the available labor farming near their settlements, and potential surpluses were lower than they had TABLE 7.7. Estimated archaeological, resource-based, and labor- based potential populations for Valley of Oaxaca subareas in Monte Alban V. Estimated Resource-based Labor-based archaeological potential potential Subarea population population* population** Etla 15,404 136,700 73,400 Central 20,839 62,300 43,900 N. Valle Grande 24,938 136,300 54,900 S. Valle Grande 23,919 119,700 58,800 W. Tlacolula 41,255 105,900 77,000 E. Tlacolula 40,119 46,500 38,400 Ejutla 19,970 112,100 52,100 Albarradas 5,416 20,100 11,800 Sola 9,168 42,300 14,000 * Based on full exploitation of all agricultural land resources in each subarea. ** Based on available labor supply and equal use of Class I and Class II land near recorded settlements. FIELDIANA: ANTHROPOLOGY TL-SLA-SLA-11 iG ial a been during the Late Classic (Balkansky, 2002, pp. 75—76, figure 2.7). Exchange networks following the transport corridor between communities in Sola and beyond likely buffered the need for populations to situate on good farmland. In addition to potential food surpluses produced in neighboring areas, proximity to these networks may have been attractive for providing access to a wider range of goods. These networks of exchange were essential during dry years when populations in both Sola and Ejutla could have suffered food shortages. In contrast, the Albarradas population was more dispersed, including numerous sites, some large like Hierve el Agua, on the flatter, well-watered eastern slopes of Cerro Guirone, where large surpluses were possible. Even in low-rainfall years, these areas could have produced some surplus food to exchange with other communities. The large communities in eastern Tlacolula continued to depend on exchange for agricultural produce grown elsewhere in the valley and likely traded for agricultural surpluses from both western Tlacolula and the Albarradas area (Fig. 7.21). Even if hostilities with Macuilxochitl in western Tlacolula during the Late Postclassic (Oudijk, 2008) interfered with exchange networks across Tlacolula, the Albarradas area alone could have grown and traded sufficient crops to populations in eastern Tlacolula to overcome potential short- Site map of TL-SLA-SLA-11. falls. Not only does the population of Albarradas in the Late Postclassic period appear to have had a different relationship with more central parts of the valley than did other edge regions (Sola and Ejutla), but the nature of flows between Albarradas and eastern Tlacolula may have shifted compared to earlier, with the former serving as a kind of “tortilla-basket” for the latter. Monte Alban V Summary When the Spanish arrived in highland Mesoamerica in AD 1520, they encountered a factionalized but heavily commercial- ized landscape. Populations had grown to their highest pre- Hispanic levels; yet, except for the late emergence of the Aztec empire centered on Tenochtitlan, most of the population had reorganized into smaller competing polities following the decline of the large Classic period centers. In the resulting power vacuum and disruption of networks, small, semiauton- omous polities proliferated across the landscape as local elites seized the opportunity to assert their independence and form new networks of interaction. These competing polities were FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 111 TL-SLA-SLA-12 2240 m Fic. 7.18. largely reliant on external resources, trade goods, and the spoils of war. These smaller polities competed with each other for followers, influence, and goods. Market exchange expanded, and by the end of the Postclassic there was more exchange of goods flowing in all directions across the highlands and all of Mesoamerica. These factionalized polities had permeable boundaries, were economically interdependent, and formed weak alliances with one another that did not always hold together for long. In the Valley of Oaxaca, the Tlacolula arm was the most densely Fiaw 19: Site map of TL-SLA-SLA-13. Site map of TL-SLA-SLA-12. settled and commercialized region. Textual accounts tell of shifting alliances throughout the Postclassic period. Yagul and Mitla were the power centers in the Early Postclassic. By later in the Postclassic period, power had been wrested from Mitla and Yagul by a new confederation of Macuilxochitl in western Tlacolula, Zaachila in northern Valle Grande, and ruling families from the Mixteca Alta. With the new confederation, there was greater Mixtec presence, especially in northern and western parts of the Valley of Oaxaca. Mixtec presence is much less evident in other parts of the valley. Archaeological population as % of subarea maximum potential population Fic. 7.20. Monte Alban V archaeological population as a percentage of the maximum potential population in each Valley of Oaxaca subarea. FIELDIANA: ANTHROPOLOGY Monte Alban V 6,400 Surplus rounded to nearest 100 Fic. 7.21. Monte Alban V potential agricultural surplus for each Valley of Oaxaca subarea. Although Mitla had somewhat declined by the Late Postclassic, it remained an important economic and religious center (Appel, 1982). Yet the foundations of Mitla’s economy were never farming; they always were based on craft production and exchange. Acrimonies with Macuilxochitl and the center of the valley may have encouraged the populations of eastern Tlacolula to focus more on exchange networks to the east. Interdependencies with the Albarradas area may have been strengthened, and that region experienced a significant prolif- eration of settlement on the well-watered eastern slopes of Cerro Guirone, where large surpluses were possible. Perhaps, as a result of these shifts, the economic ties and flows between Albarradas and neighboring eastern Tlacolula were reconfig- ured during the Postclassic period in ways not evidenced earlier in the pre-Hispanic sequence. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 113 be > CSS ter War ; 7? if ‘veth curries 9° ° Gena LY (TV) tore hy Ste vy @reh batt aay an aa — yt — : ee ae ae 7 mt yy | nie a en 7 “te ; a ' hh it (ovr bP >, 31. teh Cee Jayary ; o- fe - : 7 ; pe = - (meg utes r | an { regtaelnd aan a Chapter 8: Conclusions “A spectre is haunting our time: the spectre of the short term.” (Guldi & Armitage, 2014, p. 1) Today, the Albarradas area is demographically, politically, and economically (in terms of basic production and domestic wealth) peripheral to the Valley of Oaxaca. Although we were conducting fieldwork there in the mid-1990s, local officials ignored the directive from the federal and state governments to observe daylight savings time and conducted all local business on /a hora de dios. Tourists arrive to swim in the hot springs and enjoy the vistas of calcified waterfalls at Hierve de Agua; most travelers just pass through the area on their way to major market towns in the Mixe region and beyond. But for locals, the mountain ridge is hardly a barrier. Many residents of this area have relatives who live “on the other side of the mountain,” and there is considerable travel between communities on both sides of the rugged terrain that separates Albarradas from eastern Tlacolula. These cross-mountain communications, intermar- riages, market participation, and even basketball tournament competitions take place although the only paved road through the Albarradas area (until the recent construction of a new superhighway to the coast) traverses the mountain pass in the northern part of the survey region. After a dispute between San Lorenzo Albarradas and Roaguia, not long after our survey, precipitated difficult passage on the road between these two communities, Roaguia, on the eastern side of the ridge, got together with Xaaga, on the western side, to build a new, narrow gravel road over a saddle in the ridge south of Cerro Guirone; the new road directly connects these communities so that travelers can avoid using the contested stretch. Although imposing, the mountain ridge that physiographi- cally defines eastern Tlacolula was not an impenetrable cultural, economic, or political boundary throughout the pre-Hispanic (and more recent) history of the area. During the Middle Formative, when first San José Mogote and then Monte Alban were the largest settlements in the valley, their leaders did not exert extensive political control. Certainly, they did not tax, directly govern, or service the Albarradas region. Nevertheless, the spatial limits of these early polities were not firmly demarcated or defined, and through networks of interaction, ceramics distinctly within the Valley of Oaxaca tradition are found across the entire valley system, including far south in Ejutla and on the eastern side of the mountains in Albarradas. Only during the Early Classic period, when the large Guirun site was established on the western slopes of Cerro Guirone while the eastern slopes were largely abandoned, was there spatial overlap between the political boundary with a physio- graphic one, yet it did not prevent interaction and interchanges between the Valley of Oaxaca and areas to the east. Even today, the boundaries of modern administrative districts do not follow the top of the ridgeline, and some of the terrain on the western side of the ridge, including major archaeological ruins in eastern Tlacolula, is administratively part of San Lorenzo Albarradas, situated on the eastern side of Cerro Guirone. The ethnic boundary between Zapotec and Mixe speakers today (and probably in the past) is farther east, beyond the Hormiga Colorada and closer to the Mixe region. Across the Valley of Oaxaca, cultural and political boundaries generally do not match physiographic ones, today or in the past. Here, we have raised the key issue of why people choose to settle where they do. The people who chose to settle near Monte Alban likely perceived advantages to their well-being by residing close to the regional capital, especially a more collectively organized capital (as in Monte Alban I, IIIA, and early IIIB-IV) that provided a range of goods and services (including a relatively pacific environment) in exchange for allotments of agricultural surplus and labor. Settlements in the Albarradas area, especially early on when eastern Tlacolula was less densely settled than most of the rest of the valley, likely were established by people who perceived the benefits of proximity to principal trade routes, even as the location of their communities in defendable locations in the rough terrain near the Hormiga Colorada indicates they were concerned about potentially hostile neighbors. The quality of agricultural land does not appear to have been the overriding factor in these early settlement decisions (although clearly it must have been taken into account). The early Formative era residents of Albarradas had access to goods that were traded along the travel route, including ceramics from the Valley of Oaxaca and goods from coastal areas, such as seashell ornaments. Given the location of many of the early sites in settings with limited agricultural options, they likely made goods for exchange from local resources, such as tools made from local chert sources. Fibers (ixtle) from xerophytic plants also were a potential trade item. Into the modern era, people from villages in the Albarradas area have continued to bring such goods to markets in Tlacolula. As populations expanded during the Early Classic and Tlacolula became one of the more densely settled areas in the valley system, much of the earlier settlement in Albarradas was abandoned as people moved to the Guirun site, on the western side of the mountain ridge, far from highly productive agricultural land. Tlacolula is in a rain shadow and the driest part of the valley, so this move reflects social/political factors, possibly the increasing importance of markets. Populations in the better-watered areas of the valley, Etla and Valle Grande, focused more heavily on agricultural surpluses that could be exchanged for goods produced elsewhere in the valley. With the fall of Monte Alban, the developing landscape of competitive cacicazgos was characterized by leaders who adopted more exclusionary political strategies. With the increasing demographic centrality of eastern Tlacolula, there was a shift in settlement in the Albarradas area to locations closer to better agricultural land. There are few opportunities for anything but rainfall farming in the Albarradas area. But the area does receive considerably more rainfall than eastern Tlacolula. During the Late Postclassic period, there are areas on the lower, flatter slopes on the eastern side of Cerro Guirone where significant agricultural surpluses could have been grown as long as the labor supply was adequate. As hostilities with Macuilxochitl emerged, the occupants of large communities in FIELDIANA: ANTHROPOLOGY N.S., NO. 46, NOVEMBER 13, 2017, PP. 115-116 115 tern Tlacolula, especially Mitla, may have encouraged this economic interaction to provide more reliable sources of agricultural produce. Important Postclassic tombs (Saville, 1900, 1909) that were constructed at both Xaaga and the Guirtn site (both situated between Mitla and the Albarradas area) likely reflect the important socioeconomic networks that linked eastern Tlacolula and Albarradas at that time. These diachronic relations, which were not static over time and do not simply reflect the different physiographic bases of distinct parts of the valley, reaffirm our perspective that marketplace exchange was a key feature of the Oaxacan economy from, at least, the Middle Formative period (Beals, 1975; Feinman et al., 1984; Feinman & Garraty, 2010; Feinman & Nicholas, 2010). Material identities (most evident in pottery) were shared across landscapes far broader than polity boundaries, and settlements were not established nor did they grow in strict accord with agrarian considerations. Economic interdependence, as opposed to domestic self-sufficiency, has long been the predominant adaptation in highland Mesoamer- ica (Feinman & Nicholas, 2004b, 2012; Feinman, 2006). Throughout the pre-Hispanic sequence in highland Oaxaca, settlement continuity was greater within the Valley of Oaxaca than at (or just beyond) its physiographic limits. Fewer sites in outlying areas had long, continuous occupation compared to sites inside the valley. Settlement fluctuation over time was greater near the edges of the valley (and those areas just beyond). Networks from distant areas flowed through these edges into the Valley of Oaxaca, and communications and flows from the urban capital and other large communities dissemi- nated out. The northern, Etla arm was the most affected by events in central Mexico as first Teotihuacan and then Tenochtitlan became the largest cities in Mexico, influencing regions to their south. The small highland valleys in the Mixteca and the Sierra Norte often reflected changes that occurred in Etla. Ejutla, far to the south, was somewhat less affected by such extraregional links. Throughout most of the sequence, settle- ment trends there were similar to those in neighboring southern Valle Grande. But Ejutla did begin to recover from the population declines at the end of the Classic period earlier than did the Valle Grande. Southern trade routes into the Valley of Oaxaca pass through Ejutla, and settlement pattern changes there may reflect the increased importance of the southern trade route during the Postclassic. Sola and Miahuat- lan always were more peripheral to the Valley of Oaxaca and 116 experienced fewer large swings in population over time. Throughout most of the pre-Hispanic sequence after the initial establishment of Monte Alban, the Albarradas area, to the east, was often more densely settled than any other edge area except during the Early Classic. With the close links between Albarradas and eastern Tlacolula, the former also experienced significant settlement discontinuities. Even though the Albarradas area is small and was never as populous as the Valley of Oaxaca, our research there has informed and refined our perspective on boundaries and interaction across the larger region. The boundaries of the Monte Alban polity (or even its spheres or degrees of control) were never static or impenetrable, nor did they conform well to physiographic features or other social and cultural boundaries. Our survey of the Albarradas area, where important trade routes pass on their way to the Valley of Oaxaca, has yielded new perspectives on these interregional exchange routes and their effects on settlement decisions on both sides of the high mountain ridge at the eastern edge of the Valley of Oaxaca. At times, eastern Tlacolula appears to have been more strongly connected to Albarradas than to the center of the valley; the physiographic boundary was no barrier to interactions with settlements in the Albarradas area that were situated to serve as conduits for the flow of goods and resources from coastal and other faraway areas into the Tlacolula arm of the valley. In sum, our cross-time examination of settlement change in Albarradas has illustrated that in the Central Valleys of Oaxaca, environmental differences or constraints are not infallible predictors of population densities, political boundar- ies, or even the specific ways that a landscape is used by the people who lived there. Rather, the traditions of people (and, crucially, the relations between them), often across broad spatial realms, helped impact and craft how local inhabitants constructed and used their spaces. These people—landscape interactions were anything but static, which reflects that the people-to-people interactions (cooperative and competitive) across communities, regions, and even wider geographic domain shifted over time. Interdependence, community, mar- ketplace exchanges, collective forms of governance, and competition between neighbors were all key elements that affected how pre-Hispanic peoples in highland Oaxaca met these challenges. These institutions and ways of cooperating leave their legacy in the lifeways and adaptations of the people of Oaxaca still today. FIELDIANA: ANTHROPOLOGY Literature Cited ABASCAL, R., G. HARBOTTLE, AND E. V. Sayre. 1973. Correlation between terra cotta figurines and pottery from the Valley of Mexico and source clays by activation analysis, pp. 81-99. In Beck, C. W., ed., Archaeological Chemistry. Advances in Archaeological Chemistry 138. American Chemical Society, Washington, DC. ABRAMS, E. M. 1994. How the Maya Built Their World: Energetics and Ancient Architecture. University of Texas Press, Austin. ApvAmMs, R. McC. 1977. World picture, anthropological frame. American Anthropologist, 79: 265-279. AHLQUIST, J. S., AND M. Levi. 2010. Leadership: What it means, what it does, and what we want to know about it. Annual Review of Political Science, 14: 1—24. ANAWALT, P. R. 1981. Indian Clothing Before Cortes: Meso- american Costumes from the Codices. University of Oklahoma Press, Norman. ApPEL, J. 1982. The Postclassic: A summary of the ethnohistoric information relevant to the interpretation of Late Postclassic settlement pattern data, the central and Valle Grande survey zones, pp. 139-148. Jn Blanton, R. E., S. A. Kowalewski, G. M. Feinman, and J. Appel, Monte Alban’s Hinterland, Pt. I: The Prehispanic Settlement Patterns of the Central and Southern Parts of the Valley of Oaxaca, Mexico. Memoirs No. 15. Museum of Anthropology, University of Michigan, Ann Arbor. ARNOLD, P. J., C. A. Poot, R. R. KNEEBONE, AND R. S. SANTLEY. 1993. Intensive ceramic production and Classic-period political economy in the Sierra de los Tuxtlas, Veracruz, Mexico. Ancient Mesoamerica, 4: 175-191. BALKANSKY, A. K. 1997. Archaeological Settlement Patterns of the Sola Valley, Oaxaca, Mexico. Ph.D. dissertation, Depart- ment of Anthropology, University of Wisconsin, Madison. 1998a. Urbanism and early state formation in the Huamelulpan Valley of southern Mexico. Latin American Antiquity, 9: 37-67. 1998b. Origin and collapse of complex societies in Oaxaca (Mexico): Evaluating the era from 1965 to the present. Journal of World Prehistory, 12: 451-493. . 2002. The Sola Valley and the Monte Alban State: A Study of Zapotec Imperial Expansion. Memoirs No. 36. Museum of Anthropology, University of Michigan, Ann Arbor. . 2006. Surveys and Mesoamerican archaeology: The emerging macroregional paradigm. Journal of Archaeological Research, 14: 53-95. BALKANSKY, A. K., S. A. KOWALEWSKI, V. PEREZ RODRIGUEZ, T. J. PLUCKHAHN, C. A. SmitH, L. R. Stiver, D. BELIAEV, J. F. CHAMBLEE, V. Y. HEREDIA ESPINOZA, AND R. SANTOS PEREZ. 2000. Archaeological survey in the Mixteca Alta of Oaxaca, Mexico. Journal of Field Archaeology, 27: 365-389. BALKANSKY, A. K., V. PEREZ RODRIGUEZ, AND S. A. KOWALEWSKI. 2004. Monte Negro and the urban revolution in Oaxaca, Mexico. Latin American Antiquity, 15: 33-60. BALL, H. G., AND D. L. BROCKINGTON. 1978. Trade and travel in prehispanic Oaxaca, pp. 107-114. In Lee, T. A., and C. Navarrete, eds., Mesoamerican Communication Routes and Cultural Contacts. Papers No. 40. New World Archaeological Foundation, Brigham Young University, Provo, UT. BARBER, S. B., AND A. A. Joyce. 2006. When is a house a palace? Elite residences in the Valley of Oaxaca, pp. 211-255. In Christie, J. J., and J. Sarro, eds., Palaces and Power in the Americas: From Peru to the Northwest Coast. University of Texas Press, Austin. Barth, F. 1969. Introduction, pp. 9-38. /n Barth, F., ed., Ethnic Groups and Boundaries. Little, Brown, Boston. BEALS, R. L. 1973. Ethnology of the Western Mixe. Cooper Square Publishers, New York. . 1975. The Peasant Marketing System of Oaxaca, Mexico. University of California Press, Berkeley. . 1979. Economic adaptations in Mitla, Oaxaca, pp. 165— 193. In Dalhgren, B., ed., Mesoamérica: Homenaje al Doctor Paul Kirchhoff. Instituto Nacional de Antropologia e Historia, Mexico City. BEEKMAN, C. S., AND A. F. CHRISTENSEN. 