£ ix FIELDIANA Anthropology NEW SERIES, NO. 31 The Early Ceramics of the Inca Heartland Brian S. Bauer April 30, 1999 Publication 1501 PUBLISHED BY FIELD MUSEUM OF NATURAL HISTORY Information for Contributors to Fieldiana General: Fieldiana is primarily a journal for Field Museum staff members and research associates, although manuscripts from nonaffiliated authors may be considered as space permits. The Journal carries a page charge of $65.00 per printed page or fraction thereof. Payment of at least 50% of page charges qualifies a paper for expedited processing, which reduces the publication time. Contributions from staff, research associates, and invited authors will be considered for publication regardless of ability to pay page charges, however, the full charge is mandatory for nonaffiliated authors of unsolicited manuscripts. Three complete copies of the text (including title page and abstract) and of the illustrations should be submitted (one original copy plus two review copies which may be machine copies). No manuscripts will be considered for publication or submitted to reviewers before all materials are complete and in the hands of the Scientific Editor. Manuscripts should be submitted to Scientific Editor, Fieldiana, Field Museum of Natural History, Chicago, Illinois 60605-2496, U.S.A. Text: Manuscripts must be typewritten double-spaced on standard-weight, SV2- by 11 -inch paper with wide margins on all four sides. If typed on an IBM-compatible computer using MS-DOS, also submit text on 5V4-inch diskette (WordPerfect 4.1, 4.2, or 5.0, MultiMate, Displaywrite 2, 3 & 4, Wang PC, Samna, Microsoft Word, Volks- writer, or WordStar programs or ASCII). For papers over 100 manuscript pages, authors are requested to submit a "Table of Contents," a "List of Illustrations." and a "List of Tables" immediately following title page. In most cases, the text should be preceded by an "Abstract" and should conclude with "Acknowledgments" (if any) and "Literature Cited." All measurements should be in the metric system (periods are not used after abbreviated measurements). The format and style of headings should follow that of recent issues of Fieldiana. For more detailed style information, see The Chicago Manual of Style (13th ed.), published by The University of Chicago Press, and also recent issues of Fieldiana. References: In "Literature Cited," book and journal titles should be given in full. Where abbreviations are desirable (e.g., in citation of synonymies), authors consistently should follow Botanico-Periodicum-Huntianum and TL-2 Taxonomic Literature by F. A. Stafleu & R. S. Cowan (1976 et seq.) (botanical papers) or Serial Sources for the Biosis Data Base (1983) published by the BioSciences Information Service. Names of botanical authors should follow the "Draft Index of Author Abbreviations, Royal Botanic Gardens, Kew," 1984 edition, or TL-2. References should be typed in the following form: Croat, T. B. 1978. Flora of Barro Colorado Island. Stanford University Press, Stanford, Calif., 943 pp. Grubb, P. J., J. R. Lloyd, and T D. Pennington. 1963. A comparison of montane and lowland rain forest in Ecuador. I. The forest structure, physiognomy, and floristics. Journal of Ecology, 51: 567-601. Langdon, E. J. M. 1979. Yage among the Siona: Cultural patterns in visions, pp. 63-80. In Browman, D. L., and R. A. Schwarz, eds., Spirits, Shamans, and Stars. Mouton Publishers, The Hague, Netherlands. Murra, J. 1946. The historic tribes of Ecuador, pp. 785-821. In Steward, J. H., ed.. Handbook of South American Indians. Vol. 2, The Andean Civilizations. Bulletin 143, Bureau of American Ethnology, Smithsonian Institution, Washington, D.C. Stolze, R. G. 1981. Ferns and fern allies of Guatemala. Part II. Polypodiaceae. Fieldiana: Botany, n.s., 6: 1- 522. Illustrations: Illustrations are referred to as "figures" in the text (not as "plates"). Figures must be accompanied by some indication of scale, normally a reference bar. Statements in figure captions alone, such as "X 0.8," are not acceptable. Captions should be typed double-spaced and consecutively. See recent issues of Fieldiana for details of style. All illustrations should be marked on the reverse with author's name, figure number(s), and "top." Figures as submitted should, whenever practicable, be 8V2 by 11 inches (22 X 28 cm) and may not exceed WVi by 16V4 inches (30 X 42 cm). Illustrations should be mounted on boards in the arrangement to be obtained in the printed work. This original set should be suitable for transmission to the printer as follows: Pen and ink drawings may be originals (preferred) or photostats; shaded drawings must be originals, but within the size limitation; and photostats must be high-quality, glossy, black and white prints. Original illustrations will be returned to the corre- sponding author upon publication unless otherwise specified. Authors who wish to publish figures that require costly special paper or color reproduction must make prior arrangements with the Scientific Editor. Page Proofs: Fieldiana employs a two-step correction system. The corresponding author will normally receive a copy of the edited manuscript on which deletions, additions, and changes can be made and queries answered. Only one set of page proofs will be sent. All desired corrections of type must be made on the single set of page proofs. Changes in page proofs (as opposed to corrections) are very expensive. Author-generated changes in page proofs can only be made if the author agrees in advance to pay for them. © This paper meets the requirements ot ANSI/NISO Z39.48-1992 (Permanence ot Paper). FIELDIANA Anthropology NEW SERIES, NO. 31 The Early Ceramics of the Inca Heartland Brian S. Bauer Research Associate Department of Anthropology Field Museum of Natural History Roosevelt Road at Lake Shore Drive Chicago, Illinois 60605-2496 U.S.A. Associate Professor of Anthropology University of Illinois at Chicago 1007 West Harrison Street Chicago, Illinois 60607-7139 U.S.A.* * Address to which reprint requests and correspondence should be sent. Accepted February 12, 1998 Published April 30, 1999 Publication 1501 PUBLISHED BY FIELD MUSEUM OF NATURAL HISTORY © 1999 Field Museum of Natural History ISSN 0071-4739 PRINTED IN THE UNITED STATES OF AMERICA CK This work is dedicated to Luis Barreda Murillo, friend and teacher. (Photograph courtesy of Gordon McEwan.) Table of Contents Preface and Acknowledgments ix Chapter 1. The Inca in Cuzco 1 The Research Area and Research Meth- ods 1 Classic Inca Ceramics 8 Other Contemporary Styles 10 Chapter 2. Ceramics of the Late Inter- mediate Period (A.D. 1000-1400) 13 Killke Ceramics 13 Colcha Ceramics 20 The Relationship Between Killke and Colcha Ceramics 22 Summary and Discussion 29 Chapter 3. Ceramics of the Middle Hori- zon (A.D. 550-1000) 61 Wari and Wari-Style Ceramics 61 Ocros-Style Ceramics of the Province of Paruro 63 Arahuay Ceramics 67 Qotakalli Ceramics 70 Ccoipa Ceramics 75 Muyu Oreo Ceramics 78 Horizontally Incised Drinking Vessels 81 Summary and Discussion 83 Chapter 4. Ceramics of the Early Inter- mediate Period and the Early Hori- zon (1000 B.C.-A.D. 550) 123 Chanapata and Derived Chanapata Ceram- ics 123 Marcavalle Ceramics 125 Paruro Formative Ceramics 126 Incised, Early Tiwanaku-Related Vessels 127 Chapter 5. A Revised Ceramic Sequence for the Cuzco Region 143 Literature Cited 147 Appendix 1 . Shallow Bowls 151 Appendix 2. Radiocarbon Dates from the Cuzco Region 155 2-5. Common Colcha bowl designs 22 2-6. Common Colcha bowl designs 23 2-7. Common Colcha bowl designs 24 2-8. Distance from Cuzco of sites with Killke ceramics 26 2-9. Distance from Cuzco of sites with Colcha ceramics 26 2-10. Density of sites with Killke ceramics and distance from Cuzco 27 2-11. Density of sites with Colcha ceram- ics and distance from Araypallpa 28 3-1. Straight-sided flaring bowls from Muyu Roqo 65 3-2. Orange-ware drinking vessels from Muyu Roqo 66 3-3. Orange- ware drinking vessels from Muyu Roqo 67 3-4. Black-ware incised drinking vessels from Muyu Roqo 68 3-5. Common Qotakalli designs 72 3-6. Qotakalli rims from steep-sided bowls 73 3-7. Ccoipa straight-sided drinking ves- sels 76 3-8. Ccoipa straight-sided drinking ves- sels 77 3-9. Muyu Oreo ceramics 80 3-10. Horizontally incised drinking vessels .. 83 4-1. Paruro Formative vessel with molded face 127 4-2. Incised, early Tiwanaku-related ce- ramics 1 30 5-1. Revised ceramic sequence for the Cuzco region 144 A-l. Shallow bowls 151 List of Drawings (Ceramic Designs) List of Illustrations Frontispiece: Luis Barreda Murillo 1-1. The standard ceramic sequence for the Cuzco Valley 3 2-1. Common Killke bowl designs 15 2-2. Common Killke designs 17 2-3. Common Killke bowl designs 18 2-4. Single-strap-handle face neck jar 19 2-1. Common Killke designs 32 2-2. Common Killke designs 33 2-3. Common Killke bowl designs 34 2-4. Common Killke bowl designs 35 2-5. Common Killke bowl designs 36 2-6. Common Killke bowl designs 37 2-7. Common Killke bowl designs 38 2-8. Common Killke bowl designs 39 2-9. Common Killke bowl designs 40 2-10. Transitional Killke-Inca bowl 41 2-11. Common incurving Killke bowl de- signs 42 2-12. Common incurving Killke bowl de- signs 43 2-13. Single-strap-handle face neck jar 44 2-14. Single-strap-handle face neck jar 44 2-15. Single-strap-handle face neck jars 45 2-16. Single-strap-handle face neck jars 46 2-17. Single-strap-handle face neck jars 47 2-18. Common Colcha bowl designs 48 2-19. Common Colcha bowl designs 49 2-20. Common Colcha bowl designs 50 2-21. Common Colcha bowl designs 51 2-22. Common Colcha bowl designs 52 2-23. Common Colcha bowl designs 53 2-24. Common Colcha bowl designs 54 2-25. Common Colcha bowl designs 55 2-26. Common Colcha bowl designs 56 2-27. Common Colcha bowl designs 57 2-28. Common Colcha bowl designs 58 2-29. Common Colcha jar designs 59 2-30. Colcha straight- sided drinking ves- sels 60 3-1. Incurving bowls from Muyu Roqo 86 3-2. Straight-sided, flaring bowls from Muyu Roqo 87 3-3. Straight-sided, flaring bowls from Muyu Roqo 88 3-4. Straight-sided, flaring bowls from Muyu Roqo 89 3-5. Orange-ware drinking vessels from Muyu Roqo 90 3-6. Orange-ware drinking vessels from Muyu Roqo 91 3-7. Orange- ware drinking vessels from Muyu Roqo 92 3-8. Orange-ware drinking vessels from Muyu Roqo 93 3-9. Black-ware incised drinking vessels from Muyu Roqo 94 3-10. Black-ware incised drinking vessels from Muyu Roqo 95 3-11. Black- ware incised drinking vessels from Muyu Roqo 96 3-12. Black- ware drinking vessels from Muyu Roqo 97 3-13. Arahuay bowl designs 98 3-14. Arahuay bowl designs 99 3-15. Common Qotakalli designs 100 3-16. Common Qotakalli designs 101 3-17. Monochrome Qotakalli ceramics 102 3-18. Monochrome Qotakalli ceramics 103 3-19. Qotakalli bases from steep-sided bowls 104 3-20. Qotakalli rims from steep-sided bowls 105 3-21. Qotakalli rims from steep-sided bowls 106 3-22. Ccoipa straight- sided drinking ves- sels 107 3-23. Ccoipa straight-sided drinking ves- sels 108 3-24. Ccoipa straight-sided drinking ves- sels 109 3-25. Ccoipa straight-sided drinking ves- sels 110 3-26. Ccoipa bowls Ill 3-27. Common Ccoipa designs 112 3-28. Common Ccoipa designs 113 3-29. Ccoipa bowl 114 3-30. Muyu Oreo ceramics 115 3-31. Muyu Oreo ceramics 116 3-32. Muyu Oreo straight-sided drinking vessel with front face god motif 117 3-33. Muyu Oreo straight-sided drinking vessel with a convex, horizontal band 117 3-34. Muyu Oreo slightly concave, steep- sided bowl 118 3-35. Muyu Oreo annulated bowl with concave sides 118 3-36. Horizontally incised drinking ves- sels 119 3-37. Horizontally incised drinking ves- sels 120 3-38. Horizontally incised drinking ves- sels 121 4-1. Paruro Formative vessel rims 131 4-2. Paruro Formative vessels with short flaring rims 132 4-3. Paruro Formative vessels with long, slightly flaring necks 133 4-4. Paruro Formative vessel rims 134 4-5. Paruro Formative flaring bowls 135 4-6. Paruro Formative flaring bowl 136 4-7. Paruro Formative flaring bowls 137 4-8. Paruro Formative bowls and fig- urine 138 4-9. Paruro Formative vessels with mod- eled faces 139 4-10. Paruro Formative vessel with mod- eled face 140 4-11. Incised, early Tiwanaku-related ce- ramics 141 A-l. Shallow bowls 152 A-2. Shallow bowls 153 VI List of Maps 1-1. The Inca Empire in 1532 2 1-2. Department of Cuzco 4 1 -3. Province of Paruro 5 1-4. The research region 7 1-5. Test excavated sites 9 3- 1 . Distribution of sites with Wari-style ceramics 69 3-2. Distribution of sites with Arahuay ceramics 71 3-3. Distribution of sites with Qotakalli ceramics 74 3-4. Distribution of sites with Ccoipa ce- ramics 79 3-5. Distribution of sites with Muyu Oreo ceramics 82 3-6. Distribution of sites with horizontally incised drinking vessels 85 4-1. Distribution of sites with Paruro For- mative ceramics 128 4-2. Distribution of sites with incised, early Tiwanaku-related ceramics 129 A-l. Distribution of sites with shallow bowls 154 Back cover: Fragment of a Paruro Formative flaring bowl. This study presents a new pottery sequence for the Cuzco region in the south central Andes of Peru. The investigation was based on surface collections and test excavations con- ducted in the Cuzco region from 1984 through 1987. The archaeological survey, covering more than 600 square ki- lometers, recorded the locations of approximately 250 sites, which contained a total of 450 cultural components. A new ceramic chronology for the lnca heartland is proposed in the conclusion to the study. Preface and Acknowledgments This work explores the ancient ceramic traditions of the Cuzco region, Peru. Ceram- ic studies are critical tools in the modeling of pre- historic and historic cultural developments. Rela- tively little archaelogical research can be con- ducted without them. The first ceramic sequence for the Cuzco region was proposed by John H. Rowe in 1956. This general sequence divided the pre-Hispanic period of the region into six broad temporal-ceramic classifications. There have been no systematic attempts to update it in 40 years. Although much of the current archaeological re- search in the Cuzco region has focused on de- scribing and analyzing its pre-Inca cultures, few new ceramic styles have been formally proposed. The lack of descriptions and illustrations for the Cuzco ceramic styles has handicapped the devel- opment of archaeological research in the southern Andes of Peru. During the late 1980s and early 1990s, the Cuz- co ceramic sequence was intensively and inde- pendently studied by Mary Glowacki, Julinho Za- pata, and myself. We have each arrived at slightly different reconstructions of the sequence, which is reflective of our different research areas, ref- erence collections, and expertise. As each of these studies is published, our understanding of the re- gion's complex ceramic sequence will grow. Sim- ilarities in the three sequences will serve as areas for discussion, and differences in the sequences can serve as points for debate. That three alter- native models can be constructed is not surprising because the available data are still fragmentary. Although our understanding of the Cuzco ceramic sequence is just beginning, it is clear that the se- quence includes a remarkable combination of lo- cal production centers and complex, regional ex- change networks. Additional research will no doubt allow the redefinition of various ceramic styles and perhaps even suggest different evolu- tionary sequences. Many individuals and organizations have aided me in completing the study and the book. Most important, Luis Barreda Murillo took great care to teach me much of the Cuzco Valley ceramic sequence during my first stays in Cuzco, and I continue to learn from him today. Without his help, understanding, and support, this project would never have been started. I would also like to thank Mary Glowacki, who helped me classify and describe the Middle Horizon styles in my col- lections. Without her help, the manuscript would not have been completed. My understanding of Cuzco ceramics has been improved by conversations with members of the Universidad San Antonio Abad del Cuzco, in- cluding Raymundo Bejar Navarro, Manuel Cha- vez Ballon, Jose Gonzales Corrales, Italo Oberti Rodrigues, Alfredo Valencia Zegarra, and es- pecially Julinho Zapata. Likewise, my research has benefited from contact with members of the Instituto Nacional de Cultura in Cuzco, including Percy Ardiles Nieves, Fernando Astete Victoria, Arminda Gibaja Oviedo, Wilbert San Roman Luna, and Wilfredo Yepez Valdez. Members of the survey crews included Melissa Baker, Tamara Bray, Silva Lopez Aranguri, Mar- tina Munsters, Leslie Ranken, Nilo Torres Poble- te, Wilbert Torres Poblete, and Wilbert Vera Ro- bles. Edmundo de la Vega and Wilbert Torres Poblete helped direct the test excavations, and El- liana Gamarra Carrillo and Marlene Pinares su- pervised the processing of the artifact collections in Cuzco. I thank Michael Malpass, Dean E. Ar- nold, Mary Glowacki, Frances Hayashida, and Charles Stanish for their critical readings of the manuscript. Various chapters have also been read by Karen Mohr-Chavez, Chad Gifford, Paul Gold- stein, Sara Lunt, and Martina Munsters. Their crit- icisms and suggestions are gratefully acknowl- edged. Permission for the project was granted by the Instituto Nacional de Cultura (INC): Lima and Cuzco. Funds and support were provided by the L. J. Skaggs and Mary C. Skaggs Foundation, the Fulbright-Hays Fellowship Committee, the Orga- nization of American States, the Institute for New World Archaeology, and the University of Chi- cago Housing System. I thank them for their aid. Portions of some chapters appeared in slightly IX different versions in previous publications. Parts of tery from Cuzco, Peru, in the Field Museum of Chapter 1 appeared in Bauer, The Development of Natural History (Field Museum of Natural History, the Inca State (University of Texas Press). Parts of Chicago). Grateful acknowledgment is made to Chapter 2 appeared in the same publication and in these publishers for permission to reuse material. Bauer and Stanish, Killke and Killke -Related Pot- The Inca in Cuzco The Cuzco region of the south central Andean highlands of Peru is famous as the heartland of the Inca Empire (ca. a.d. 1400-1532). This em- pire, perhaps the largest to develop in the New World, was the product of thousands of years of cultural development (Map 1-1). Prior to the Inca, the Cuzco region underwent various periods of foreign influence and indigenous development.1 The Ayacucho-based Wari Empire controlled parts of the region from about a.d. 550 to 900, and recent research indicates that the Tiwanaku Empire, and other earlier polities from the Lake Titicaca region, may have also had an impact on the region. Before, during, and after these periods of outside influence, Cuzco was home to thriving and diverse local societies. John H. Rowe began systematic archaeological research in the Cuzco region in the early 1940s with test excavations at several sites in and near the city of Cuzco and exploratory visits to many other sites in the region. Building on this work and on excavation data generated by the Cuzco archaeologist Manuel Chavez Ballon, in 1956 Rowe began to develop a pre-Hispanic ceramic sequence for the region that would include seven broad temporal-ceramic classifications. These classifications span from latest to earliest: Classic Inca, Killke, Wari (Huari) and contemporary re- gional styles, Huaru, Derived Chanapata, Chana- pata, and Marcavalle (Fig. 1-1). Over the past 40 years, a number of researchers have conducted large-scale exploration, survey, 1 The term "Cuzco region" is defined as the area ap- proximately 60 aerial kilometers in radius from the city of Cuzco (Bauer, 1992a). and excavation projects in the Cuzco region. These projects have improved our understanding of known ceramic types and have recovered evi- dence of several new ceramic styles. In addition, a host of radiocarbon dates have been recovered that are critical for updating the Cuzco ceramic sequence. It is the goal of this work to establish a ceramic chronology for the Province of Paruro (Department of Cuzco) and to use this chronolo- gy, along with the results of other recently com- pleted projects, to revise the ceramic sequence for the Inca heartland. The Research Area and Research Methods In this work, the prehistoric ceramic sequence for the Province of Paruro is described and ana- lyzed. The investigation has direct implications for the cultural history of the Inca because the Province of Paruro is located immediately south of the city of Cuzco (Map 1-2). Surface collec- tions and test excavations were conducted in the Province of Paruro from 1984 through 1987 under the auspices of the Pacariqtambo Archaeological Project. The archaeological survey covered more than 600 square kilometers. The locations of ap- proximately 250 sites, which included a total of 450 occupations, were recorded during the pro- ject. The sites ranged from small, isolated scatters of ceramics to the large nucleated center of Mau- kallaqta, which measures nearly 6 hectares (Bauer, 1992a, pp. 65-108, 1992b). Surface col- The Inca in Cuzco Map 1-1. The Inca Empire in 1532. lections were made at each site to determine their occupational history. Test excavations were con- ducted at 15 sites during 1986 and 1987. The Pa- ruro surface and test excavation collection is one of the largest systematically collected samples of ceramics so far recovered in the Cuzco region. The Province of Paruro is separated from the Cuzco Valley in the north by the Huanacauri ridge, which rises over 4,000 meters above sea level (masl) and forms the watershed between the Urubamba and Apurimac rivers (Map 1-3). The southeastern and southwestern borders of the province are defined by the Apurimac and Santo Tomas rivers, respectively. The southern bound- ary is drawn across a high (more than 4,000 masl) puna region to the north of Livitaca. The province is divided into nine districts: Yaurisque, Huanoquite, Pacariqtambo, and Paru- ro, which are north of the Apurimac River, and Colcha, Pillpinto, Accha, Ccapi, and Omacha, 2 Chapter One AD 1600 Colonial x -j * 1500 1400 Classic Inca 3 Early Inca 08 1300 (Killke) a> E 1200 3 J: O 1100 ffl _J 1000 900 Wari and local styles o N 800 O X 2 600 500 400 300 Huaru 0 ro TJ 200 Q) E a> 100 JZ ro LLi 100 200 300 400 Derived Chanapata 500 Chanapata C 600 O N b 700 X P3 800 Marcavalle 111 900 1000 BC *LH = Late Horizon Fig. 1-1. 1 fhe s andard ceramic 5 equence for the Ci zco Valley which are south of the Apurimac River. This river and its major tributaries form the internal political boundaries for these nine districts. The Apurimac River crosscuts the province, with one-third of the territory lying to the north and two-thirds lying to the south of the river. The northern third of the province is subdivided by the Molle Molle River and the Paruro River, two tributaries of the Apu- rimac. The Molle Molle River enters the Apuri- mac near the village of Nayhua (2,400 masl) and forms the boundary between the districts of Huan- oquite and Pacariqtambo. The Paruro River is lo- cated east of the Molle Molle River and enters the Apurimac River near the village of Cusibamba The Inca in Cuzco UCAYALI JUNIN >5s ( BRAZIL 7 \. 0 \ "* \ 0 \ < 0\ 0 * J s 7 T ' "^ CHILI MADRE DE DIOS AYACUCHO .-0N^-' APURIMAC S' w>* V. PAUCARTAMBO . URUBAMBAsx "^ , V Urubamba* L.^ i ( ■*V.—*>_JL. ^-WCalca . Paucartambo ! Chi i7rn.' V J ,S"7 '? "x DEPARTMENT CAPITAL *w PROVINCE CAPITAL • DEPARTMENT BOUNDARY PROVINCE BOUNDARY LAKE C^ 0 10 20 30 40 50 km QUISPICANCHI / ~C £UZCO. , S f^ ^tUrcos \Paruro.^ \\^ s VAeomlyo. \yj| /^ ^f \ PS ACOMAYQ^ I / PARUl(0^i%CANCHIS ) m **~»-^~Yanaoca»\ >,. s~ f I \ \ Sicuani / ZANAS.A. f Santo Tomas S *A\ f>J O-y — / ESPINAR (T A/ ^7 -Yauri I ^ ^ \ AREQUIPA S. / ^ CHUMBIVILCASCANAS. PUNO Y Map 1-2. The Department of Cuzco. 4 Chapter One • cuzco Huanacauri aurisque / ^ , I QUISPICANCHI * Maukallaqta/ ^,\ yPacariqtambo ^^^ Paruro . COTABAMBA^ Si AREQUIPA ^ DEPARTMENT CAPITAL © COMMUNITY • PROVINCE BOUNDARY — SURVEY LIMIT ... ROADS — RIVERS - Omacha , Jullpayi Ccoyaneio\ Antapallpa* Tahuipampa* ] v ( Tica Mayo • \ .^— - C_^ SANTO TOMAS 0 5 10 Contour lines every 1,000 i 20 KM Map 1-3. The Province of Paruro. (2,775 masl). The Paruro River Valley is wider than the Molle Molle valley and contains the most productive agricultural land in the province. Now as in antiquity, the Paruro River Valley contains the highest population concentration of the prov- ince. The southern two-thirds of the province is di- vided in half by the Velille River. The isolated district of Ccapi is situated south of the Apurimac River and west of the Velille River. Between the Velille and the Apurimac rivers lie the districts of Colcha, Accha, Pillpinto, and the southernmost district of Omacha. The largest community in the southern two-thirds of the province is the town of Accha. The modern population distribution of the prov- ince is largely localized into the district capitals of Yaurisque, Huanoquite, Pacariqtambo, Paruro, Colcha, Accha, Ccapi, Omacha, and Pillpinto. This settlement pattern is a direct reflection of the Spanish reduction (reduction) policy, which be- gan in 1571 (Gade & Escobar Moscoso, 1982). In an effort to more efficiently extract tribute, land, and labor, as well as to promote religious indoc- trination of native peoples, Viceroy Francisco de Toledo implemented a systematic reorganization The Inca in Cuzco of the Andean demographic landscape, forcing the local inhabitants of the Andes to abandon their traditional settlements. The scattered populations of the highlands were resettled into newly created towns called reducciones, and the former settle- ments were frequently destroyed to prevent their reoccupation. This resettlement policy marked the end of indigenous systems of settlement place- ment in the Andes. As a result, the current pop- ulation distribution and settlement pattern of the Province of Paruro does not reflect indigenous systems of spatial, social, or economic organiza- tion, and the prehistoric regional settlement pat- terns of the Province of Paruro are best discerned through archaeological investigations. Survey Boundaries and Research Methods The survey area represents approximately one- third of the Province of Paruro. It extends from the ruins of Huanacauri near Cuzco to the town of Accha, 42 aerial kilometers to the south. The area is delineated on the west by the Velille and Molle Molle rivers. On the east it includes the drainage system of the Paruro River, and further south it is bounded by the Apurimac River (Map 1-4). Systematic survey work conducted in this portion of the Province of Paruro covered most of the territory thought to have been controlled by the Masca, Chillque, and Tambo ethnic groups (Bauer, 1992a). The archaeological survey followed guidelines provided by Parsons and Hastings (1977) for re- gional survey work in the Andes. The goal of the survey was to identify the locations of all prehis- toric habitation sites and support facilities in the research zone. To conduct the survey, teams of two to three persons, spaced at 50 to 150-m in- tervals, walked assigned areas and identified the locations of prehistoric settlements and related features, such as roads, terraces, and bridges. The general state of site preservation in the Province of Paruro is not as outstanding as in many other regions of the Department of Cuzco. For example, the Urubamba River Valley, some 60 kilometers north of Cuzco, contains Inca and pre-Inca sites with well-preserved stone masonry (Fejos, 1944; Kendall, 1974, 1976, 1984, 1985). In contrast, the majority of the Inca and pre-Inca sites in the Province of Paruro are represented only by concentrations of ceramic fragments on the surface of plowed fields or on the sides of eroding ridges. Only a few archaeological sites in the region contain visible architectural remains. When a site was found, its location was recorded on enlarged aerial photographs (approximate scale 1:10,000), and on topographic maps (scale 1: 25,000), survey forms were completed, and pho- tographs were taken. If the site had structural re- mains, they were mapped with the aid of Brunton pocket transits and 25-meter measuring tapes. Regional Coverage The survey was designed for 100% coverage of the study region. As in the work of Parsons and Hastings (1977, p. 11) in the Upper Mantaro re- gion, this survey of Paruro systematically covered the valley floors and lower valley slopes, the ridg- es and mountaintops and their upper adjacent slopes, and the low- to medium-gradient slopes of the region. The steep slopes, which were danger- ous to cross, were not systematically examined, except where access could be gained through trails. The exclusion of these areas from the field- work probably does not present a significant bias in the data collection, because the steepness of the slopes, which discouraged survey work, would have also limited prehistoric activities or occu- pations. The implementation of a systematic regional survey in the Andes is complicated by variations in surface visibility in different ecological zones. Territory in grain-producing zones generally pro- vides good to excellent conditions for surveys. The land is relatively free of ground-covering plants such as grass, and large areas are cultivated with scratch plows, which bring artifacts to the surface. Other ecological zones present very dif- ferent surface conditions that complicate locating and dating sites. For example, ridges and moun- taintops are rarely cultivated, and their surface ar- tifacts are constantly exposed to the weather (Par- sons & Hastings, 1977, p. 12). The eroded nature of surface ceramics at these locations frequently makes cultural identification of the sites difficult. Furthermore, much of the high puna area of the region is covered with thick wild grasses, leaving few ground areas visible. While researching in high ecological zones for evidence of prehistoric occupation and land use, surveyors frequently de- viated from their survey lines to examine stream- cuts and to inspect scattered patches of earth where the grass had eroded away and the soil was exposed. Another difficulty in achieving a true 100% 6 Chapter One B.-*'1'* /"^ "''X. I c 1. MM. 0 10 20 i i i Contour lines every 400 m Map 1-4. The research region. coverage of the study region was the presence of several relatively large towns and numerous small villages. As a result of modern construction and land use, only limited areas within the commu- nities could be surveyed. To support the limited coverage of these locations, interviews were held concerning archaeological finds in each commu- nity. In some of the towns, such as Accha, and in The Inca in Cuzco several of the small villages, such as Ccoipa, ar- chaeological sites were identified and surface col- lections were made, although the exact dimen- sions of these sites were difficult to estimate. Be- cause the areas covered by the modern commu- nities represent a small proportion of the total research region, and because limited surface sur- veys were conducted in the majority of them, their presence does not seriously affect the overall data base for the region. Ceramic Collections Ceramic collections were made at each site and then analyzed to determine the periods of site oc- cupation. During the collection process, a team of surveyors systematically walked over the surface of the site in parallel lines, approximately 5 me- ters apart. If the site contained architectural units or field boundaries, separate collections were made in each unit. The collections were then ex- amined at the site. Diagnostic shards were re- tained, while nondiagnostic fragments were left at the site. This preliminary sort was necessitated by the remoteness of the research area and the diffi- culties in transporting the collections by horse to the nearest road and then by trucks to Cuzco. Ex- ceptions were made at sites that contained a rel- atively small number of surface fragments. In these cases, all shards were retained. At the close of the project, all artifacts were deposited at the Instituto Nacional de Cultural in Cuzco for per- manent storage. Excavations During the course of the investigations in the Paruro region, test excavations were also con- ducted at 14 other sites, including Bandera Moqo (206), Aqachkar (233), Muyu Roqo (432), Roqo Cassa (450), Qachir (463), Sutic (467), Tejahuasi (496), Castilla Moqo (513), Cruz Moqo (531), Muyu Oreo (536), Llactaccata (575), Marcapata (624), Mollerada (655), and Ccorpina (659). The excavations consisted of 2 X 2-m or 2 X 4-m pits in fields or along ridge tops, and these were tested for subsurface remains (Map 1-5). These test ex- cavations, like those at Maukallaqta, used 10-cm levels and were terminated when sterile subsoil was reached. All artifacts recovered during the ex- cavations were transported to Cuzco for labora- tory analysis. Ceramic Analysis The classification scheme used in this study to characterize the different ceramic styles of the Cuzco region is a type-variety system. The de- velopment of the Paruro ceramic sequence began with the systematic collection of surface ceramics from sites in the region and their transportation to Cuzco for analysis. In the laboratory, the deco- rated shards were separated into homogeneous groups based on wares, design elements, pigment colors, and surface treatments. These groups were then further subdivided according to vessel forms. Later excavations at both single-component and multicomponent sites provided information on the relative dates of the styles. Carbon and bone sam- ples recovered from reliable cultural contexts dur- ing the excavations provided absolute dates for some styles. To provide complementary information to the regional survey data, a test excavation program was conducted in the Province of Paruro. In 1 986, several rooms were excavated at the site of Mau- kallaqta (l)2 to collect information concerning the room functions and to test for pre-Inca remains (Bauer, 1991, 1992a, 1992b). These excavations were conducted using arbitrary 10-cm levels until stratigraphy was identified and excavations con- tinued to the level of the sterile subsoil. A 1 X 1- m grid system was laid out in large structures, and a 50 X 50-cm grid system was used in the smaller rooms. 2 Numbers beside site names refer to site numbers as catalogued during the regional survey. Classic Inca Ceramics The best-known ceramic style from the Cuzco region is Classic Inca (also called Inca, Late Inca, and Cuzco Inca). Although this style is poorly dated, it is generally believed that its production began around a.d. 1400 and that Classic Inca ce- ramics continued to be produced until shortly af- ter the Spanish Conquest (a.d. 1532). Rowe (1946, pp. 246-247) suggests that the production of Classic Inca ceramics took place within the Cuzco Valley. The most likely locus of imperial ceramic production is between the modern com- munities of San Sebastian and San Jeronimo, in 8 Chapter One 0 10 20 i i i Contour lines every 400 m Map 1-5.,' Test excavated sites. The Inca in Cuzco the adjacent areas of Larapa and Sanyo (clay).1 Archival sources support this suggestion by not- ing that settlements of olleros (potters) lived in this area of the valley (Archivo General de la Na- cion, Archivo Agrario, Miscelanea: Hacienda La- rapa, 1596). Many of the world's major museums obtained substantial collections of Classic Inca ceramics during the 19th century (Seler, 1893; Valencia Ze- garra, 1979; Bauer & Stanish, 1990). The system- atic reporting and analysis of this ceramic style began, however, with the earliest archaeological expeditions into the south central highlands of Peru in the early 20th century. Perhaps the most famous collection was made during Hiram Bing- ham's work at Machu Picchu. Utilizing materials recovered from numerous cave burials surround- ing Machu Picchu and, to a lesser extent, finds recovered during his excavations, Bingham (1915) developed a vessel typology that is still widely used today. The physician on Bingham's expedition, George Eaton (1916, Plates V-XIV), also provided a large number of illustrations of the Machu Picchu material. There are also several early reports on Classic Inca ceramics found in the Cuzco Valley. For example, Luis E. Valcarcel (1934, 1935) supplies a discussion of the Classic Inca ceramics recovered at Sacsahuaman, and Luis A. Pardo (1938, 1939, 1957) outlines various museum pieces. In addition, Rowe's landmark work on the archaeology of Cuzco contains a de- tailed discussion of Classic Inca ceramics, includ- ing classifications of different substyles (Rowe, 1944, pp. 47-49). The post-World War II era has witnessed an increasing number of references to Classic Inca ceramics by researchers working in the Cuzco re- gion. This is not surprising, because Classic Inca ceramics are present at most sites in the region. Recent reports that incorporate significant descrip- tions and illustrations of Classic Inca ceramics in- clude Ann Kendall's (1976) studies in the lower Urubamba River Valley. Her work (Kendall, 1974, 1985, p. 347) also provides one of only three radiocarbon dates published for Classic Inca ceramics from Cuzco (see Appendix 2, p. 155). Excavations by Kendall at the site of Ancasmarca yielded Inca ceramics within a context dating to 482 ± 91 b.p. ([BM 930] a.d. 1468 ± 91). Sara Lunt's (1984, 1987, 1988) study of Inca and Killke ceramics from this same region offers the 3 This region continues to be exploited for its clay resources today. first detailed examination of late-prehistoric ce- ramic wares from the Cuzco region. Dean Ar- nold's 1972-1973 excavations at the site of Qata Casallacta (Liu, Riley, & Coleman, 1986, p. 108), on a mountain shelf near the city of Cuzco, yield- ed examples of Classic Inca ceramics as well as a radiocarbon date of 370 ± 80 b.p. ([ISGS 545] a.d. 1580 ± 80). A report by Alcina Franch and colleagues on excavations at Chinchero, a large Inca site north of Cuzco, contains many illustra- tions of Classic Inca ceramics (Alcina Franch et al., 1976). Heffernan (1989) furnishes numerous examples of this style found in the Limatambo area, located west of Cuzco. Unfortunately, a car- bon sample recovered with Classic Inca ceramics near Limatambo provided a