2003. Controlling for doubt and uncertainty through multiple lines of evidence: A new look at the Mesoamerican Nahua migrations. Journal of Archaeological Method and Theory, 10: 111-164. BERDAN, F. F. 1977. Distributive mechanisms in the Aztec economy, pp. 91-101. Jm Halperin, R., and J. Dow, eds., Peasant Livelihood: Studies in Economic Anthropology and Cultural Ecology. St. Martin’s Press, New York. . 1985. Markets in the economy of Aztec Mexico, pp. 339— 367. In Plattner, S., ed., Markets and Marketing. University Press of America, Lanham, MD. . 1987. Cotton in Aztec Mexico: Production, distribution, and uses. Mexican Studies/Estudios Mexicanos, 3: 235-262. . 2003a. Borders in the eastern Aztec empire, pp. 73-77. In Smith, M. E., and F. F. Berdan, eds., The Postclassic Mesoamerican World. University of Utah Press, Salt Lake City. . 2003b. The economy of Postclassic Mesoamerica, pp. 93— 95. In Smith, M. E., and F. F. Berdan, eds., The Postclassic Mesoamerican World. University of Utah Press, Salt Lake City. BERDAN, F. F., S. KEpEcs, AND M. E. Situ. 2003. A perspective on Late Postclassic Mesoamerica, pp. 313-317. Jn Smith, M. E., and F. F. Berdan, eds., The Postclassic Mesoamerican World. University of Utah Press, Salt Lake City. BERDAN, F. F., AND M. E. Smitu. 2003. The evolution of a core zone: The Basin of Mexico, pp. 238-242. Jn Smith, M. E., and F. F. Berdan, eds., The Postclassic Mesoamerican World. University of Utah Press, Salt Lake City. BERNAL, I. 1965. Archaeological synthesis of Oaxaca, pp. 788- 813. In Willey, G. R., ed., Handbook of Middle American Indians, Vol. 3: Archaeology of Southern Mesoamerica, Pt. 2. University of Texas Press, Austin. BERNAL, I., AND L. GAmio. 1974. Yagul, el palacio de los seis patios. Serie Antropologica 16. Instituto de Investigaciones Antropologicas, Universidad Nacional Autonoma de México, Mexico City. BLANTON, R. E. 1972. Prehispanic Settlement Patterns of the Ixtapalapa Peninsula Region, Mexico. Occasional Papers No. 6. Department of Anthropology, Pennsylvania State Universi- ty, University Park. . 1978. Monte Alban: Settlement Patterns at the Ancient Zapotec Capital. Academic Press, New York. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 11y . 1983. Urban Monte Alban during period III, pp. 128— 131. Jn Flannery, K. V., and J. Marcus, eds., The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civilizations. Academic Press, New York. ——. 1989. Continuity and change in public architecture: Periods I through V of the Valley of Oaxaca, Mexico, pp. 409— 447. In Kowalewski, S. A., G. M. Feinman, L. Finsten, R. E. Blanton, and L. M. Nicholas, Monte Alban’s Hinterland, Pt. II: Prehispanic Settlement Patterns in Tlacolula, Etla, and Ocotlan, the Valley of Oaxaca, Mexico. Memoirs No. 23. Museum of Anthropology, University of Michigan, Ann Arbor. . 1998. Beyond centralization: Steps toward a theory of egalitarian behavior in archaic states, pp. 135-172. Jn Feinman, G. M., and J. Marcus, eds., Archaic States. School of American Research Press, Santa Fe, NM. . 2005. The achievements of and prospects for survey archaeology, pp. 295-301. Jn Blanton, R. E., ed., Settlement, Subsistence, and Social Complexity: Essays Honoring the Legacy of Jeffrey R. Parsons. Cotsen Institute of Archaeology, University of California, Los Angeles. . 2015. Theories of ethnicity and the dynamics of ethnic change in multiethnic societies. Proceedings of the National Academy of Sciences of the United States of America, 112: 9176-9181. . 2016. How Humans Cooperate: Confronting the Challenge of Collective Action. University Press of Colorado, Boulder. BLANTON, R. E., AND L. FARGHER. 2008. Collective Action in the Formation of Pre-Modern States. Springer, New York. . 2010. Evaluating causal factors in market development in premodern states: A comparative study, with critical comments on the history of ideas about markets, pp. 207-226. Jn Garraty, C. P., and B. L. Stark, eds., Archaeological Approaches to Market Exchange in Ancient Societies. University Press of Colorado, Boulder. . 2011. The collective logic of pre-modern cities. World Archaeology, 43: 505-522. BLANTON, R. E., L. F. FARGHER, AND V. Y. HEREDIA ESPINOZA. 2005. The Mesoamerican world of goods and its transforma- tions, pp. 260-294. In Blanton, R. E., ed., Settlement, Subsistence, and Social Complexity: Essays Honoring the Legacy of Jeffrey R. Parsons. Cotsen Institute of Archaeology, University of California, Los Angeles. BLANTON, R. E., AND G. M. FEINMAN. 1984. The Mesoamerican world-system: A comparative perspective. American Anthro- pologist, 86: 673-682. BLANTON, R. E., G. M. FEINMAN, S. A. KOWALEWSKI, AND L. M. NicHo as. 1999. Ancient Oaxaca. Cambridge University Press, Cambridge. BLANTON, R. E., G. M. FEINMAN, S. A. KOWALEWSKI, AND P. N. PEREGRINE. 1996. A dual-processual theory for the evolution of Mesoamerican civilization. Current Anthropology, 37: 1-14. BLANTON, R. E., S. A. KOWALEwski, G. M. FEINMAN, AND J. AppEL. 1982. Monte Alban’s Hinterland, Pt. I: The Prehispanic Settlement Patterns of the Central and Southern Parts of the Valley of Oaxaca, Mexico. Memoirs No. 15. Museum of Anthropology, University of Michigan, Ann Arbor. BLANTON, R. E., S. A. KOWALEWSKI, G. M. FEINMAN, AND L. M. FINSTEN. 1993. Ancient Mesoamerica: A Comparison of Change in Three Regions, 2nd ed. Cambridge University Press, Cambridge. BRASWELL, G. E. 2003. Obsidian exchange spheres, pp. 131-158. In Smith, M. E., and F. F. Berdan, eds., The Postclassic 118 Mesoamerican World. University of Utah Press, Salt Lake City: Bray, W. 1972. The city-state in central Mexico at the time of Spanish conquest. Journal of Latin American Studies, 4: 161— 185. Brito DE Marti, E. 1982. Almanaque de Oaxaca 1982. Almanaque de Mexico, Mexico City. BROCKINGTON, D. L. 1973. Archaeological Investigations at Miahuatlan, Oaxaca. Publications in Anthropology No. 7. Vanderbilt University, Nashville, TN. ByLAND, B. 1980. Political and Economic Evolution in the Tamazulapan Valley, Mixteca Alta: A Regional Approach. Ph.D. dissertation, Department of Anthropology, Pennsylvania State University, State College. ByLAND, B. E., AND J. M. Pour. 1990. Mixtec landscape perception and archaeological settlement patterns. Ancient Mesoamerica, 1: 113-131. . 1994. In the Realm of 8 Deer: The Archaeology of the Mixtec Codices. University of Oklahoma Press, Norman. CaRBALLO, D. M. 2016. Urbanization and Religion in Ancient Central Mexico. Oxford University Press, Oxford. CarBALLO, D. M., AND T. PLUCKHAHN. 2007. Transportation corridors and political evolution in highland Mesoamerica: Settlement analyses incorporating GIS for northern Tlaxcala, Mexico. Journal of Anthropological Archaeology, 26: 607-629. Carter, N. P. 2017. Epigraphy and empire: Reassessing textual evidence for Formative Zapotec imperialism. Cambridge Archaeological Journal, 27: 433-450. CarTIER, C. 2002. Origins and evolution of a geographical idea: The macroregion in China. Modern China, 28: 79-142. Caso, A., I. BERNAL, AND J. R. Acosta. 1967. La ceramica de Monte Alban. Memoria 13. Instituto Nacional de Antropologia e Historia, Mexico City. CHANCE, J. K. 1976. The urban Indian in colonial Oaxaca. American Ethnologist, 3: 603-632. Cuerry, J. F. 2003. Archaeology beyond the site: Regional survey and its future, pp. 137-159. In Papadopoulos, J. K., and R. M. Leventhal, eds., Theory and Practice in Mediterranean Archae- ology: Old World and New World Perspectives. Cotsen Institute of Archaeology, University of California, Los Angeles. CHISHOLM, M. 1968. Rural Settlement and Land Use: An Essay in Location, 2nd ed. Hutchinson University Library, London. Cook, S. 1983. Mestizo palm weavers among the Zapotec: A critical re-examination of the Albarradas enigma, pp. 39-46. Jn Simmen, E., ed., Notas Mesoamericanas. Universidad de las Americas, Cholula, Puebla, Mexico. CorREA-ASCENCIO, M., I. G. ROBERTSON, O. CABRERA-CORTES, R. CABRERA-CASTRO, AND R. P. EvERSHED. 2014. Pulque produc- tion from fermented agave sap as a dietary supplement in prehispanic Mesoamerica. Proceedings of the National Acad- emy of Sciences of the United States of America, 111: 14223- 14228. CowaiLL, G. L. 1975. On causes and consequences of ancient and modern population changes. American Anthropologist, 77: 505-525. . 1979. Teotihuacan, internal militaristic competition, and the fall of the Classic Maya, pp. 51-62. Jn Hammond, N., and G. R. Willey, eds.. Maya Archaeology and Ethnohistory. University of Texas Press, Austin. . 1992. Toward a political history of Teotihuacan, pp. 87— 114. In Demarest, A. A., and G. W. Conrad, eds., Ideology and Pre-Columbian Civilizations. School of American Research Press, Santa Fe, NM. FIELDIANA: ANTHROPOLOGY . 1997. State and society at Teotihuacan, Mexico. Annual Review of Anthropology, 26: 129-161. . 2000. The central Mexican highlands from the rise of Teotihuacan to the decline of Tula, pp. 250-317. In Adams, R. E. W., and M. J. McLeod, eds., The Cambridge History of the Native Peoples of the Americas, Vol. Hl: Mesoamerica, Pt. 1. Cambridge University Press, Cambridge. . 2004. Origins and development of urbanism: Archaeo- logical perspectives. Annual Review of Anthropology, 33: 525- 532. . 2008. An update on Teotihuacan. Antiquity, 82: 962-975. . 2015. Ancient Teotihuacan: Early Urbanism in Central Mexico. Cambridge University Press, Cambridge. D’A.trRoy, T. N., AND T. K. EARLE. 1985. Staple finance, wealth finance, and storage in the Inka political economy. Current Anthropology, 26: 187-206. De ATLEY, S. P., AND F. J. FINpLow, eds. 1984. Exploring the Limits: Frontiers and Boundaries in Prehistory. BAR Interna- tional Series 223. Archaeopress, Oxford. De Lucia, K., AND C. T. MoreHART. 2015. Surplus and social change: The production of household and field in pre-Aztec central Mexico, pp. 73-96. In Morehart, C. T., and K. De Lucia, eds., Surplus: The Politics of Production and the Strategies of Everyday Life. University Press of Colorado, Boulder. DeMarralis, E. 2016. Making pacts and cooperative acts: The archaeology of coalition and consensus. World Archaeology, 48; 1-13. Dennis, P. A. 1987. Intervillage Conflict in Oaxaca. Rutgers University Press, New Brunswick, NJ. DieHL, R. A., AND J. C. BERLO, eds. 1989. Mesoamerica after the Decline of Teotihuacan: AD 700-900. Dumbarton Oaks, Washington, DC. DRENNAN, R. D. 1989. The mountains north of the valley, pp. 367-384. In Kowalewski, S. A., G. M. Feinman, L. Finsten, R. E. Blanton, and L. M. Nicholas, Monte Alban’s Hinterland, Pt. II: The Prehispanic Settlement Patterns in Tlacolula, Etla, and Ocotlan, the Valley of Oaxaca, Mexico. Memoirs No. 23. Museum of Anthropology, University of Michigan, Ann Arbor. DRulIVEN, P., AND A. KRruiTHoF. 1992. Tlacolula Valley, Oaxaca, Mexico, pp. 132-152. In Reitsma, H., T. Dietz, and L. de Haan, eds., Coping with Semiaridity: How the Rural Poor Survive in Dry-Season Environments. Netherlands Geographic Studies 146, Amsterdam. Dupatx, G. 1834. Relation des trois expeditions du capitaine Dupaix, ordonnées en 1805, 1806, et 1807 pour la recherche des antiquites du pays. Jn Baraderé, H., ed., Antiquités mexicaines. Bureau des Antiquités Mexicaines, Paris. ELAM, J. M. 1989. Defensible and fortified sites, pp. 385—407. In Kowalewski, S. A., G. M. Feinman, L. Finsten, R. E. Blanton, and L. M. Nicholas, Monte Alban’s Hinterland, Pt. I: Prehispanic Settlement Patterns in Tlacolula, Etla, and Ocotlan, the Valley of Oaxaca, Mexico. Memoirs No. 23. Museum of Anthropology, University of Michigan, Ann Arbor. Extson, C. M. 2007. Excavations at Cerro Tilcajete: A Monte Alban II Administrative Center in the Valley of Oaxaca. Memoirs No. 42. Museum of Anthropology, University of Michigan, Ann Arbor. Etson, C. M., AND R. A. Covey. 2006. Intermediate elites in New World states and empires, pp. 3—20. Jn Elson, C. M., and R. A. Covey, eds., Intermediate Elites in Pre-Columbian States and Empires. University of Arizona Press, Tucson. Etson, C. M., AND R. J. SHERMAN. 2007. Crema ware and elite power at Monte Alban: Ceramic production and iconography in the Oaxaca Valley, Mexico. Journal of Field Archaeology, 32: 265-282. E._ton, H. 1996. Frontiers of the Roman Empire. Indiana University Press, Bloomington. EMBERLING, G. 1997. Ethnicity in complex societies: Archaeolog- ical perspectives. Journal of Archaeological Research, 5: 295— 344. Esparza, M. 1983. Mitla: Lo que contarian sus paredes si pudieran hablar. Oaxaca, Nuestra Causa Comun, 4(30): 36-41, 5(31): 56-61, 5(32): 48-51. Direccion de Informacion del Gobierno del Estado, Oaxaca, Mexico. Evans, S. T. 1990. The productivity of maguey terrace agriculture in central Mexico during the Aztec period. Latin American Antiquity, 1: 117-132. FAHMEL Beyer, B. 2005. El camino de Tehuantepec. Anales de Antropologia, 39-II: 9-24. . 2007. Arquitectura prehispanica de la comarca situada entre San Dionisio Ocotepec and San Baltazar Chichicapan, Oaxaca. Boletin del Consejo de Arqueologia, Instituto Nacio- nal de Antropologia e Historia, Mexico City, http://www.inah. gob.mx. . 2010. El Preclasico de San Dionisio Ocotepec, Oaxaca, visto a traves de sus figurillas. Anales de Antropologia, 44: 9— 30. . 2012. La produccion céramica en los valles de San Dionisio Ocotepec y Chichicapan, Oaxaca, pp. 235-240. In Acosta Ochoa, G., ed., Arqueologias de la vida cotidiana: Espacios domésticos y areas de actividad en el México antiguo y otras zonas culturales. VII Coloquio Pedro Bosch Gimpera. Instituto de Investigaciones Antropologicas, Universidad Na- cional Autonoma de Mexico, Mexico City. Fana, H., G. M. FEINMAN, AND L. M. Nicuo as. 2015. Imperial expansion, public investment, and the long path of history: China’s first political unification and its aftermath. Proceedings of the National Academy of Sciences of the United States of America, 112: 9224-9229. FaAnc, H., A. UNDERHILL, G. M. FemnmMAn, L. M. NICHOLAS, F. Luan, H. Yu, AnD F. Car. 2012. Archaeological Report on Regional Systematic Survey in the Coastal Area of Southeast- ern Shandong, China. Cultural Relics Press, Beijing. FARGHER, L. F. 2016. Corporate power strategies, collective action, and control of principals: A cross-cultural perspective, pp. 309-326. In Fargher, L. F., and V. Y. Heredia Espinoza, eds., Alternative Pathways to Complexity: A Collection of Essays on Architecture, Economics, Power, and Cross-Cultural Analysis. University Press of Colorado, Boulder. FARGHER, L. F., V. Y. HEREDIA ESPINOZA, AND R. E. BLANTON. 2011. Alternative pathways to power in Late Postclassic highland Mesoamerica. Journal of Anthropological Archaeol- ogy, 30: 306-326. FautseiT, R. K. 2013. Cerro Danush: Excavations at a Hilltop Community in the Eastern Valley of Oaxaca, Mexico. Memoirs No. 54. Museum of Anthropology, University of Michigan, Ann Arbor. FAuLseIT, R. K., G. M. FEINMAN, AND L. M. NICHOLAS. in press. Household economic strategies and social resilience in the Valley of Oaxaca, Mexico. Jn Faulseit, R. K., N. Xiuhtecutli, and H. Holt-Mehta, eds., City, Craft and Residence in Mesoamerica: Research Papers Presented in Honor of Dan M. Healan. Middle American Research Institute, Tulane University, New Orleans. FAuLsEIT, R. K., C. D. GarciA Rios, J. Pink, A. G. Lopez RIVERA, C. J. Royas Ortiz, E. AUSEL, AND H. LAPHAM. 2016. Proyecto FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 119 rqueologico de Dainzu-Macuilxochitl, Oaxaca, México, 2015: Levantamiento topografico, mapeo y excavaciones en las terrazas O8 y 825. Final report for the Consejo de Arqueologia, Instituto Nacional de Antropologia e Historia, Mexico City. FEINMAN, G. M. 1980. The Relationship between Administrative Organization and Ceramic Production in the Valley of Oaxaca, Mexico. Ph.D. dissertation, Graduate Center, City University of New York, New York. ———. 1985. Changes in the organization of ceramic production in pre-Hispanic Oaxaca, Mexico, pp. 195-223. Jn Nelson, B. A., ed., Decoding Prehistoric Ceramics. Southern Illinois Univer- sity Press, Carbondale. 1986a. On the cutting edge or out of bounds? Archaeology approaches the frontier. Anthropology, 10: 43—54. . 1986b. The emergence of specialized ceramic production in Formative Oaxaca, pp. 347-373. In Isaac, B. L., ed., Research in Economic Anthropology, Supplement 2: Economic Aspects of Prehispanic Highland Mesoamerica. JAI Press, Greenwich, CT. 1991. Demography, surplus, and inequality: Early political formations in highland Mesoamerica, pp. 229-262. In Earle, T., ed., Chiefdoms: Power, Economy, and Ideology. Cambridge University Press, Cambridge. 1994. Social boundaries and political change: A comparative perspective, pp. 225-236. In Stein, G., and M. S. Rothman, eds., Chiefdoms and Early States in the Near East: The Organizational Dynamics of Complexity. Prehistory Press, Madison, WI. . 1995. The emergence of inequality: A focus on strategies and processes, pp. 255-279. In Price, T. D., and G. M. Feinman, eds., Foundations of Social Inequality. Plenum Press, New York. ———. 1996. The changing structure of macroregional Meso- america: With focus on the Classic-Postclassic transition in the Valley of Oaxaca. Journal of World-Systems Research, 2: 275— 290. . 1998. Scale and social organization: Perspectives on the archaic state, pp. 85-133. Jn Feinman, G. M., and J. Marcus, eds., Archaic States. School of American Research Press, Santa Fe, NM. . 2001la. Mesoamerican political complexity: The corpo- rate-network dimension, pp. 151-175. Jn Haas, J., ed., From Leaders to Rulers. Kluwer Academic/Plenum Publishers, New York. . 2001b. Postclassic southern Mexican highlands: Postclas- sic Oaxaca, pp. 361-377. Jn Peregrine, P. N., and M. Ember, eds., Encyclopedia of Prehistory, Vol. 5: Middle America. Kluwer Academic/Plenum Publishers, New York. . 2006. The economic underpinnings of prehispanic Zapotec civilization: Small-scale production, economic interde- pendence, and market exchange, pp. 255-280. In Marcus, J., and C. Stanish, eds., Agricultural Strategies. Cotsen Institute of Archaeology, University of California, Los Angeles. . 2007. The last quarter century of archaeological research in the Central Valleys of Oaxaca. Mexicon, 29: 3-15. . 2012a. Comparative frames for the diachronic analysis of complex societies, next steps, pp. 21-43. Jn Smith, M. E., ed., The Comparative Archaeology of Complex Societies. Cam- bridge University Press, Cambridge. . 2012b. A dual—processual perspective on the power and inequality in the contemporary United States: Framing political economy for the present and the past, pp. 255-288. Jn Price, T. D., and G. M. Feinman, eds., Pathways to Power: New Perspectives on the Emergence of Social Inequality. Springer, New York. 120 . 2016. The emergence of collectively organized cities in highland Mesoamerica: Innovations in living together. Paper presented at Plenums-Tagung des Cluster 2 “Innovationen: Technisch, Sozial,” Museums fiir Asiatische Kunst, Berlin. . 2017. Reframing ancient economies: New models, new questions, pp. 139-149. Jn Fernandez-Gotz, M., and D. Krausse, eds., Eurasia at the Dawn of History: Urbanization and Social Change. Cambridge University Press, Cambridge. . 2018. A diachronic perspective on the prehispanic ceramic tradition of the Valley of Oaxaca. In Park, Y., J. Minich, and D. Arnold, eds., Pre-Columbian Pottery: A Thematic Approach to New World Ceramics. University Press of Florida, Gainesville, in press. . In press. The prehispanic Mesoamerican world: Framing interaction. Jn Englehardt, J. D., and M. D. Carrasco, eds., New Perspectives on Interregional Interaction in Ancient Mesoamerica. University Press of Colorado, Boulder. FEINMAN, G. M., R. E. BLANTON, AND S. A. KOWALEwWSKI. 1984. Market system development in the prehispanic Valley of Oaxaca, Mexico, pp. 157-178. In Hirth, K. G., ed., Trade and Exchange in Early Mesoamerica. University of New Mexico Press, Albuquerque. FEINMAN, G. M., AND C. P. GARRATY. 2010. Preindustrial markets and marketing: Archaeological perspectives. Annual Review of Anthropology, 39: 167-191. FEINMAN, G. M., M. Go iTKo, AND L. M. Nicuo ras. 2015. A network analysis of pre-Hispanic obsidian exchange: Implica- tions for macro-regional dynamics and ancient economies. Paper presented at the International Workshop on Digging a Vertex, Finding the Edges: Approaches to Social Network Analysis in Archaeology, University of Cologne, Cologne, Germany. FEINMAN, G. M., S. A. KOWALEWSKI, L. FINSTEN, R. E. BLANTON, AND L. M. NICHOLAS. 1985. Long-term demographic change: A perspective from the Valley of Oaxaca, Mexico. Journal of Field Archaeology, 12: 333-362. FEINMAN, G. M., AND L. M. Nicuotas. 1990. At the margins of the Monte Alban state: Settlement patterns in the Ejutla Valley, Oaxaca, Mexico. Latin American Antiquity, 1: 216-246. . 1992. Pre-Hispanic interregional interaction in southern Mexico: The Valley of Oaxaca and the Ejutla Valley, pp. 75— 116. In Schortman, E. M., and P. A. Urban, eds., Resources, Power, and Interregional Interaction. Plenum Press, New York. . 1996. Defining the eastern limits of the Monte Alban state: Systematic settlement pattern survey in the Guirtn area, Oaxaca, Mexico. Mexicon, 18: 91-97. . 1999. Reflections on regional survey: Perspectives from the Guirun area, Oaxaca, Mexico, pp. 172-190. Jn Billman, B. R., and G. M. Feinman, eds., Settlement Pattern Studies in the Americas: Fifty Years since Vira. Smithsonian Institution Press, Washington, DC. . 2004a. Hilltop Terrace Sites of Oaxaca, Mexico: Intensive Surface Survey at Guirun, El Palmillo, and the Mitla Fortress. Fieldiana: Anthropology, New Series 37. Field Museum of Natural History, Chicago. . 2004b. Unraveling the prehispanic highland Mesoamer- ican economy: Production, exchange, and consumption in the Classic period Valley of Oaxaca, pp. 167-188. Jn Feinman, G. M., and L. M. Nicholas, eds., Archaeological Perspectives on Political Economies. University of Utah Press, Salt Lake City. . 