radiocarbon date of 200 ± 80 b.p. ([ANU 5838] a.d. 1750 ± 80]), and two carbon samples (ANU 5839 and ANU 5840) from the fill of an Inca terrace in the same area provided equally unhelpful dates (Heffernan, 1989, p. 539). There are also several descriptions of Inca ceramics found outside the Inca heartland, ranging from Ecuador to Argentina (see Jijon y Caamano & Larrea 1918; Jijon y Caamano, 1934; Meyers, 1975; Baca, 1974, 1989; D'Altroy & Bishop, 1990; Calderari, 1991; Calderari & Wil- liams, 1991; D'Altroy 1992). Since the begin- nings of a broad-based literature on Classic Inca ceramics have been established and the basic ves- sel forms for this style have been identified, no further discussion of Classic Inca ceramics will be made in this report. This is not, however, to suggest that additional research on Inca ceramics is unnecessary. On the contrary, a large, system- atic study of Classic Inca ceramics from Cuzco is an important project that remains to be done and that would be an enormous asset to researchers working in all parts of the former Inca empire. Other Contemporary Styles We know that other types of ceramics were im- ported into the Cuzco region during the period of imperial Inca rule. Fine black Chimu ceramics have been found in excavations at Sacsahuaman (Valcarcel, 1946, p. 181), at Qotakalli (P. Lyon, pers. comm. 1992), and in the city of Cuzco itself (Carmen Farfan Delgado, pers. comm. 1994). Fur- thermore, Bingham recovered a non-Inca vessel during his excavations at Machu Picchu, which, based on its stirrup handle, appears to be imported from the north coast (Eaton, 1916, plate XIV, fig- 10 Chapter One ure 1). The recovery of north coast materials in Cuzco is not surprising, because the Inca domi- nated that area for several generations. Other areas of the Inca domain, especially the Lake Titicaca region, are also represented in sur- face collections from Paruro. Fragments of Sillus- tani (Tschopik, 1946), Urcusuyu (Rowe, 1944, p. 49; Tschopik, 1946), and Pacajes (Ryden, 1957; Albarracin- Jordan & Mathews, 1990) ceramics have been recovered at several sites. Sillustani ce- ramics are composed of a white paste that has been called "kaolin" and were fabricated in the Department of Puno in the Lake Titicaca basin. Urcosuyu vessels are believed to have been pro- duced in the same general region. Pacajes ceram- ics were made in the Tiwanaku region on the southern end of Lake Titicaca. There is no doubt that other imported styles from other parts of the Inca Empire will be recovered as additional re- search is conducted at sites in the Cuzco region. Although numerous research projects have been conducted in the Cuzco region over the past 40 years, its pre-Inca ceramic styles are still poor- ly understood. Because adequate definitions, de- scriptions, and illustrations of ceramic styles are essential tools in modeling and assessing prehis- toric cultural developments through time, it is im- portant that attempts be made to update the Cuzco ceramic sequence. This investigation into the pre-Inca ceramics of the Inca heartland is divided into five chapters. The following three chapters describe the differ- ent ceramic styles found in the Province of Paruro according to time periods. Chapter 2 examines the Late Intermediate period (a.d. 1000-1400) styles of Killke and Colcha. In Chapter 3, the styles of the Middle Horizon (a.d. 600-1000), including Wari, Wari-style, Arahuay, Qotakalli, Ccoipa, and Muyu Oreo, are presented. The Early Intermediate period (200 b.c.-a.d. 600) and the Early Horizon (1400-200 B.C.) styles of Derived Chanapata, Chanapata, and Marcavalle are described in Chap- ter 4. The period chronology used in this study is based on that developed by Rowe (1967). To make it of practical use for the Cuzco region, the dates of the periods, originally developed for the prehistoric occupations of the lea Valley, have been slightly altered. In Chapter 5 the ceramic sequence for the Prov- ince of Paruro and the ceramic findings from re- search projects conducted elsewhere in the Cuzco region are compared and integrated to produce a revised ceramic chronology. This revised chro- nology builds on Rowe's initial ceramic sequence and on the results of projects that have been con- ducted since then (Rowe, 1956). Research in the Province of Paruro supports the general temporal- ceramic classifications advanced by Rowe for the Cuzco region. The systematic recovery of a larger ceramic sample and the publication of other re- search findings since that time, however, permit the delineation of a more complex ceramic se- quence. This revised Cuzco sequence is, by ne- cessity, tentative, because there are still broad spans of time for which we have little or no in- formation. Critical examinations of the ceramic styles and ceramic sequence introduced here will result in a better defined ceramic classification for the Inca heartland and will enable archaeologists to better understand the development of complex societies in this important area of the Andes. The Inca in Cuzco 1 1 Ceramics of the Late Intermediate Period (a.d. 1000-1400) The development of the early Inca state in the Cuzco Valley is generally associated with the development of Killke ceramics. This ceramic style spans the time between a.d. 1000 and 1400, or what can be called the Late Intermediate peri- od. Despite the importance of Killke ceramics in the cultural history of the Cuzco region, few de- tailed descriptions or illustrations of them have been published. The surface collections and test excavations from the Province of Paruro provide a wealth of new information on this style. Re- search in this area has also documented the wide- spread presence of a new Killke-related style named Colcha. Killke Ceramics Among the earliest known scientific excava- tions conducted in the Department of Cuzco were those of Max Uhle (1912), at the site of Q'atan in the Urubamba River Valley. The work carried out by Uhle in the Cuzco region produced ceram- ics of a different style from that typically associ- ated with the Inca. Because little was known of the pre-Inca ceramics of the Department of Cuz- co, Uhle could only suggest that this new style dated to a pre-Inca but post-Tiwanaku period, and he proposed a time frame of broadly a.d. 800 to 1400. Soon after Uhle's discovery, the research of Jijon y Caamano and Larrea (1918) reproduced Uhle's findings. Later, Jijon y Caamano (1934) again reproduced some of Uhle's material and presented additional examples of similar ceramics that he had found in the Cuzco region and in mu- seum collections. Like Uhle, Jijon y Caamano suggested a broad pre-Inca, post-Tiwanaku time period for the production of this style. These iso- lated finds by Uhle and Jijon y Caamano were later classified as Killke-related ceramics and were dated to the immediate pre-Classic Inca pe- riod of the Cuzco region (Rowe, 1944, pp. 61- 62). In 1941, Rowe began his work in the Cuzco Valley with a series of test excavations. Although earlier research in the region had focused on ex- plorations and site descriptions, Rowe's work rep- resented the first systematic archaeological re- search project within the Department of Cuzco. One goal of his research was to identify and de- scribe the pre-Inca ceramic sequence for the Cuz- co region (Rowe, 1944, p. 61). He conducted test excavations in a courtyard of the monastery of Santo Domingo, near the Inca "Temple of the Sun" (Coricancha), and these excavations re- vealed an undisturbed deposit containing ceramic materials similar to those previously found by Uhle and Jijon y Caamano (Rowe, 1944, pp. 61- 62).' Surface collections made by Rowe later that year at a number of sites showed that this new ceramic style was widely distributed throughout the Cuzco Valley. 1 Additional excavations conducted by Barreda Mu- rillo (pers. comm. 1992; Lynch, 1973, p. 241) and by Valencia Zegarra (pers. comm. 1993) have confirmed the presence of Killke ceramics at the Coricancha. Ceramics of the Late Intermediate Period 1 3 Rowe performed test excavations in and around the city of Cuzco during 1942 and 1943 to further investigate the new style, which by then had been named Killke.2 The recovery of large quantities of Killke ceramics at the site of Sacsahuaman, just north of the city, was especially important in this research. Through the use of the Sacsahuaman materials, Rowe developed a broad stylistic ty- pology for what he called the "Killke Series" (Rowe, 1944, pp. 60-62). Although he did not find stratified Killke and Inca deposits, he in- ferred, on the basis of surface collections, that Killke ceramics were the antecedents to Inca ce- ramics in the Cuzco Valley (Rowe, 1944, p. 61). The identification of an Early Inca ceramic type in the Cuzco Valley had a profound impact on the study of the Inca. Classic Inca ceramics of the Cuzco region had long been recognized (Bing- ham, 1915; Eaton, 1916; Valcarcel Vizquerra, 1934, 1935; Pardo, 1938, 1939). The discovery of a precursor to this style provided a means to iden- tify sites occupied during the early development of the Inca state. Future surface collections and excavations at sites containing Killke ceramics would yield information on the social and eco- nomic conditions in which state development took place (Dwyer, 1971; Bauer, 1990, 1992a). Soon after Rowe's identification of the Killke style, Muelle led an expedition into the Province of Paruro. Near the Hacienda of Ayusbamba in the District of Pacariqtambo, Muelle identified three sites that contained ceramics similar to the Killke materials identified by Rowe in the Cuzco Valley (Muelle, 1945). Muelle's recovery of Killke materials outside the Cuzco Valley sug- gested that they were distributed throughout the entire region. Despite this discovery and the im- mediate academic acceptance of Killke as the Ear- ly Inca ceramic style in the Cuzco region, an ex- tensive study of Killke ceramics was not begun for another 20 years. From 1966 to 1968, Edward Dwyer conducted excavations at three sites in the Cuzco region: Mi- nas Pata in the Lucre Basin, Pucara Pantillijlla, near Pisac, and Sacsahuaman. The purpose of Dwyer's research was to further investigate the Killke Series as earlier defined by Rowe. Of the three sites selected for excavation, Sacsahuaman 2 The ceramic style found by Rowe in his excavations in the monastery of Santo Domingo was first named "Canchon" (Rowe, 1944, p. 46). After his 1942-1943 fieldwork, Rowe renamed the style "Killke" after a site located on the outskirts of the city (Rowe, 1944, pp. 60- 61). again provided the largest sample of Killke ce- ramics, and carbon extracted from a hearth in a Killke context yielded a radiocarbon age of 770 ± 140 b.p. ([Gak 2958] a.d. 1180 ± 140]). Dwyer (1971, p. 140) used this radiocarbon date to set the beginning of Killke ceramic production slight- ly earlier than Rowe's date, writing, "Killke cul- ture was probably dominant in the Cuzco Valley from around 1100 a.d. until the establishment of the Inca empire." Since Rowe's (1944) initial work and Dwyer's (1971) detailed study of Killke ceramics, other re- searchers have recorded the presence of Killke and Killke-related ceramics in the Cuzco region. Rivera Dorado (1971a, 1971b, 1972, 1973), for example, describes Killke and Killke-related ma- terials recovered in excavations at the sites of Cancha-Cancha and Chacomoqo in the Chinchero area. Heffernan (1989), working in the Limatam- bo area west of Cuzco, has found Killke and Killke-related materials at a large number of sites. Kendall (1974, 1976, 1985) and Lunt (1983, 1987, 1988) present examples of Killke and Killke-related ceramics recovered during excava- tions and surface collections in the Cusichaca Val- ley, in the lower drainage system of the Urubamba River. Kendall's excavations at the site of Ancas- marca provide a carbon sample that yielded a ra- diocarbon date for Killke materials of 660 ± 60 b.p. ([UCLA 1676M] a.d. 1290 ± 60). In addi- tion, Barreda Murillo (1973, pp. 70-71), Gibaja Oviedo (1983), and McEwan (1983, pp. 239-243; 1984, p. 215; 1987) have reported Killke ceramics in the Lucre Basin, southeast of Cuzco, as well as the presence of a poorly defined, Killke-related style named "Lucre." Gibaja Oviedo (pers. comm. 1992) also reports finding Killke ceramics at Pisac and Ollantaytambo, in the Urubamba Riv- er Valley, while Gonzales Corrales (1984a, 1984b) has described finding Killke materials in the city of Cuzco. Furthermore, Valencia Zegarra, Chavez Ballon, and Oberti Rodriguez have each identified Killke remains in the Cuzco region (pers. comm. 1987). From these various studies, it can be concluded that Killke and Killke-related ceramic styles are widely distributed across the Cuzco region; however, the nature of the distri- bution remains to be investigated. Description Bauer and Stanish (1990, pp. 1-2) have out- lined the basic ware, design elements, and surface 14 Chapter Two Fig. 2-1. Common Killke bowl designs. treatment criteria used to identify Killke ceramics. These definitional characteristics, built on descrip- tions provided by earlier researchers, including Rowe (1944), Dwyer (1971), Kendall (1976), and Lunt (1987, 1988), are presented below, along with new information on Killke vessel forms and design compositions. The definitional character- istics of Killke ceramics are then compared with those of Colcha, a Killke-related style of the Prov- ince of Paruro. Ware — Killke pottery is composed of a medi- um-coarse fabric containing a moderate quantity of nonplastic inclusions varying in size from 0.01 to 0.25 mm. These inclusions vary in color from an ashy white to a dull, dark gray. The inclusions appear to be high in feldspar, although the pres- ence of quartz and chert, as well as of andesite, amphibole, arkose, and syenite has also been not- ed (Lunt, pers. comm. 1987). The clean clay ma- trix suggests that the clay was washed, and a bi- modal grain size distribution of the nonplastic in- clusions suggests that temper was added (Lunt, pers. comm. 1989). The ware is medium hard, and the surface of the vessel frequently fires to a buff or salmon-pink color. (For a detailed discussion of the ceramic wares of the Cuzco region, see Lunt [1987].) Design Elements, Color, and Surface Treatment The exterior and interior surfaces of Killke bowls and the exterior of other Killke vessels are covered with a slip of smoothed body clay. The surfaces are then burnished, producing a medium- gloss effect (Dwyer, 1971, p. 87). The decorations on Killke vessels are generally geometric in form and composition. Narrow lines, thicker bands, tri- angles, and diamonds are the most common ele- ments. Black is the most frequently used color, followed by red and, rarely, white. Dwyer (1971, p. 104) writes, "The white and black colors are uniformly consistent, and the red varies from deep purple to pink depending upon conditions of ap- plication and firing. There are never two shades of red on one vessel. These colors are all painted on unpigmented buff slip background. The only Ceramics of the Late Intermediate Period 15 exception to this rule is the occasional use of white as a background." Among the wide variety of design motifs used in Killke ceramics, the most frequent are broad red or occasionally black bands outlined by one to three narrow black lines. Other motifs include sets of nested triangles that often alternate in color from red to black, linked ovals with central dots, linked rectangles with solid interior ovals, large areas covered with black cross-hatching, cross- hatched diamonds, and pendant rows of solid or cross-hatched triangles (Figs. 2-1 and 2-2, Draw- ings 2-1 and 2-2; Dwyer, 1971; Rowe, 1944, Fig. 19, pp. 11-21; Kendall, 1976). Designs on Killke ceramics frequently display low to medium color-tone contrast. This appears to result from the use of watery pigments. In ad- dition, the edges of Killke designs are often blurred, a characteristic that may be caused by a slight absorption of the thin pigments by the sur- face of the vessel (Lunt, pers. comm. 1988). There is also what Rowe has called a certain "charac- teristic carelessness of execution" compared to the Classic Inca ceramics that most likely devel- oped out of the Killke style (Rowe, 1944, p. 49). This is most apparent in the broad and apparently quickly executed brush strokes of the designs, which often leave undulating bands, as well as in the frequent overlapping of adjacent designs. Vessel Forms Forms of Killke vessels have been presented by Dwyer (1971) and Bauer and Stanish (1990). Killke bowl forms include both straight- and curved-sided as well as incurving bowls. Three different types of medium-sized, single-handled jars have also been identified: jars with high-arch- ing handles, jars with faces portrayed on their necks (face neck jars), and jars with conical necks (Dwyer, 1971; Bauer & Stanish, 1990). Large Killke jars frequently have ovoid bodies, concave necks, and paired handles. In addition, straight- sided drinking vessels (also called "tumblers" [Dwyer, 1971, p. 100]) are found in Killke col- lections. To aid future research on Killke ceramics, I have provided a description of common design compositions found in the vessel form categories, as reflected in the Paruro collections. This is not to suggest, however, that these common design compositions appear exclusively on the described vessel forms or that the decoration of the de- scribed vessels is strictly limited to these designs. An examination of Killke ceramics in Rowe (1944), Dwyer (1971), Rivera Dorado (1971a, 1971b, 1972, 1973), Kendall (1976), Lunt (1984, 1987), as well as in Bauer and Stanish (1990) sug- gests that a wide range of design compositions exist within most vessel form categories. Straight- and Curved-Sided Bowls — The variation in Killke bowl forms is considerable. The most common bowls have straight or convex- curved sides that flare upward from a flat base. The diameter of these vessels is greatest at the rim. They range from having nearly flat sides and almost no depth, in which case they resemble plates, to having sides that extend sharply upward and that have a depth of 10 centimeters or more (see also Dwyer, 1971, pp. 88-93, 99). The exteriors of the bowls are rarely decorated. Rim decorations are, however, common. The most frequent rim decoration consists of a simple black band that covers the rim and its immediate bor- ders or a similar band containing a series of small pendant triangles. There are a large number and a wide range of interior bowl designs (Dwyer, 1971, Figs. 32-60, 226-228, 239-246). One of the most common in- terior design arrangements, especially among the straight-sided bowls in the Paruro collections, consists of four large pendant triangles that are evenly spaced around the interior rim. The bor- ders of the triangles are generally defined by thick lines, whereas the interior spaces of the triangles are filled with various geometric designs made with finer lines. Frequently, the large triangles will contain a closely drawn cross-hatched design (Fig. 2-3, Drawing 2-3; Rowe, 1944, Fig. 19, pp. 11, 16). Occasionally, circles are suspended from the apexes of these