2005. More than alluvial land and water: The late pre- Hispanic emergence of eastern Tlacolula, Oaxaca, Mexico, pp. 229-259. In Blanton, R. E., ed., Settlement, Subsistence, and Social Complexity: Essays Honoring the Legacy of Jeffrey R. FIELDIANA: ANTHROPOLOGY Parsons. Cotsen Institute of Archaeology, University of California, Los Angeles. . 2009. El] Palmillo: Una perspectiva doméstica del periodo clasico en el Valle de Oaxaca, temporadas 1999-2008. Final project report prepared for the Instituto Nacional de Antro- pologia e Historia de México, Mexico City. . 2010. A multiscalar perspective on market exchange in the Classic-period Valley of Oaxaca, pp. 85-98. Jn Garraty, C. P., and B. L. Stark, eds., Archaeological Approaches to Market Exchange in Ancient Societies. University Press of Colorado, Boulder. . 2011a. Monte Alban: Una perspectiva desde los limites del Valle de Oaxaca, pp. 241-284. In Robles Garcia, N. M., and A. I. Rivera Guzman, eds., Monte Alban en la encrucijada regional y disciplinaria: Memoria de la Quinta Mesa Redonda de Monte Alban. Instituto Nacional de Antropologia e Historia, Mexico City. .2011b. La Fortaleza de Mitla: Una perspectiva doméstica de los periodos clasico y posclasico en el Valle de Oaxaca, temporada de 2011. Technical report prepared for the Instituto Nacional de Antropologia e Historia, Mexico City. . 2011c. The ballcourt at El Palmillo: Implications for Late Classic Oaxaca, Mexico. Mexicon, 33: 98-104. . 2012. The late prehispanic economy of the Valley of Oaxaca, Mexico: Weaving threads from data, theory, and subsequent history, pp. 225—258. In Matejowsky, T., and D. C. Wood, eds., Political Economy, Neoliberalism, and the Prehistoric Economies of Latin America. Research in Economic Anthropology, Vol. 32. Emerald, Bingley. . 2013. Settlement Patterns of the Ejutla Valley, Oaxaca, Mexico: A Diachronic Macroscale Perspective. Fieldiana: Anthropology, New Series 43. Field Museum of Natural History, Chicago. . 2015. Excavaciones en el area al sur de la Zona Arqueologica de Lambityeco, tercera temporada. Technical report prepared for the Instituto Nacional de Antropologia e Historia, Mexico City. . 2016a. Framing the rise and variability of past complex societies, pp. 271-289. Jn Fargher, L. F., and V. Y. Heredia Espinoza, eds., Alternative Pathways to Complexity: A Collection of Essays on Architecture, Economics, Power, and Cross-Cultural Analysis. University Press of Colorado, Boul- der. . 2016b. After Monte Alban in the Central Valleys of Oaxaca: A reassessment, pp. 43-69. Jn Faulseit, R. K., ed., Beyond Collapse: Archaeological Perspectives on Resilience, Revitalization, and Transformation in Complex Societies. Occasional Paper No. 42. Center for Archaeological Investiga- tions, Southern Illinois University, Carbondale. . 2016c. Excavaciones en el area al sur de la Zona Arqueologica de Lambityeco, cuarta temporada. Technical report prepared for the Instituto Nacional de Antropologia e Historia, Mexico City. . 2017. Re-envisioning prehispanic Mesoamerican econo- mies: Modes of production, fiscal foundations of collective action, and conceptual legacies, pp. 253-281. Jn Rosenswig, R. M., and J. J. Cunningham, eds., Modes of Production and Archaeology. University Press of Florida, Gainesville. FEINMAN, G. M., L. M. NIcHOLAs, AND M. Go itTKo. 2013. Exchange in the Valley of Oaxaca: Late Classic obsidian procurement. Mexicon, 35: 60-68. FEINMAN, G. M., L. M. NicHOLAs, AND H. R. Haines. 2002. Houses on a hill: Classic period life at El Palmillo, Oaxaca, Mexico. Latin American Antiquity, 13: 251-277. . 2007. Classic period agricultural intensification and domestic life at El Palmillo, Valley of Oaxaca, Mexico, pp. 23-61. In Thurston, T., and C. Fisher, eds., Seeking a Richer Harvest: The Archaeology of Subsistence Intensification, Innovation, and Change. Springer, New York. FEINMAN, G. M., L. M. Nicuoras, H. A. LAPHAM, R. HIGELIN PONCE DE LEON, J. Rios ALLIER, AND C. MoreEHART. 2016. Broadening the context for Classic period Lambityeco, Oaxaca: New discoveries from 2013-2015. Mexicon, 38: 46-55. FEINMAN, G. M., L. M. NicHoLas, N. M. Rosies GArciA, M. GoLiTKo, C. ELson, R. FAuLseit, E. GONZALEZ LICcON, I. Oxtcuin, L. D. PAcHEco ArtAs, V. PEREZ RODRIQUEZ, G. RAMON, AND J. Rios ALLier. 2017. Prehispanic obsidian exchange in the Valley of Oaxaca, Mexico. Unpublished manuscript, Integrative Research Center, Field Museum of Natural History, Chicago. FesLer, J. W. 1968. Centralization and decentralization, pp. 370— 379. In Sills, D. L., ed., International Encyclopedia of the Social Sciences, Vol. 2. Macmillan, New York. FINSTEN, L. 1983. The Classic-Postclassic Transition in the Valley of Oaxaca, Mexico: A Regional Analysis of the Process of Political Decentralisation in a Prehistoric Complex Society. Ph.D. dissertation, Department of Sociology and Anthropol- ogy, Purdue University, West Lafayette, IN. . 1995. Jalieza, Oaxaca: Activity Specialization at a Hilltop Center. Publications in Anthropology 48. Vanderbilt Univer- sity, Nashville, TN. . 1996. Periphery and frontier in southern Mexico: The Mixtec Sierra in highland Oaxaca, pp. 77-95. Jn Peregrine, P. N., and G. M. Feinman, eds., Pre-Columbian World Systems. Prehistory Press, Madison, WI. FLANNERY, K. V. 1968. The Olmec and the Valley of Oaxaca: A model for inter-regional interaction in Formative times, pp. 119-130. Jn Benson, E. P., ed., Dumbarton Oaks Conference on the Olmec. Dumbarton Oaks, Washington, DC. FLANNERY, K. V. 1983a. Precolumbian farming in the Valleys of Oaxaca, Nochixtlan, Tehuacan, and Cuicatlan: A comparative study, pp. 323-339. In Flannery, K. V., and J. Marcus, eds., The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civilizations. Academic Press, New York. 1983b. The legacy of the early urban period: An ethnohistoric approach to Monte Alban’s temples, residences, and royal tombs, pp. 132-136. Jn Flannery, K. V., and J. Marcus, eds., The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civilizations. Academic Press, New York. FLANNERY, K. V., ed. 1970. Preliminary archaeological investiga- tions in the Valley of Oaxaca, Mexico: A report to the National Science Foundation and the Instituto Nacional de Antropolo- gia e Historia. Mimeograph. Museum of Anthropology, University of Michigan, Ann Arbor. . 1976. The Early Mesoamerican Village. Academic Press, New York. FLANNERY, K. V., A. V. Kirksy, M. J. KirkBy, AND A. W. WILLIAMS, JR. 1967. Farming systems and political growth in ancient Oaxaca. Science, 158: 445-454. FLANNERY, K. V., AND J. MArRcus. 1994. Early Formative Pottery of the Valley of Oaxaca, Mexico. Memoirs No. 27. Museum of Anthropology, University of Michigan, Ann Arbor. . 2005. Excavations at San José Mogote 1: The Household Archaeology. Memoirs No. 40. Museum of Anthropology, University of Michigan, Ann Arbor. . 2015. Excavations at San Jose Mogote 2: The Cognitive Archaeology. Memoirs No. 58. Museum of Anthropology, University of Michigan, Ann Arbor. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 12] LLEGOoS Ruiz, R. 1978. El Senor 9 Flor en Zaachila. Universidad Nacional Autonoma de México, Mexico City. Garcid Cook, A. 1994. Cantona. Arqueologia Mexicana, 2(10): 60-65. —_—, 2003. Cantona: The city, pp. 311-343. In Sanders, W. T.., A. G. Mastache, and R. H. Cobean, eds., El urbanismo en Mesoamerica/Urbanism in Mesoamerica, Vol. 1. Instituto Nacional de Antropologia e Historia, Mexico City, and Pennsylvania State University, University Park. GarciA Cook, A., AND B. L. Merino Carrion. 1998. Cantona: Urbe prehispanica en el altiplano central de México. Latin American Antiquity, 9: 191-216. GarciA Cook, A., AND D. TENoRIO. 2010. Estudio de procedencia de obsidiana arqueologica de Cantona, Puebla. Arqueologia, 43: 217-229. Go.itko, M., AND G. M. FEINMAN. 2015. Procurement and distribution of pre-Hispanic Mesoamerican obsidian 900 BC-— AD 1520: A social network analysis. Journal of Archaeological Method and Theory, 22: 206-247. Go.iTko, M., J. MEIERHOFF, G. M. FEINMAN, AND P. R. WILLIAMS. 2012. Complexities of collapse: The evidence of Maya obsidian as revealed by social network graphical analysis. Antiquity, 86: 507-523. GOMEZ CHAVEZ, S., AND M. W. SPENCE. 2012. Interaction among the complex societies of Classic-period Mesoamerica, pp. 283— 297. In Nichols, D. L., and C. A. Pool, eds., The Oxford Handbook of Mesoamerican Archaeology. Oxford University Press, Oxford. GONGALVES DE LIMA, O. 1956. El maguey y el pulque en los codices mexicanos. Fondo de Cultura Economica, Mexico City. GONZALEZ LIcON, E. 2015. Procesiones en Oaxaca. Arqueologia Mexicana, 22(131): 42-47. Goopsy, R. G. 1998. Technological patterning and social boundaries: Ceramic variability in southern New England, pp. 161-182. Jn Stark, M. T., ed., The Archaeology of Social Boundaries. Smithsonian Institution Press, Washington, DC. GREEN, S. W., AND S. M. PERLMAN, eds. 1985. The Archaeology of Frontiers and Boundaries. Academic Press, Orlando, FL. GuLp!, J., AND D. ArmitTaGe. 2014. The History Manifesto. Cambridge University Press, Cambridge. Haccetrt, P. 1966. Locational Analysis in Human Geography. St. Martin’s Press, New York. Ha tt, T. D. 2009. Puzzles in the comparative study of frontiers: Problems, some solutions, and methodological implications. Journal of World-Systems Research, 15: 25—47. Hassan, F. A. 1981. Demographic Archaeology. Academic Press, New York. HeGmon, M. 1998. Technology, style, and social practices: Archaeological practices, pp. 264-279. In Stark, M. T., ed., The Archaeology of Social Boundaries. Smithsonian Institution Press, Washington, DC. Hewitt, W. P., M. C. WINTER, AND D. A. PETERSON. 1987. Salt production at Hierve el Agua, Oaxaca. American Antiquity, 52: 799-816. HIGELIN PONCE DE LEON, R., A. MARTINEZ TUNOoN, N. M. Rosies GarciA, AND N. Tuross. 2013. San Miguel Albarradas: Interpretaciones culturales de un espacio funerario. Estudios de Antropologia Biologica, 16: 95-118. Hirtn, K. G. 1978. Teotihuacan regional population administra- tion in eastern Morelos. World Archaeology, 9: 320-333. . 1980. Eastern Morelos and Teotihuacan: A Settlement Survey. Publications in Anthropology No. 25. Vanderbilt University, Nashville, TN. . 1987. Formative period settlement patterns in the Rio Amatzinac Valley, pp. 343-367. In Grove, D. C., ed., Ancient Chalcatzingo. University of Texas Press, Austin. . 1995. Urbanism, militarism, and architectural design: An analysis of Epiclassic sociopolitical structure at Xochicalco. Ancient Mesoamerica, 6: 237-250. Hirtn, K. G., AND J. PILLsBury, eds. 2013. Merchants, Markets, and Exchange in the Pre-Columbian World. Dumbarton Oaks, Washington, DC. Homes, W. H. 1897. Archaeological Studies among the Ancient Cities of Mexico: Pt. I], Monuments of Chiapas, Oaxaca, and the Valley of Mexico. Anthropological Series 1(1). Field Columbian Museum, Chicago. HorcasitTAs, F., AND R. GeorGE. 1955. The Relacion de Tlacolula y Mitla: Notes and translations. Mesoamerican Notes, 4: 13-24. Hutson, S. R. 2014. Artefactos prehispanicos de la Sierra Mixe, pp. 267-278. In Winter, M., and G. Sanchez Santiago, eds., Panorama arqueologico: Dos Oaxacas. Instituto Nacional de Antropologia e Historia, Oaxaca, Mexico. Jones, S. 1997. The Archaeology of Ethnicity: Constructing Identities in the Past and Present. Routledge, New York. Joyce, A. A. 2010. Mixtecs, Zapotecs, and Chatinos: Ancient Peoples of Southern Mexico. Wiley-Blackwell, Malden, MA. . 2014. Warfare in Late/Terminal Formative-period Oaxaca, pp. 117-141. Jn Scherer, A. C., and J. W. Verano, eds., Embattled Bodies, Embattled Places: War in Pre- Columbian Mesoamerica and the Andes. Dumbarton Oaks, Washington, DC. Joyce, A. A., AND S. B. BARBER. 2016. Alternative pathways to power in Formative Oaxaca, pp. 41—58. Jn Fargher, L. F., and V. Y. Heredia Espinoza, eds., Alternative Pathways to Complexity: A Collection of Essays on Architecture, Econom- ics, Power, and Cross-Cultural Analysis. University Press of Colorado, Boulder. Joyce, A. A., H. Nerr, M. S. THEME, M. WINTER, J. M. ELAM, AND A. WoRKINGER. 2006. Ceramic production and exchange in Late/Terminal Formative period Oaxaca. Latin American Antiquity, 17: 579-594. Joyce, A. A., A. G. WorKINGER, B. HAMANN, P. KROoEFGES, M. OLAND, AND S. M. Kina. 2004. Lord 8 Deer “Jaguar Claw” and the Land of the Sky: The archaeology and history of Tututepec. Latin American Antiquity, 15: 273-297. KapreL, W. 1977. Alternative Adaptive Strategies in Three Mexican Towns. Ph.D. dissertation, Department of Anthro- pology, University of Arizona, Tucson. Kepecs, S. 2003. Salt sources and production, pp. 126-130. In Smith, M. E., and F. F. Berdan, eds., The Postclassic Mesoamerican World. University of Utah Press, Salt Lake City. Kina, M. E. 1986. Preceramic cordage and basketry from Guila Naquitz, pp. 157-161. Jn Flannery, K. V., ed., Guila Naquitz: Archaic Foraging and Early Agriculture in Oaxaca, Mexico. Academic Press, Orlando, FL. Kino, S. M. 2008. Interregional networks of the Oaxacan Early Postclassic: Connecting the coast and the highlands, pp. 255— 291. In Blomster, J. P., ed., After Monte Alban: Transforma- tion and Negotiation in Oaxaca, Mexico. University Press of Colorado, Boulder. . 2012. Hidden transcripts, contested landscapes, and long- term indigenous history in Oaxaca, Mexico, pp. 230-263. In Oland, M., S. M. Hart, and L. Frink, eds., Decolonizing Indigenous Histories: Exploring Prehistoric/Colonial Transi- tions in Archaeology. University of Arizona Press, Tucson. FIELDIANA: ANTHROPOLOGY Kirxsy, A. V. 1973. The Use of Land and Water Resources in the Past and Present Valley of Oaxaca. Mexico. Memoirs No. 5. Museum of Anthropology, University of Michigan, Ann Arbor. Kirxsy, M. J., A. V. WHyTE, AND K. V. FLANNERY. 1986. The physical environment of the Guila Naquitz cave group, pp. 43— 61. Jn Flannery, K. V., ed., Guila Naquitz: Archaic Foraging and Early Agriculture in Oaxaca, Mexico. Academic Press, Orlando, FL. KwnappeTt, C., ed. 2013. Network Analysis in Archaeology: New Approaches to Regional Interaction. Oxford University Press, Oxford. KowaA.ewskl, S. A. 1976. Prehispanic Settlement Patterns of the Central Part of the Valley of Oaxaca, Mexico. Ph.D. dissertation, Department of Anthropology, University of Arizona, Tucson. 1980. Population-resource balances of period I of Oaxaca, Mexico. American Antiquity, 45: 151-165. . 1982. Population and agricultural potential: Early I through V, pp. 149-180. Jn Blanton, R. E., S. A. Kowalewski, G. M. Feinman, and J. Appel, Monte Alban’s Hinterland, Pt. I: The Prehispanic Settlement Patterns of the Central and Southern Parts of the Valley of Oaxaca, Mexico. Memoirs No. 15. Museum of Anthropology, University of Michigan, Ann Arbor. . 1990. The evolution of complexity in the Valley of Oaxaca. Annual Review of Anthropology, 19: 39-58. . 1991. Penoles: Archaeological survey in the Mixtec Sierra, Mexico. Report to the National Science Foundation and National Geographic Society, Washington, DC. . 2003a. Scale and the explanation of demographic change: 3,500 years in the Valley of Oaxaca. American Anthropologist, 105: 313-325. . 2003b. What is the community? The long view from Oaxaca, Mexico. Social Evolution & History, 2(1): 4-24. . 2003c. Backcountry pots. Ancient Mesoamerica, 14: 65— Tey. . 2004. The new past: From region to macroregion. Social Evolution & History, 3(1): 81-105. . 2008. Regional settlement pattern studies. Journal of Archaeological Research, 16: 225—285. . 2012. A theory of the ancient Mesoamerican economy, pp. 187-224. In Matejowsky, T., and D. C. Wood, eds., Political Economy, Neoliberalism, and the Prehistoric Econo- mies of Latin America. Research in Economic Anthropology, Vol. 32. Emerald, Bingley. . 2013. The work of making a community, pp. 201-218. In Birch, J., ed., From Prehistoric Villages to Cities: Settlement Aggregation and Community Transformation. Routledge, New York. Kowa .ewsk], S. A., A. K. BALKANSKY, L. R. STIVER WALSH, T. J. PLUCKHAHN, J. F. CHAMBLEE, V. PEREZ RopriGuEz, V. Y. HerepIA EspINozaA, AND C. A. SmiTH. 2009. Origins of the Nuu: Archaeology in the Mixteca Alta, Mexico. University Press of Colorado, Boulder. Kowa .ewskI, S. A., L. BARBA PINGARRON, J. BLANCAS, M. Y. Cortes Vitcuis, G. GARCIA AYALA, L. LOpEzZ ZARATE, A. ORTIZ, T. J. PLUCKHAHN, B. A. STEERE, AND B. VILCHIS FLoREs. 2008. Proyecto urbanismo temprano y tardio en Coixtlahuaca, Oaxaca: Informe técnico final. Report prepared for the Instituto Nacional de Antropologia e Historia, Mexico City. KowALeEwskl, S. A., L. BARBA PINGARRON, G. GARCIA AYALA, B. A. STEERE, J. BLANCAS vAsQuEZ, M. Y. Cortés VILcuis, L. Lopez ZARATE, A. Ortiz ButTroOn, T. J. PLUCKHAHN, AND B. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO | VitcHis Fiores. 2010. La presencia azteca en Oaxaca: La provincia de Coixtlahuaca. Anales de Antropologia, 44: 77— 103. KowA_ewskI, S. A., R. E. BLANTON, G. M. FEINMAN, AND L. FINSTEN. 1983. Boundaries, scale, and internal organization. Journal of Anthropological Archaeology, 2: 32—S6. KowALEwsKI, S. A., G. M. FEINMAN, L. FINSTEN, R. E. BLANTON, AND L. M. NicHoLas. 1989. Monte Alban’s Hinterland, Pt. II: The Prehispanic Settlement Patterns in Tlacolula, Etla, and Ocotlan, the Valley of Oaxaca, Mexico. Memoirs No. 23. Museum of Anthropology, University of Michigan, Ann Arbor. KowALewskI, S. A., G. M. FEINMAN, L. M. NICHOLAS, AND V. Y. HerepiA. 2006. Hilltowns and valley fields: Great transforma- tions, labor, and long-term history in ancient Oaxaca, pp. 197— 216. In Durrenberger, E. P., and J. E. Marti, eds., Labor in Cross-Cultural Perspective. Rowman & Littlefield, Lanham, MD. KowaLewskl, S. A., AND L. FINSTEN. 1983. The economic systems of ancient Oaxaca: A regional perspective. Current Anthropol- ogy, 24: 413-441. KowaLewskl, S. A., C. S. SPENCER, AND E. REDMOND. 1978. Description of ceramic categories, Appendix 2. /n Blanton, R. E., Monte Alban: Settlement Patterns at the Ancient Zapotec Capital. Academic Press, New York. Kowa.skl, J. K., AND C. KRISTAN-GRAHAM, eds. 2007. Twin Tollans: Chichen Itza, Tula, and the Epiclassic to Early Postclassic Mesoamerican World. Dumbarton Oaks, Washing- ton, DC. Kuropa, E. 1984. Under Mt. Zempoaltepetl: Highland Mixe Society and Ritual. National Museum of Ethnology, Osaka, Japan. Levi, M. 1988. Of Revenue and Rule. University of California Press, Berkeley. Levine, M. N. 2011. Negotiating political economy at Late Postclassic Tututepec (Yucu Dzaa), Oaxaca, Mexico. American Anthropologist, 113: 22-39. Levine, M. N., A. A. Joyce, AND M. D. Giascock. 2011. Shifting patterns of obsidian exchange in Postclassic Oaxaca, Mexico. Ancient Mesoamerica, 22: 123-133. LiguTFoot, K. G., AND A. MArtinez. 1995. Frontiers and boundaries in archaeological perspective. Annual Review of Anthropology, 24: 471-492. Linp, M. D. 1991-1992. Unos problemas con la cronologia de Monte Alban y una nueva serie de nombres para las fases. Notas Mesoamericanas, 13: 177-192. . 1994. Monte Alban y el Valle de Oaxaca durante la fase Xoo, pp. 99-111. In Winter, M., ed., Monte Alban: Estudios recientes. Contribucion 2. Proyecto Especial Monte Alban 1992-1994, Oaxaca, Mexico. . 2000. Mixtec city-states and Mixtec city-state culture, pp. 567-580. In Hansen, M. H., ed., A Comparative Study of Thirty City-State Cultures. Royal Danish Academy of Sciences and Letters, Copenhagen. Linp, M. D., AND J. Urcip. 1983. The lords of Lambityeco and their nearest neighbors. Notas Mesoamericanas, 9: 78-111. . 2010. Lords of Lambityeco: Political Evolution in the Valley of Oaxaca during the Xoo Phase. University Press of Colorado, Boulder. Lipp, F. J. 1991. The Mixe of Oaxaca: Religion, Ritual, and Healing. University of Texas Press, Austin. MacNetisu, R. S. 1967. Summary of subsistence, pp. 290-309. Jn Byers, D. S., ed., Prehistory of the Tehuacan Valley, Vol. 1: bo Ww 1vironment and Subsistence. University of Texas Press, Ausuon. MacNelisu, R. S., F. A. PETERSON, AND J. A. NEELY. 1972. The archaeological reconnaissance, pp. 341-495. In MacNeish, R. S., M. L. Fowler, A. Garcia Cook, F. A. Peterson, A. Nelken- Terner, and J. A. Neely, The Prehistory of the Tehuacan Valley, Vol. 5: Excavations and Reconnaissance. University of Texas Press, Austin. MANZANILLA, L. R., X. BoKHIMI, D. TENORIO, M. JIMENEZ-REYES, E. Rosaces, C. MARTINEZ, AND M. Winter. 2017. Procedencia de la mica de Teotihuacan: Control de los recursos suntuarios foraneos por las élites gobernantes. Anales de Antropologia, 51: 23-38. Marcus, J. 1976. The iconography of militarism at Monte Alban and neighboring sites in the Valley of Oaxaca, pp. 123-139. In Nicholson, H. B., ed., The Origins of Religious Art and Iconography in Preclassic Mesoamerica. Latin American Center, University of California, Los Angeles. . 1980. Zapotec writing. Scientific American, 242: 50-64. . 1983a. The conquest slabs of Building J, Monte Alban, pp. 106-108. Jn Flannery, K. V., and J. Marcus, eds., The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civilizations. Academic Press, New York. . 1983b. Changing patterns of stone monuments after the fall of Monte Alban, AD 600-900, pp. 191-197. Jn Flannery, K. V., and J. Marcus, eds., The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civilizations. Academic Press, New York. . 1983c. The style of the Huamelulpan stone monuments, pp. 125-126. In Flannery, K. V., and J. Marcus, eds, The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civilizations. Academic Press, New York. . 1983d. Monte Alban II in the Macuilxochitl area, pp. 113-115. Jn Flannery, K. V., and J. Marcus, eds., The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civilizations. Academic Press, New York. . 1983e. Teotihuacan visitors on Monte Alban monuments and murals, pp. 175-181. Jn Flannery, K. V., and J. Marcus, eds., The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civilizations. Academic Press, New York. . 1989. Zapotec chiefdoms and the nature of Formative religions, pp. 148-197. Jn Sharer, R. J., and D. C. Grove, eds., Regional Perspectives on the Olmec. Cambridge University Press, Cambridge. . 