cross-hatched triangles (Draw- ing 2-3E; Dwyer, 1971, Fig. 60). In other exam- ples, a loose cross-hatch design, made with sets of three to four fine lines, fills the triangles (Draw- ing 2-4). The large triangle motifs on the bowls may also contain two to eight thin lines that in- tersect at the apex of the triangle. These triangles occasionally contain curved lines (Drawing 2-5B- C) or linked ovals (Drawing 2-6). A different design motif found on Killke bowls consists of two to three parallel red bands running diagonally across the interior of the vessel. The red bands are outlined with thin black lines. Curv- ing black lines are drawn either on the red bands or between them (Drawing 2-7). Designs may also run parallel to the rim of the vessel. One common design contains a red band 16 Chapter Two Fig. 2-2. Common Killke designs. outlined by thin black lines. Between the rim and the red band are wavy lines (Drawing 2-8). A sim- ilar motif includes sets of thin vertical lines run- ning between the vessel rim and the red band (Drawing 2-9).3 The center of these vessels fre- quently displays a row of llamas. The llamas can be so stylized that they simply appear as X's (Drawings 2-9 and 2-10). A related motif is de- scribed by Dwyer: "One common motif is to have 3 Other research collections show that the area be- tween the rim and the outlined red band can be filled with cross-hatching or nested triangles (Rivera Dorado, 1971b). the bowl divided by one or two parallel red bands which are outlined in black, and which pass from the rim down over the bottom and continue to the opposite rim. The areas on either side of these bands are filled with geometric designs formed by thin black lines usually filling outline triangles with cross-hatching, or various combinations of thin black lines and wider bands" (Dwyer, 1971, p. 89). These central bands, which are also com- mon motifs in Classic Inca ceramics and transi- tional Killke-Inca pieces (Drawing 2-10), are fre- quently extensions of the rim design (Drawing 2- 9E). Ceramics of the Late Intermediate Period 17 Fig. 2-3. Common Killke bowl designs. Incurving Bowls — This vessel form category consists of shallow bowls with vertical to incurv- ing lips. The incurving bowls in the Paruro col- lections are decorated with a limited range of de- sign motifs (Drawings 2-11 and 2-12). Their ex- teriors generally possess a thick black band on the rim and lip. One to two centimeters below the lip and running parallel to it is a second black band, normally of equal width. The area between the two bands is filled with tightly clustered, finely drawn geometric designs. These designs usually consist of triangles separated by rows of thin lines, although cross-hatched diamonds and wavy lines have been noted. The interior of the vessels are not generally decorated, except for the interior rim, which is frequently painted black (see also Rowe, 1944, Fig. 15; Dwyer, 1971, Figs. 61-64, 284-287). One specimen in the Paruro collections however, contains two black lines, with a red line between them, across the interior of the bowl (Drawing 2-1 ID). As observed by Dwyer (1971, p. 93), these bowl types are "consistently finer in terms of surface finish, wall thickness and vessel symmetry than other bowls. Decoration was ap- plied with more precision than usual on any other Killke vessels." Large Jars — The large jar forms for Killke ce- ramics are difficult to reconstruct from the frag- ments in the Paruro collections. It appears that most had a flat base, ovoid body, and concave neck. Paired handles were set either horizontally on the lower body of the vessel or vertically be- tween the vessel rim and shoulder. A thick, dark line or a series of pendant triangles is commonly found on the rim of the large jars. The exterior, upper necks of the vessels are frequently painted with simple red or black designs characteristic of other geometric motifs found on Killke ceramics (see Dwyer, 1971, Figs. 74-110). Single-Strap-Handle Jars — These vessels have globular bodies that gradually swell outward from a flat base. A single vertical handle runs from the rim of the vessel to its lower neck or shoulder (Lunt, 1987, p. 25). From observations of museum collections, it appears that a wide va- riety of subtypes exist in this general vessel form classification (Dwyer, 1971, Figs. 288-292; Bauer & Stanish, 1990). Surface collections and exca- 1 8 Chapter Two Fig. 2-4. Single-strap-handle face neck jar. vations of Killke ceramics from the Province of Paruro support these observations. Two major subtypes are described here: face neck jars and jars with conical necks. Face Neck Jars — One of the most distinctive single-handled jar subtypes in Killke ceramics is the face neck jar (Dwyer, 1971, Figs. 291, 292). Numerous examples were found in the Paruro col- lections. These jars have a flaring rim with a rounded lip and a straight to slightly convex neck. A face is portrayed on the neck of the jar, opposite the strap handle. Occasionally the cheeks of the face are decorated with sets of nested triangles or linked diamonds. Above the face, on the vessel's rim and upper border section, is either a headdress or a cap (chullo); (Bauer & Stanish, 1990, pp. 8, 9, 12, 14). The bodies of the jars frequently con- tain broad bands of red outlined by black lines, which divide the exterior surface into panels. These panels are sometimes decorated with de- tailed geometric figures and other times left blank. The headdress is depicted by a series of wide, evenly spaced bands that run vertically down from the rim (Fig. 2-4, Drawings 2-13 through 2- 17). The bands are outlined on each side by two or three lines. The headdress terminates in a head- band motif, which may be portrayed by a row of nested triangles or solid diamonds. The chullo de- sign differs from the headdress in the depiction of ear flaps that extend down from the cap on both sides of the face and in the absence of a headband (Seler, 1893, Lam. 6, Fig. 6; Jijon y Caamano, 1934, Fig. 54; Bauer & Stanish, 1990, pp. 9, 12, 14). Jars with Conical Necks and Single Handles — This jar subtype is represented by jars with necks that are straight to slightly convex on the sides; the necks slope inward from the throat to the rim. The rims of the vessels flair outward, and the han- dle is attached to the jar just above the midsection line and to the rim. Only a few examples of this jar type were identified in the Paruro collections, although complete examples are known from mu- seum collections (for examples, Dwyer, 1971, pp. 97-98, Figs. 289, 290; Bauer & Stanish, 1990, pp. 9, 15). Straight-Sided Drinking Vessels — These ce- ramics are narrow, flat-bottomed drinking vessels Ceramics of the Late Intermediate Period 19 with straight sides and slightly flared lips. Frag- ments of straight-sided drinking vessels were rel- atively rare in the Paruro collections. Common designs on these vessels include horizontal bands and vertical rows of finely cross-hatched dia- monds. The rim and base of the vessel are fre- quently painted with a single black line (for ex- amples, see Dwyer, 1971, Figs. 1 18-120 and 1 lo- ll?). Dating Understanding the time period during which Killke ceramics were produced is critical in as- sessing prehistoric cultural developments in the Cuzco region. Rowe (1944, p. 61) has suggested that the transition from Killke (or Early Inca) to Classic Inca ceramic production occurred during a period of rapid state development. This transi- tion period has been equated with the rule of Pa- chacuti Inca Yupanqui, which is generally be- lieved to have begun in a.d. 1438. Employing the dynastic list of Inca kings provided by Cabello Balboa (1952 [1586]), Rowe (1944, p. 57) has also written that the mythical founding of Cuzco by Manco Capac took place around a.d. 1250. He then proposes this a.d. 1250 date as the beginning of the Killke ceramic tradition. The advent of radiocarbon dating now offers a method to date Killke ceramic production inde- pendent from information presented in the Span- ish chronicles. Currently, three radiocarbon sam- ples exist from reliable Killke contexts in the Cuz- co region: one sample from Dwyer's (1971) work at Sacsahuaman, another sample from Kendall's (1974, 1985) excavations at the site of Ancas- marca, and finally one sample from Bauer's (1990, 1992a) excavations at the site of Tejahuasi. The radiocarbon dates for these three carbon sam- ples fall between a.d. 1010 ± 140 and 1290 ± 60 (see Appendix 2, p. 155). Despite the small sample size and the large standard deviations of some samples, it appears that Killke ceramic production began earlier than the a.d. 1250 estimate currently suggested in the literature. After all, this a.d. 1250 date was first established in 1944 to mark the mythical founding of Cuzco by Manco Capac (see Bauer 1992a, pp. 36-48). It is reasonable to reassign the beginning of Killke ceramic production to approximately a.d. 1000. Furthermore, until additional carbon work can be conducted concerning the transition between Killke and Classic Inca ceramics, and un- til more dates are published from Killke and Clas- sic Inca contexts, a.d. 1400 may be selected as the date for the termination of Killke and the be- ginning of Classic Inca production in the Cuzco region. Colcha Ceramics While we were conducting archaeological sur- veys in the Province of Paruro, it became apparent that besides Killke ceramics a second style, close- ly related to Killke ceramics, was present in the region. As the District of Colcha was surveyed, the density of sites with this new style increased; thus, the style was named "Colcha." Description The new ceramic style of Colcha is defined as a Killke-related style because it shares many sty- listic similarities with Killke. Colcha ceramics can, however, be distinguished from Killke ce- ramics by ( 1 ) a much coarser paste than was used in the Killke ceramics, (2) a chalky, white slip, and, to a lesser extent, (3) the use of design motifs that have not been identified in Killke ceramics. The differences and similarities between Killke and Colcha ceramics are outlined below. Ware Killke ceramics have been described as con- taining a medium-coarse fabric with nonplastic in- clusions that vary in size from 0.01 to 0.25 mm. Colcha ceramics contain a sharply contrasting paste with a coarse fabric and a large quantity of nonplastic inclusions. Granitic inclusions are the most frequent in Colcha ceramics and range in size from 0.01 to 2.0 mm. The paste of Colcha ceramics is of moderate hardness and contains scattered single grains of black and gold mica, as well as free quartz, feldspar, and biotite. Many casts are altered, and tiny clay pellets are present (Lunt, pers. comm. 1989). The vessel surface fre- quently fires to a red or orange color. Design Elements, Color, and Surface Treatment Geometric design elements predominate in Col- cha ceramics. The most common elements are 20 Chapter Two wavy and straight lines, thicker bands, triangles, diamonds, and cross-hatching. These design ele- ments are similar to those found on Killke ceram- ics, but Colcha ceramics also exhibit the frequent use of dots, a design that is rarely seen on Killke ceramics. While Killke and Colcha ceramics share com- mon inventories of basic geometric design ele- ments, there are differences in the manner in which they are applied to the surface of the vessel. Both Rowe (1944, p. 60) and Dwyer (1971, p. 104) note that Killke designs are generally applied to unpigmented, frequently burnished surfaces. The exception to this rule is the occasional use of a white slip. In contrast to Killke ceramics, a dom- inant feature of Colcha ceramics is the presence of a white slip. Because of this slip, the surfaces of Colcha vessels, unlike those of Killke, are not burnished. The use of a white slip and thicker pigments in Colcha ceramics produces stronger tonal contrasts than are found in Killke ceramics. In addition, the slip (which lends a very chalky surface to Colcha ceramics) seals the porous sur- face of the vessels and prevents the absorption of the decorative designs and the edge blurring that are frequently observed in Killke ceramics. The two styles also vary in their selective use of colors. In Killke, for example, narrow black lines arranged in geometric designs and applied directly to a buff or pink surface are common. In Colcha, however, black is rarely used. Far more common in Colcha is the use of medium- to dark- brown lines painted on white slip. The Paruro samples also indicate that the broad red bands out- lined by narrow black lines, which are a hallmark of Killke ceramics, are not present in Colcha ce- ramics. Red, when it is used on Colcha vessels, simply appears in narrow, straight, or wavy lines. Vessel Forms No complete Colcha vessels were found during the work south of Cuzco. As a result, reconstruc- tion of vessel forms is a difficult and highly spec- ulative task. From the recovered Colcha shards it appears, however, that many of the Colcha vessel forms closely resemble those of Killke ceramics. Straight- and Curved-Sided Bowls — Frag- ments of straight- and curved-sided Colcha bowls are common in the Paruro collections. The rims of these vessels are generally decorated with a brown or occasionally a red line. The interiors exhibit a variety of motifs. Frequently, one or two curving lines running parallel to the rim are used (Drawing 2-18). Another common rim motif is a thick brown line located 1-2 cm below the paint- ed rim of the vessel and running parallel to it. The space between the rim and the line can be filled with cross-hatchings (Drawing 2-19A and C-F), pendant triangles (Drawing 2-19B), a wavy red line (Drawing 2-20), or nested triangles (Drawing 2-21 A and C). The space may also contain sets of thin lines running perpendicular to the rim (Drawings 2-2 ID and 2-22). Colcha bowls, like those of Killke, may also contain stylized llamas below these border decorations (Drawing 2-21). Figures 2-5 and 2-6 show some combinations of these elements. One of the designs most frequently observed on Colcha bowls is similar to, although generally less well executed than, a design found on Killke bowls. The design includes four large pendant tri- angles that are evenly spaced around the interior of the vessel. The pendant sides of the triangles are formed with relatively thick lines, and the in- terior of the triangle may contain a series of thin secondary lines running parallel to their sides and intersecting at the apex (Drawing 2-23). Wavy lines are occasionally painted within or between the pendant triangles. When the wavy lines occur in pairs, they are generally executed in red and brown (Drawings 2-24A-C and 2-25). The large triangle motifs of the bowls may be filled with cross-hatching (Drawing 2-24E-F). Another decoration frequently found on Colcha bowls, but one not widely found among the pub- lished examples of Killke bowls, consists of a painted rim followed by one or two wavy lines, which are often painted red and brown. Below the wavy lines is a thin, brown band running parallel to the rim. Below this thin band is a series of large, linked diamonds. The borders of the dia- monds are made with relatively thick lines, whereas their interiors are filled with thin cross- hatching. A second thin brown band, running par- allel to the rim, appears to have been drawn be- neath the linked diamonds (Drawing 2-26). A similar band and linked diamond design may be found on other bowls, but rather than running par- allel to the rim of the vessel, the bands and dia- monds cross from rim to rim through the center of the bowl. Dots, a design element rarely found in Killke ceramics, are a strong component of the Colcha design repertoire. A single row of dots or, more frequently, double lines of dots are found running parallel to the painted rim of certain bowls (Fig. Ceramics of the Late Intermediate Period 2 1 mup.i,npiMLi....,i ", y* Fig. 2-5. Common Colcha bowl designs. 2-7, Drawing 2-21 A, E-F). Double lines may also be seen descending toward the center of the bowl (Drawings 2-27B and 2-28), or they are used as space fillers (Drawing 2-27C-D). Incurving Bowls — No fragments of incurving bowls were identified in the Colcha collections. Large Jars — Several Colcha jar rims were re- covered during the course of research in the Prov- ince of Paruro. As on the Killke jars, the interior rims of the Colcha vessels frequently display a thick dark line, or a series of small pendant tri- angles. The exterior rims of the vessels are usually painted with a thick dark line, and the necks are covered with various geometric designs, including cross-hatched diamonds, triangles filled with al- ternating red and brown lines, ovals with dots in the center, and many other designs (Drawing 2- 29). Single-Strap-Handle Jars — No single-strap- handle jars occur in the Colcha collections. The lack of these jars may be the result of the small fragments recovered rather than of the absence of this vessel form in the Colcha ceramic inventory. Further investigation is necessary to clarify this point. Nevertheless, it is important to point out that no Colcha fragments are recorded that resem- ble the Killke face neck jars. Since this single- strap-handle subtype is readily identifiable in Killke collections, it appears likely that no parallel subtype exists within the Colcha assemblage. Straight-Sided Drinking Vessels — Several straight-sided drinking vessel fragments have been identified among the Colcha shards. The most common motif on these vessels is composed of horizontal bands of linked, cross-hatched dia- monds. Below and above these diamonds run thick brown lines (Drawing 2-30). The Relationship Between Killke and Colcha Ceramics The temporal and spatial relations between Killke and Colcha ceramics need to be examined to determine whether these two styles represent an evolutionary sequence or two ceramic styles produced during the same period. Either one of 22 Chapter Two Fig. 2-6. Common Colcha bowl designs. these possible relationships between Killke and Colcha ceramics can have important implications for archaeological research in the Cuzco region. If Colcha ceramics are found to predate Killke ceramics, new evidence might be presented for a developmental Killke style. The discovery of an antecedent to Killke ceramics south of Cuzco might, in turn, support literal readings of the chronicles, which suggest that ancestors of the Inca migrated to the Cuzco Valley from the Prov- ince of Paruro (Brundage, 1963 pp. 15-18). It is also possible that Colcha ceramics developed after Killke ceramics and that they represent a relative- ly late-prehistoric style in the Cuzco region, per- haps concurrent with Inca ceramic production. Alternatively, Killke and Colcha ceramics may not represent an evolutionary sequence but may instead be two contemporaneous styles. Killke ce- ramics have a long association with the Cuzco Valley, and the assumption is that they were pro- duced there. It is possible that Colcha ceramics represent a southern style that was made in the Province of Paruro contemporaneously with Killke ceramics. Identifying a second production center in the Cuzco