1992a. Political fluctuations in Mesoamerica: Dynamic cycles of Mesoamerican states. National Geographic Research & Exploration, 8: 392-411. . 1992b. Mesoamerican Writing Systems: Propaganda, Myth, and History in Four Ancient Civilizations. Princeton University Press, Princeton, NJ. . 2002. Carved stones from the Sola Valley, Oaxaca, pp. 103-121. In Balkansky, A. J., The Sola Valley and the Monte Alban State: A Study in Zapotec Imperial Expansion. Memoirs No. 36. Museum of Anthropology, University of Michigan, Ann Arbor. . 2006. Identifying elites and their strategies, pp. 212-246. In Elson, C. M., and R. A. Covey, eds., Intermediate Elites in Pre-Columbian States and Empires. University of Arizona Press, Tucson. . 2009. How Monte Alban represented itself, pp. 77-110. In Fash, W. L., and L. Lopez Lujan, eds., The Art of Urbanism: How Mesoamerican Kingdoms Represented Themselves in Architecture and Imagery. Dumbarton Oaks, Washington, DC. 124 Marcus, J.. AND K. V. FLANNERY. 1990. Science and science fiction in Postclassic Oaxaca: Or, “Yes, Virginia, there is a Monte Alban IV,” pp. 191-205. In Marcus, J., ed., Debating Oaxaca Archaeology. Anthropological Papers 84. Museum of Anthropology, University of Michigan, Ann Arbor. . 1996. Zapotec Civilization: How Urban Society Evolved in Mexico’s Oaxaca Valley. Thames & Hudson, London. MarkKens, R. J. 2004. Ceramic Chronology in the Valley of Oaxaca, Mexico during the Late Classic and Postclassic Periods and the Organization of Ceramic Production. Ph.D. disserta- tion, Department of Anthropology, Brandeis University, Waltham, MA. MarkeENs, R. J., AND C. MARTINEZ Lopez. 2009. El sistema de produccion ceramica en Monte Alban durante el Preclasico Tardio y el Clasico Tardio, pp. 123-152. In Robles Garcia, N., ed., Bases de la complejidad social en Oaxaca: Memoria de la Cuarta Mesa Redonda de Monte Alban. Instituto Nacional de Antropologia e Historia, Mexico City. MarkKENSs, R. J., AND M. Winter. 2014. La tumba 1 de Chuxnaban, Quetzaltepec, Mixes, pp. 279-291. Jn Winter, M., and G. Sanchez Santiago, eds., Panorama arqueologico: Dos Oaxacas. Instituto Nacional de Antropologia e Historia, Oaxaca, Mexico. MARKENS, R. J., M. WINTER, AND C. MArTiNEZ LOpEz. 2008. Ethnohistory, oral history, and archaeology at Macuilxochitl: Perspectives on the Postclassic period (800-1521 CE) in the Valley of Oaxaca, pp. 193-215. Jn Blomster, J. P., ed., After Monte Alban: Transformation and Negotiation in Oaxaca, Mexico. University Press of Colorado, Boulder. MARKMAN, C. W. 1981. Prehispanic Settlement Dynamics in Central Oaxaca, Mexico: A View from the Miahuatlan Valley. Publications in Anthropology No. 26. Vanderbilt University, Nashville, TN. MARTINEZ Lopez, C., AND R. MARKENS. 2004. Analisis de la funcion politico-economica del conjunto Plataforma Norte lado poniente de la Plaza Principal de Monte Alban, pp. 75-99. In Robles Garcia, N., ed., Estructuras politicas en el Oaxaca antiguo: Memoria de la Tercera Mesa Redonda de Monte Alban. Instituto Nacional de Antropologia e Historia, Mexico City. MartTiNEZ Y OyEDA, E. 1996. Guia ilustrada de las plantas de Yagul. Proyecto Yagul 96: Conservacion de los recursos ecologicos. Centro INAH Oaxaca, Oaxaca, Mexico. Masson, M. A., AND H. Orr. 1998. The role of genealogical records in late Precolumbian Valley of Oaxaca political history. Mexicon, 20: 10-16. McGuire, R. H. 2002. The meaning and limits of the Southwest/ Northwest: A perspective from northern Mexico, pp. 173-183. In Villalpando, M. E., ed., Boundaries and Territories: Prehistory of the U.S. Southwest and Northern Mexico. Anthropological Research Paper No. 54. Arizona State University, Tempe. McNEILL, W. H. 1998. World history and the rise and fall of the West. Journal of World History, 9: 215-236. MiIpp.eTon, W. D., G. M. FEINMAN, AND L. M. NicHo.as. 2001. An investigation of the use of xerophytic plant resources in the economy and subsistence of El Palmillo, Oaxaca, Mexico. Project report submitted to the Heinz Family Foundation, Pittsburgh. Mitton, R. 1981. Teotihuacan: City, state, and civilization, pp. 198-243. In Sabloff, J. A., ed., Supplement to the Handbook of Middle American Indians. University of Texas Press, Austin. Minc, L. D., R. J. SHERMAN, C. ELtson, M. Winter, E. M. REDMOND, AND C. S. SPENCER. 2016. Ceramic provenance and the regional organization of pottery production during the later FIELDIANA: ANTHROPOLOGY Formative periods in the Valley of Oaxaca, Mexico: Results of trace-element and mineralogical analyses. Journal of Archaeo- logical Science: Reports, 8: 28—46. Montero GuzMAn, M. D. 2004. Huehuetéotl: Origen e inter- pretacion de una deidad tutelary durante el Formativo Terminal en Cantona, Puebla. Tesis de Licenciatura, Escuela Nacional de Antropologia e Historia, Mexico City. NEELy, J. A., CARAN, S. C., AND B. M. WinsporouGu. 1990. Irrigated agriculture at Hierve el Agua, Oaxaca, Mexico, pp. 115-189. In Marcus, J., ed., Debating Oaxaca Archaeology. Anthropological Papers No. 84. Museum of Anthropology, University of Michigan, Ann Arbor. NicuHo as, L. M. 1989. Land use in prehispanic Oaxaca, pp. 449-— 505. In Kowalewski, S. A., G. M. Feinman, L. Finsten, R. E. Blanton, and L. M. Nicholas, Monte Alban’s Hinterland, Pt. II: The Prehispanic Settlement Patterns in Tlacolula, Etla, and Ocotlan, the Valley of Oaxaca, Mexico. Memoirs No. 23. Museum of Anthropology, University of Michigan, Ann Arbor. Nicuots, D. L. 1996. An overview of regional settlement pattern survey in Mesoamerica: 1960-1995, pp. 59-95. In Mastache, A. G., J. R. Parsons, R. S. Santley, and M. C. Serra Puche, eds., Arqueologia mesoamericana: Homenaje a William T. Sanders, Vol. 1. Instituto Nacional de Antropologia e Historia, Mexico City. . 2016. Teotihuacan. Journal of Archaeological Research, 24: 1-74. Oper, F. A. 1887. Travels in Mexico and Life among the Mexicans. Rev. ed. Estes and Lauriat, Boston. O’Brien, M. J., R. D. MAson, D. E. LEwWARCH, AND J. A. NEELY. 1982. A Late Formative Irrigation Settlement below Monte Alban: Survey and Excavation on the Xoxocotlan Piedmont, Oaxaca, Mexico. Institute of Latin American Studies, Univer- sity of Texas, Austin. Orson, M. 1993. Dictatorship, democracy, and development. American Political Science Review, 87: 567-576. Ortiz Diaz, E. 2002. Los zapotecos de la Sierra de Juarez: jAntiguos orfebres? Anales del Instituto de Investigaciones Esteéticas, 81: 141-149. Oupuk, M. 2000. Historiography of the Benizaa: The Postclassic and Early Colonial Periods (1000-1600 AD). Research School CNWS, Leiden University, Leiden. . 2002. The Zapotec city-state, pp. 73-90. Jn Hansen, M. H., ed., A Comparative Study of Six City-State Cultures: An Investigation Conducted by the Copenhagen Polis Centre. Royal Danish Academy of Sciences and Letters, Copenhagen. . 2008. The Postclassic period in the Valley of Oaxaca: The archaeological and ethnohistorical records, pp. 95-118. In Blomster, J. P., ed., After Monte Alban: Transformation and Negotiation in Oaxaca, Mexico. University Press of Colorado, Boulder. Pappock, J. 1983a. The rise of the Nuifie centers in the Mixteca Baja, pp. 208-211. Jn Flannery, K. V., and J. Marcus, eds., The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civilizations. Academic Press, New York. . 1983b. Monte Alban II in the Yagul—Caballito Blanco area, pp. 115-117. Jn Flannery, K. V., and J. Marcus, eds., The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civilizations. Academic Press, New York. . 1983c. Lord 5 Flower’s Family: Rulers of Zaachila and Cuilapan. Publications in Anthropology No. 29. Vanderbilt University, Nashville, TN. . 1983d. Lambityeco, pp. 197-204. Jn Flannery, K. V., and J. Marcus, eds., The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civilizations. Academic Press, New York. PALERM, A., AND E. R. Wo r. 1957. Ecological potential and cultural development in Mesoamerica. Pan American Union Social Science Monograph, 3: 1-37. PARKER, B. J. 2006. Toward an understanding of borderland processes. American Antiquity, 71: 77-100. Parsons, E. C. 1936. Mitla: Town of the Souls and Other Zapotec-Speaking Pueblos of Oaxaca, Mexico. University of Chicago Press, Chicago. Parsons, J. R. 1971. Prehistoric Settlement Patterns in the Texcoco Region, Mexico. Memoirs No. 3. Museum of Anthropology, University of Michigan, Ann Arbor. —. 2009. Reflections on my life in archaeology. Ancient Mesoamerica, 20: 3-13. . 2010. The pastoral niche in pre-Hispanic Mesoamerica, pp. 109-136. Jn Staller, J. E., and M. Carrasco, eds., Pre- Columbian Foodways: Interdisciplinary Approaches to Food, Culture, and Markets in Ancient Mesoamerica. Springer, New York. Parsons, J. R., E. M. BrumMFiEL, M. H. Parsons, AND D. J. WILSON. 1982. Prehispanic Settlement Patterns in the Southern Valley of Mexico: The Chalco-Xochimilco Region. Memoirs No. 14. Museum of Anthropology, University of Michigan, Ann Arbor. Parsons, J. R., K. W. KintiGH, AND S. A. GrecG. 1983. Archaeological Settlement Pattern Data from the Chalco, Xochimilco, Ixtapalapa, Texcoco, and Zumpango Regions, Mexico. Technical Reports No. 14. Museum of Anthropology, University of Michigan, Ann Arbor. PEREZ RopriGuez, V., A. MARTINEZ TUNON, L. Minc, L. STIVER WALSH, AND M. NAvArRRO RosALEs. 2017. Chronological trends in the use of Valley of Oaxaca ceramics and ceramic styles at Cerro Jazmin, Mixteca Alta, Oaxaca. Journal of Archaeolog- ical Science: Reports, 12: 580-590. Peterson, D. A., M. C. WINTER, AND W. P. Hewitt. 1989. Reply to Doolittle. American Antiquity, 54: 847-850. Pires-FERREIRA, J. W. 1975. Formative Mesoamerican Exchange Networks with Special Reference to the Valley of Oaxaca. Memoirs No. 7. Museum of Anthropology, University of Michigan, Ann Arbor. . 1976. Obsidian exchange in Formative Mesoamerica, pp. 292-306. In Flannery, K. V., ed., The Early Mesoamerican Village. Academic Press, New York. PLoc, S. 1976. Measurement of prehistoric interaction between communities, pp. 255-272. In Flannery, K. V., ed., The Early Mesoamerican Village. Academic Press, New York. PLUNKET, P. S. 1983. An Intensive Survey in the Yucuita Sector of the Nochixtlan Valley, Oaxaca, Mexico. Ph.D. dissertation, Department of Anthropology, Tulane University, New Or- leans. PLUNKET, P. S., AND G. URUNUELA. 2005. Recent research in Puebla prehistory. Journal of Archaeological Research, 13: 89— 2a . 2012. Where East meets West: The Formative in Mexico’s Central Highlands. Journal of Archaeological Re- search, 20: 1—S1. Pout, J. D. 1999. The lintel paintings of Mitla and the function of the Mitla palaces, pp. 177-197. In Kowalski, J. K., ed., Mesoamerican Architecture as a Cultural Symbol. Oxford University Press, New York. . 2003a. Creation stories, hero cults, and alliance building: Confederacies of central and southern Mexico, pp. 61-66. Jn Smithy) M"E:, and F: F: Berdan, eds.; The” Postclassic Mesoamerican World. University of Utah Press, Salt Lake City. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 125 . 2003b. Ritual ideology and commerce in the southern Mexican highlands, pp. 172-177. In Smith, M. E., and F. F. Berdan, eds., The Postclassic Mesoamerican World. University of Utah Press, Salt Lake City. ———. 2003c. Royal marriage and confederate building among the Eastern Nahuas, Mixtecs, and Zapotecs, pp. 243-248. In Smith, Me. Eseend »P.ehayBerdanseeds.5 Lhe Postelassic Mesoamerican World. University of Utah Press, Salt Lake City. Pounps, N. J. 1969. The urbanization of the classical world. Annals of the Association of American Geographers, 59: 135— 157. Prescott, J. R. 1965. The Geography of Frontiers and Boundaries. Aldine, Chicago. ReDMoND, E. M. 1983. A Fuego y Sangre: Early Zapotec Imperialism in the Cuicatlan Canada, Oaxaca. Memoirs No. 16. Museum of Anthropology, University of Michigan, Ann Arbor. REDMOND, E. M., AND C. S. SPENCER. 2006. From raiding to conquest: Warfare strategies and early state development in Oaxaca, Mexico, pp. 336-393. In Arkush, E. N., and M. W. Allen, eds., The Archaeology of Warfare: Prehistories of Raiding and Conquest. University Press of Florida, Gainesville. RENFREW, C. 1982. Polity and power: Interaction, intensification and exploitation, pp. 264-290. In Renfrew, C., and M. Wagstaff, eds., An Island Polity: The Archaeology of Exploi- tation in Melos. Cambridge University Press, Cambridge. . 2003. Retrospect and prospect: Mediterranean archaeol- ogy in a new millennium, pp. 311-318. Jn Papadopoulos, J. K., and R. M. Leventhal, eds., Theory and Practice in Mediterra- nean Archaeology: Old World and New World Perspectives. Cotsen Institute of Archaeology, University of California, Los Angeles. Rice, P. M. 1998. Contexts of contact and change: Peripheries, frontiers, and boundaries, pp. 44-66. Jn Cusick, J. G., ed., Studies in Culture Contact: Interaction, Culture Change, and Archaeology. Occasional Paper No. 25. Center for Archaeo- logical Investigations, Southern Illinois University, Carbon- dale. Rosies GArciA, N. M. 1994. Las canteras de Mitla, Oaxaca: Tecnologia para la arquitectura monumental. Publications in Anthropology No. 47. Vanderbilt University, Nashville, TN. Rosies Garcia, N. M., AND A. JUAREZ OSNAYA. 2009. Com- plejidad arquitectonica en Mitla: Descubrimientos recientes y pasados, pp. 379-397. In Robles Garcia, N. M., ed., Bases de la complejidad social en Oaxaca: Memoria de la Cuarta Mesa Redonda de Monte Alban. Instituto Nacional de Antropologia e Historia, Mexico City. RopsetH, L., AND B. J. PARKER. 2005. Introduction: Theoretical considerations in the study of frontiers, pp. 3-21. Jn Parker, B. J., and L. Rodseth, eds., Untaming the Frontiers in Anthro- pology, Archaeology, and History. University of Arizona Press, Tucson. Roscore, P. B. 1996. War and society in Sepik New Guinea. Journal of the Royal Anthropological Institute, 2: 645-666. . 2013. War, collective action, and the “evolution” of human polities, pp. 57-82. Jn Carballo, D. M., ed., Cooperation and Collective Action. University Press of Colorado, Boulder. SABLOFF, J. A., AND W. AsHMoreE. 2001. An aspect of archae- ology’s recent past and its relevance in the new millennium, pp. 11-32. In Feinman, G. M., and T. D. Price, eds., Archaeology at the Millennium: A Sourcebook. Springer, New York. SANCHEZ Lopez, A. 1989. Oaxaca tierra de maguey y mezcal. Instituto Tecnologico de Oaxaca, Oaxaca, Mexico. 126 SANDERS, W. T. 1965. The cultural ecology of the Teotihuacan Valley: A preliminary report of the results of the Teotihuacan Valley Project. Manuscript on file, Department of Anthropol- ogy, Pennsylvania State University, University Park. SANDERS, W. T., A. Kovar, T. CHARLTON, AND R. A. DIEHL. 1970. The Teotihuacan Valley Project Final Report, Vol. 1: The Natural Environment, Contemporary Occupation, and 16th Century Population of the Valley. Occasional Papers in Archaeology No. 1. Pennsylvania State University, University Park. SANDERS, W. T., J. R. PARSONS, AND R. S. SANTLEY. 1979. The Basin of Mexico: Ecological Processes in the Evolution of a Civilization. Academic Press, New York. SANDERS, W. T., AND B. Price. 1968. Mesoamerica: The Evolution of a Civilization. Random House, New York. SAVILLE, M. H. 1900. Cruciform structures near Mitla. American Museum of Natural History, Bulletin, 13: 201-218. . 1909. The cruciform structures of Mitla and vicinity, pp. 151-190. In Boas, F., ed., Anthropological Essays Presented to Frederic Ward Putnam in Honor of his Seventieth Birthday. Stechert and Company, New York. SCHMIEDER, O. 1930. The Settlement of the Tzapotec and Mije Indians: State of Oaxaca, Mexico. University of California Press, Berkeley. SCHOENWETTER, J. 1974. Pollen records of Guila Naquitz cave. American Antiquity, 39: 292-303. ScHON, R., AND M. L. Gataty. 2006. Diachronic frontiers: Landscape archaeology in highland Albania. Journal of World- Systems Research, 7: 231-262. SCHORTMAN, E. M., AND P. A. URBAN. 1992. Current trends in interaction research, pp. 235-255. Jn Schortman, E. M., and P. A. Urban, eds., Resources, Power, and Interregional Interac- tion. Plenum Press, New York. . 2012. Networks, cores, and peripheries: New frontiers in interaction studies, pp. 471-481. Jn Nichols, D. L., and C. A. Pool, eds., The Oxford Handbook of Mesoamerican Archae- ology. Oxford University Press, Oxford. SHELDON, S. 1980. Ethnobotany of Agave lecheguilla and Yucca carnerosana in Mexico’s zona ixtlera. Economic Botany, 34: 376-390. SHERMAN, R. J. 2005. Settlement Heterogeneity in the Zapotec State: A View from Yaasuchi, Oaxaca, Mexico. Ph.D. dissertation, Department of Anthropology, University of Michigan, Ann Arbor. SHERMAN, R. J., L. Minc, C. M. Exson, C. S. SPENCER, AND E. M. REDMOND. 2011. Trace-element analysis on pottery from Oaxaca. American Museum of Natural History online, accessed October 24, 2011. SmiTH, C. A. 2002. Concordant Change and Core-Periphery Dynamics: A Synthesis of Highland Mesoamerican Archaeo- logical Survey Data. Ph.D. dissertation, Department of Anthropology, University of Georgia, Athens. SmiTH, C. E., Jr., AND T. KERR. 1968. Pre-conquest plant fibers from the Tehuacan Valley, Mexico. Economic Botany, 22: 354— 358. SmitH, M. E. 2002. The earliest cities, pp. 3-19. In Gmelch, G., and W. P. Zenner, eds., Urban Life: Readings in the Anthropology of the City. Waveland Press, Prospect Heights, IL. —. 2003. Aztec city-states in the Basin of Mexico and Morelos, pp. 58-60. Jn Smith, M. E., and F. F. Berdan, eds., The Postclassic Mesoamerican World. University of Utah Press, Salt Lake City. FIELDIANA: ANTHROPOLOGY . 2004. The archaeology of ancient state economies. Annual Review of Anthropology, 33: 73-102. . 2005. City size in Late Postclassic Mesoamerica. Journal of Urban History, 31: 403-434. SMITH, M. E., AND F. F. BERDAN. 2000. The Postclassic Mesoamerican world system. Current Anthropology, 42: 283- 286. SmitH, M. E., G. M. FEINMAN, R. DRENNAN, T. EARLE, AND I. Morris. 2012. Archaeology as a social science. Proceedings of the National Academy of Sciences of the United States of America, 109: 7617-7621. SmitH, M. L. 2005. Networks, territories, and the cartography of ancient states. Annals of the Association of American Geographers, 95: 832-849. . 2007. Territories, corridors, and networks: A biological model for the premodern state. Complexity, 12(4): 28-35. . 2012. Boundaries in action: The past and present of delimitation. Anthropology News, 53(2): 3, 5. SPENCER, C. S. 1982. The Cuicatlan Canada and Monte Alban: A Study of Primary State Formation. Academic Press, New York. . 2003. War and early state formation in Oaxaca, Mexico. Proceedings of the National Academy of Sciences of the United States of America, 100: 11185—11187. . 2007. Territorial expansion and primary state formation in Oaxaca, Mexico, pp. 55-72. In Chacon, R. J., and R. G. Mendoza, eds., Latin American Indigenous Warfare and Ritual Violence. University of Arizona Press, Tucson. SPENCER, C. S., AND E. M. REDMoNnpD. 1997. Archaeology of the Canada de Cuicatlan, Oaxaca. Anthropological Papers No. 80. American Museum of Natural History, New York. . 2001. Multilevel selection and political evolution in the Valley of Oaxaca, 500-100 BC. Journal of Anthropological Archaeology, 20: 195-229. . 2003. Militarism, resistance, and early state development in Oaxaca, Mexico. Social Evolution & History, 2(1): 25—70. . 2006. Resistance strategies and early state formation in Oaxaca, Mexico, pp. 21-43. Jn Elson, C. M., and R. A. Covey, eds., Intermediate Elites in Pre-Columbian States and Empires. University of Arizona Press, Tucson. SPENCER, C. S., E. M. REDMOND, AND C. M. ELson. 2008. Ceramic microtypology and the territorial expansion of the early Monte Alban state in Oaxaca, Mexico. Journal of Field Archaeology, 33: 321-341. Spores, R. 1965. The Zapotec and Mixtec at Spanish conquest, pp. 962-986. In Willey, G. R., ed., Handbook of Middle American Indians, Vol. 3: Archaeology of Southern Meso- america, Pt. 2. University of Texas Press, Austin. . 1967. The Mixtec Kings and Their People. University of Oklahoma Press, Norman. 1972. An Archaeological Settlement Survey of the Nochixtlan Valley, Oaxaca. Publications in Anthropology No. 1. Vanderbilt University, Nashville, TN. . 1974. Marital alliance in the political integration of Mixtec kingdoms. American Anthropologist, 76: 297-311. . 1993. Tututepec: A Postclassic-period Mixtec conquest state. Ancient Mesoamerica, 4: 167-174. STARK, M. T. 1998. Technical choices and social boundaries in material culture patterning: An introduction, pp. I-11. Jn Stark, M. T., ed., The Archaeology of Social Boundaries. Smithsonian Institution Press, Washington, DC. Stiver, L. R. 2001. Prehispanic Mixtec Settlement and State in the Teposcolula Valley of Oaxaca, Mexico. Ph.D. dissertation, Department of Anthropology, Vanderbilt University, Nash- ville, TN. Stocker, T. L., AND R. H. CoBEAN. 1984. Preliminary report on the obsidian mines at Pico de Orizaba, Veracruz, pp. 83-95. In Ericson, J. E., and B. A. Purdy, eds., Prehistoric Quarries and Lithic Production. Cambridge University Press, Cambridge. STODDART, S. K. 2010. Boundaries of the state in time and space: Transitions and tipping points. Social Evolution & History, 9(2): 28-52. SwyNGEDOouw, E. 1997. Neither global nor local: “Glocalization” and the politics of scale, pp. 137-166. Jn Cox, K. R., ed., Spaces of Globalization: Reasserting the Power of the Local. Guilford Press, New York. TAYLor, W. B. 1972. Landlord and Peasant in Colonial Oaxaca. Stanford University Press, Stanford, CA. TERRELL, J. E. 2010. Language and material culture on the Sepik Coast of Papua New Guinea: Using social network analysis to simulate, graph, identify, and analyze social and cultural boundaries between communities. Journal of Island and Coastal Archaeology, 5: 3—32. TOKOVININE, A. 2013. Place and Identity in Classic Maya Narratives. Dumbarton Oaks, Washington, DC. Totstoy, P. 1988. Comment on “Ecological theory and cultural evolution in the Valley of Oaxaca.” Current Anthropology, 29: 66-67. TRINKHAUus, K. M. 1984. Boundary maintenance strategies and archaeological indicators, pp. 35—49. Jn De Atley, S. P., and F. J. Findlow, eds., Exploring the Limits: Frontiers and Bound- aries in Prehistory. BAR International Series 223. Archaeo- press, Oxford. Tweepig, E. 1911. Mexico as I Saw It. Thomas Nelson and Sons, London. Urap, J. S. 1992. La Tumba 5 del Cerro de la Campana, Suchilquitongo, Oaxaca: Un analisis epigrafico. Arqueologia, 8: 73-112. —. 