region, but outside of the Cuz- co Valley, during Killke times would lend insight into the production and distribution of regional styles during the Late Intermediate period. It would also help to answer questions concerning the Incas' relationships with other regional groups during the period of state formation. Test Excavation Results During 1987, test excavations were conducted in the Province of Paruro to further develop the ceramic sequence for the Cuzco region. One par- ticularly important issue was the chronological re- lationship between Killke and Colcha ceramics. Test excavations were conducted at multicompo- nent sites in the areas of Yaurisque, Paruro, Pa- cariqtambo, and Colcha. Unfortunately, like other researchers working in the Cuzco region, I found that most of the sites we sampled were badly dis- turbed (Kendall, 1974). Inca construction activi- ties, centuries of plowing, and extensive erosion had reduced the contents of the upper soil levels Ceramics of the Late Intermediate Period 23 Fig. 2-7. Common Colcha bowl designs. to undifferentiated mixtures containing Inca, Killke, and Colcha ceramics. The disturbed nature of these and other sites in the Cuzco region con- tinues to make interpretations of the late-prehis- toric ceramic sequence difficult. The most conclusive evidence concerning the relationship between Killke and Colcha ceramics was found during test excavations at the site of Tejahuasi (492). This site is located on the prop- erty of the Paruro-based kin group (ayllu) Cucu- chiray and is located at an altitude of 3,200 masl on a long, narrow ridge near the town of Paruro. Although no structural remains are visible on the surface of the ridge, its slopes are covered with dense concentrations of ceramics. A series of test excavations, each measuring 2 X 4 m, was carried out along the western side of the site. The excavations proceeded first through a 20-cm-deep plow zone and then through a de- posit of compacted earth that was approximately 30 cm deep. The remains of a low-standing stone and clay wall were found in one of the test units at a depth of 56 cm. On the northern side of the wall foundation, a floor of packed earth, small cobbles, and flat stones was identified. A deposit of compact dark brown soil, 30 cm deep, was en- 24 Chapter Two countered beneath the floor and above the natural subsoil of the ridge. The deposit contained both Killke and Colcha ceramics. A carbon sample from this compact layer of dark brown soil pro- vided a radiocarbon age of 940 ± 140 b.p. ([B- 27494] a.d. 1010 ± 140), a date similar to that obtained by Dwyer (1971) during his excavations of Killke materials at the site of Sacsahuaman near Cuzco. The stratigraphic sequence at Tejahuasi pro- vides information for establishing provisional dates for Killke and Colcha production. The stone and clay wall and the adjacent floor represent the last occupational phase of the site, which dates to the Inca period. Most important to this analysis is the absence of Inca materials below the structure and the sealing of a pre-Inca deposit, one con- taining both Killke and Colcha ceramics, by the construction of a packed earth and stone floor. It should be noted, however, that these observations come from test excavations that, because of their limited coverage, do not always provide definitive results. In addition, the single radiocarbon sample from Tejahuasi has yielded a large standard de- viation. Nevertheless, the presence of both Killke and Colcha ceramics in a context beneath a floor indicates that these two related styles may have been both used and produced during the Late In- termediate period. Surface Survey Evidence During surveys in the Province of Paruro, 102 sites were found that contained Killke or Colcha ceramics. Of these sites, 56 contained Killke ce- ramics and 79 contained Colcha ceramics. Both Killke and Colcha ceramics were found in 33 of the 102 sites. Because a large number of Killke and Colcha sites were found during the survey work, statistical analyses can be conducted to in- vestigate the spatial relationship of these two styles. The preliminary excavation data from Teja- huasi indicate that Colcha and Killke ceramics may represent two contemporary styles. Given this provisional finding, we might ask whether these two styles are the products of a single center of ceramic production or of two separate centers. If the two styles were produced in two widely separate centers (e.g., one in the Cuzco Valley and another in the Province of Paruro), then it is prob- able that their distribution patterns south of Cuzco would not be identical. If, on the other hand, Killke and Colcha ceramics were produced by a single production center, the distribution patterns of the two styles might appear similar in the ar- chaeological record. To compare the distribution patterns of these two styles in the Province of Paruro, the distance of each site containing Colcha or Killke ceramics from Cuzco was calculated (Figs. 2-8 and 2-9).4 The Killke and Colcha site number — distance ob- servations were then mathematically compared in a Kolmogorov-Smirnov test to determine if the two styles present similar or different patterns in the Province of Paruro. The results of the test re- jected the null hypothesis (i.e., that there is no significant difference between the two samples) at a 0.05 level of confidence. In other words, the Kolmogorov-Smirnov test found a significant dif- ference between the distribution patterns of Killke and Colcha ceramics. This finding supports the suggestion that these two styles represent the products of two separate centers of production. The suggestion that Killke and Colcha ceramics were produced in two separate centers can be fur- ther explored by comparing the density of sites containing these styles in the survey region. A number of researchers have examined the distri- bution of various styles from their centers of dis- tribution or production, and various studies indi- cate that the spatial occurrence of most styles de- clines as a function of distance from their source (Hodder, 1974, 1980; Hodder & Orton, 1976; Renfew, 1975; Orton, 1980; Arnold, 1980; Rice, 1987, pp. 198-199). The mathematical relationship between the density of sites containing Killke and Colcha ce- ramics and their distance from Cuzco was exam- ined in a series of regression analyses. If the pro- duction of Killke or Colcha ceramics took place in the Cuzco Valley, then the density of sites with these ceramic styles should be greatest in the northern part of the survey region and should de- crease further south, and as the distance from Cuzco increases. The Killke ceramic analysis in- dicates that the Yaurisque region has the highest 4 Because many of the sites in the Province of Paruro are small and yielded limited collections of surface ce- ramics, I have selected to use a "presence-absence" measure of Killke and Colche ceramics in this study rather than to discuss the relative quantities of these ce- ramic styles (or specific attributes) present at each site. Thus, until additional research is conducted in the re- gion, to include a large-scale excavation program at a number of different sites and an expanded program of surface collections, these results should be considered as provisional. Ceramics of the Late Intermediate Period 25 12 (/) 3 15 19 23 27 31 35 39 43 47 51 55 59 km Di sta 1 ice f r< 1 5m C uzc 0 fj > o E EC 3 6- E ro u .j 3 >. E Q. 0. < < o Q. Fig. 2-8. Distance from Cuzco of sites with Killke ceramics. 15 19 23 27 31 35 39 43 47 51 55 59 km E ist£ ince f i om ( "uz 1 8 > o E (X CO u Q. » o E ro CL ro o ro CJ fj 3 o E ro 0. ro Q_ X < o O ro £ Fig. 2-9. Distance from Cuzco of sites with Colcha ceramics. 26 Chapter Two Fig. 2-10. Density of sites with Killke ceramics and distance from Cuzco. density of sites with Killke ceramics (Fig. 2-10). This particular section of the survey area is clos- est to Cuzco. The density of Killke sites drops steadily as the distance from Cuzco increases.5 Since the discovery of Killke ceramics near Cuz- co by Rowe in the early 1940s, it has been as- sumed that their ceramics were produced some- where in the Cuzco Valley. The inverse relation- ship between the density of sites having Killke ceramics and their distance from Cuzco implies that the Cuzco Valley was indeed the production area of Killke ceramics. Colcha ceramics, however, present a very dif- ferent distribution pattern from that of Killke ce- ramics. Although sites with Colcha ceramics can be found in the northernmost reaches of the re- search area, near the town of Yaurisque, as well as at the southern limits, near the town of Accha, the density of sites with Colcha ceramics is great- est near the community of Araypallpa. If the pro- duction of Colcha ceramics took place in or near the village of Araypallpa, the density of sites con- taining this style should decrease as a function of distance from this suggested manufacturing locus. To test this hypothesis, the distance of Colcha sites from the Araypallpa area was calculated and submitted to a regression analysis (Fig. 2-11). The results suggest that the density of sites having Colcha ceramics declines rapidly as the distance from Araypallpa increases.6 These findings sup- port the hypothesis that the Araypallpa area was the center of Colcha ceramic production during the Late Intermediate period. As discussed below, ethnographic and historical evidence also supports this conclusion. Ethnographic and Historical Data on Ceramic Production in the Province of Paruro Ethnographic data from the Province of Paruro indicate that the community of Araypallpa, locat- ed just south of the Apurimac River, is the only village in the survey area that has retained a tra- dition of ceramic production. Currently, the vil- lagers of Araypallpa are best known for their pro- duction of large, undecorated chicha (corn beer) fermentation jars that stand approximately 1 m high. These items, like other vessels manufactured 0.4. The coefficient of correlation (R) for this sample is 6 The coefficient of correlation (R) for this sample is 0.7. Ceramics of the Late Intermediate Period 27 Fig. 2-11. Density of sites with Colcha ceramics and distance from Araypallpa. in Araypallpa, are frequently called "Chillque."7 Chicha vessels of this kind, manufactured in Aray- pallpa, were observed in every community within the research region. Villagers in communities as far away as Mollebamba and Yaurisque describe these vessels as the preferred storage containers for chicha, and many informants remarked that their vessels were three or more generations old. In addition, older informants recalled arduous 2- to 3-day journeys that they undertook as youths to transport these large vessels from Araypallpa to their community. These ethnographic data in- dicate that the village of Araypallpa was a widely recognized regional center for ceramic manufac- ture until the middle of the 20th century and sup- port the suggestion that it may also have been an important production center in the more distant past. Further evidence that the villagers of Araypall- pa were important potters before and during the Spanish Conquest is preserved in colonial docu- 7 Poole, during her ethnographic studies in Paruro and Colcha, discusses another vessel produced in the village of Araypallpa that is still widely used in the Paruro re- gion during festivals (Poole, 1984, p. 293). She de- scribes this vessel as "A round, unpainted ceramic bowl with one very small handle," manufactured only in the community of Araypallpa, and used only during August and Carnival." When Poole (1984, p. 468) asked why the vessel was called Chillque (or Ch'ellqe), she was told, "because it was made in Araypallpa." ments. The name Chillque was applied to the group that occupied the region south of the Apu- rimac River during the late prehistoric and early historic periods. The center of this group was the community of Araypallpa (Poole, 1984, p. 468). The name Chillque is currently applied to the ves- sels that are manufactured in this same village. Thus, there appears to be a recognizable link be- tween the production of ceramics in the area south of the Apurimac River and the Chillque who in- habited that region. This link is made explicit by the indigenous chronicler Guaman Poma de Ayala (1980, p. 96 [1615:118]) in a description of the nature of various groups in the Cuzco region: In the law of the Inca they ordained to be King, "Capac Apu Inca". Inca does not mean king. But as Inca there are low status people like Chillque Inca potter; Acos Inca cheater; Uaroc Inca Llulla Uaroc liar, . . . [emphasis added]8 In sum, based on the preliminary results of ceram- ic distribution studies in the Province of Paruro, and in light of ethnographic and historical data that complement the archaeological findings, there is strong evidence to suggest that Araypallpa was 8 En la ley de los Yngas se ordenave para ser rrey, Capac Apo Ynga. Ynga no dezir rrey cino que ynga ay gente uaja como Chilque ynga ollero; Acos ynga en- bustero; Uaroc ynga Llulla Uaroc mentiroso . . . [Gua- man Poma de Ayala 1980:96 (1615 f. 118)]. 28 Chapter Two the center of Colcha ceramic production during the Late Intermediate period. Summary and Discussion Archaeological investigations in the Province of Paruro have documented the widespread pres- ence of Killke ceramics and a new, Killke-related style named Colcha. The two styles are stylisti- cally similar, sharing a common repertoire of geo- metric design elements and vessel forms. They differ, however, in paste composition and slip type, as well as in some designs, colors, and ap- plication techniques.9 Test excavations were conducted in the survey region to determine the chronological relationship between these two styles. Excavations at the site of Tejahuasi found both styles in a sealed context, with a radiocarbon date of 940 ± 140 b.p. The recovery of both styles within a single, undis- turbed context is conditional evidence that Killke and Colcha ceramics were used contemporane- ously. Sites containing Killke ceramics are concen- trated in the northern reaches of the study zone nearest Cuzco, and their numbers decrease further south. Killke ceramics have, however, been found near the community of Araypallpa, as well as near the southern limits of the survey zone, in the Ac- cha region. The presence of Colcha ceramics in surface collections decreases with distance from the village of Araypallpa. Examples of Colcha ce- ramics have nevertheless been identified in sur- face collections as far north as Yaurisque and in the southern reaches of the survey zone, in Ac- cha.10 From the archaeological survey data, it is possible to suggest that Killke ceramics were pro- duced in the Cuzco Valley, and it appears likely, based on archaeological, ethnographic, and his- torical data, that Colcha ceramics were manufac- tured in the region of Araypallpa. However, the social mechanisms through which the ceramics were traded or exchanged or the potters them- selves moved remain to be investigated once fur- ther archaeological and historical work has been completed in the region. 9 For additional information on Killke ceramics and regional ethnic groups, as well as on the emergence of the Inca state, see Bauer (1992a, pp. 89-94). 10 Luis Barreda Murillo (pers. comm. 1990) recovered two nearly complete Colcha vessels during excavations at the site of Wimpillay on the edge of the city of Cuzco. Killke Period Ceramic Styles and Regional Ethnic Groups Historical evidence suggests that the Cuzco Valley and the region immediately south of the Valley were inhabited by at least four separate ethnic groups. The Inca controlled the northern end of the Cuzco Valley. The Chillque occupied the region south of the Apurimac River near the present-day communities of Araypallpa and Col- cha and may have controlled parts of Paruro and Ccochirhuay. The Masca and Tambo were located in the areas of Yaurisque and Pacariqtambo, be- tween the Chillque to the south and the Incas to the north. The Masca were concentrated around Yaurisque and perhaps dominated areas as far south as Paruro and as far west as Huanoquite, whereas the Tambo appear to have been centered in and around the Pacariqtambo area (Bauer, 1992a). The Chillque, Masca, and Tambo were simply three of the many groups that surrounded the Inca capital, which collectively can be called Incas de Privilegio (Inca of Privilege). The identification of two contemporaneous, re- gionally produced, Killke period ceramic styles in the area directly south of Cuzco raises an impor- tant question: Can the exact boundaries of the re- gion's ethnic groups be identified through the dis- tribution of these pottery styles? If the territorial boundaries of the Inca, Chillque, Masca, and Tambo were important factors in the distribution of Killke and Colcha pottery, then the distribution curves of the Killke and Colcha pottery types could be expected to display a series of "plateaus and kinks" reflecting the trade of ceramic vessels in and across ethnic territories (Hodder, 1980, p. 152). On the other hand, if the various ethnic boundaries of the region did not affect the distri- bution of Killke or Colcha ceramic materials, then the frequency of sites containing these ceramic types might simply decline as the distance from their sources of production increases. An analysis of preliminary archaeological survey data from the Province of Paruro indicates that the frequen- cies of sites containing Killke and Colcha pottery styles display relatively smooth and uninterrupted fall-off curves from their suggested centers of production. These findings suggest that both the Inca of the Cuzco Valley and the Chillque of the Araypallpa area primarily, although certainly not exclusively, used pottery produced within their own territory. The archaeological survey has re- corded both Killke and Colcha ceramic styles dis- tributed across the territories of the Masca and Ceramics of the Late Intermediate Period 29 Tambo, and to a lesser extent, these two ceramic styles appear to have entered into each other's re- gion of production. The recovery of Killke and Colcha pottery in the territories of all four ethnic groups living in Cuzco and immediately to the south implies that the boundaries of these groups were not critical features in the distribution of Killke or Colcha pottery, and that archaeological- ly, ceramic styles in the Cuzco region do not ap- pear to reflect elements from which ethnic identity might be inferred. Killke Period Ceramic Styles and the Emergence of the Inca State Although the distribution of Killke and Killke- related pottery styles south of Cuzco cannot be used to identify the boundaries of separate groups of Incas de Privilegio, such as the Chillque, Mas- ca, and Tambo, their distribution patterns and methods of manufacture may provide important insights into the chronology and mode of state emergence in the Cuzco region. The use of nearly identical design elements and motifs in Killke and Killke-related pottery styles, such as Colcha, sug- gests that strong social contacts existed between the various centers of pottery production. In ad- dition, the overlapping distribution networks of Colcha and Killke pottery imply that a high level of trade and exchange existed between various In- cas de Privilegio groups during the Killke period. From these overlapping ceramic distribution pat- terns and their associated lines of communication, it may be tentatively proposed that the Killke pe- riod in the Cuzco region was typified by regional accordance and exchange. Although Killke and Colcha pottery appear to be stylistically very similar, and although they share overlapping distribution networks, the dis- tribution ranges of the two pottery styles from their separate centers of production are markedly different. The distribution radius of Colcha pot- tery appears to be approximately 25 aerial kilo- meters from its area of production (Fig. 2-9). The distribution radius of Colcha pottery is very lim- ited when compared to that of Killke pottery, which extends more than 60 aerial kilometers from the Cuzco Valley (Fig. 2-8). The greater dis- tribution range of Killke pottery than of Colcha pottery may be of some importance. From these observations, it is possible to suggest that regional exchange relationships had already begun to de- velop a Cuzco-centric focus during the Killke pe- riod and that the Cuzco Valley may have been emerging as a regional center of production dur- ing this same period. Further evidence indicative of regional central- ized authority developing in the Cuzco Valley during the Killke period may be observed in tech- nological attributes of Killke pottery production. For example, Killke pottery is slightly more so- phisticated in decoration complexity and in wear manufacture than other Cuzco regional styles. In addition, the widespread distribution of Killke pottery suggests that it was produced in far great- er quantities than any other Killke-related style. The emergence of what appears to be large-scale craft production in the Cuzco Valley may reflect the development of full-time specialists in ceram- ic production during the Killke period. Although superior quality and greater quantities of craft production do not necessarily indicate that a high level of sociopolitical organization existed in the Cuzco Valley, when this evidence is combined with the distribution information of Killke pot- tery, it suggests that the Cuzco Valley was a re- gional center for exchange, and was perhaps the central authority, by the Killke period. The ceramic distribution patterns south of Cuz- co may show further evidence of regional ex- change relationships during the Killke period, in- dicative of the unification of the region under a single Cuzco-based political authority. Systematic regional surveys conducted in Accha, 52 aerial kilometers from Cuzco, revealed a light presence of Killke pottery. Four days of survey work in the more southern region of Omacha, approximately 70 aerial kilometers from Cuzco, produced no ev- idence of Killke pottery, suggesting that the area between Accha and Omacha forms the frontier for Killke pottery distribution to the south of Cuzco. The chronicles suggest that this same region also represented the outer limits of the Incas de Pri- vilegio. The apparent correlation between the dis- tribution of Killke pottery and the distribution of groups absorbed into the Inca state as Incas de Privilegio indicates that Cuzco-centric distribu- tion networks may have united Cuzco with sur- rounding ethnic groups during the period of state development. If Cuzco emerged during the Killke period as a dominant power in the region, with a level of social and political organization unsurpassed by other regional ethnic groups, it is possible that its elevated status might be reflected in various pot- tery designs produced by the Incas. There is pro- visional evidence to suggest that this may have 30 Chapter Two occurred. In the Killke ceramic tradition, a highly stylized figure is frequently depicted on the sin- gle-handled jars (Drawing 2-13). This figure can be identified through a set of standardized motifs, including (1) a headdress or cap (chullo) depicted by a series of wide, evenly spaced bands outlined on each side by two to three narrow lines located on the vessel's rim and upper border section, (2) a headband motif, which is most frequently de- picted as a row of nested triangles, and (3) a series of linked diamonds or triangles on the figure's cheeks. The human qualities of the figure stand in contrast to the pantheon of mythoreligious figures represented in earlier ceramic traditions, such as Wari, which have been found in the Cuzco region. The standardized headdress, the elaborate head- band, and the painted cheeks are suggestive of emblems of power and authority. The appearance of this stylized figure during the Killke period, and the conspicuous absence of other figures on Killke pottery vessels, suggest that it may be a symbolic representation of an institutionalized rul- er or elite class. The appearance of this figure on pottery thought to have been manufactured in the Cuzco Valley hints at the possibility that the in- stitution of the "Inca" as a paramount ruler or a dominant social class had begun to coalesce by the Killke period. Although the very suggestion that certain pottery designs may in some way re- flect the sociopolitical order that produced them is extremely speculative, the appearance of this kinglike figure on Killke pottery during the period of early Inca state development does concur with other preliminary conclusions drawn from the ce- ramic distribution data. Under the traditional model of state formation in the Cuzco region, presented in the Spanish chronicles and accepted by many Andean ethno- historians and archaeologists, the Killke period was a time of fierce regional conflict: a volatile world of raids, competing polities, and political fragmentation. The traditional model suggests that political and economic unity was achieved in the Cuzco region only after Pachacuti Inca Yupan- qui's victory in the mythohistorical Chanca war. Archaeological data presented in this chapter con- cerning the production of pottery in the Province of Paruro during the Killke period does not, how- ever, conform to our expectations of a fragmen- tary and competing social landscape. Instead of being a time of regional conflict, the Killke period may have been characterized by widespread re- gional exchange. Rather than competing with oth- er ethnic groups, the Incas, may have already dominated the local social and political organi- zations of the region during the Killke period. In addition, the Inca may have unified the Cuzco area, and Cuzco may have already become the center of economic, and perhaps political, influ- ence during this same period. In other words, it is possible that unification of the Cuzco region did not simply occur during the Killke period but came about specifically under Inca dominance. In addition, it is possible that during this period the hierarchical relationships that the rulers of Cuzco would hold over the members of other regional ethnic groups at the time of the Spanish Conquest were in formation or were already present. Ceramics of the Late Intermediate Period 31 012345 CM ^BLACK OR BROWN HIRED Drawing 2-1. Common Killke designs. 32 Chapter Two 9— LL345CM (black or brown s«ed Drawing 2-2. Common Killke designs. Ceramics of the Late Intermediate Period 33 B / 0 12 3 4 5CM Iblack OR BROWN i|RED Drawing 2-3. Common Killke bowl designs. 34 Chapter Two 9— L2345CM IBLACK OR BROWN Drawing 2-4. Common Killke bowl designs. Ceramics of the Late Intermediate Period 35 0 12 3 4 5CM Jblack or brown IHred Drawing 2-5. Common Killke bowl designs. 36 Chapter Two B 0 12 3 4 5CM |black or brown Ireo Drawing 2-6. Common Killke bowl designs. Ceramics of the Late Intermediate Period 37 B 0 12345CM Jblack or brown Hred Drawing 2-7. Common Killke bowl designs. 38 Chapter Two 0 12 3 4 5CM |black or brown Ired Drawing 2-8. Common Killke bowl designs. Ceramics of the Late Intermediate Period 39 0 12 3 4 5CM |black OR BROWN Ired Drawing 2-9. Common Killke bowl designs. 40 Chapter Two 012345CM Jblack or brown Hred Drawing 2-10. Transitional Killke-Inca bowl. Ceramics of the Late Intermediate Period 41 B i I I 0 1 2 3 4 5CM ^BLACK OR BROWN ■red Drawing 2-11. Common incurving Killke bowl designs. 42 Chapter Two 1 B 1 \ 0 12 3 4 5CM ^black or brown Ired Drawing 2-12. Common incurving Killke bowl designs. Ceramics of the Late Intermediate Period 43 0 12 3 4 5CM |black or brown Ired Drawing 2-13. Single-strap-handle face neck jar. 0 1 4 5CM |black or brown Ired Drawing 2-14. Single-strap-handle face neck jar. 44 Chapter Two I B 0 12 3 4 5CM |black OR BROWN Ired Drawing 2-15. Single-strap-handle face neck jars. Ceramics of the Late Intermediate Period 45 012345 CM (black or brown Bred Drawing 2-16. Single-strap-handle face neck jars. 46 Chapter Two B 0 12 3 4 5CM (BLACK OR BROWN ilRED |ORANGE Inatural Drawing 2-17. Single-strap-handle face neck jars. Ceramics of the Late Intermediate Period 47 I / 0 12 3 4 5CM (black or brown ■red Drawing 2-18. Common Colcha bowl designs. 48 Chapter Two B 012345 CM ^|BLACK OR BROWN IHred Drawing 2-19. Common Colcha bowl designs. Ceramics of the Late Intermediate Period 49 7 B 0 12 3 4 5CM V J|black or brown ■red Drawing 2-20. Common Colcha bowl designs. 50 Chapter Two B 0 12 3 4 5CM Iblack OR BROWN Drawing 2-21. Common Colcha bowl designs. Ceramics of the Late Intermediate Period 5 1 7 B 7 7 7 0 12 3 4 5CM (black or brown Hred Drawing 2-22. Common Colcha bowl designs. 52 Chapter Two B 012345 CM Kblack OR BROWN Hred Drawing 2-23. Common Colcha bowl designs. Ceramics of the Late Intermediate Period 53 B 7 / / / 0 12 3 4 5CM Iblack OR BROWN Hred Drawing 2-24. Common Colcha bowl designs. 54 Chapter Two 0 12 3 4 5CM Iblack OR BROWN S*ed Drawing 2-25. Common Colcha bowl designs. Ceramics of the Late Intermediate Period 55 B 0 12 3 4 5CM |black or brown Ireo Drawing 2-26. Common Colcha bowl designs. 56 Chapter Two 01234 5CM ^BLACK OR BROWN Hred Drawing 2-27. Common Colcha bowl designs. Ceramics of the Late Intermediate Period 57 B 0 12 3 4 5CM |black or brown !red Drawing 2-28. Common Colcha bowl designs. 58 Chapter Two B 0 12 3 4 5CM / U If |black or brown Ired Drawing 2-29. Common Colcha jar designs. Ceramics of the Late Intermediate Period 59 J 0 12 3 4 5CM Iblack OR BROWN RED Drawing 2-30. Colcha straight-sided drinking vessels. 60 Chapter Two Ceramics of the Middle Horizon (a.d. 550-1000) The Middle Horizon encompasses a broad span of time during which much of the central and south central Andean highlands came under the control of two great states, Wari and Tiwa- naku. Current research suggests that the Wari be- gan to expand from their traditional power base in the Ayacucho region of Peru around a.d. 550. Excavations at Wari sites outside the Ayacucho region indicate that state expansion continued through at least a.d. 850, after which the state appears to have collapsed. Less is known con- cerning the development of Tiwanaku; however, it seems that by a.d. 300 the city of Tiwanaku, near the southwestern shore of Lake Titicaca in Bolivia, was of considerable importance. Expan- sion of Tiwanaku may have begun around a.d. 500 and waned, like Wari, near the end of the first millennium (Kolata, 1993, pp. 85-86). Researchers have long noted the presence of Wari-style materials in the Cuzco region, and it is generally recognized that the Wari controlled the Cuzco Valley for several centuries (McEwan, 1987, 1989, 1991). Furthermore, when Rowe es- tablished the first Cuzco ceramic sequence in 1956, he realized that local cultures in the region must have produced ceramics during the Wari oc- cupation and that some of these styles would have been influenced by Wari ceramic traditions. One of these local, Wari-related styles was identified and called "Lucre" by Chavez Ballon (Rowe, 1956, p. 142; McEwan, 1989, p. 55). This style remains poorly understood, and there are only a few published examples of it (McEwan, 1984, pp. 13; 15, 1987, pp. 58, 102-104).1 Ceramics from the Province of Paruro have helped to define other styles that were manufac- tured in the Cuzco region during the Middle Ho- rizon, including Arahuay, Qotakalli, and perhaps Ccoipa. Furthermore, ceramics from the Province of Paruro indicate that some degree of Tiwanaku influence was felt in the Cuzco region during this period. Wari and Wari-Style Ceramics Wari influence in a region can be inferred from the presence of Wari, Wari-style, and Wari-related ceramics in the archaeological record. Definitions and explanations of these three terms are neces- sary before an analysis of the Paruro materials can begin. For the purposes of this study, "Wari ce- ramics" are characterized as ceramics that were actually produced in the Ayacucho area and were later imported into the Cuzco region. These in- clude the Ayacucho styles of Chakipampa, Ocros, Vinaque, and Rubles Moco (Knobloch, 1991; Glowacki, 1996). "Wari-style ceramics" are de- 1 It should be noted that both Chavez Ballon and Bar- reda Murillo have named different ceramic styles of the Cuzco region "Lucre." Chavez Ballon "s "Lucre," as re- ported by Rowe (1956, p. 142), is a Wari-related style, where Barreda Murillo's (1973, 1991) "Lucre" is a Killke-related style. Ceramics of the Middle Horizon 61 fined as styles of ceramics that imitate the style of one of the above-mentioned Wari ceramics but were not produced in the Ayacucho region. An example is the Ocros-style ceramics of Pikillacta (Knobloch, 1991, pp. 253-254), which imitate the Ocros ceramics of the Wari homeland but were presumably produced in the Cuzco region. Final- ly, "Wari-related ceramics" are ceramic styles that were influenced by Wari ceramics but do not directly imitate them. The clearest case in the Cuzco region is that of Arahuay ceramics, a Cuz- co ceramic style that reflects features of Wari ce- ramics and Wari-style ceramics. Wari influence in the Cuzco region was first suggested by Theodore D. McCown and Rowe in the early 1940s (Rowe, 1944, p. 53). It was con- firmed by Chavez Ballon and Rowe in the early 1950s during their examination of excavation and surface collections from a number of archaeolog- ical sites east of the city of Cuzco (Rowe, 1956, p. 142). The most important information came from the site of Batan Oreo, a small knoll that juts out into the Vilcanota River Valley near the town of Huaro (Reichlen, 1954; Rowe, 1956, p. 142). The discovery by looters of an elite tomb at Batan Oreo in 1952 brought the site to the at- tention of Cuzco officials as well as of the general public (Comercio, 1952a-k; Reichlen, 1954). Chavez Ballon (pers. coram. 1990) dug at the site that same year, but his collections were destroyed before he could finish his analysis. Additional ex- cavations were conducted at Batan Oreo by Bar- reda Murillo in 1952 (Barreda Murillo, 1973), and surface collections were made there by Patterson and Rowe in the 1960s (Patterson, 1967). Initially the finds at Batan Oreo were classified as Tiwa- naku related (Reichlen, 1954). Further examina- tion of the materials by Chavez Ballon and Rowe suggested, however, a closer relationship to the Ayacucho state, Wari, than to Tiwanaku. This conclusion is supported by recent work at Batan Oreo by Zapata (pers. coram. 1992, 1997), which revealed an elite cemetery with numerous Wari- related vessels. The center of Wari influence for the Cuzco re- gion was the site of Pikillacta, located in the Lu- cre Basin approximately 30 km east of the city of Cuzco. This well-preserved site was visited by Cieza de Leon (1976, p. 261 [1551, Pt. I, Ch. 97]) soon after the European invasion, and cen- turies later by Squier (1877, pp. 419-422) who commented on its large size and apparent antiq- uity.2 Among Pikillacta's many notable features are walls that once stood 5 m high and a central zone of structures approximately 500,000 m2 in area. The site is constructed on an impressive grid system containing more than 700 individual buildings (McEwan, 1987, p. 24, 1991, p. 93). Looting at the site in the early 1920s produced two caches of turquoise figurines, the subject of a recent study by Anita Cook (1992). Several indi- viduals have conducted research at the site, in- cluding Harth-Terre (1959), Sanders (1973), and Barreda Murillo (1973, 1991). Recent work at Pi- killacta by McEwan (1983, 1984, 1987, 1991) has produced a site map. According to McEwan, Pi- killacta most likely functioned as the southern provincial administrative center for the Wari em- pire and was supported by satellite communities in the Lucre Basin. Excavations by McEwan in Pikillacta and in trash middens outside the mas- sive city walls indicate that its occupants used several different Wari, Wari-style, and Wari-relat- ed ceramics (Knobloch, 1991, p. 253; Glowacki, 1996) as well as a few exotic vessels imported from Nazca and Cajamarca (McEwan, 1990; Glo- wacki, 1996). Most of the decorated vessels of the site can, however, be classified as Ocros style, which is characterized by a light orange slip (Knobloch, 1991, p. 253; Glowacki, 1996). A detailed analysis of the ceramics from Piki- llacta has just been completed by Mary Glowacki (1996). It provides a ceramic baseline with which the Wari-style and Wari-related ceramics from other sites in the Cuzco region can be compared. Within her study, Glowacki also discusses many of the styles that are presented in this study and presents her views on their evolutionary sequence. Pikillacta is not the only site in the Cuzco re- gion to yield Wari and Wari-style materials. The site of Chokepuquio, located some 2 km west of Pikillacta, contains a series of massive structures and Wari-style ceramics (Gibaja Oviedo, 1973, 1983; McEwan, 1983, 1984, 1987). McEwan (1984, 1987) has also found a number of small sites in the Lucre Basin that contain Wari and Wari-style ceramics. Other researchers have re- ported finding Wari-style materials elsewhere in the Cuzco region. These include the sites of Batan Oreo, 12 km southeast of Pikillacta, and Ccoto- 2 Both Cieza de Leon and Squier called Pikillacta "Muyna." Squier (1877, pp. 419-422) referred to the Inca gateway of the Cuzco Valley, currently known as "Rumi Colcha," as "Piquillacta." Cieza de Leon (1976, p. 261 [1551, Pt. I, Ch. 97]) mentions Rumi Colcha, but does not give it a name. 62 Chapter Three cotuyoc, near Urcos. There are other sites con- taining Wari-style materials in the nearby Anda- huaylillas and Huaro Valleys (Zapata, pers. comm. 1992, 1997). Moreover, Wari-style ceram- ics have been found near the city of Cuzco at Coripata (Zapata, pers. comm. 1991), Arahuay (Torres Poblete, 1989), Aqomoqo, Qotakalli, and Wimpillay (Espinoza Martinez, 1983), west of Cuzco in the Plain of Anta,3 and to the south in the Province of Paruro (Bauer, 1992a; p. 155 n. 16). Wari and Wari-style materials have also been reported elsewhere in the Department of Cuzco. Rowe and his colleagues found Wari-style ceram- ics at several sites near Sicuani (Rowe, 1956, pp. 142-144). Sergio Chavez (1985, 1987) reported on an elaborate collection of 141 Wari-style metal objects from the Pomacanchi area. Within this collection were also two ceramic vessels, one of