2003. A Zapotec slab in Santiago Matatlan, Oaxaca. Mesoamerican Voices, 1: 65-90. Urcip, J. S., M. WINTER, AND R. MATADAMos. 1994. Nuevos monumentos grabados en Monte Alban, Oaxaca, pp. 2-52. In Winter, M., ed., Escritura zapoteca prehispanica: Nuevas aportaciones. Contribucion 4. Proyecto Especial Monte Alban 1992-1994, Oaxaca, Mexico. VARNER, D. 1974. Prehispanic Settlement Patterns in the Valley of Oaxaca, Mexico: The Etla Arm. Ph.D. dissertation, Depart- ment of Anthropology, University of Arizona, Tucson. WALLERSTEIN, I. 1974. The Modern World-System. Academic Press, New York. Wuitr, D. A., AND S. B. BARBER. 2012. Geospatial modeling of pedestrian transportation networks: A case study from Precolumbian Oaxaca, Mexico. Journal of Archaeological Science, 39: 2684-2696. WuiteEcoTTon, J. W. 1977. The Zapotecs: Princes, Priests and Peasants. University of Oklahoma Press, Norman. WIiLLEy, G. R. 1953. Prehistoric Settlement Patterns in the Viru Valley, Peru. Bureau of American Ethnology, Bulletin 155. Smithsonian Institution, Washington, DC. WiLuiaMs, H., AND R. F. Heizer. 1965. Geological notes on the ruins of Mitla and other Oaxacan sites, Mexico. Contributions of the University of California Archaeological Research Facility, 1: 40-54. Winter, M. C. 1974. Residential patterns at Monte Alban, Oaxaca, Mexico. Science, 186: 981—987. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 127 1989. From Classic to Postclassic in pre-Hispanic Oaxaca, pp. 123-130. In Diehl, R. A., and J. C. Berlo, eds., Mesoamerica after the Decline of Teotihuacan, AD 700-900. Dumbarton Oaks, Washington, DC. —. 1997. Classic to Postclassic in the Valley of Oaxaca: A view from Monte Alban, pp. 21-42. Jn Jones, C., and T. Jones, eds., U Mut Maya IV: Reports and Readings Inspired by the Advanced Seminars Led by Linda Schele at the University of Texas at Austin 1994-1996. Eureka Printing Company, Bay- side, CA. . 1998. Monte Alban and Teotihuacan, pp. 153-184. In Rattray, E. C., ed., Rutas de intercambio en Mesoamerica. Universidad Nacional Autonoma de México, Mexico City. . 2001. Palacios, templos y 1300 anos de vida urbana en Monte Alban, pp. 277-301. Jn Ciudad Ruiz, A., M. J. Iglesia Ponce de Leon, and M. del C. Martinez Martinez, eds., Reconstruyendo la ciudad maya: El urbanismo en las socie- dades antiguas. Sociedad Espanola de Estudios Mayas, Madrid. Winter, M. C., R. MARKENS, C. MARTINEZ LOPEZ, AND A. HERRERA Muzco T. 2007. Shrines, offerings, and Postclassic continuity in Zapotec religion, pp. 185-212. Jn Gonlin, N., and J. C. Lohse, eds., Commoner Ritual and Ideology in Ancient Mesoamerica. University Press of Colorado, Boulder. WoLr, E. R. 1982. Europe and the People without History. University of California Press, Berkeley. WorkINGER, A. 2002. Coastal/Highland Interaction in Prehis- panic Oaxaca, Mexico: The Perspective from San Francisco de Arriba. Ph.D. dissertation, Department of Anthropology, Vanderbilt University, Nashville, TN. WorkKINGER, A., AND A. Joyce. 2009. Reconsidering warfare in Formative period Oaxaca, pp. 3-38. Jn Orr, H., and R. Koontz, eds., Blood and Beauty: Organized Violence in the Art and Archaeology of Mesoamerica and Central America. Cotsen Institute of Archaeology Press, Los Angeles. ZEITLIN, J. F. 1978. Community Distribution and Local Economy on the Southern Isthmus of Tehuantepec: An Archaeological and Ethnohistorical Investigation. Ph.D. dissertation, Depart- ment of Anthropology, Yale University, New Haven, CT. ZEITLIN, R. N. 1979. Prehistoric Long-Distance Exchange on the Southern Isthmus of Tehuantepec, Mexico. Ph.D. dissertation, Department of Anthropology, Yale University, New Haven, CT. . 1993. Pacific coastal Laguna Zope: A regional center in the Terminal Formative hinterlands of Monte Alban. Ancient Mesoamerica, 4: 85-101. FIELDIANA: ANTHROPOLOGY Appendix I: Site Descriptions This appendix includes all sites in the Guirtn survey region that have at least one pre-Hispanic mound, terrace, or other significant feature. Sites are listed alphabetically by the assigned field site name. For each site, we provide general information on the environmental setting and location relative to modern villages, roads, rivers, and other topo- graphic features. We describe the pre-Hispanic architecture and present the number of mounds and/or terraces, the phases of occupation (phase component numbers are given in parentheses), size (in hectares), and significant artifacts, including obsidian and other chipped or ground stone, shell, greenstone, and jade. Ceramics were found on most sites, so only the absence of ceramics is noted. Mention is made, however, of special or unusual ceramics, such as figurine and urn fragments, kilnwasters, spindle whorls, sherd disks, or unusual concentrations of specific types, such as poly- chromes. TL-MAT-MAT-95 (Nueve Puntas) This single-component Monte Alban V site (V-57) site is located on the top of Nueve Puntas, the high mountain ridge that defines the southeastern edge of the Valley of Oaxaca. The site extends over an area of 9.4 ha and consists of 115 residential terraces and several long defensive walls on the top and western side of three of the nine points. Several of the walls are 2-3 m high. Part of the eastern side of the mountain is a rock cliff that drops more than 500 m. Because they have never been farmed, many of the terraces are very well preserved, with visible house foundations and ramped entries. On one very well preserved terrace, the house foundation was built directly on bedrock, with preserved walls still standing to a height of 0.5 m. Because of heavy leaf cover, there are few visible artifacts; the only diagnostics are G-3Ms. Although the site was recorded during the Guirtn/ Albarradas regional survey, this settlement is in the physiographic area of eastern Tlacolula (Figs. 2.23—2.25 and 7.11—7.14). TL-MIT-CC-3 This small (1.1 ha) single-component site is located in the low piedmont 1.2 km northeast of Corral del Cerro (now called San Juan del Paso), in the small mountain valley in the northern part of the survey area. The only ceramics are Monte Alban V (V-47), mostly G-3Ms. There is a poor-quality chert outcrop approximately 500 m west of the site, and we found chipping and other chert debris on the site. TL-MIT-CC-5 This multicomponent site is located on two flat ridgetops in the middle piedmont below a hill known as El Sombrero, approximately 1.5 km north of Corral del Cerro (now called San Juan del Paso), in the small mountain valley in the northern part of the survey area. We mapped 62 terraces on the two ridges; as many as 10 more were too destroyed to map. Remains of house foundations are visible on many of the terraces. There is no central architectural complex, but several low platforms are dispersed across the site. One of the platforms had been looted in the past. The site was first settled and was largest in Monte Alban II (I-10); Monte Alban II pottery is present on all terraces and in nearby areas, covering an area of 10 ha. More than half of the Monte Alban II pottery on the surface is a local variety called Black and White (T5100), and it is possible that this ware was produced on or near the site. Consider- able chert debris indicates stone working in Monte Alban II. Special artifacts include nine pieces of obsidian and one Oliva shell. Early Classic pottery is present on only one terrace on the main ridge (IIIA-11). Monte Alban V ceramics are present on eight terraces, in two clusters on the main ridge, covering a total area of 1 ha (V-60) (Figs. 4.5—4.7, 4.10, and 4.15). TL-MIT-CC-7 This small (0.25 ha) multicomponent site is located in the middle piedmont just below a source of white chert, approximately 2 km northwest of Corral del Cerro (now known as San Juan del Paso), in the small mountain valley in the northern part of the survey area. Monte Alban Late I (LI- 4) and II (II-3) ceramics were present at this isolated residence. Two broken projectile points and other chert debris from the source indicate possible stone working. TL-MIT-CC-8 (El Nopal) This multicomponent site covers an area of 1.85 ha on the top of a high piedmont hill called El Nopal, approximately 5.5 km northwest of Corral del Cerro (now known as San Juan del Paso), in the extreme northwestern part of the survey area. The hill is very steep and heavily vegetated. The site is very defendable, with a compact complex of two platforms and five small structures arranged around a small plaza on the very top of the hill, surrounded on three sides by a stone defensive wall. On the slopes below the architectural complex are terraces that we could not map or measure, but we estimate that there were as many as 30 terraces and a few other possible houses, plus remnants of defensive walls on the north and west. Most of the collected ceramics are Monte Alban Late I (LI-9) or II (II-7). Most of the Monte Alban II pottery is the local variety of Black and White (T5100), and it is likely that this ware was made on or near the site. We used an estimate of 30 houses at the site to calculate a population of 150-300 people for those phases. Monte Alban IIIA (IIJA-7) ceramics are present but in much lower quantities; we instead used very light density to estimate the population between 19 and 46. Chert tools (especially blade/unifaces, some made from nonlocal material) and debris from diverse sources are abundant on the surface, indicating that chert working was an important economic activity (Figs. 3.32, 3.33, 4.13, and 4.16). FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 129 L-SLA-SLA-8 (El Calvario) This multicomponent site extends over 7.8 ha on the top of a steep-sided hill in the middle piedmont 250 m west of the main road through San Lorenzo Albarradas. The hill is known locally as El Calvario. There is a small architectural complex of six mounds and two plazas on the top of the hill. One plaza is large and open, whereas the second plaza is much smaller and enclosed on three sides by small structures. This architectural complex is surrounded by 51 small terraces on the upper slopes of the hill. The site was largest in Monte Alban Late I (LI-11) and II (II-9), when all of the terraces were occupied. The dominant pottery in Monte Alban II is a local variety called Black and White (T5100). Monte Alban Early I (EI-2) and HIA (IIIA-10) ceramics are present only on the top of hill and one terrace in an area of 0.5 ha. The mounds and plazas appear to have been constructed during Monte Alban Late I and II. Monte Alban V (V-59) ceramics are present only in the smaller plaza. There is a black chert source on the site, and there are dozens of pieces of worked chert on the surface. There also is a high quantity of ground stone tool fragments (almost two dozen), including manos, metates, and axes. These high quantities are indicative of chert working and the production of ground stone tools. Although these materials are found in parts of the site with ceramics from Monte Alban Early I- IIIA, they most likely pertain to Monte Alban Late I and II, when the site was largest. Ten obsidian blades and one piece of marine shell (Oliva sp.) were collected in areas where most of the pottery is Late I and II (Figs. 3.13, 3.21, 3.23, 3.24, 4.11, and 5.5). TL-SLA-SLA-11 (El Portazuelo) This small (<1 ha) single-component site is located on top of a low mountain called El Portazuelo on the northern slopes of Cerro Guirone. The mountain is on the south side of the saddle where the modern highway between Mitla and the Mixe region passes, approximately 3 km east of Corral del Cerro (as the crow flies). We mapped one terrace, but the presence of cut stone foundations of three isolated small structures indicates that there were more. There are several short defensive walls on the southern slope of the peak. The site had an excellent view in many directions, including to Mitla to the west. The site may have been a small redoubt used for monitoring the pass. The only diagnostic ceramics are Monte Alban V G-3Ms (V-3). The only other artifacts were a chert biface and a rectangular piece of onyx (Fig. 7.17). TL-SLA-SLA-12 (El Portazuelo) This small site (<1 ha) is located on top of the southern peak of El Portazuelo, on the northern side of Cerro Guirone approximately 1.25 km from Guirone’s peak. We mapped a linear arrangement of 11 terraces along the ridgetop, including four small platforms and structures containing cut stone blocks. Because of heavy leaf cover in this mountainous area, we did not see any diagnostic ceramics. The character of the site is very similar to the remains we mapped on the northern peak of El Portazuelo (TL-SLA-SLA-11), 400 m to the north, 130 where we collected G-3Ms diagnostic of Monte Alban V, and other Monte Alban V sites in mountainous areas of the valley, so we designated this site as V-51 (Fig. 7.18). TL-SLA-SLA-13 This small (<1 ha) single-component site is located in a small flat area on a high piedmont ridge descending the southern peak of El Portazuelo, just above the modern highway, approximately | km west of the turn off to San Lorenzo Albarradas. We mapped three small terraces. There is a 35-m-long defensive wall below the three terraces. The few sherds on the surface are Monte Alban V (V-4). A white chert point was found on the slope above the site. The site is situated above a black chert source (Fig. 7.19). TL-SLA-SLA-15 This small (<1 ha) single-component site is located on the top of a middle piedmont hill southeast of the intersection of the main highway and the road to San Lorenzo Albarradas. Ceramics include a dozen G-3M sherds and two café sahumador handles, dating the site to Monte Alban V (V-6). We observed several obsidian blades. The most significant aspect of the site is the very high density of chert-working debris in a very small area. TL-SLA-SLA-17 (Guirtn and La Cuadrada) This large (98 ha) multicomponent site is located on a series of low mountains and high piedmont ridges on the western slopes of Cerro Guirone, at the extreme eastern end of the Tlacolula Valley, approximately 5 km southeast of San Pablo Villa de Mitla. The site is just north of the mountain pass traversed by the Camino Real. The site was visited by early explorers in the 19th century (Dupaix, 1834; Ober, 1887) and has long been in the literature, known for two cruciform tombs, standing adobe walls, and a two-mound temple associated with a ballcourt complex. Marshall Saville (1900, 1909) carried out early excavations of the tombs. The size of the site was not known until it was mapped during the 1980 Tlacolula Valley survey (Kowalewski et al., 1989). During the 1995 Guirin survey project, we discovered that the site extends beyond the 1980 survey boundary, also encompassing the La Cuadrada stone quarry to the north (Holmes, 1987; Robles Garcia, 1994), one of the quarries from which large stone blocks were cut for the palaces in Mitla. The Guirun site combines two field site numbers: TL-SLA-SLA-2 (1980) and TL-SLA-SLA-17 (1995). We completed a more complete map of the site during an intensive survey in 1996. We have published our more recent findings (Feinman & Nicholas, 2004a) and present only a brief description here. The site consists of a minimum of 330 residential terraces dispersed across three major ridges between Cerro Guirone and the La Cuadrada quarry. The best-preserved terraces are on several lower, steep-sided ridges. The upper, flatter reaches of the site have seen much destruction by farming, and many terraces in those areas were too destroyed to be mapped. FIELDIANA: ANTHROPOLOGY Complexes of platforms or mounds are present on virtually every natural ridgetop and flat saddle of the ridge system descending the western face of Cerro Guirone. Groups of terraces are associated with most of these complexes. The most formal architecture, including the aforementioned tombs, is located on the top of two lower ridges closer to the valley floor. The earliest occupation at the site was very small; Monte Alban Late I ceramics are present only on two nearby terraces (LI-13). The site remained small in the subsequent occupation; Monte Alban II ceramics are present on five terraces and one platform widely dispersed across all three major ridge systems (11-13-18). The site grew tremendously in the Classic period, covering 47 ha in the Early Classic (IIIA-13) and 80 ha in the Late Classic (IV-13). Approximately 85% of the mapped terraces were occupied by the Early Classic; virtually all were occupied by the Late Classic. The extent of the area with Postclassic pottery is 98 ha, including the highest reaches of the occupied area and La Cuadrada quarry. We did not find Monte Alban V (V-62) on all the terraces (about 83%), and except for the El Paderon part (the area with the standing adobe walls) of the site, Postclassic ceramics are sparse in many other areas, indicating a more dispersed settlement in Monte Alban V. Exotic materials include obsidian from a range of sources and a small greenstone pendant carved in the form of a human face. We observed extensive working of chert and other stone materials into a range of diverse objects; chert debris and tools were spread across all sectors of the site. There are at least three chert sources on the site and one basalt source nearby (Figs. 2.14—2.20, 5.9-5.13, and 7.4). TL-SLA-SLA-19 This multicomponent site is located on a middle piedmont slope in the mountains at the eastern edge of Tlacolula, approximately 300 m west of the mountain pass south of Cerro Guirone, near the route of the Camino Real. The site is on the southern side of the Camino Real, less than 300 m east of Guirtn (which is north of the route). Together, these two sites could have monitored traffic entering and leaving the Valley of Oaxaca. The site was heavily plowed in the past; we could see remains of more than 15 small house foundations, but they were too destroyed to map. We mapped only one terrace but suspect that there were more. There are few artifacts but one piece of obsidian. The site was largest (4.9 ha) in Monte Alban IIIA (IITA-2) and IV (IV-3). The site was a much smaller, isolated residence (<.25 ha) in Monte Alban V (V-9). TL-SLA-SLA-20 (Loma de las Tumbas) This multicomponent site covers an area of 5.6 ha on the top of a defendable knob near the end of long narrow piedmont ridge that descends to the Hormiga Colorada from the eastern side of El Tigre, the high mountain ridge in eastern Tlacolula between Cerro Guirone and Nueve Puntas. The site 1s located approximately half the way between the road to Roaguia and the Hormiga Colorada and has an excellent view in all directions. The Camino Real passes just south of this site. There are two small, adjacent plazas on the top of the knob; the larger plaza has two mounds at opposite ends. Just west of the small plazas are the remains of 12 small house foundations; many have visible tombs, several of which were looted. Descending the slopes below on the west, south, and east are 28 terraces. We found Monte Alban Late I (LI-7) and II (II-5) ceramics across the entire site, including dozens of Black and White (T5100) sherds and half a dozen cremas or imitation cremas. One terrace near the top also was occupied in Monte Alban IIIA (IIIA-5) and V (V-55). We recorded an unusually high number of special artifacts, including seven shell ornaments, two stone beads, and lapidary debris. The shell ornaments include a fragment of Pinctada mazatlanica, a Thais triangularis whole-shell bead, two Olivella sp. whole- shell beads, an Oliva porphyria pendant, a Spondylus sp. tubular bead, and a gastropod pendant cut in the shape of an animal head. Three dozen pieces of chert, including bifaces, cores, and debris, indicate stone working in Monte Alban Late I and II. There also are at least 10 manos and hammerstones but only two pieces of obsidian (Figs. 3.27, 3.28, 4.8, 4.9, and 4.12). TL-SLA-SLA-22 (Sitio de las Tunas) This multicomponent site covers an area of 5.2 ha on a middle piedmont slope on the eastern side of El Tigre about half the way between San Lorenzo and Roaguia, approxi- mately 600 m east of the road. The site is known locally as Sitio de las Tunas. There are six small structures and a small plaza in the center of the site, which are surrounded by at least eight terraces on the west and south. The site has been plowed, destroying additional terraces and house foundations. The entire site is Monte Alban Late I (LI-8) and II (JI-6). All of the structures and most of the terraced slope also were occupied in Monte Alban V (V-56). Only two terraces were occupied in Monte Alban IIIA (IIIA-6). We noted a dozen pieces of worked chert and flakes including four cores, nine pieces of obsidian, one ground stone ax, and three pieces of marine shell (Fusinus dupetitthouarsi). The chert working could pertain to Monte Alban Late I, II, and/or V. Although there is an onyx source 25 m to the south, we did not find any of this material on the site (Fig. 3.29). TL-SLA-SLA-23 This small (<0.5 ha) single-component Monte Alban IJIB- IV (IV-4) site is located on a middle piedmont knob on the eastern side of El Tigre, less than 100 m east of the road to Roaguia, approximately 1.5 km north of the village. The site consists of one small structure and several destroyed, unmapped terraces. Ceramics are sparse, but high amounts of white chert debris on the north side of the structure indicate chert working. TL-SLA-SLA-24 (Hierve el Agua) This large multicomponent site is located on the eastern side of the Nueve Puntas mountain ridge near the modern village of Roaguia, in a middle piedmont area that flattens out at the FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 131 ise of the mountain ridge before plunging down to the deeply incised Hormiga Colorada to the east. The area around Roaguia has some of the best agricultural resources in the region. Although very accessible from the north, the promontory where the site is located is extremely defendable from the south and has an excellent view in all directions. The site is well known for its natural springs, terraces, and fossilized canals (Flannery et al., 1967). It has long been debated whether these features were used primarily for agriculture (Flannery et al., 1967; Neely et al., 1990) or for salt production (Hewitt et al., 1987; Peterson et al., 1989). Today, part of the site is an official archaeological zone that is open to the public. The site has been heavily impacted by modern farming and tourism, and there is very little preserved architecture. We mapped several small buildings in the northern, flatter part of the site, where we found only Monte Alban IIIB-IV and V ceramics. We found ceramics from all phases near one small mound on the tip of a long, steep ridge that extends south of the archaeological zone. We mapped three terraces that appear to be residential; many others were too destroyed to map or clearly were agricultural. The site was first occupied during Monte Alban Late I (LI- 5), covering 13 ha in the southern part of the site near the springs and the archaeological zone. It was about half that size (6.5 ha) in Monte Alban II (II-4). We did not find any diagnostic Early Classic pottery. The site grew substantially in Monte Alban HIB-IV (IV-10, 86 ha), expanding onto the more gentle slopes to the north, and then reached its greatest extent in Monte Alban V (V-53), when it covered an area of 97 ha. We did not see many artifacts other than pottery, mostly chert tools and debris and half a dozen pieces of obsidian. There is an onyx quarry nearby (Figs. 2.21, 2.22, and 6.9- 6.13). TI-SLA-SLA-27 (Cerro del Rio) This multicomponent site covers an area of 3.8 ha on the top of a bedrock outcrop on a small knob at the end of a long middle piedmont ridge descending from E] Tigre. The knob is encircled on three sides by a sharp bend in the Hormiga Colorada and is known locally as Cerro del Rio. The site is clearly defendable. The highest part of the site consists of a large plaza with one small structure at either end. There is a compact grouping of 12 terraces on the southern side of the plaza, although plowing may have destroyed as many as a dozen more on both sides of the knob. The lowest terrace has a very high, defensive retaining wall. Monte Alban Late I ceramics (LI-6) are present across the entire site. Only one terrace also had Monte Alban V ceramics (V-54). The site is near a white/orange chert source, and chert debris is visible on most terraces. There is a source of ground stone on the site and many manos and metate fragments. Both materials may have been worked on the site in Monte Alban Late I (Figs. 3.30 and 3.31). TL-SLA-SLA-29 This single-component site covers 5.3 ha of agricultural fields on a broad middle piedmont slope approximately | km northeast of Hierve el Agua and 500 m east of the road. All 132 ceramics were Monte Alban V (V-14), including several G-3M sherds. Worked chert debris was present across the site. TL-SLA-SLA-34 This large (37 ha) single-component site is located on broad, flat terrain in the middle piedmont on the eastern side of Cerro Guirone, approximately 600 m east of San Lorenzo. The site consists of a large scatter of Monte Alban V ceramics (V-19). We mapped one large mound constructed on top of a natural rock outcrop near the center of the sherd scatter. The mound had been looted. There are eight modern agricultural terraces but no clear evidence of ancient ones. There is very little chert on the site, which is located in one of the better agricultural areas in the survey region. Of the three obsidian blades we observed, two are green, most likely from the Pachuca source in Hidalgo. TL-SLA-SLA-36 This single-component site covers an area of 9 ha on gently sloping terrain in the middle piedmont on the eastern side of San Lorenzo. The site is heavily impacted by modern activities, including some terracing. The site consists of a scatter of Monte Alban V ceramics (V-21) and five green obsidian blades. TL-SLA-SLA-41 (El Trigo) This multicomponent sites covers an area of 2.8 ha on the top and southern slope of a small piedmont knob on a middle piedmont ridge descending the eastern slope of Cerro Guirone, 500 m east of the road to San Lorenzo Albarradas. The site is known locally as El Trigo. The northern and eastern sides of the hill are very steep and defendable. We mapped 23 terraces, two small structures, and one plaza. The terraces surround the architecture on the west, south, and east. While pottery from Monte Alban Late I through HIB-IV is dispersed across the entire site (LI-12, H-11, IIA-12, IV-12, V-61), we observed much more Monte Alban V pottery. All the terraces appear to have been occupied in Monte Alban V, and on seven terraces we also observed HIB-IV pottery. Ceramics on some terraces are undiagnostic. It is not clear whether earlier use of the terraces is hidden by the higher density of later materials on the surface or whether earlier, possibly unterraced deposits were exposed during the con- struction of the terraces in Monte Alban V. We conservatively used very light density to estimate Monte Alban Late I-HIA populations at the site; populations could be somewhat higher if any of the terraces were occupied then. We used terrace density for Monte Alban IV and the full number of terraces for Monte Alban V. There are more than two dozen pieces of chert-working debris, including several cores, and 10 pieces of ground stone, including manos, hammerstones, abraders, and one basalt ax. Much of this activity may pertain to Monte Alban V. We also collected 14 pieces of obsidian including one gray core, two pieces of polished onyx, two pieces of marine FIELDIANA: ANTHROPOLOGY shell, and several drilled sherd disks made from G-3M body sherds (Figs. 3.25, 3.26, and 7.8). TL-SLA-SLA-43 This small (<1 ha) multicomponent site is located on the top of a long narrow ridge in the middle piedmont, approximately 1 km north of the road to the Mixe region near its intersection with the road to San Lorenzo. The defendable ridgetop is located between two barrancas; the northern one is very deep, called the Barranca Honda. We mapped one clear terrace and eight small house foundations dispersed on the ridgetop. At least one house had been looted. The site was occupied in Monte Alban IIIB-IV (IV-9) and V (V-50). The only artifacts other than sparse pottery are a few chert tools, one chert core, and one piece of unworked onyx. There are several onyx sources on lower slopes of the ridge (Fig. 6.8). TL-SLA-SLA-50 This small (<0.25 ha) single-component site is located on the top of a mountain ridge on the northern side of the saddle (between Cerro El Portazuelo and Cerro El Mogote) where the modern road to the Mixe region passes. The site’s location is very defendable and has a great view of Cerro Guirone to the south and El Trigo to the southeast. The site may have been a lookout or outpost. We mapped two small stone house foundations. Sparse ceramics are Monte Alban V (V-32). TL-SLA-SLA-51 This small site (<0.25 ha) is located on the easternmost tip of the mountain ridge on the north side of the pass (between Cerro El] Portazuelo and Cerro El Mogote), approximately 200 m southeast of TL-SLA-SLA-50. The site is very defendable, with great views in all directions—west into the Tlacolula Valley and east toward the Mixe region. The only visible remains are the cut stone foundation walls of a small building, possibly a military outpost. Leaf cover made finding sherds difficult, but the similarity of the architectural remains and nearness to TL-SLA-SLA-50 indicate Monte Alban V (V- 33, TL-SLA-SLA-52 This multicomponent site covers 2.9 ha of low terrain between the confluence of the Hormiga Colorada and a small stream that flows into the San Luis tributary of the Tehuantepec River, along the route of the Camino Real. Most of the ceramics on the site are early. We found Monte Alban Early I (EI-1) ceramics over the whole site; the extent of earlier material (Rosario phase, R-1) is smaller, covering an area of 1.9 ha. This is the earliest site we recorded in the survey region. There is a smaller area (<1.0 ha) with Monte Alban V ceramics (V-34) and also some colonial pottery around a structure that appears to also be colonial. The site is littered with chert debris, utilized flakes, and several cores; chert working could pertain to both Monte Alban Early I and V. We observed only two pieces of obsidian (Figs. 3.4 and 3.12). TL-SLA-SLA-53 (Llego Yiachi) This large (92 ha) multicomponent site is located on the top and slopes of a steep natural rock cliff on the eastern bank of the Hormiga Colorada. Its local name is Llego Yiachi. Because of the steepness of many of the slopes around Llego Yiachi and significant vegetation, we could not check all of the surrounding area, but where we could we mapped tall stone defensive walls and 55 residential terraces. The tallest, best- preserved stone defensive walls are on the eastern side of Llego Yiachi, near the easiest approach to the top of the hill. In addition to the terraces we could map, dozens of others were too poorly preserved to map. The total number of terraces at the site could easily be two or three times the number we were able to map. There are several small architectural complexes dispersed across the lower slope between Llego Yiachi and the Hormiga Colorada, but the principal architectural complex is on the very top of the rock cliff that towers above the rest of the site. This complex consists of a linear arrangement of three small patio-mound groups, a larger plaza surrounded by a low wall, and a large platform on which stood a long rectangular building with adobe walls, similar to the adobe walls in several palaces at the Mitla Fortress (Feinman & Nicholas, 2004a, figure 3.4). Like the palaces at Mitla, the building with the adobe walls may not have been constructed until near the end of the occupation. A small tomb in one of the small plazas had been recently looted. There is a small I-shaped ballcourt (approximately 20 X 20 m) at the base of the western side of the rock cliff, where we observed only Monte Alban V pottery. The site is not just a military outpost. The site was occupied from Monte Alban Late I through V. Monte Alban Late I (LI-10) and II (II-8) ceramics are present only on the top of Llego Yiachi. The site was much larger in the Classic period (IIIA-8, IV-11) covering approximately 37 ha on the top and lower eastern slopes. At the western edge of the ridge system, right above the Hormiga Colorada, is another small area of Monte Alban IIIA pottery associated with several small structures (IIIA-9). Monte Alban V (V-58) ceramics are the most widespread, extending over an area of 92 ha, including all of the mapped terraces and structures. Artifacts are sparse but include several mano and metate fragments, a dozen chert tools and debris, six obsidian blades and one core, and five crystals. Most of the obsidian is green (Pachuca). There are several G-3M kilnwasters on the eastern slope of the hill, so there is possible ceramic production in Monte Alban V (Figs. 2.26, 2.27, 5.6—-5.8, and 7.5—7.7). TL-SLA-SLA-56 This small (<0.25 ha) site is located in a low rockshelter on the northern bank of the Hormiga Colorada, below Llego Yiachi. The roof of the cave is heavily sooted. Recent activities and looting exposed pottery from Monte Alban II (II-1), HIB- IV (IV-7), and V (V-37). There is chert debris in front of the cave but very little inside. We found one cut piece of conch shell inside the rockshelter. The cave may have been a sacred FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 133 location for a longer duration than we could document from the exposed surface artifacts. TL-SLA-SLA-59 (Buenavista) This single-component site is located in eastern Tlacolula and covers an area of 2.9 ha on the top of a steep-sided peak (called Buenavista) on the narrow ridgeline between El Tigre Appendix II: Grid Square Maps We extended the system of 4 X 4-km grid squares that was established for the Valley of Oaxaca east into the Albarradas area. This appendix consists of maps for each grid square that show the location and boundaries of each phase component for every site that was identified during the E47 and Nueve Puntas. The site is very defendable, with great views in all directions. There is a small platform and structure on the very top and another small platform and structure approximately 100 m down the southern slope (both areas had looter’s pits). We mapped 21 small, dispersed terraces on the northern and southern slopes of the peak. Stone house foundations are visible on many of the terraces. Other possible terraces were too destroyed to map. The only pottery is Monte Alban V (V-52) (Figs. 7.15 and 7.16). settlement pattern survey of the Guirun/Albarradas region. Each grid square measures 4 X 4 km and is numbered in the lower left corner. All sites are labeled with the field site number and all phase component numbers assigned to each site during analysis. TL-MAT-MAT-95 FIELDIANA: ANTHROPOLOGY N6 E18 Ve8 V-42 TL-MITAA-36 TL-MIT-XA-35 Valley of Oaxaca Survey Area TL-MAT-MAT-95 E17 FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 135 Valley of Oaxaca Survey Area TL-SLA-SLA-60 ° V-39 TL-MAT-MAT-95 TL-SLA-SLA-24 LI TL-SLA-RO-1 TL-SLA-SLA-59 V-52 Oo IIB-IV-4¢ LI-8™ TL-SLA-SLA-22 TL-SLA-SLA-23 II-6 V-56 TL-SLA-SLA-29 V-14 TL-SLA-SLA-26 IA V-12 Vv Qrsiasinas Le O IIIB-IV-1 Wa V-14 \ II-4 IIB-IV O TL-SLA-RO-2 IB-IV-10 ay } V-53 TL-SUA-RO-3 E19 IIA-4 V-1 TL-SLA-SLA-27 LI-6 V-54 FIELDIANA: ANTHROPOLOGY TL-SLA-SLA-55 V-36 O TL-LA-SLA-54 OIIA-4 V-35, TL-SLA-SLA-57 O undat ed I-14 V-13Q B-iv-79 TL-SLA-SLA-56 TL-SLA-SLA-28 \-37 V-38 9° TL-SLA-SLA-58 TL-SLA-SLA-53 LI-10 Il II-8 INA IA-8 IIB-IV IIB-IV-11 oy V-58 E20 Valley of Oaxaca Survey Area Edy, FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO Valley of Oaxaca Survey Area E18 TL-SLA-SLA-17 V-62 TL-SLA-SLA-7 ° V-44 » IN|A-481 (1980) OTL-SLA-SLA-61 IIIB-IV-8 V-40 Es N-14 TL-SLA-SLA-17 IIB-IV-13 V-62 V-9 L-SLA-SLA-19¥ IA-2 iIIB-IV-3 Guirtn Survey Area V-8 °o TL-SLA-SLA V- TL-SLA-SLA-10 9 : v2 LI-ll NIB-Iv-2 ‘I-98 TL-SLA-SLA-9 -18 TL-SLA-SLA-30 — IIIB-IV-5 V-15 Vv i TL-SLA-SLA-8 El-2 IA-10 V-21 IIB-IV- TL-SLA-SLA-36 TL-SLA V-18 Vv -SLA-33 TL-SLA-SLA-34 TL-SLA-SLA-21 oO V-10 TL-SLA-SLA-22 FIELDIANA: ANTHROPOLOGY TL-SLA-SLA-32 TL-SLA-SLA-31 TL-SLA-SLA-34 V-16 TL-SLA-SLA-35 V-20 9 TL-SLA-SLA-20 LI-7 i1-5 INA-5 V-55 E20 Guirtn Survey Area Valley of Oaxaca Survey Area Bly, FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 140 Valley of Oaxaca Survey Area E18 TL-SLA-SLA-50 x g V-32 V-33 TL-MIT-CC-3 a, V-47 TL-MIT-CC-1 °o V-45 TL-MIT-CC-2 = V-46 ao TL-SLA-SLA-46 9° V-29 TL-SLA-SLA-45 o V-28 TL-SLA-SLA-47 I-12 i. La Cuadrada V TL-SLA-SLA-17 IH43© jya_43 W-13 eee 82 ° TL-SLA-SLA-51 TL-SLA-SLA-11 { V-3 TL-SLA-SLA-12 S V-51 V-314 ° TL-SLA-SLA-49 TL-SLA-SLA-48 °o V-30 TL-SLA-SLA-17 TL-SLA-SLA-42 ° V-26 TL-SLA-SLA-13 a TL-SLA-SLA-15 v6 9 TL-SLA-SLA-14 2° V-5 TL-SLA-SLA-16 o V-7 FIELDIANA: ANTHROPOLOGY | TL-SLA-SLA-43 oS iIB-IV-9 v-50 ° on er TL-SLA-SLA-44 vey TL-SLA-SLA-40 y aE Soe TL-SLA-SLA-39 V-25 TL-SLA-SLA-38 V-23 v-22 Q)LSLA-SLA-37 LI-12 W-44 INA-12 IIB-IV-12 V-61 Q|_LTL:SLA-SLA-41 E20 LI-9 Q Il-7 INA-7 TL-MIT-CC-8 Valley of Oaxaca Survey Area E47. FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 14] TL-MIT-CC-5 TL-MIT-CC-4 ° TL-MIT-CC-70 1 WA-11 a) V-48 v-49 eS {> | TL-MIT-CC-6 Ce’) vee ||-2 I-10 N10 E18 Guirtn Survey Area E19 142 FIELDIANA: ANTHROPOLOGY Appendix III: Site Survey Data The coding system devised for the Valley of Oaxaca regional settlement surveys was used in organizing the Guirtn/ Albarradas site data to facilitate comparisons between the two regions. We have made modifications to some categories and have changed data presentation to make the files easier to read and use. Data for all phase components were entered into Excel files and are presented below in four parts. The first part (a) includes site size (in ha), population, and occupational continuity; the second part (b) contains artifact information; the third part (c) presents architectural and other cultural features; and the fourth part (d) presents environmental and modern land use information. Each phase component is on a separate line. General descriptions of all field-numbered sites with at least one mound, terrace, or other significant feature are included in Appendix I. Population, site size, and architectural features have been updated for all components of the Guirtn site, but artifact categories include only those collected during the regional survey (for materials collected during the intensive site survey, see Feinman & Nicholas, 2004a, pp. 81-86). Coding Scheme Each file begins with five columns that are used to identify each site. The first column (area) is the area prefix. The area prefix for the Guirun/Albarradas survey region is 7 (0-6 were used to identify survey regions in Oaxaca and Ejutla). The second column (phase) is the phase number. We used the Oaxaca numbering system and omitted the numbers for phases that we did not identify in the Guirtn/Albarradas survey region (4 = Rosario, 5 = Monte Alban Early I, 6 = Monte Alban Late I, 7= Monte Alban IJ, 8= Monte Alban IIIA, 10= Monte Alban IV, and 11 = Monte Alban V). The third column (no.) is the site number. This numbering system restarts at one for each phase. When referred to in the text, the site number is preceded by the phase abbreviation for that phase. The fourth column (subarea) designates the subarea of the valley where the site is located (AL for Albarradas and ET for eastern Tlacolula). The fifth column (grid) is the grid square in which the site is located. Each grid square number consists of a north component (6-10) and an east component (17-20). When site boundaries extend beyond the limits of a grid square, the site is assigned to the square containing the largest proportion of the site. The grid numbering system is an extension of the grid employed in the Valley of Oaxaca regional survey. Column Variable Values Site area Site area in ha 000.0 ha Density Density of population 0 = 5-10, 1 = 10-25, 2 = 25—S0, 3 = other, ? = missing data Lopop Lower population estimate = Hipop Upper population estimate — Avpop Average population estimate — Ros Rosario 0 = absent, | = present EI Monte Alban Early I 0 = absent, 1 = present LI Monte Alban Late I 0 = absent, | = present II Monte Alban II 0 = absent, | = present IA Monte Alban IIIA 0 = absent, | = present IV Monte Alban IV 0 = absent, | = present Vv Monte Alban V 0 = absent, | = present B. Artifacts. Column Variable Values OCI Obsidian color 0 = none, | = gray, 2 = green, 3 = black, 4 = green and gray, 5 = green and black, 6 = gray and black, 7 = gray, green, and black, 8 = other OP Number of obsidian points - OB Number of obsidian blades — OF Number of obsidian flakes _ OCr Number of obsidian cores — CP Number of chert points — CB Number of chert blades — CF Number of chert flakes — FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO A. Population data. B. Continued. Column Values 0 = absent, 1 = present 0 = absent, | = present 0 = absent, | = present 0 = absent, 1 = present 0 = absent, 1 = present = absent, 1 = present Variable Values CG Number of chert cores _- Ci Total number of chert tools -- CO Other chert 0 = absent, 1 = present QF Quartz flakes/other 0 = absent, | = present ChO Other chipped stone 0 = absent, | = present Mn Number of manos — Mt Number of metates = Hm Number of hammerstones aa Ax Number of axes — GO Other ground stone 0 = absent, 1 = present BS Number of burnishing stones — Sh Number of shell = GS Number of greenstone — C. Architecture. Column Variable A Adobes BS Building stones PF Plaster floors B Evidence of burning CS Carved stones B Visible burials T Number of visible tombs SCst <1 m Number of single component structures < 1 m high SCst >1 m Number of single component structures > or = to 1 m high SCst. vol Total volume of single component mounds (m°*) Tst <1 m Total number of structures < 1 m high Tst >1 m Total number structures > or = to | m high Tst. vol Total volume of mounds (m*) Ter Number of residential terraces W Number of defensive walls D. Environmental data. Column Variable Values Elev Elevation above sea level 0000 m Zone Environmental zone 1 = low alluvium, 2 = high alluvium, 3 = lower piedmont, 4 = middle piedmont, 5 = high piedmont, 6 = mountain Slp Slope 1 = flat, 2 = shallow, 3 = moderate, 4 = steep HT Hilltop or ridgetop 0 = no, | = yes ST Soil texture 1 = mostly sand, 2 = sandy loam, 3 = loam, 4 = clayey loam, 5 = dark clay, 6 = sandy clay SD Soil depth 0 = 0-20 cm, 1 = 20 cm-1 m,2=>1m SCl Soil color 0 = white, 1 = light brown, 2 = reddish, 3 = dark brown, 4 = dark gray/black Er Soil erosion 0 = none, | = light, 2 = moderate, 3 = heavy Alv Alluviation 0 = insignificant, 1 = significant H20 Distance to nearest water 0.0 km WT Type of water source 0 = main river, | = tributary stream, 2 = arroyo, 3 = spring, 4 = well Irr Type of irrigation 0 = none, 1 = well, 2 = well with pump, 3 = valley floor canal, 4 = piedmont canal (now or in recent past), 5 = floodwater, 6 = portable pump from stream, 7 = terracing only Pl Recency of plowing 0 = never, | = this year, 2 = sometime in past Pot Pot hunting 0 = none, 1 = some, 2 = heavy Sci Scientific excavation 0 = none, | = present Veg Predominant vegetation Lnd Modern land use Crop Predominant crops Spe Special resources AG Ancient agricultural