which was classified by S. Chavez (1987, p. 8) as Chakipampa B; the other appears to be a local style. Excavations conducted by Wilbert San Ro- man Luna (1979, 1983) at the site of K'ullupata in Pomacanchi have yielded various Wari-style ar- tifacts. Other isolated Wari-style finds have been recovered still further to the southeast of Cuzco in the Province of Chumbivilcas by S. Chavez (1987, 1988) and Lantaron Pfoccori (1988) and in the Province of Espinar by Meddens (1989). Despite the growing awareness of the impor- tance of the Wari influence in the Cuzco region, few examples of Wari or Wari-style artifacts found in this area have been published. Among those currently available are samples from Mc- Ewan's (1983, 1984, 1987, 1991) and Barreda Murillo's (1973, 1991) excavations at Pikillacta, Espinoza Martinez's (1983) material from Aqo- moqo, Torres Poblete's (1989) work at Arahuay, as well as Chavez's (1987, 1988) and San Roman Luna's (1979, 1983) work in Pomacanchi. It should be noted, however, that S. Chavez's (1987, p. 17) study of a large collection of Wari-style metal objects recovered in the Pomacanchi area suggests that these objects may also reflect some Tiwanaku influence. Furthermore, S. Chavez in- dicates that he has found a provincial Wari-style beaker from Chumbivilcas that also exhibits some Tiwanaku features. These findings emphasize that there were most certainly people living in transi- tional zones, such as the high altiplano regions south of Cuzco, who were not directly controlled by either Wari or Tiwanaku but were influenced by both of these emerging states. Ocros-Style Ceramics of the Province of Paruro Research in the Province of Paruro recorded examples of various Wari-style ceramics at nine sites. At all but one of these sites, the Wari-style materials represented an extremely small percent- age of the ceramic inventories (generally between one and four Wari-style shards per site). The ex- ception was the site of Muyu Roqo (432), located on a mountain slope immediately west of the town of Paruro. Surface collections from this small site, which measures approximately 50 X 50 m, pro- vided a large number of Ocros-style fragments. Test excavations at the site yielded hundreds more, as well as nearly 2,000 camelid bone frag- ments. Because this site is exceptional, its ceramic remains are discussed here.4 Ware The ware of the Ocros-style ceramics recovered at Muyu Roqo is made up of a fine fabric that includes a moderate quantity of minute inclusions ranging in color from white to dark gray. The ware varies from a light orange to a dark brick red, or sometimes is a dull black as a result of firing in an oxygen-reducing atmosphere. Vessel Forms, Design Elements, Color, and Surface Treatment The interior and exterior of most of the Ocros- style materials at Muyu Roqo are covered with a thin slip of smooth body clay. The interior and exterior of the bowls are carefully burnished. The exterior and approximately the upper third of the interior of the drinking vessels are also burnished. The interior, and occasionally the exterior, of the bowls and the exterior and the upper lip portions of the painted drinking vessels are colored red (Munsell 2.5YR 5/6), or what can also be called light orange. The designs drawn over this red 3 In 1992, the Casa de Garcilaso had several fragments of Wari ceramics collected from the Plain of Anta on display. 4 For a comparison of Muyu Roqo ceramics and those found at Pikillacta; see Glowacki (1996). Ceramics of the Middle Horizon 63 (light orange) background were placed on the ves- sels with considerable care. As will be discussed in greater detail below, the painted Ocros-style pieces of Paruro exhibit a lim- ited number of design elements and motifs, and the designs and vessel shapes in the collection are uniform. The most frequent design elements on drinking vessels are horizontal bands, which may be either painted with one solid color or contain a latticework that is normally white. Bowls, on the other hand, are frequently decorated with rect- angle pendants in a variety of colors, looping de- signs, or dependent cross-hatched triangles. A wide range of colors is used on the Ocros- style materials, including cream (or white), yellow (or orange), gray, and burgundy-maroon. Black is also used, but only in thin lines that either outline bands and pendants or pass through the center of the pendant rectangles in the bowls. Beside painted fragments, surface collections and excavations at Muyu Roqo provided a large number of black, incised drinking vessel frag- ments decorated with horizontal bands and incised circles. Similarly incised vessels have not been reported in large quantities at other Cuzco area sites yielding Wari, Wari-style, or Wari-related materials. Test Excavation Results The surface materials collected at Muyu Roqo were markedly different from those recovered at other archaeological sites in the survey region. Rather than providing just a few or, more com- monly, no Wari-style artifacts, the slopes of this small mountain shelf yielded an unprecedented number of relatively high-quality Ocros-style ma- terials. Test excavations were dug there during 1987 to better understand the nature of this unusual site. These excavations provided no evidence of stratified remains, although they did yield a large number of ceramic fragments (more than 1,000) and large mammal (most likely camelid) remains (more than 1,900).5 Although some isolated Chaki- pampa-style pieces were identified in the excava- tion collection, the vast majority of the fragments were from Ocros-style vessels. The vessel shapes identified in the Muyu Roqo inventory are limited. A small number of incurv- ing bowls, made of either orange or black ware 5 Bone fragments over 2 cm in length are included in this count. (Drawing 3-1), were found, as were the remains of some earthenware jars. The rest of the recon- structible vessels were from straight-sided, flaring bowls and tall drinking vessels, the latter of which were either painted or incised. The straight-sided, flaring bowls of Muyu Roqo have flat bases. The angle of the rims are fairly constant and contain finely rounded but slightly thinned lips. Occasionally, on what appear to be large bowls, the lips are flattened. The exterior of the bowls are not decorated, other than being oc- casionally painted orange. The interiors of the bowls display a variety of designs. The most com- mon are pendant rectangles, outlined in black, which can contain a single black curving or straight line through their center. The pendants are generally colored white or yellow (or orange), al- though black and gray are occasionally used. While the majority of the pendants are mono- chrome, examples of bi- and polychrome pen- dants have been noted (Fig. 3-1, Drawings 3-2 through 3-4). Other designs found on the interiors of the straight-sided, flaring bowls include a white, den- tric, split- V pattern (Drawing 3-2B) and a series of white U-shaped curves (Drawing 3-3B) that border the interior rim. The U-shaped curves ap- pear alone or on vessels with the pendant rect- angle motif. A large cross-hatched pendant trian- gle, generally painted in white, has also been not- ed on several bowl fragments. Besides straight- sided, flaring bowls, work at Muyu Roqo recovered another vessel type in large numbers that has not been described elsewhere. These are drinking vessels, at least 15 cm high, with slightly flaring rims. The rims of the vessels sometimes narrow at the top and are rounded. Their bases are flat, and their sides tend to be straight to slightly concave. The exterior surface is always slipped, and the upper interior surface of the vessel is frequently slipped. The drinking vessels recovered from Muyu Roqo were divided into two general ware cate- gories, orange and black. The upper two-thirds of the orange-ware drinking vessels is decorated with geometric designs. Although a variety of or- ange-ware drinking vessels designs were noted in the collection, there is a dominant decorative pat- tern (Drawings 3-5 through 3-8). This pattern in- cludes a narrow, horizontal gray band, outlined in black approximately 0.5 cm below the undecor- ated rim. Below this gray band is a wide horizon- tal band painted with a repetitive "X-and-dot" de- sign, which is generally executed in white (Fig. 64 Chapter Three Fig. 3-1. Straight-sided flaring bowls from Muyu Roqo. 3-2, Drawing 3-5C). The lower, decorated section of the drinking vessel is covered by two to four bands, outlined in black (Fig. 3-3). A number of colors are used in these bandings, including white, gray, yellow, orange, and burgundy-maroon. The black-ware drinking vessels are similar in shape to the orange- ware ones, although their rims appear to be slightly more flared (Drawings 3-9 through 3-11). The use of decorative space on the black-ware and orange-ware drinking vessels is similar; the upper two-thirds of the exterior is dec- orated, while the lower third remains unmodified. The exteriors of the black-ware drinking vessels are incised with fine lines. These drinking vessels exhibit a wide, dominant horizontal band 1-2 cm from the rim. This band contains a series of in- cised circles, concentric circles, half circles, tri- angles, and dots (Fig. 3-4). The composition of this band varies. A narrow band is found below the dominant one. This lower band may be free of decorations or it may have an additional series of designs. In a few cases, the dominant band of these vessels is decorated in white with the same X-and-dot design found on the orange-ware ves- sels (Drawing 3-12). In still rarer cases, the dom- inant band is actually raised from the surface of the vessel (Drawing 3-9B). The Muyu Roqo ceramic collection suggests that the site may have had an unusual function. The proportion of finer-ware (i.e., orange and black) to earthenware vessels is different from what is found at most domestic sites, where earth- enware vessels greatly outnumber the finer-ware vessels. At Muyu Roqo, only 16% of the shards were earthenware, with the remaining 84% divid- ed between the finer orange-ware (60%) and black-ware (24%). Furthermore, bowls and drink- ing vessels dominated the finer-ware vessel forms. Of the finer ware, 19% of the fragments could be attributed to bowls and 49% to drinking vessels. The remaining 32% included pieces that were ei- ther too small to determine their original vessel form or were other vessel shapes, such as incurv- ing bowls. The high percentage of finer-ware items at Muyu Roqo suggests that Muyu Roqo was a nondomestic site. The high percentage of bowls and drinking vessels implies that extensive eating and drinking activity took place there. This Ceramics of the Middle Horizon 65 Fig. 3-2. Orange-ware drinking vessels from Muyu Roqo. interpretation is reinforced by the great number of large mammal bones recovered at the site. In sum, although the exact nature of Muyu Roqo cannot be determined until additional excavations take place there, the current evidence suggests that this site was used for ritual activities during the Mid- dle Horizon. Dating Dating at Muyu Roqo is not yet possible owing to the lack of carbon samples. McEwan's (1987, pp. 42, 43, 80, 89) work at Pikillacta, however, does provide a number of radiocarbon samples to date the Wari occupation of the Cuzco Valley. Two samples from the lowest excavation levels of Pikillacta yielded radiocarbon dates of 1430 ± 90 b.p. ([TX 4751] a.d. 520 ± 90) and 1350 ± 60 b.p. ([TX 4750] a.d. 600 ± 90). As noted by Knobloch (1991, p. 253) and Glowacki (1996), these excavations also provided examples of Oc- ros-style ceramics. Two additional samples from vines within the matrix of walls yielded radiocar- bon dates of 1140 ± 60 b.p. ([TX 4247] a.d. 810 ± 90) and 1 100 ± 60 b.p. ([TX 3996] a.d. 850 ± 60). These radiocarbon dates suggest that initial occupation of Pikillacta may have occurred be- 66 Chapter Three Fig. 3-3. Orange-ware drinking vessels from Muyu Roqo. tween a.d. 600 and 650, and construction contin- ued until at least a.d. 800 or 850. McEwan (1987, p. 80) believes that the abandonment of Pikillacta might have occurred as late as a.d. 1000. Accord- ingly, until more dates become available, the pe- riod of Wari influence in the Cuzco region is ten- tatively established at a.d. 600-1000. Distribution Only nine sites in the survey area provided Oc- ros-style ceramics, and with the exception of Muyu Roqo, each yielded an extremely small number of such fragments (Map 3-1). Neverthe- less, the distribution pattern of sites with Ocros- style ceramics is unlike that found for any other ceramic style in the Province of Paruro. Six of the nine sites are located in the Paruro Valley, while the other three are situated to the south of the Apurimac River. The Lucre Basin, which holds the Wari center of Pikillacta, is located approximately 20 km northeast of the Paruro Valley. These two areas are connected by several large trails; the journey on foot takes 6-8 hours. Although truck transpor- tation between the community of Paruro and the city of Cuzco now dominates the regional ex- change networks, some trading still takes place between the inhabitants of Paruro and Lucre. The close clustering of sites containing Ocros-related materials in the Paruro Valley suggests that the Paruro-Lucre exchange networks were more ex- tensive in prehistory. If this is the case, then we can expect to find Paruro-produced styles, such as Colcha and perhaps Ccoipa (see below), at sites in the Lucre Basin. Arahuay Ceramics Arahuay ceramics take their name from the site of Arahuay, located approximately 2 kilometers south of Cuzco on the valley slope. The site was excavated by Nilo Torres Poblete in 1989, and the style was defined in his Licenciado thesis (Draw- Ceramics of the Middle Horizon 67 Fig. 3-4. Black-ware incised drinking vessels from Muyu Roqo. ing 3-13).6 Further examples of Arahuay ceramics have been identified in the ceramic collections of Pikillacta (McEwan, 1990; Glowacki, 1996)7 and by Zapata (pers. comm. 1994, 1997) in tombs with Wari-related vessels at Batan Oreo. Although additional research is needed to further define this new style, enough is known to conclude that there is a light presence of Arahuay ceramics in the Province of Paruro. A variety of vessel forms have been identified in Arahuay collections, but the most common fragments are from straight-sided, flaring bowls (Drawings 3-13 and 3-14; Torres Poblete 1989, pp. 59-61, 94-95, Fig. 16). Some of the bowls are decorated with broad horizontal red bands, outlined with black lines, running several centi- meters below the rim. Areas between the horizon- tal red band and the rim have been marked off with vertical red bands outlined with black lines. 6 Excavations near Arahuay at the site of Tarawi were conducted by Rowe and Lyon in 1975 (K. Chavez 1980, p. 215; Lynch 1975, p. 230). 7 Glowacki (1996) prefers to call this style Arahuay/ Wamanga. Small decorative motifs can also be found on Ar- ahuay ceramics. Checks are common; however, curving lines and X's have also been noted. The interior of the bowls can also be decorated with pairs of vertical straight black lines with a curving red line between them. Both the design compositions found on Arahuay ceramics and the vessel shapes are reminiscent of Ocros-style ceramics, particularly the straight-sid- ed, flaring bowls. There are also some similarities between Arahuay and Killke ceramics. One of the most notable characteristics of Killke ceramics is the broad red band outlined in black. This same feature also appears to be a distinctive character- istic of Arahuay ceramics. Outlined bands are found on both Arahuay jars and bowls. Like their counterparts in Killke vessels, these bands run hor- izontally and vertically and at times are used to define panels that are filled with geometric designs. Dating Arahuay ceramics have been recovered from two sites in the Cuzco region in Wari contexts: 68 Chapter Three 0 10 20 i i i Contour lines every 400 m Map 3-1. Distribution of sites with Wari-style ceramics. Pikillacta (McEwan, 1990; Glowacki, 1996) and B atari Oreo (Zapata, pers. coram. 1994, 1997). The stylistic similarities that are shared by Ocros- style and Arahuay materials imply that Arahuay ceramics were influenced by Wari ceramic traditions and that they were produced during the Wari occupation of the Cuzco region. Arahuay ceramics introduced to the Cuzco re- gion include vessels decorated with broad red bands outlined with black lines. The use of out- Ceramics of the Middle Horizon 69 lined red bands continues in later Killke as well as in Classic Inca ceramics. Based on these ob- servations, I propose the following hypothesis: (1) the production of Arahuay ceramics began after the establishment of a Wari presence in the Cuzco region, and (2) Arahuay ceramics influenced, or could be the direct antecedent of, Killke ceramics. Furthermore, if Killke ceramics did develop from Arahuay traditions, I would then propose that the transition between Arahuay and Killke ceramics occurred after the abandonment of Pikillacta in the early part of the Late Intermediate period. Distribution Fragments of Arahuay ceramics have been re- covered in surface and excavation contexts from eight sites in the Paruro region (Map 3-2). Most notable is the recovery of Arahuay ceramics from beneath Inca remains at Maukallaqta (Bauer, 1992b; Drawing 3-17A-B). The number of sites in the Province of Paruro with Arahuay ceramics may increase as this new style becomes better un- derstood. Because Arahuay is a newly identified ceramic style, little is known concerning its distribution in the Cuzco region. The type site of Arahuay, lo- cated just outside the city of Cuzco, contains a large number of fragments, and additional re- mains have been recovered at Pikillacta and Batan Oreo. Systematic regional surveys in the Province of Paruro have, however, provided only a few sites with this ceramic style. Although too few sites have been found with Arahuay ceramics to define the style's pattern of distribution with cer- tainty, current evidence suggests that the largest sites with this ceramic style are located in the Cuzco Valley and the Lucre Basin. Accordingly, it is possible, although highly speculative, to pro- pose that Arahuay ceramics were produced some- where within the valley area that connects Cuzco and Pikillacta and from there were traded south- ward into the Province of Paruro. Qotakalli Ceramics Qotakalli ceramics, another Middle Horizon ce- ramic style, were first identified by Lyon and Rowe (1978). The type site is situated south of the modern airport across the Huatanay River. A description of Qotakalli ceramics was published by Barreda Murillo in 1982. Since that time, Qot- akalli ceramics have been found at several sites in the Cuzco region (Espinoza Martinez, 1983; Valencia Zegarra, 1984; McEwan, 1984, p. 12, 1987, pp. 19, 58, 98, 99, 1990; Torres Poblete, 1989, pp. 56-59, 92-93, Fig. 14; Zapata, pers. comm. 1994, 1997; Glowacki, 1996). Description Lyon and Rowe (1978), and later Barreda Mu- rillo (1982), have described the Qotakalli ceram- ics collected during their investigations