terraces AQ Ancient quarries 144 0 = cleared, 1 = chamizo/willow (riverine), 2 = mesquite, 3 = cactus, 4 = grass, 5 = oak, 6 = pine, 7 = other 0 = collecting/hunting, 1 = town, 2 = farming, 3 = industrial, 4 = grazing, 5 = garbage, 6 = other none, | = milpa, 2 = alfalfa, 3 = castor beans, 4 = tobacco, 5 = truck gardening, 6 = sugarcane, 0 = 7 = carrizal, 8 = maguey (= 7 = other ores, 8 = onyx 0 = absent, 1 = present 0 = absent, | = present none, | = quarryable stone, 2 = clay for pottery, 3 = salt, 4 = chert, 5 = mica, 6 = magnetite-hematite, FIELDIANA: ANTHROPOLOGY APPENDIX III.A. Population data. eS Die Air CASS IV < Area Phase No. Subarea Grid Site area Density Lopop Hipop Avpop Ros 7 4 l AL 720 1.89 1 19 47 33 l ] 0 0 0 0 | 7 5 ] AL 720 2.83 1 28 71 50 ] ] 0 0 0 0 ] 7 5 2 AL 819 0.49 1 5 12 9 0 | l ] ] 0 ] 7 6 l AL 719 0.13 1 5 10 8 0 0 ] 0 | 0 l 7 6 2 AL 920 0.04 0 5 10 8 0 0 ] 0 0 0 1 a 6 3 AL 918 1.56 0 8 16 12 0 0 ] ] 0 0 0 7 6 4 AL 1018 0.25 0 5 10 8 0 0 ] ] 0 0 0 ul 6 5 AL 719 13.16 l 132 329 230 0 0 ] ] 0 1 ] yf 6 6 AL 719 3.78 3 120 240 180 0 0 ] 0 0 0 ] i) 6 7 AL 820 4.17 3 235 470 B53 0 0 | l ] 0 l au, 6 8 AL 819 B25 5 126 Sy) 189 0 0 ] l ] 0 | io 6 9 AL 1017 1.85 3 150 300 22s) 0 0 1 l 0 0 7 6 10 AL 720 2.74 3 29 Sy 44 0 0 ] ] | ] ] 7 6 11 AL 819 7.78 3 295 590 443 0 1 1 ] ] 0 ] 7 6 12 AL 920 Zoi 0 23 58 40 0 0 1 ] | ] ] a 6 13 ET 818 0.17 3 10 20 15 0 0 ] ] ] ] l d. i l AL 720 0.04 ] 5 10 8 0 0 0 ] 0 ] ] ft 7 2 AL 1018 0.50 0 > 10 8 0 0 0 ] 0 0 ] 7 7 3 AL 1018 0.25 0 » 10 8 0 0 ] ] 0 0 0 7 7 4 AL 719 6.48 0 32 65 49 0 0 ] ] 0 ] l 7 7 5 AL 820 4.17 3 235 470 353 0 0 1 ] ] 0 l 7 7 6 AL 819 Biz 3 126 DD 189 0 0 ] ] 1 0 1 7 7 7 AL 1017 1.85 3 150 300 225 0 0 ] ] ] 0 0 i 7 8 AL 720 3.90 3 65 129 2 0 0 ] ] ] ] l 7 7 9 AL 819 7.78 3 Dos 590 443 0 ] ] 1 ] 0 l 7 7 10 AL 1018 9.96 5 389 801 595 0 0 0 ] ] 0 ] 7 i 11 AL 920 2.31 0 UE 58 40 0 0 ] ] ] ] ] 7 i 12 AL 918 0.75 1 8 19 13 0 0 1 1 0 0 0 7 7 1S ET 918 0.07 3 5 10 8 0 0 0 ] | ] ] 7 a 14 ET 818 0.04 3 5 10 8 0 0 ] ] ] ] ] 7 dl 15 ET 818 0.04 3 > 10 8 0 0 0 ] ] ] 1 7 v 16 ET 818 0.04 3 ) 10 8 0 0 0 ] | ] | Hf 7 17 ET 818 0.04 3 5 10 8 0 0 0 ] ] ] 1 7 i, 18 ET 818 0.13 3 10 20 15 0 0 0 ] | ] | 7 8 ] AL 719 0.13 0 5 10 8 0 0 0 0 ] 0 ] if 8 2 ET 818 4.85 3 85 170 128 0 0 0 0 ] l ] 7 8 3 AL 719 1.00 ] 10 25 18 0 0 1 0 ] 0 ] I 8 4 AL 720 0.32 0 5 10 8 0 0 0 0 ] 0 ] U 8 5 AL 820 0.04 3 5 10 8 0 0 | | 0 l 7 8 6 AL 819 0.06 3 15 30 28 0 0 ] l | 0 l 7 8 | AL 1017 1.85 1 19 46 32 0 0 ] 1 ] 0 0 7 8 8 AL 720 35.83 3 528 1,236 882 0 0 ] 1 ] ] l 7 8 9 AL 720 5.26 3 61 123 2 0 0 ] | ] ] ] 7 8 10 AL 819 0.49 3 5 10 8 0 ] | ] ] 0 | 7 8 1] AL 1018 0.11 3 5 10 8 0 0 0 1 ] 0 | 7 8 12 AL 920 2.31 ] 23 58 40 0 0 ] 1 ] ] ] 7 8 13 ET 818 47.25 3 1,650 3,300 2,475 0 0 ] ] ] 1 1 7 10 1 AL 719 0.06 0 5 10 8 0 0 0 0 0 ] 0 a 10 Z AL 819 1.5] l 15 38 26 0 0 0 0 0 ] 0 a 10 3 ET 818 4.85 3 85 170 128 0 0 0 0 ] ] ] di 10 4 AL 719 0.44 3 11 DD 17 0 0 0 0 0 ] 0 i 10 5 AL 819 0.88 0 > 10 8 0 0 0 0 0 ] ] 7 10 6 AL 819 0.75 0 5) 10 8 0 0 0 0 0 ] 1 i 10 Zi AL 720 0.04 ] 5 10 8 0 0 0 ] 0 1 | i 10 8 ET 819 0.13 0 > 10 8 0 0 0 0 0 ] l 7 10 9 AL 920 0.61 3 45 90 68 0 0 0 0 0 ] l a 10 10 AL mL 64.69 3 TSS) 1,833 1,294 0 0 ] ] 0 ] l 7 10 1] AL 720 35.83 3 528 1,236 882 0 0 ] l | ] | 7 10 12 AL 920 DES ] 53 116 87 0 0 ] l l ] ] u, 10 13 ET 818 79.50 3 1,945 3,890 2,918 0 0 ] | l ] ] yf 11 1 AL 719 0.13 1 5 10 8 0 0 0 0 l 0 ] 7 11 2 AL 819 1.00 ] 10 25 18 0 0 0 0 0 0 | a 1] 3 AL 919 0.82 3 20 40 30 0 0 0 0 0 0 ] 7 11 4 AL 919 0.57 3 15 30 23 0 0 0 0 0 0 | 7 11 5 AL 919 0.25 ] 5 10 8 0 0 0 0 0 0 ] 7 1] 6 AL 919 0.69 1 7 17 12 0 0 0 0 0 0 l 7 11 7 AL 919 0.10 0 5 10 8 0 0 0 0 0 0 l 7 1] 8 ET 818 0.06 0 5 10 8 0 0 0 0 0 0 l i 11 9 ET 818 0.13 0 5 10 8 0 0 0 0 ] ] l 7 1] 10 AL 819 0.03 ] 5 10 8 0 0 0 0 0 0 l 7 1] 1] AL 719 1.00 ] 10 25 18 0 0 | 0 ] 0 l FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 145 APPENDIX III.A. Continued. e & Il WA IV << Area Phase No. Subarea Grid Site area Density Lopop Hipop Avpop Ros 7 1] 12 AL 719 0.31 0 5 10 8 0 0 0 0 0 0 l 7 1] 1S AL 720 0.13 0 5 10 8 0 0 0 0 0 0 1 7 1] 14 AL 719 Sy ] 53 132 93 0 0 0 0 0 0 1 7 1] 15 AL 819 3.64 0 18 36 27 0 0 0 0 0 ] 1 7 1] 16 AL 820 1.82 ] 18 46 82 0 0 0 0 0 0 ] Gi 1] 17 AL 820 2.20 0 1] 22 17 0 0 0 0 0 0 1 a 1] 18 AL 819 2.01 0 10 20 iS 0 0 0 0 0 ] ] fl 1] 19 AL 820 36.97 0 185 370 Dil 0 0 0 0 0 0 ] i 1] 20 AL 820 0.13 0 5 10 8 0 0 0 0 0 0 ] 7 1] 2] AL 819 9.07 ] 91 227 159 0 0 0 0 0 0 1 7 1] 2 AL 920 0.44 0 15 30 23 0 0 0 0 0 0 ] ( 1] 23 AL 920 7.50 ] 75 188 131 0 0 0 0 0 0 ] 7 1] 24 AL 920 0.13 0 5 10 8 0 0 0 0 0 0 ] il 1] 25 AL 920 1.89 0 9 19 14 0 0 1 0 0 0 l 7 1] 26 AL 919 0.13 0 5 10 8 0 0 0 0 0 0 1 7 1] i AL 920 0.25 ] 10 20 15 0 0 0 0 0 0 1 7 1] 28 AL 918 0.04 0 5 10 8 0 0 0 0 0 0 1 7 1] 29 AL 918 0.04 0 5 10 8 0 0 0 0 0 0 ] 7 1] 30 AL 919 0.06 0 5 10 8 0 0 0 0 0 0 ] 7 1] 31 AL DG 0.03 0 5 10 8 0 0 0 0 0 0 ] 7 1] 32 AL 919 0.10 5 10 20 15 0 0 0 0 0 0 ] 7 1] 33 AL 919 0.04 3 5 10 8 0 0 0 0 0 0 ] 7 1] 34 AL 720 0.95 1 10 24 17 1 1 0 0 0 0 1 7 1] 35 AL 720 0.32 ] > 10 8 0 0 0 0 ] 0 l 7 1] 36 AL 720 0.02 0 5 10 8 0 0 0 0 0 0 ] 7 1] 37 AL 720 0.04 ] 5) 10 8 0 0 0 ] 0 ] ] 7 1] 38 AL 720 0.04 0 2») 10 8 0 0 0 0 0 0 ] if 1] 39 ET 718 0.04 0 5 10 8 0 0 0 0 0 0 ] 7 1] 40 ET 819 0.13 0 5 10 8 0 0 0 0 0 ] ] 7 1] 4] ET 818 0.25 0 5 10 8 0 0 0 0 0 0 ] i, 1] 42 ET ie 0.28 0 5 10 8 0 0 0 0 0 0 ] 7 1] 43 ET 7 0.01 ] 5 10 8 0 0 0 0 0 0 l 7 1] 44 ET 819 0.40 0 5 10 8 0 0 0 0 0 0 ] 7 11 45 AL 918 0.25 1 5 10 8 0 0 0 0 0 0 ] 7 1] 46 AL 918 1.07 0 10 20 15 0 0 0 0 0 0 ] 7 1] 47 AL 918 1.13 0 6 1] 8 0 0 0 0 0 0 1 7 1] 48 AL 1018 0.09 0 5 10 8 0 0 0 0 0 0 ] i 1] 49 AL 1018 32/5 ] 38 94 66 0 0 0 ] 0 0 ] 7 11 50 AL 920 0.61 3 45 90 68 0 0 0 0 0 1 ] 7 1] 51 AL 919 0.58 3 5) 110 83 0 0 0 0 0 0 1 7 1] 52 ET 718 2.89 0 135 270 203 0 0 0 0 0 0 ] q 1] 53 AL 719 72.9] 3 851] 2,066 1,458 0 0 1 ] 0 ] ] q 1] 54 AL 719 0.03 3 5 10 8 0 0 ] 0 0 0 ] 7 1] 5) AL 820 0.04 3 5 10 8 0 0 1 1 1 0 ] 7 1] 56 AL 819 Daag 3 106 212 159 0 0 1 ] 1 0 ] 7 1] 57 ET 717 9.38 3 575 1,150 863 0 0 0 0 0 0 ] 7 1] 58 AL 720 83.59 3 1,321 3,060 2,191 0 0 ] ] ] ] 1 7 1] 59 AL 819 0.06 0 5 10 8 0 l ] 1 1 0 1 7 1] 60 AL 1018 1.07 3 40 80 60 0 0 0 ] ] 0 ] 7 1] 61 AL 920 Hdl 3 125 250 188 0 0 | 1 ] ] 1 7 1] 62 ET 818 119.13 3 1,857 3,800 2,828 0 0 ] ] l ] ] 7 12 ] AL 720 a an = an a ? Q ? e 2 ? ? 146 FIELDIANA: ANTHROPOLOGY APPENDIX III.B. Artifacts. Area Phase No. Subarea Grid OC! OP OB OF OCr CP CB CF CC CT CO QF ChO Mn Mt Hm Ax GO BS Sh GS SRS GS Get i eet het GC COCO CG art et CE CS COCO COE CC SS CCS Festi SS St OOS OOO Ori OS Ores Oren CCG oe eGo SS CQ Cort GCS rt COCO CCG COE et GS OC COCO SSS) Sr Sra en OS) Se ONS OS SS SOS SSeS GS So SOS SS SSS eae Seo 268286) 09 2892) Shoe Ne SS SS 2S NY SN SS SS SO OO OS SOS) SN SS SS SOS) OS SS SS SSS SS SSS NS OO SS SS SS SS S SS 22] SoS SSS SS SI SOS SS) SS SS OS SSIS) SOO OSS SS) OS SISOS SS SS) SS OOO SS SSS) SOS SS SS SS SS SS SS SSS SS SS SS S) men arn tte Gal ret NYS et CC CO SO OS OS SOS GUIS Seat Set es OO SOE ONO CO a eS Eee a CNG tle GONG ttt GEG IC CO CO CI rs SS te COC Ce CS CO SC GS CS GS Ge Ge) Se I ret COCO NS CS SCO OSG (CO OCS OS SOS OS SOS SO SOS OS SO nS SO OE SCO SO COG SSeS SS © SS) SF SoS S29 SF OS OS OO sy OS 2 OS OE SS Se OS SO OS S92 SSS Sa) dO SQN OSS OS OS] SS OSS SS) EQ OO OOo SSS — OS) CRC GG EC CCRC Cr COC CG CC et OCC COE COC CCS SS SOS SSE COS ee OOS et ORS Fr OC 8 CEG EGG) 2222S SS] SI (SI!"_ SI OS OS 2 SS 3] BO SS SO) SO OS S| OO SS SS SS SS SO SS SS SS OS SI SS SO SS SS SS SO SO OS SS OS SS SS OS] SS OS] SS SS SS OS SO SS SS SS OS SS SS © Sees OOnOn eae eae Owens n Onn ee aes coqoocococoooeonsae st One ae aa seas = COQonmoeon nt Oeeaeseoocooonancocooe {4 GIGI GI ECO CON CONC Oe CO a GOs Sta eC COCO CC ea Sy Ca OOO Sir Srna ao GO OOS OO Cor ncn Ca) Coe GOOG ONE oe COC Conc = = = = = = _ 2X oS 2S] SS SS SN 9 NY OS OS eh) SO OS SO Og QQ SE SOS SSO eS Soe eS SON Se ee OSS SOS SS Se Va SS OS SS SS SS SS SS SS Nee en eee eee aceon ere ee ee eee eee eee ea eee eee NN ’ eo d em = v4 —_ ee CO CO OE OO SS SGN SSE OES SSS SO SS OES SS SS et OOS SCOR EC SOI SE Se CO CC CC aC SOR GE Se OS SOS SSNS SS SS SOS SSS SS SSS SSS SS SSS SS SS SS SS SSS SSS SSS SS SSS oS eG Gn CC COE OG OR OO CO RS OG ESS OOOO SS OS On OES st SOOO GNI SS OS SCOR ONO GN COCO SC GG CECE RSCG CS GS SR SCS EO GC COC COO NO GNC Si COR CO EC O ES S O SS eS SS COOnen SS tS OSS SIO ONS carn (Ste Sr SSS SoS == 1O OOD STOAAONTAOOC BOON AONTOMAOOOCOOCO OOOO as OOO OCU OMNMNOMNOAtM AB OOF OOo oOMOoOMONnNneoeooosaNnNoCoooCceo Se CSCS SS SO OO OES SS OS SS SS SOS OOS 1S SOO OS SS Or SiS OS SOS Sa SS StS SS StS Sh SOS Si seo coce et SSS at OS Ss OO oo ST or ST TSE KM COCO OC CoO SOM OCCA CoS oT SS MOS ACOSO CO Ott Ee oS) oS Sat St CS SoS So SOND OSAAADHAOGAMr-OAHACAAHSAAHAGAHAm- ODN WMG WOWW WOW ODROA COA DAA ADA ANARAWDAAHADHRGDAHAGAGTATAAAADAAARAOOAD ANA Ses N ems aN HSH eA BN Se NMP ee N Bee NAN Se eA Se Se Se Se Ss Ss Sse Ss aN NA SB eA KN See A Bae Se Se Se Se Beas N BN BHAA Se ee ae ae ae ae Sa Be Be eS m~mr-OMmr-NDHDOKRrr-AAMHOM-ODAAOrOOChMH WOOO Mm™-AOAAAD CO CO OO OM OMmMMm- OO OM ODD OM CO COM CO OM ODM HA OhM ODA A AAD C CO OC = = = = = = = Ce ee ee On ee en ee DOOR OOP OOOO OOOO RO OR ORE RRR REE RRR EER RE RRR REO RE RRR ERE RRR RRR RRR RR EERE REE EE 147 FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO APPENDIX III.B. Continued. Area Phase No. Subarea Grid OClL OP OB OF OCr CP CB CF CC CT CO QF ChO Mn Mt Hm Ax GO BS Sh GS et a a rent eet CC) 8 C1: SS) 1 1) OS OS SO OO OS St OS OS OOO OS SO SOS OOS St OS SS ens So SS Oo OS OOO Oo So Ooo a SS SS SO SOOO OO OOCO COCO CO oO OSC SOO SOO mn oOo Oa NS CS CS CO tS OOOO SOS SOS OS OS OSS OSS 1S SS SSS SS SS SSA SS ooo coo ooo Qoocnoocnoooooo ooo oocoqoqoqoqoqoocaaaqqocncoonsoaocda=Cc Co =o re Co SOS OS SOS oS Ooo oo Oooo oo oO oo ooo 2 oc coo oo Soo So oo 0 Qo Oa Oo So So aS 4 oo ooo ooo cna cocoon oaoooeoeoooooroocnoocooooooonocfaconoqoood=a foo AS See eC CoC Co ore OOo SS SS SSS SS S'S SIS) SS SSS Oro SS SS SS SSS SS SS = SS Sn ONC CO COC CS Co On SC So So Onl OS OOS SOO eS) OS OLS SSS SO SS SS SS SS OS St Ort CaS SC BO COC a CO CO CO CS CO OS ys SO SSS SOS ES OCS OS Os (SS So oo So oo coo aoc aaa qonaooooaoaaaaoqocooeqcoaqococcoocoocooaooco=s ooo o fo So oo I a a et da a te med lS me red ge te rm cet I rd eel ad eh a ered So I a Se Cr eC a) (SGN OC SSS SC SS SC SIG) 1 (CSC NO Ga COS Se a or COE SS OO SS OOS SS SES OSS SS SOO St OS SEO Ca St Sec ee OE ORO StS SS SO SSS StS ISS OS OS SSS SSS SS YG) (SS) NOS SS RS OOS SOE OOO SOS OSS SSO EOS OE SSO Orr SSS OOS SS SS SOS OS Oh SoS SR Ge CO CG GG COS CO SES Ot OOhrs SO St Se EC OC OE CO SCO TC OE SCO ORS CS OS SESS OS OO SO COS SS OS SOO SO rte (aS SC COC COO Sst CO COT CO SO SoS OOO Orr Or Ore tOiO1S GSS ON OS) eS oS Sa GS Geo OSC CO CO St eC CS CON i IS GNIS IS SS Sa SGU GS Ses 1S 1S 00 1S NOs Oe) Ga = RGR Ge Ge OC IC ECON OO OOO OI SS SES OS OOS ES SRO OOS OOS (SSS tS ee SCOR Cet COGN CCC IC OE rt OSC COR OOO) GSO OO Oath OO Osh eS ONO (ONS SON SS On SS SS On SS! 60) 00 ON ON ICN CN OS OS S SCO ON) CO T= TG CO) 60) GO CO GO: CO S CN 00) GN CN © Oh Se (CO SO COS SAM MANANAMAATM@AAQANAA TMA Geass = GT GG eres er ree ree teen ee eee le Cs Ge et GN treet Gal ™m™ ~ ~~ C0 CO CO WW WMHODAATAADADAATADAADAAAY & Mm-m~-M- COMmm-ONnADAACCA A & ™ lm CO OM OOD CO fr See ae AAA AAA ABABA ABABA ABH BB HP BBB RBBB AAA AAA BHAA AHAB HAHAH eH <<< moeeteeetetettortetertetortetetstoc ANTFMNOMODOWANANMNTNOM WHA GANMNATHOMODAGANMTHNOMODeANMNTHNHORPMODK GAA = SSeS See CAAA GAA GOA AAS ion oF on cn en on em en em SF St SP SP oS SP OSS OS OST HWY IH WH WH WH WH OH WO’ WH WO ee See ee eee eee eA a mmm Bm mm mm mm mm meme eee MmOO ORO RO ROP PO OOOO OOOO ORO ERO RR ER RRR RE RRR RR ER RR RR RO Ee ee FIELDIANA: ANTHROPOLOGY 148 APPENDIX III.C. Architecture. SCst. Tst Tst Tst. >1m SCst SCst Ter vol >I1m vol Sora SSS SSS SOS SS Oo Oo Ooo oC os OOO ooo oOo Oo ooo So SO So So SOS Soo So So So So So = a or ee Ce Cs COC CC CO CO it CO OE CS CO COS COS COS SS SE SCS OSG Oe SS Sr SS SS SS et ree ire eee ei cemel SS ct ree ret te eed RS SS ret repre eet et eet) rd) ie SR SES SOND SAOMARHRDCAKrSOHACWSOWDOAHACAHAr-OCAHAWSAOA AD AOD OAD WH ODAONHDANGCOAr- CDA AODAAADAAA SO AAS SHAS See NSS N SN SHAN Se See N Se HN Se eA eee Se a BeBe See eS ANH eNA SH HA See eae ee QI m™m-OoOrnnodnrnr-WAOWOOMr-AODAOr-OOCr- AD OMm-MOOAAACWOMHOMOCOhMm OMmrOMOOM- me OOD CO CO COM CO CO = = =o = = —_ = TFHUNUOWOUOUOUOUOUOOOOOOOM- rr e------~—-—--~-K~--K- KOM MDAMMMAMAMDMDMAAMDADOODOCCC oO TS eS eS eS et est oe POOR OOOO ROO ROO ROE ERR RRR ERR RRR RRR REE RR RE RE ERE ER RE ER RE EE OS ON ’ Sea Cet VA SC ie CoC Ge Gol GIG nC on A om Se Yaga) aS) SSS) SS SSS OMN amon ‘Oo ot es N DEM S'S SS Sooo aa S = SEMAN) SS OS] OS QS a] QS oo ooo ooo oo So So oo oS I OE Oe SSO ROIS OOS ROOT OG CG Go Ge) SSS eS GN COS COS COS OR SE ES IS SOS SS SOS OS SS SS SOS SS SS sO OOS RCO SE OES EC SEO OS et CO OCS Oo CoC Cone SS rest east ead SOO SSS SOS SS) SSS SSS) SS QS SS Os BHAA HHH HHH HY met1m vol >1m Phase No. Subarea_ Grid Area ae a OS SS 2 SS ES OS OS EO OSI OOAANS SS TS — See NOS OCD Om nm om Se OSS SS a Oo Oo ooooaqcococoooCccCoOoCc CC COCO OCC COC CSO ONO NCO N oS CoS = ee SN = =o 79 0 on om a ooo 397 1,388 SS OSES SE CS SOS SSS SOS OSS SOS SESS SI SS SS ISI SIS SSS Orn eee eee ene nee ON Cea (GO ees CS OS oO co oo acooocnrocaocoaqqcqooqoqoqooqoocoCcmcCcmCcCccCcCccCcCC CSc CC coco Se SoS OSs Co AS See eC NC OS (ron OOS Oooo Co CoC So SCooocaocoeooocooocsocCcCcCo So oS So SS OO Re So So So Sc Oe So a eS So Oo oo aco ocoaocaeocooococoocoococCcmCcmOClcCCSO oO oCoocCoCcoCeo SoC CcOoCSoC CCC Cos oa oS oe et er ea GC GS (CS OO SO SO Oo OCC CO COCoCoCooCcCOCO CoCo SC Oo Ss oo So SC Soo SSC OS SSO OOS OO OSS SSO OSS OSES ISIS OSS OSS SO SISOS SES OSS SO Sra sO I tI SS) re re i rer et) ra ret ed ret Se eo ene a QO I 1 OO ea Ss OS So OO CO OC COC oS OC OOo Co oo Ss So oo oo Clo oS DANDOADSGOASGCSOASSSSASDADAAAAARSSSSSOMDMAAKrrADDDADAMDONRDANARAODArOHRAMSWS SSNS SNA SANSHAANNAS ASS Sees AANNAAS See eee ee sS Aaeseseaeaseeasesana mMHmOHMOMDAWDMDMODARDARDARDADADADRLCLL ELE WDE RWDWAAHRCSCADRLC LOOK MOSOHOE Seed HAH AAA Be Ae A eA A AA HA AB BAB BB HB HBB BBB BHAA HAHA HHH pH HA HAH Bp AHHH BH SG KG se se ee eee eee a ea SBSNNTNHN ORM ODOANANMNTNHNOMWDHOGANANMNHNTNHNOMODHOCANMNTNOPMADOANMDTHOM-ODOAA ed det reed reed eet) reed ret ANANANANANATAAIMNMNAMNNAMNNMNNM TAT T TTT FTF TFTNMNMHMHMNMNMMNMNNNOW!O’O Re Rm RR RB RB RR RR RR RR eR Rm mem em me mem ne ne DOOR OBO POOP OOO OOO RE RR ER RR ERE EE EEE ER EEE EE EE EE EE ee Ee FIELDIANA: ANTHROPOLOGY 150 APPENDIX III.D. Environmental data. Area Phase No. Subarea Grid Eley Zone Slp HT ST SD SCI Er Aly H2O WT Irr Pl Pot Sci Veg Lnd Crop Spe AG AQ Ses ee see Ce CC CNC CONC CCC Cor (SSS SSS SS SS SS SSS SS OS SOS SSS SS SS) St OSes SS SOOO Sa SOc Sos Osorno So OOS Ore OS OSS OOOO StS SSO 1S OSS SSS SO OO SS SS SOS SS SSS OOS ORS SS ei SC SSP SS) Soa SS SS So SS SSS SS es SS Se CS Sea a = SS OSS SSS SS) SSS OS SO a Se Se SS SS SSO) SS SS OOS SOS ey CO OS OS SS FINAN TAANANANANANINCOANAINUOANON SS SOS OO SOS 1S SS aS —=OOn SB enna ees ae eS oonmn COCO ae a a = NDWAANANANANAIAIAIAIAIAITANQNCOONANAN Sa CC SO Oo SO OOS OS tt SrA Steet ee FANNTMOOANOAANANNTOUONANANOCAA ST OSS OS Ano COR CO ENO COR EC SRG CC CS rst COS SOS OS (OSS SSS 1S SS OS Ss OS SiS a SS Ss SS SS SS SSO SSS SO SS eS SS SSS SS SS SS ore Set SS Sv YS SS SS SS OTS) CNC COGN ret ict net eet ret vo SS GCS rte GN ied Sere QS (Vas ea eS Sea SS Ss =) Vay as = SVOAIVAA NAAM ANANAAINAN SH AANAIAAN 0.01 0.01 CORA MAS eS SCONES CD) Wa CON) =O) SSS 1G) NOCSOCOSASCSOYSOsaASoSSoSsS MOMS OOS 1S OOOO OOO SISO Seo one Sooo oe OS SS SS Ore SS re UCC re ered ret GY GN) GSN ret isnt SYD re Ve) eI OS re a SS re eet te) ANANAANANAANNAANRANNANAANIAN LOO SCO CS SS Oo eS St st cocoons NANNAMNMNMANANNMNATAMNANAARAANM Ant tytHtt ttt Hon ttonttet st See eee en EGON CE ECO CEC COE CO SS SS RS SS SS SESS SESE ORS SSS SS OSS Ot OS SS OES SS Fe SESS SESS SSS SO SS Oe Ge Bs PFAMNMNOSATONANONA TTONDATANVNONATMHNATANTGNAHONANUAADAMWANHNAOAANATMNATOD DHA TAPAGANONOnNT nA Hu TnONToannnenAa AAAMCSAN LAME ME ARSAARARMEMCASCHARSSCAAOMSMAMEMEAASCHANSCOSHMBAAIE ME BAM AAIAKRBANVSH St oe ee ee a ia a ke ie Le ee ee he ee ee ee ee SONDNNDODAWAWANAHAOGAM-OHACAGDAAAGATr- OA WOHOWWOH WOOD WOOD WODHGOAr DAA WADOAUADAWAIAAADGDAGTAHACPTAAADAAARAARACWHA ADA ANS eA eee aeNese a NaN eA Bee Be N Se eA ae BAe Se ee ae ae eae aS ANB eHNA A Se A Se See Se Se Se aN BAN BeN AS See ee SG ee) m™-m-Oornnonrr-nr-dwdmOoraonw- OOM OOO OOADND DN CO CO CO CO CO CO MO CO OM OOD COL C CO C COR OD MD COM ODD AHA DN A CO C C CF = S eo e = 2 = eH He Adee Ae eee A A A Rp A A A A A A A HB BBB BRR RB RB HB BAA AA Ae Ae Ae AB AR A ABR Ae A A Ap Ae Ae A Ae Ae A AA Ae AA Bee ft4<444¢4¢eeeee ture e eee e teeta tote ete te eee totter etter ence SEIN NMTN OM ODOC — IM AANA MNMOMODOCAAMNAMNHM OM WOANMNANOMWODOANMANMNTNHNHOM ODO IMABNMNMTFNHNOMm- OAC eI ee SiS yale SiS eS SO OI Na CS CSCO) OO 10000) 0000300 00 COROO C000 00) SSS SOS SSS a ceed eat eet rt set eet SS ot SmNmrTtOCOCOCOCCCoCCoCoocCoNOoOnANOoOTWOOs SOO RA ASP ANANNANANNR ARAN R ROCCO R A Ae oo ee oftOooett-eooeceoe ee ANAANSEMANANAANAAANAANANAAABANANAN S Vey Ve) Ya) =) Lai Yay = = aa) —) 6a) win Sie = Che aS Sooo oro Sere) CO La re Noro Sere eaemnowoooenn eenooenrtenna VIS Clore oie So SS SiS SOS SO SS SiS a Soa oS SSS SSS a SS ANN mM BI MONNN Ve AN RM DANN OR ADNAN ee OO =—Oooon Ce SB SB See SB So Sa = Oe onmoooconeo = ANANANAANANAANAANANMNANNANRANAANN Onmrennt On Onn eee oqooononne enone ANAFNNAMNAANTFTONMNNMNAAENARTIANANTTAANAIMNAM ToOHtTHTNMHHNHHH TTT NTH OHN TTT ITH GSS SS SOOO et rt GN GN et nC CoG GEG GGG Ee) News NR CoCo Omen BN COR Be BNA Se SSS So oa SS oo SS] Soa SS OO SS lS S&S NANANANPFIANNAANAIAIAIAIANA ANNAN NnonocweennN NNN OS SSNs SOON Atom SOE SSS Seo osco eecsceonecns SSazSounSotoonn SSS Soro oro ono 0.05 ee ea SC a GN tt GN Gs rt GN et rete oe i og Maw i i ee ie a I NANANMNMPANNWIANAAIAMNANAAAIAMNAAN SSS OO srt se et SS SS Sc ANMANNAANAATAMNRFPAREMNANANANARAIAIMNNA tots ston t tH tot tont+oHtt st RR RR RR RR we a a MORO OOP POOP OOOO OPO OOO ROR RRR RR ERE RRR ERO RRR RE RRR REE ERE ER RR RRR RE RRR EEE EE EE EE 15] FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO APPENDIX III.D. Continued. Area Phase No. Subarea Grid Eley Zone Slp HT ST SD SCI Er Aly H20 WT Irr Pl Pot Sci Veg Lnd Crop Spe AG AQ SE a eG an a ECC COE CCS CS Co | on Co SS oS SS aS SOOO SS SSS rear tS ES Sn SC SS SC SO SS 1S OS So SOS SSS SS SSS OS SSS SiS SOS Sr Sao SS Soc oo ocaoaanaocoooztothooococo yTooooooocoqoccoetft sas OoCcoocostoCooMnm stot a= tS I Ga ed det st eee saa a trl ete ANAFANNAAANANATANTANTACCOCOTOOCAATOCCONTUOUTANNONATOONANNCONANAAS SSeS SOS OOS Or on aS a OOS ayaa an On anionic) B= Ona RO a) COCO NO) SSS SC CCC CC OC SCO OE ORCI aS SO OSES SOS OSS SO OE SOS SS Orr aic Sn ee GC Oa SC OOS et SiS OS SS Oe SOS eS NO NOG Go) a ir 1 NN ca et eG) iG GS GN (GN SS GG QS (SS GS St GS tS SS tt tt tS Se Ban Gere ee CC CE COE OL SS SS or SoS SS Sor SS SO EO Se to ANAMANANANANANIAAIAAAINATIAAIAIAIAANANIANMIAN TF APFAIMNAAAIATAAIANAIANTATIAIAIMNMANANIANIAIAIAARWAN 0.15 0.10 ASSSCMHSONSHMNNNNDSCONSOMNANHNNHNTDONHNSSOMOSDOrMNMNANSSSCOSSSOMACSS FSO SONASSOCSOSOSOSCSCOSOSCHFSSSGHANANGCHASCaSHMNSASHASOMNSANKHSATSHNSNSCTSA Meee oS OS OOo oS OOOO ooo SOOO oO Oooo Sooo oo OSS > a eh ene EG Co ECC Ce eS EC eC One Sore SSS SS Soe i CS Sc re SS SN deer NA GA rad em tel ree) NUMNMNANANANMIMNAAANMAANANANINAMWIAAAAANMNA TAINAN MAMWIANMAPANANANANAAAIAIAINANARAAIAIAIAN a a SS eS tS SG AI ee a Sell MBS ANANANCAMNAAIMAAPIANANVIANANMWAAANARIAAAIMNAAATTAAIAANVIAAAIMAMNAATIFEAMNAAN tod te ett IH T TO NHOONMMMtTOHMTTHAMMHtHTOoOTTTTOHMTtTTtTHT SS SS S_ ?_ SY O SOS”_ SS SS SSL OOOQPL{_OooS SSS SS SSS SOS SSS SS SESS SS SSS —=) DPOMADMmMAATHMVOMNAMNDWONMNADROFASOMNANKRMODADAADADADTNDAADANCAHNHWAONNHNADTAMHANO BO eee es ey he SN Oe ON OL ee NGL OLN Oy OL 00, Sy S00 OY NOL OO SEN 08. tO. SOO RR BR RR BR RR Se Re SO NANA Se me mem BN N HS BNA ee Se NS ANN eee BN Re BRNO Ne NSO Cn Soros SS SOS Gy S60) 60! ON CN ON ON OO S © ©’ 60 ON CO I= = Oy CO 00 160 60) 100) © Oy CO!) ON ON © ON F= SON CO GS OO OS SNP APANDANDPANANNAN RBA De eee NANNA KH eet e eee soe oO Aerts es Qe en Ne BAAN mr MH WOWMDDWDHADWAWOAYANANAAAAAAADAAAAKrKeKe-K-KWOOMr-r-OHDAHACCAArerOKOMrr- ODD Or = = = Soe 4A AHA A BH AAA HB AHA BA AB AB BH DD BRB BB BBA HAA BA WBA AAW Bee Ae Hee qatdtatstetetttetstttetrtetteteteeetttoe eters ANTFNOMADOANMANOMWOHOANMNANHNHOMAODOANMHTANHNOFADOANMNTANHNOrAODAOAA aso eee se ANAANAANAIAANIMNMMNAMNNMANN TAT AT rT y~TyFFTNNnNNMNHMNMNNNNOO’O ee a a Oe ee DOO RO ORO POO OOOO OER ERE EEE ROE REE EEE EE EEE ERE ER EE Re Ee EE FIELDIANA: ANTHROPOLOGY 152 Appendix IV: Terrace and Structure Summaries This appendix includes information on terraces and structures that were recorded during the Guirtn/Albarradas regional survey. Appendix IV.A presents information on each mapped terrace, 1V.B on each mapped structure (mound) and platform, IV.C on each mapped plaza, and IV.D on each mapped house foundation that was not part of a defined terrace (unless noted otherwise). The terrace and structure information for the Guirun site was updated during the intensive site survey in 1996 and has been published (Feinman & Nicholas, 2004a, appendix 5), so we do not include the APPENDIX IV.A. Field site no. Subarea Terrace no. Length Width Area TL-MAT-MAT-95 ET 1 7 4 28 TL-MAT-MAT-95 ET 2 8 4 32 TL-MAT-MAT-95 ET 3 i i 49 TL-MAT-MAT-95 ET 4 7 4 28 TL-MAT-MAT-95 ET 5 5 5 BS TL-MAT-MAT-95 ET 6 8 3 24 TL-MAT-MAT-95 ET 7 9 3 27 TL-MAT-MAT-95 ET 8 6 4 24 TL-MAT-MAT-95 ET 9 6 4 24 TL-MAT-MAT-95 ET 10 7 3 21 TL-MAT-MAT-95 ET 11 8 4 28 TL-MAT-MAT-95 ET 12 6 4 24 TL-MAT-MAT-95 ET 13 8 3 24 TL-MAT-MAT-95 ET 14 i 3 21 TL-MAT-MAT-95 ET 15 9 4 36 TL-MAT-MAT-95 ET 16 10 4 40 TL-MAT-MAT-95 ET ia 7 4 28 TL-MAT-MAT-95 PA 18 5 3 15 TL-MAT-MAT-95 ET 19 11 3 33 TL-MAT-MAT-95 ET 20 18 3 45 TL-MAT-MAT-95 ET 2] 6 3 18 TL-MAT-MAT-95 ET oe) 18 3 54 TL-MAT-MAT-95 ET 23 8 7 56 TL-MAT-MAT-95 ET 24 13 5 65 TL-MAT-MAT-95 ET 25 9 5 45 TL-MAT-MAT-95 ET 26 17 4 68 TL-MAT-MAT-95 ET 27 13 8 108 TL-MAT-MAT-95 ET 28 10 5 50 TL-MAT-MAT-95 ET 29 12 4 48 TL-MAT-MAT-95 aly 30 i 4 28 TL-MAT-MAT-95 ET 31 13 5 65 TL-MAT-MAT-95 ET 32 6 6 36 TL-MAT-MAT-95 ET 33 8 3 24 TL-MAT-MAT-95 ET 34 6 4 24 TL-MAT-MAT-95 ET 35 5 5 25 TL-MAT-MAT-95 ET 36 7 6 42 TL-MAT-MAT-95 ET Sy 7 4 28 TL-MAT-MAT-95 ET 38 14 6 84 TL-MAT-MAT-95 ET 39 15 6 90 TL-MAT-MAT-95 ET 40 7 4 28 TL-MAT-MAT-95 ET 4] 15 5 Is TL-MAT-MAT-95 ET 42 9 4 36 TL-MAT-MAT-95 ET 43 12 > 60 TL-MAT-MAT-95 ET 44 12 5 60 TL-MAT-MAT-95 ET 45 15 3 TS TL-MAT-MAT-95 ET 46 6 5 30 TL-MAT-MAT-95 ET 47 17 5 85 TL-MAT-MAT-95 ET 48 7 4 28 TL-MAT-MAT-95 ET 49 6 4 24 Guirun site in these appendices. All other sites are listed by field site number. Phases of occupation are indicated by phase component numbers. The subarea column denotes whether the site is in the Albarradas area (AL) or in eastern Tlacolula (ET). Length and width measurements are listed in meters, terrace area is listed in square meters, and mound volume is listed in cubic meters. The final column includes special items found on the features beyond the usual pottery sherds. Terraces. EI LI UW WA IV V Artifacts and comments 57 house foundation 57 house foundation 57 house foundation 57 house foundation (3 X 2 m) 57 house foundation (5 X 3 m), visible terrace entrance 57 house foundation, entrance on NE Si 57 _ road below terrace FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 153 APPENDIX IV.A. Continued. Field site no. Subarea Terrace no. Length Width Area EI LI Il JIA IV V Artifacts and comments TL-MAT-MAT-95 ET 50 6 3 18 57 road below terrace TL-MAT-MAT-95 ET 5] 5 3 15 57 TL-MAT-MAT-95 ET 52 fl 4 28 57 TL-MAT-MAT-95 ET 53 20 3 60 i TL-MAT-MAT-95 ET 54 8 4 32 57 TL-MAT-MAT-95 ET 55 5 3 15 57 TL-MAT-MAT-95 ET 56 15 2 30 57 house foundation TL-MAT-MAT-95 ET 57 12 3 30 57 circular structure (1-m diameter) TL-MAT-MAT-95 ET 58 5 2 10 OM TL-MAT-MAT-95 ET 59 14 3 42 Si TL-MAT-MAT-95 ET 60 4 4 16 57 house foundation TL-MAT-MAT-95 ET 61 16 4 64 57 TL-MAT-MAT-95 ET 62 8 4 32 i TL-MAT-MAT-95 ET 63 7 3 2] Sy TL-MAT-MAT-95 ET 64 8 3 24 57 TL-MAT-MAT-95 ET 65 12 4 48 57 TL-MAT-MAT-95 ET 66 5 4 20 57 TL-MAT-MAT-95 ET 67 5 3 15 57 TL-MAT-MAT-95 ET 68 7 3 21 57 TL-MAT-MAT-95 ET 69 8 4 32 57 TL-MAT-MAT-95 ET 70 5 3 15 of TL-MAT-MAT-95 ET 71 4 4 16 57 TL-MAT-MAT-95 ET 2 9 8 2 i TL-MAT-MAT-95 ET Ws 6 4 24 57 road below terrace TL-MAT-MAT-95 ET 74 10 2 20 Dy TL-MAT-MAT-95 ET TS 6 5 30 57 TL-MAT-MAT-95 ET 76 16 S 80 Di TL-MAT-MAT-95 ET i 13 4 Sy Di TL-MAT-MAT-95 ET 78 16 4 64 57 TL-MAT-MAT-95 ET 79 6 4 24 57 TL-MAT-MAT-95 ET 80 6 5 2, Si TL-MAT-MAT-95 ET 81 5 4 20 57 TL-MAT-MAT-95 ET 82 5 4 20 oi TL-MAT-MAT-95 ET 83 12 4 48 Sy TL-MAT-MAT-95 ET 84 if 4 28 Dil TL-MAT-MAT-95 ET 85 10 5 50 Sil TL-MAT-MAT-95 ET 86 7 3 21 57 TL-MAT-MAT-95 ET 87 7 4 28 Sil TL-MAT-MAT-95 ET 88 9 3 OF) 57 TL-MAT-MAT-95 ET 89 12 4 48 Si TL-MAT-MAT-95 ET 90 12 9 108 57 TL-MAT-MAT-95 ET 9] 13 5 65 57 TL-MAT-MAT-95 ET 92 12 4 48 S7/ TL-MAT-MAT-95 ET 93 9 4 36 Si TL-MAT-MAT-95 ET 94 15 5 75 57 TL-MAT-MAT-95 ET 95 9 5 45 57 TL-MAT-MAT-95 ET 96 9 4 36 Sy TL-MAT-MAT-95 ET 97 7 4 28 Sil TL-MAT-MAT-95 ET 98 2D 4 88 Si TL-MAT-MAT-95 ET 99 18 4 iD oH) TL-MAT-MAT-95 ET 100 16 5 80 Si TL-MAT-MAT-95 ET 101 9 4 36 i] TL-MAT-MAT-95 ET 102 16 5 80 Si TL-MAT-MAT-95 ET 103 25 4 100 oy TL-MAT-MAT-95 ET 104 14 4 56 Si TL-MAT-MAT-95 ET 105 9 3 27 57 TL-MAT-MAT-95 ET 106 7 3 21 Si TL-MAT-MAT-95 ET 107 7 3 18 Sy) TL-MAT-MAT-95 ET 108 6 4 24 57 TL-MAT-MAT-95 ET 109 8 6 44 Ot TL-MAT-MAT-95 ET 110 19 5 95 i, TL-MAT-MAT-95 ET 111 7 3 21 57 TL-MAT-MAT-95 ET 112 8 5 40 i TL-MAT-MAT-95 ET 113 15 4 60 i TL-MAT-MAT-95 ET 114 20 4 80 Si TL-MAT-MAT-95 ET 115 30 4 120 57 TL-MIT-CC-5 AL 1 5) 2 9 — — 10 — —_—- — TL-MIT-CC-5 AL 2 4 4 16 — — 10 — —_— — TL-MIT-CC-5 AL 3 ess i Ie — — 10 — — — chert TL-MIT-CC-5 AL 4 16 4 64 — — 10 — —- — TL-MIT-CC-5 AL 5 5 5 20 — — 10 — — — _ walls on 4 sides, possible house foundation 154 FIELDIANA: ANTHROPOLOGY APPENDIX IV.A. Continued. Field site no. TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, Subarea Terrace no. Length Width — SANANNAATHAAYTANAAADANADAIOMHAHPWUNHPTWODeK HAPAARPHAADAWNHNOHA Area 28 198 418 198 400 420 350 300 324 324 115 60 80 LI 11 1] Artifacts and comments green obsidian blade chert tool orange-painted sherd chert, bone fragments chert flake hammerstone chert flakes 2 house foundations wall below terrace chert flakes chert core 2 house foundations 2 house foundations (1 on NW, 1 on SE), gray obsidian flake, chert biface chert uniface chert flake house foundation, chert 3 house foundations, chert core gray obsidian flake, lots of chert, including core house foundation? chert flakes building stone, partially preserved wall, grinding stone, chert flake house foundation, lots of chert, 2 green obsidian blades house foundation (6 X 4 m) with tomb, stairway up to Terrace 36, mano fragment, shell fragment (Oliva), chert core, 2 green obsidian blades house foundation, lots of chert, including point tip chert flakes chert core 3 house foundations, chert gray obsidian flake, hammerstone, lots of worked chert chipped stone mano, abrasion stone, chert MEXICO ISS APPENDIX IV.A. Continued. Field site no. TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-11 TL-SLA-SLA-12 TL-SLA-SLA-12 TL-SLA-SLA-12 TL-SLA-SLA-12 TL-SLA-SLA-12 TL-SLA-SLA-12 TL-SLA-SLA-12 TL-SLA-SLA-12 TL-SLA-SLA-12 TL-SLA-SLA-12 TL-SLA-SLA-12 TL-SLA-SLA-13 TL-SLA-SLA-13 TL-SLA-SLA-13 156 Subarea Terrace no. SCO WAAHAN _ — STO MOAINADNARWNR ra pest WN Length 9 17 19 20 9 108 21 72 36 13 Sil 1] 20 58 89 Width 4 Ww wn — ‘on OWWDH ek Area 36 82 133 580 1,068 28 109 135 132 230 305 EI i) Il © 0 0 0 © © © 0 OS © \O \o © © 90 OO OO ODO OOOO OOOO OO OO OO OO OO OO 0 OOO IIA IV Artifacts and comments chert worked chert 3 house foundations, building stone, lots of worked chert, including cores, mano and ax fragments, human bone chert flakes, gray obsidian flake chert flakes, tube or ceramic mold metate fragment, chert, gray obsidian flake lots of chert chert, including biface small road 14 m long (2 m wide) on E, chert biface 4 house foundations, building stone, lots of chert, 2 cut building stones, basalt fragment, hammerstone, 2 manos chert worked chert worked chert looted ax fragment lots of chert 2 entrances, incised fecomate ramp down to Terraces 40 and 41 worked chert chert above chert source chert source large stones in wall no wall, but artificially flattened no wall, but artificially flattened stone retaining wall cut bedrock retaining wall stone retaining wall stone retaining wall stone retaining wall small stone retaining wall below Structure 1 stone retaining wall on 1 side, on bedrock outcrop stone wall wall below wall below FIELDIANA: ANTHROPOLOGY APPENDIX IV.A. Continued. Field site no. Subarea Terraceno. Length Width Area EI LI II WIA IV V Artifacts and comments TL-SLA-SLA-19 ET ] 17 2 344 —-— — — 2 3 — black obsidian blade TL-SLA-SLA-20 AL ] 37 5 185 — 7 5 — —- — TL-SLA-SLA-20 AL 2 12 4 48 — 7 5 — — — chert, worked red stone TL-SLA-SLA-20 AL 3 20 6 1200 — 7 5 — —- — TL-SLA-SLA-20 AL 4 24 5 120. — 7 5 — —- — TL-SLA-SLA-20 AL 5 17 5 85 — i 5 _ —- — TL-SLA-SLA-20 AL 6 30 1] 330 = — 7 5 — — — chert, polished carnelian TL-SLA-SLA-20 AL i 24 10 240 — 7 5 — —- — TL-SLA-SLA-20 AL 8 13 8 104 — 7 5 a —- — TL-SLA-SLA-20 AL o 9 3 27. — i BS == —- — TL-SLA-SLA-20 AL 10 17 i Ho — i 5 — —- — TL-SLA-SLA-20 AL 11 7 5 350 — i 5 — —- — TL-SLA-SLA-20 AL 12 17 7 119 — 7 5 5 — 55 mano fragment TL-SLA-SLA-20 AL 13 8 4 320 — 7 5 — — — chert TL-SLA-SLA-20 AL 14 70 6 420 — q iy — — — chert TL-SLA-SLA-20 AL 15 21 4 84 7 5 — —- — TL-SLA-SLA-20 AL 16 44 ia 308 = 7 5 — —- — TL-SLA-SLA-20 AL 17 50 8 400 = — 7 5 — —- — TL-SLA-SLA-20 AL 18 26 4 104 — 7 5 — —_— — TL-SLA-SLA-20 AL 19 1] 4 44, — d > = —- — TL-SLA-SLA-20 AL 20 8 4 320 — 7 5 — —_—- — TL-SLA-SLA-20 AL 21 8 4 320 — 7 5 _— —- — TL-SLA-SLA-20 AL 22 42 4 168 — 7 5 — — — chert biface, hammerstone TL-SLA-SLA-20 AL 23 1 6 720 — fi 5 ae — — chert core, polishing stone (for ceramics) TL-SLA-SLA-20 AL 24 38 7 266 — 7 5 — — — chert core, mano fragment, above source of reddish stone TL-SLA-SLA-20 AL 25 47 5 2350 — 7 5 _ — — chert, good wall TL-SLA-SLA-20 AL 26 13 5 65 — if 5 — — — __ good wall TL-SLA-SLA-20 AL OT 12 3 36 7 5 — —- — TL-SLA-SLA-20 AL 28 19 3 57 7 5 — — — above riverbed TL-SLA-SLA-22 AL 3 20 8 160 — 8 6 = —- — TL-SLA-SLA-22 AL 4 10 4 40, — 8 6 ~- — — . stone retaining wall TL-SLA-SLA-22 AL > 10 5 50. — 8 6 — — — . stone retaining wall TL-SLA-SLA-22 AL ] 10 5 5300 8 6 6 — 56 wall on natural bedrock TL-SLA-SLA-22 AL p 18 5 90 — 8 6 6 — 56 2 house foundations, green obsidian flake, 3 chert cores TL-SLA-SLA-22 AL 6 14 6 84 8 6 — 56 chert TL-SLA-SLA-22 AL a 20 5 100 — 8 6 — — 56 chert, good wall TL-SLA-SLA-22 AL 8 15 i 105 — 8 6 — — 56 chert TL-SLA-SLA-22 AL 9 ? z 2. — 8 6 — — 56 no measurements possible TL-SLA-SLA-22 AL 10 2 ie > — 8 6 — — — _ no measurements possible TL-SLA-SLA-24 AL ] 47 34 1598 — — — — 10 53 cave just above with possible habitation TL-SLA-SLA-24 AL DZ, 80 16 12830 — — — — 10 53 greenstone ax, trace of canal leading to terrace TL-SLA-SLA-24 AL 3 20 8 60 — 5 — _— 10 53 chert core TL-SLA-SLA-27 AL 76 4 304. — 6 — — — small chert tool, greenstone disk, mano fragment TL-SLA-SLA-27 AL 2» 45 3 135 — 6 _ — — — _ mano fragment TL-SLA-SLA-27 AL 3 75 5 375 — 6 — — — —_ human bone, mano fragment TL-SLA-SLA-27 AL 4 19 6 14 — 6 — _ — — chert TL-SLA-SLA-27 AL 5 50 10 500. — 6 — — —_- — TL-SLA-SLA-27 AL 6 12 4 48 — 6 — ~- —- — TL-SLA-SLA-27 AL 7 40 3 1200 — 6 — a — — _ mano fragment (of red bedrock) TL-SLA-SLA-27 AL 8 7 4 28 — 6 — — —_- — TL-SLA-SLA-27 AL 9 10 6 60 — 6 — — —_- — TL-SLA-SLA-27 AL 10 40 8 320 — 6 — -- —- — TL-SLA-SLA-27 AL 1] 110 14 1540 — 6 — — — — high retaining wall TL-SLA-SLA-27 AL 12 31 4 124 — 6 — ~ — 54 TL-SLA-SLA-4] AL ] 18 10 180 — 12 11 12 12. 61 chert fragments, including core, good wall TL-SLA-SLA-41 AL 2 23 12 276 — 12 #11 12 12. 61. chert, including core, drill, gray obsidian drill, flake, blade, drilled sherd, good retaining wall, entrance on N TL-SLA-SLA-41 AL 3 36 3 108 =— 12 «11 12 12. 61 large foundation stones, stone abrader, chert core, large flake TL-SLA-SLA-41 AL 4 7 5 32 —— 8125 71) 12 12 61 2 house foundations (3 X 2.5 m, 2 X 2.5 m), chert TL-SLA-SLA-41 AL 5 i 5 350° =— 12 11 12 12 61 house foundation, retaining wall partly natural FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 157 Field site no. TL-SLA-SLA-41 TL-SLA-SLA-41 TL-SLA-SLA-4] TL-SLA-SLA-41 TL-SLA-SLA-41 TL-SLA-SLA-4]1 TL-SLA-SLA-41 TL-SLA-SLA-4]1 TL-SLA-SLA-41 TL-SLA-SLA-41 TL-SLA-SLA-4] TL-SLA-SLA-41 TL-SLA-SLA-4] TL-SLA-SLA-4] TL-SLA-SLA-4] TL-SLA-SLA-41 TL-SLA-SLA-4] TL-SLA-SLA-4] TL-SLA-SLA-43 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-S59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 158 Subarea AL AL AL AL AL ET ET ET ET Terrace no. 6 OAANIDNAARWNK Kw WNre WO WwW — NN I ONWA OG WAI RW OW W NN WwW A BON KW YNYNDWDNAWON NY W Area 90 45 45 650 280 45 24 112 120 12 12 12 12 12 12 12 12 12 12 12 APPENDIX IV.A. Continued. Il 11 1] 11 1] 1] 1] 1] 11 — Nw NV vy tw 61 61 61 61 61 61 61 61 a2 52 a2 52 52 52 52 52 ay a2 52 52 a2 Artifacts and comments gray obsidian core, green obsidian flake, goundstone ax fragment, hammerstone chert flakes, wall includes natural bedrock house foundation (2 X 4 m), wall on natural bedrock mano, hammerstone, gray obsidian flake, chert chunk gray obsidian blade, chert flakes, polished onyx, sherd disk green obsidian core, retaining wall house foundation, worked bedrock 2 chert cores, polished onyx, good retaining wall chert flakes mano chert flake wall partly natural mano fragment, chert core, good retaining wall building stone good retaining wall good retaining wall eroded retaining wall attached to House 2 chert house foundation house foundation house foundation house foundation, not possible to measure terrace, mano good wall stone wall partially preserved house foundation (2 X 2 m), good stone wall well-preserved stone wall well-preserved stone wall round structure (2.5 X 3.5 m) on terrace well-preserved stone wall house foundation on W side house foundation (4 X 4 m) on SE corner house foundation (4 X 4 m) on back edge house foundation on back edge 2 house foundations (both 4 X 3 m) L-shaped terrace, stone retaining wall FIELDIANA: ANTHROPOLOGY Field site no. TL-MAT-MAT-95 TL-MAT-MAT-95 TL-MAT-MAT-95 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-8 TL-MIT-CC-8 TL-MIT-CC-8 TL-MIT-CC-8 TL-MIT-CC-8 TL-MIT-CC-8 TL-MIT-CC-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-12 TL-SLA-SLA-12 TL-SLA-SLA-12 TL-SLA-SLA-12 TL-SLA-SLA-20 TL-SLA-SLA-20 TL-SLA-SLA-20 TL-SLA-SLA-22 TL-SLA-SLA-22 TL-SLA-SLA-22 TL-SLA-SLA-22 TL-SLA-SLA-22 TL-SLA-SLA-22 TL-SLA-SLA-23 TL-SLA-SLA-24 TL-SLA-SLA-24 TL-SLA-SLA-24 TL-SLA-SLA-24 TL-SLA-SLA-27 TL-SLA-SLA-27 TL-SLA-SLA-34 TL-SLA-SLA-41 TL-SLA-SLA-41 Subarea ET ET ET AL AL AL AL AL AL AL Structure Base Base Top Top EW Height Volume Ros no. l l Dy l yy v0 'D Jv J ES) BWP — NE nRWNYe A bo NN 3 '0 3 So —_ 3 — DOS S&S WH be pl NS z 10 20 10 40 15 APPENDIX IV.B. Structures and platforms. EW NS 2 2 Ww & 10 15 8 8 30-30 oe on 2 8 3 6 5 De on +8 10-6 ‘i 8 oat ale 8 6 8 6 16 8 i 5 148 yy 9 5 eS ii i0 Caen 5 7 5 230 1) yea gh ee 2 2 Cian? 10 84 ga 25 3 12 7 10 5 Omnis Sed 4. 6 11 8 14 19 Saeed 32 3 nN WD CO CO 3 Nn — _ AmDnokhwon nwo fh RW WwW ? 0.3 0. 0. 0. tw iw WwW oS wo TSS Se ae AwnAnnNo dw 4.6 0.3 2.0 ee ee ae NOW aw 0.3 ? 19 38 15 229 NY 32 29 59 24 89 534 58 359 171 46 162 Dy) 195 1] 90 345 EI LI HW WIA IV ee =F) ee Se hs =e i a ae (he 24 le ae nee en ee 2 al jf ite eo ae an a —— ee ee he == Wo Se et ee en aes a UN ee ee LOG 2 eS S110 ee ee ee ae) aE ge Eo = = E75 $5 =) eg se ee eS et ie a ca ar a ee o=0 1. we ae ee omen Or An) ees = SO oS Ga ee LSS —- — — — 4 en eee ee ee |) Sg ee SD eet eee 1) ees Ge ee = Wee ee 2 Ue ee 2 ao 12 I 12 12 Vv 37 57 57 5] 51 51 61 61 Comments not possible to measure large hammerstone arrowhead shape, looter’s pit 2 rooms (10 X 6m, 5 x 5 m), stairway (11 steps, 6 X 11 m) rock cobbles bedrock base on Platform 2 cobbles and stone blocks, looter’s pit 2 metate fragments, worked chert nodules flattened stones, metate stairway, stone cobble construction stairway on E (8-10 steps, 25 X 18.5 m), green obsidian blade, mano composed of stone rock wall composed of stone (cut from bedrock) walls maintain flattened space double line of rocks on W side, possible stone floor chert uniface 2 shell beads (1 is Thais triangularis) ground stone ax, chert destroyed by agriculture house foundation, human bones, chert, 3 shell fragments (Fusinus dupetitthouarsi) stone foundation built on natural bedrock, chert flakes room on E (4 X 3.5 m) adobe present, looted, stone cobble construction building stone building stone good wall large rocks define base, built on natural bedrock, looted built on natural bedrock supports Structure 1, part is natural bedrock FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 159 Field site no. TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-S3 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-S3 TL-SLA-SLA-53 TL-SLA-SLA-S3 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-5 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-S53 TL-SLA-SLA-53 TL-SLA-SLA-58 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 Ww Field site no. TL-MIT-CC-8 TL-SLA-SLA-8 TL-SLA-SLA-8 TL-SLA-SLA-20 TL-SLA-SLA-22 TL-SLA-SLA-41 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 160 Subarea AL Subarea AL AL AL Structure Base no. S rp WN Qw 22509 NH e Ho eS ace SO WAAYDAMAHW lo — peak a et fee WN — Nn 16 18 19 20 21 Dy) Plaza A A mOA wWrrrpy NS 1 26 37 APPENDIX IV.B. Continued. Base Top Top NS EW NS EW Height Volume Ros EI LI Il WA IV V Comments Di 8 0 — — 10 8 8 11 58 adobe wall, green stone bead, crystal 6 10 3 6 0.5 18 — — 10 8 8 11 58 cobble foundation 13 7 6 3 0.5 25 — — 10 8 8 11 58 cobble foundation 9 9 4 8 0.8 43 (es 8 11 58 1] 8 9 6 LS 104 — — 10 8 8 11 58 26 9 24 a IES 297 —- — 10 8 8 11 58 5 16 3 14 1.5 89 | (es 8 115s 19 5 17 3 1S 107 — — 10 8 8 11 58 6 IS 4 13 1.5 104 — — 10 8 8 11 58 6 1] 5. 8 0.3 12 — — 10 8 8 11 58 10 > 9 3 1.3 47 — — 10 8 8 11 58 i 16 5 14 1.3 112 — — 10 8 8 11 58 12 4 10 ”) 0.8 po — — 10 8 8 11 58 green obsidian blade 5 5 3 5 0.5 6 — — 10 8 8 11 58 adjacent to adoratorio 1] 6 9 4 0.3 12 — — 10 8 8 11 58 stucco, chert point, looted 3 4 0.0 0 58 worked chert 3 4 2. 3 0.5 5 58 20 fi 5 2 3.0 186 58 side mound of ballcourt (20 X 7 m) 20 7 5 3.0 186 58 side mound of ballcourt (20 X 7 m) 2 5 2 4 0.2 2 58 chert 5 p, 4 2 0.2 2 58 5 7 4 5 0.2 5 58 3 6 2 5 0.5 6 —- —- — — 9 — 58 1-m entrance on S, stone cobble foundation, 5 walls, chert 7 3 5 2 0.3 4 —- —- — — 9 — 58 entrance on E y) 2 1 1 0.3 0 — — 10 8 8 11 58 34 28 34 28 3.0 2,827.44 — — 10 8 8 11 58 looted, stucco floor 24 17 18 13 0.5 156 — — 10 8 8 11 58 4 9 3 6 0.5 13 58 30 5 32 3 1.5 181 — — 10 8 8 11 58 32 5 32 3 LS 188 — — 10 8 8 11 58 ] l ] l 0.0 0 38 outpost 5 6 5 6 0.3 q oy 5 5 3 3 0.5 8 52 12 8 12 8 0.5 48 52 entry on S, looter’s pit 1] 1] 1] 11 0.3 30 2 APPENDIX IV.C. Plazas. No. of EW Area _ structures EI LI IT HWIA IV V Comments 14 238 3 — 9 7 7 — — 35 910 3 2 11 9 10 — 59 worked chert, ground stone, gray obsidian blade 70 ~=—-.2,590 2) Delt 9 10 — — _ building stone, tomb?, human bone, mano, ax, 2 stone molcajetes, 2 gray obsidian flakes, 2 blades, green obsidian flake, Oliva shell 18 144 ] — 7 5 — — — mano fragment, chert core 15 330 6 — 8 6 6 — 56 35 560 3 — 12 11 12 12 61 gray obsidian blade, black obsidian blade 14 98 1 — 10 8 8 11 58 7 49 5 — 10 8 8 11 58 looted tomb, building stone, chert hammerstone, biface 17 289 5 — 10 8 8 11 58 9 81 ] —- — — —~— — $58 4 20 3 — 10 8 8 11 58° chert FIELDIANA: ANTHROPOLOGY Field site no. TL-MAT-MAT-95 TL-MAT-MAT-95 TL-MAT-MAT-95 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-MIT-CC-5 TL-SLA-SLA-1]1 TL-SLA-SLA-11 TL-SLA-SLA-11 TL-SLA-SLA-20 TL-SLA-SLA-20 TL-SLA-SLA-20 TL-SLA-SLA-20 TL-SLA-SLA-20 TL-SLA-SLA-20 TL-SLA-SLA-20 TL-SLA-SLA-20 TL-SLA-SLA-20 TL-SLA-SLA-20 TL-SLA-SLA-20 TL-SLA-SLA-20 TL-SLA-SLA-20 TL-SLA-SLA-22 TL-SLA-SLA-22 TL-SLA-SLA-22 TL-SLA-SLA-22 TL-SLA-SLA-22 TL-SLA-SLA-22 TL-SLA-SLA-22 TL-SLA-SLA-22 TL-SLA-SLA-41 TL-SLA-SLA-41 TL-SLA-SLA-41 TL-SLA-SLA-43 TL-SLA-SLA-43 TL-SLA-SLA-43 TL-SLA-SLA-43 TL-SLA-SLA-43 TL-SLA-SLA-43 TL-SLA-SLA-43 TL-SLA-SLA-43 TL-SLA-SLA-43 TL-SLA-SLA-S0 TL-SLA-SLA-50 TL-SLA-SLA-51 TL-SLA-SLA-S53 TL-SLA-SLA-S53 TL-SLA-SLA-S53 TL-SLA-SLA-53 TL-SLA-SLA-53 Subarea ET AL AL AL AL AL AL AL House foundation no. —e ee Nn RB WN WN Nw 11 12 patio Ne atio BRWNKHKWNKDTANADUNAHW aon aD Nn oO 2 a ° OMB WNHN KH KN eS APPENDIX IV.D. House foundations. Length Width 3 2 5 3 1 ] 3 4 4 5 5 3 10 6 > 5 6 4 2 ? 4 7 2 3 i) 4 5 5 5 > 5 SI 5 3 6 5 4 ? 2 5 5 4 8 5 5 3 5 7 8 3 4 10 12 3 5 2 2 Z ? ? 2 ? ? 12 8 3 3 3 2 4 eZ 5 3 3 5 3 3 5 5 4 4 3 2 4 3 6 4 4 9 4 5 4 4 2 2 2 ? 2 ? 2 2 2 3 2 5 Area 6 15 | 12 18 14 55 25 24 : 25 6 30 \o — MONO WM WD +9 +--+ tr LI Wo WA Iv = #10 = = =e 10 = = eC at = =) £10 a = =e 10 ot — — aes ae (oe a (i (a ch i ee (AE Sap. co liees5 2 a TES ee TS ee a TS ae (a 5 0s 8 6 SS a a oe S64 ee. ee S066 a Dali eel? ae lees 22 eri? Det! peel? a 9 Seay te 9 Se 9 reel 2 ne 9 a 9 ete Se 9 = = = 9 a 9 Se 82S 9 = 9.8 SaueeL Comments on Terrace 38 on Terrace 39 on Terrace 57 metate, Houses 1-4 are connected metate, Houses 1-4 are connected metate, Houses 1-4 are connected metate, Houses 1-4 are connected 3 chert flakes, chert cobble on Terrace 38 adjacent patio cut bedrock blocks, on bedrock hearth basalt flake, chert cores, flaking debris tomb, mano, chert biface, hammerstone, nacreous shell, shell animal head pendant chert flake Oliva porphyria pendant, chert biface chert, green obsidian blade ax fragment, chert flake, 5 hammerstones chert flakes, hammerstone, 12 cores, human bones tomb, human bones, greenstone bead, polishing stone (for ceramics), too destroyed to measure tomb, greenstone fragment with abraded edge tomb, human bones, clear obsidian blade, chert cores, debris tomb, human bones, chert flake stone wall visible 3 green obsidian blades, 4 gray obsidian blades worked stone not measured not measured not measured not measured adjacent to Structure 6 on Terrace 4 on Terrace 4 on Terrace 8 entrance on N, elaborate stone foundation attached to patio, chert uniface, looter’s pit in house part of foundation wall is bedrock outcrop chert biface, core, unworked onyx stone retaining wall rough circle of stones rough circle of stones clear corner, 2 walls on S and E FEINMAN AND NICHOLAS: SETTLEMENT PATTERNS IN ALBARRADAS, OAXACA, MEXICO 161 Field site no. TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-53 TL-SLA-SLA-S9 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 TL-SLA-SLA-59 Subarea House foundation no. AYANDNABRPWNKHK OWOAD Length BERR KR RWW Ow APPENDIX IV.D. Continued. Width Ww BBW NY W do -D -D W Area Oo BO +9 + + SO 16 16 12 EI LI I IA IV 11 1] V Comments 52 circular house, looter’s pit 52 on Terrace 2 52 on Terrace 9 $2 on Terrace 17 $2. _ on Terrace 18 52 on Terrace 20 52. _on Terrace 20 FIELDIANA: ANTHROPOLOGY = : 7 n , ™ , oh . ; cP 7 i. ae cd ae : % 4 a Seeger * mes. 4 ad 5 > =~ is his 7 ee eae > ae Aan gies aa 2 ees + f4 ’ psa eee oes Ge A. oe Ci a-é A paiec oo re hoe or 7 oe ‘ ao ‘ | mane as? — y AS as a —_— ft a aa oe a > re a eer ae | P— eo eas ee, ee ae =. ee i 7 em ine 7 oes oan . . “eq ae + eR Oee oes. “| MS 0g 65 4 “ese Cie Sele Oca 5 ae mat | ois -s . ae eee . a) ee ReGen . Te a van _. es. € oe ? aa os Pea. ers i y : ‘ Reins 72 a re * <5. 2 eo’ 2 5 ee . a i Jagth s Dae a /— 2) ue a a , Oe ee a “ae a — ‘Seer. 4 tin ~~? S @% 7 a - > F oy ° » be °”« e448 : Pete se 14 oe Par ee ' =~ °° Su oan 7) @ me ee * .-"-@ 7 e¢ es? i . a Ts Ste 7 . aero. ta a : oro o> a ae <- 7996s 7 ee i: Oe oe 3@4 ve S % Chicago, Illinois 60605-2496 — eum.org — : iel Fieldiana@f . Field Museum of Natural History mai 1400 South Lake Shore Drive rE