in the Cuzco region. Building on these works, and on the large sample of Qotakalli materials recovered during archaeological research in the Paruro re- gion, I have provided an expanded description of this style. Ware The ware of Qotakalli ceramics varies in that it tends to be composed of a fine fabric with minute white and gray inclusions. The fine nature of the clay paste suggests that a ground temper was add- ed (Lunt, pers. comm. 1989). It is a medium-hard ware that frequently fires to a light orange or pink. Design Elements, Color, and Surface Treatment Qotakalli ceramics contain a limited variety of geometric decorations (Drawings 3-15 and 3-16). The most common elements are straight horizon- tal and vertical lines, zigzagging lines, diamonds, triangles, and dots (Fig. 3-5). Several variations of Qotakalli ceramics have been identified, in- cluding monochromes (black on cream or, more rarely, red on cream [Drawings 3-17 and 3-18]) and bichromes (black and red on cream). The ma- jor designs are generally defined in black, while the minor ones are executed in either red or black. The straight and zigzagging lines tend not to ap- pear singularly but in parallel groups of two to five lines. In the bichrome vessels, these lines fre- quently alternate between black and red. The in- terior of the diamonds is commonly cross-hatched or contains nested diamonds. The interior of the triangles is at times painted solid, at times filled with nested triangles or dots. The exteriors of most currently recognized Qot- 70 Chapter Three 0 10 20 i i i Contour lines every 400 m Map 3-2. Distribution of sites with Arahuay ceramics. akalli vessels have a cream or white slip and are well burnished. The slip provides a flat surface on which the decorations are painted and ensures rel- atively strong color-tone contrasts. Although the color of the black pigment does not differ greatly between fragments, the color of the red pigment varies from pink to red to dark orange. The lines used to form the designs can be as wide as 0.5 Ceramics of the Middle Horizon 7 1 Fig. 3-5. Common Qotakalli designs. cm, but they are more typically 0.2-0.3 cm in width. In general, the bichrome examples tend to have thinner lines than the monochrome ones and are more extensively burnished. Many of the de- signs on Qotakalli ceramics are executed with considerable care, and there is little overlapping of lines or design motifs. Vessel Forms Lyon and Rowe (1978) and Barreda Murillo (1982) indicate that frequently found Qotakalli vessels include bowls, deep dishes, and flat-based pitchers with round bodies and straight necks. Perhaps the most distinct vessel form is a steep- sided, tripodal bowl with legs that vary in length from 1 to 3 cm (Drawing 3-19A-C). The base of these bowls is rounded and the sides are slightly flared. Another common vessel is similar to the steep-sided, tripodal bowl except that it contains a flat base and stands on a rim that is approxi- mately 0.5 cm high (Drawing 3-19D-E). These steep-sided, rim-based bowls are interesting be- cause this form is relatively rare in Wari ceramic collections but is present in many Tiwanaku col- lections. If broken, the upper portions of the two steep- sided bowl forms cannot be distinguished (Draw- ings 3-20 and 3-21). The rims of both steep-sided bowl forms may continue the gentle outward slope of the vessel side or may display a more pronounced flair. The lips generally terminate in a fine point, like the lips of most Qotakalli vessels. The exteriors of the steep-sided bowls tend to be covered with decorations. The interior rims are frequently decorated with nested or cross-hatched triangles, which may alternate in color (Fig. 3-6). More rarely, the entire interior surface of the steep-sided bowls is covered with geometric de- signs. Another common bowl form is gourd shaped, with straight to inward-curving sides and a small, flat or slightly curved base (Drawing 3- 17 A). The exterior of this type of bowl is frequently deco- rated on the upper half, while the interiors are not decorated. Qotakalli ceramic inventories can also include 72 Chapter Three Fig. 3-6. Qotakalli rims from steep-sided bowls. large vessels with modeled and applied human- istic features that are painted in two colors with straight or curving lines (Barreda Murillo, 1982, p. 21). The ware of these vessels tends to be coarser than that of other Qotakalli vessels. Fur- thermore, these modeled vessels are generally not covered with a cream or white slip. Dating amples of both monochrome and bichrome Qota- kalli ceramics (Glowacki, 1996). Additional Qota- kalli finds have been recovered by Zapata at Batan Oreo in burials that also contained Wari-related materials. The recovery of Qotakalli fragments in the structures and refuse middens of Pikillacta and in burials at Batan Oreo suggests that this style was produced during the Wari occupation of Cuz- co. More precise dating awaits further research on this important style. Although Qotakalli ceramics have been found at numerous sites in the Cuzco region, there is little information on their production dates. Bar- reda Murillo (1982, p. 14) indicates that he re- covered Qotakalli ceramics while excavating at Pikillacta. The presence of Qotakalli materials at Pikillacta has been confirmed by McEwan (1984, p. 12, 1987, pp. 98, 99, 1990) and Glowacki (1996). In the course of his work at this large Wari center, McEwan found examples of monochrome (black on white) Qotakalli ceramics on the floor of a few structures. Excavations in trash middens on the edge of the ancient city also produced ex- Distribution Qotakalli ceramics were found at 19 sites in the Province of Paruro (Map 3-3). The density and distribution of sites containing Qotakalli-ceramics are markedly different from those of sites con- taining other Middle Horizon pottery styles. The density of the Qotakalli sites is greatest in the northern sectors of the research region, near Yaur- isque, and declines as one progresses to the south. Sixteen of the Qotakalli sites are located north of the Apurimac River, which divides the survey re- Ceramics of the Middle Horizon 73 0 10 20 i i i Contour lines every 400 m Map 3-3. Distribution of sites with Qotakalli ceramics. gion in half. The other three sites are situated along the southern bank of the Apurimac. No ex- amples of Qotakalli ceramics were found in the southern sections of the survey region, near Ac- cha or Pillpinto. Although the number of Qotaka- lli sites is too small for statistically valid density- distance tests to be conducted, the concentration of Qotakalli sites in the northern part of the region 74 Chapter Three suggests that ceramics of this style may have been traded into the Province of Paruro from the Cuzco Valley. Ccoipa Ceramics Another possible Middle Horizon ceramic style is Ccoipa. Ccoipa ceramics were first identified in 1984 at a site immediately south of the village of Ccoipa (District of Pacariqtambo) and were named after the village (Bauer, 1989). The Ccoipa type site (312) is situated on a ridge south of the church of Ccoipa. This small site, measuring less than 2,500 m2, is cut by the only road that passes through the village. The northern slope of the ridge rises up from the edge of the village. This slope, currently under cultivation, provided the best area for surface collections during the 1984- 1985 field seasons in the region. The top of the ridge and its other slopes are covered with grass and eucalyptus trees. These areas yielded few fragments. Description A definition of Ccoipa ceramics is developed from surface collections made at the village of Ccoipa and from other samples of Ccoipa ceram- ics recovered during survey and excavation work in the Province of Paruro. Ware Ccoipa ceramics are composed of a medium- coarse fabric containing some fine- to medium- sized (<1.0 mm) nonplastic inclusions. Most of these inclusions consist of white or cream grains, although black, matte red, or lustrous grains, as well as gold mica, are also present. Occasional large examples of these grains, not exceeding 2.0 mm, have been noted. The oxidized paste color is generally a reddish yellow. Vessel Forms, Design Elements, Color, and Surface Treatment Most of the Ccoipa material so far collected consists of straight- sided drinking vessels with nonflaring rims (Figs. 3-7 and 3-8). However, fragments of drinking vessels with flaring rims, deep, straight-sided flaring bowls, shallow, curve- sided bowls, and global pots have also been iden- tified. Both the interiors and the exteriors of Ccoi- pa vessels are generally burnished. Only three colors appear on Ccoipa ceramics — red, black, and white. Of these, red and black are considerably more frequent than white. There is also a limited variety of designs on Ccoipa ves- sels. The most common designs are sets of nested rectangles alternating in color between red and black. The nesting terminates with a red dot with- in the innermost rectangle (Drawings 3-22A-B and 3-23A-B). At times this design is rotated 90°, forming sets of diamonds (Drawings 3-23B, 3- 24D, 3-27B, and 3-28A, E, and I). The designs on the bowls appear to be more varied than the designs on the drinking vessels (Drawing 3-26). The lip of Ccoipa vessels is generally painted with a red or black line, and black inverted tri- angles may be suspended from the rim on the ex- terior (Drawings 3-24A and C and 3-25A and C). On some of the finest made vessels, the bottoms of the black inverted triangles are outlined in white (Drawings 3-22B and 3-24C). In some ex- amples, a red band, varying greatly in width, is found on the interior rim of the vessel (Drawings 3-22A, 3-23C, 3-24C, and 3-25E), while in others the entire interior surface is painted red. Another common design is a row of linked, horizontal diamonds (or ovals) formed by the in- tersecting of two zigzagging lines. The linked di- amonds are painted black. Red dots are placed in the centers of these linked diamonds, and occa- sionally other red dots are placed on their exteri- ors near the intercession points (Drawings 3-27 A, C, and E and 3-28C). The area of linked diamonds is highlighted by wide black or red bands that run above and below them. On a few samples the di- amonds are painted on a slightly raised band of clay. Although nonlinear designs are rare, they do occur (Drawing 3-23C). Animal motifs are also rare; however, one drinking vessel, with red and black llamalike animals encircling its base and with red and black birds (condors?) that hover above the animals (Drawing 3-23D), has been re- covered. The decorations on Ccoipa ceramics are infor- mally applied: the widths of lines vary as they undulate across the vessel surface, corners formed by two intercepting perpendicular lines are often poorly executed, and there is frequent overlapping of design motifs. Ceramics of the Middle Horizon 75 Fig. 3-7. Ccoipa straight-sided drinking vessels. Test Excavation Results Test excavations at two sites in the Province of Paruro, Rokeccasa (450) and Ccorpina (659), pro- vided important information on Ccoipa ceramics. Rokeccasa — The site of Rokeccasa is located at the end of a small, steep ridge west of the town of Paruro. There is a terraced area, measuring ap- proximately 20 X 8 m, near the summit of the ridge on its northern slope. Surface collections were made at the site in 1985 and 1987. These collections provided a few Inca and Killke frag- ments as well as numerous (more than 60) ex- amples of Ccoipa ceramics. In 1987 test excavations were conducted at Ro- keccasa. This site was selected for excavation over the Ccoipa village site because much of Ro- keccasa has never been cultivated, and the site is generally better preserved. Several test squares dug on the slopes of Rokeccasa provided addi- tional examples of Ccoipa materials but no strat- ified deposits. Two test excavations were also dug near the center of the terrace area of Rokeccasa, because this appeared to be the most likely place to find architectural remains. One of these 2 X 2 m ex- cavations yielded significant information. The first 40 cm of this test pit provided no cultural arti- facts. However, at a depth of 41-42 cm scattered ceramic fragments were found, and at slightly lower levels the tops of several stones began to appear. At a depth of 50 cm it was clear that the northeast quarter of this excavation contained a wall foundation. The test excavation was then ex- panded 1 m north and 1 m east. The excavations revealed an oval structure (ap- proximately 3X5 m), perhaps with a doorway on its northern side.8 A number of burnt earth patches containing flecks of carbon were found on the floor, along with several small ceramic con- centrations. One of these concentrations contained the remains of a straight-sided bowl (Drawing 3- 8 A shallow pit was found approximately 10 cm below the floor level. This pit contained part of a human pelvis, an articulated right hand, and two articulated feet (the left placed over the right). These remains appear to rep- resent an earlier burial that was partly destroyed with the construction of the oval structure above it. 76 Chapter Three Fig. 3-8. Ccoipa straight-sided drinking vessels. 29), another yielded two decorated Ccoipa ceram- ic fragments, and a third concentration contained a spindle whorl. A number of deer bone fragments were also recovered. The largest fragment, found near the northeast "corner" of the building, at the intersection of the wall and the floor, was sub- mitted for radiocarbon dating (accelerator mass spectrometry). The sample provided a radiocar- bon age of 730 ± 55 b.p. ([AA 8936] a.d. 1220 ± 55). The modest construction of the structure, the light scatter of bone and ceramic fragments across its floor, and the concentrations of burnt earth suggest that it was a domestic dwelling. Ccorpina — The site of Ccorpina is situated on a steep ridge that descends to the Apurimac River just west of the community of San Lorenzo. The ridge contains the low foundations of several cir- cular structures and a dense scatter of ceramics. Surface collections at the site in 1987 provided examples of Inca, Killke, Colcha, and Ccoipa fragments. Excavations were conducted there dur- ing the same year. These excavations yielded Ccoipa ceramics in a stratum of earth, at a depth of 61-75 cm, directly below the floor of a struc- ture containing Colcha ceramics. Dating The period of Ccoipa manufacture is not well established. As noted above, a bone sample from Rokecassa provided a radiocarbon age of 730 ± 55 b.p., suggesting that Ccoipa ceramics were be- ing produced during the Late Intermediate period. I am uneasy with this relatively late date. Exca- vations at Ccorpina produced Ccoipa material in a stratum of earth below a structure with Colcha ceramics, suggesting that Ccoipa production be- gan before that of Colcha ceramics. Furthermore, the design motifs of Ccoipa ceramics, especially the nested rectangles of alternating color, support a Middle Horizon classification. Although signif- icantly cruder, the designs of Ccoipa ceramics are more closely related to those displayed on Qota- kalli ceramics, dated by excavations at Pikillacta and Batan Oreo to the Middle Horizon, than to Ceramics of the Middle Horizon 77 the Late Intermediate period styles of Colcha and Killke. From its stylistic qualities and its strati- graphic position in the Ccorpina excavations, I be- lieve that Ccoipa ceramics were produced during the Middle Horizon. This hypothesis needs to be tested through additional excavations and radio- carbon dating. mit of a prominent, round hill.10 The site is ap- proximately 30 minutes on foot north of Yaur- isque along the Royal Road of Cuntisuyu, in the direction of Cuzco. A definition of Muyu Oreo ceramics has been developed from materials col- lected on the surface and in test excavations at this site. Distribution Ccoipa ceramics are relatively unsophisticated in their design and manufacture features. They have been found at 35 sites in the research region (Map 3-4), the majority of which are located in the areas of Yaurisque, Pacariqtambo, Paruro, Colcha, and Accha. These areas contain the best agricultural land in the region and have always supported the largest populations. The widespread presence of this ceramic style in the Province of Paruro suggests that it may have been a dominant style for the region immediately south of Cuzco during the Middle Horizon.9 Ware The composition of Muyu Oreo ceramics is similar to — although slightly finer than — that of Ccoipa ceramics. They are composed of a medi- um-coarse fabric containing some nonplastic in- clusions. As in Ccoipa ceramics, these inclusions consist of fine- to medium-sized (<1.0 mm) white to cream grains. Occasionally, however, large ex- amples of these grains, not exceeding 2.0 mm, are present. Also as in Ccoipa ceramics, fine inclu- sions of black, matte red, or lustrous grains or gold mica are occasionally present. The paste gen- erally fires to a dark red color. Muyu Oreo Ceramics During survey work in Paruro it became evi- dent that a new ceramic style, characterized by bright white, black, and orange paints over a pol- ished, dark red pigment, was present in the col- lections. The style was named Muyu Oreo, after the site where it was most common (Bauer, 1989). At the close of the project, Muyu Oreo ceramics had been recovered from 14 sites in the research region. Since that time, similar ceramics have been recovered from other archaeological sites in the Cuzco region. For example, fragments of what appear to be Muyu Oreo ceramics were collected by Hector Espinoza Martinez (1983) during ex- cavations at Acomoqo (Pueblo Joven, Primero de Mayo), by Torres Poblete (1989, pp. 59-61, 93- 94, Fig. 15) during his work at Arahuay, and by Zapata (pers. comm. 1994, 1997) at Batan Oreo. Description The site of Muyu Oreo (536) is located between the villages of Yaurisque and Pumate at the sum- 9 In 1987, I identified a site with Ccoipa material on the Huanoquite plain just west of the study region, and in 1991 and 1994, I found Ccoipa fragments at several sites in the Cuzco Valley. Design Elements, Color, and Surface Treatment Muyu Oreo ceramics contain a distinct range of pigment colors, including dark red, black, white, and orange. In addition, one fragment has patches of gray. The dark red is used as a background on which other colors are applied (Drawings 3-30 and 3-31). The interior and the exterior surfaces of Muyu Oreo ceramics are heavily burnished. The black, white, and orange pigments are rela- tively thick, often with little absorption into the vessel (Fig. 3-9). Vessel Forms Few recovered fragments of this style were large enough to suggest specific vessel forms. The vessels forms that have been reconstructed in- clude straight-sided drinking vessels, steep-sided bowls, and annulated bowls. A number of complex Muyu Oreo fragments were recovered that deserve individual descrip- tions. One piece, the remnant of a straight- sided drinking vessel, is especially interesting because 10 The name Muyu Oreo (round mountain) is a com- mon toponym in the Cuzco region. 78 Chapter Three 0 10 20 — - — i — Contour lines every 400 m Map 3-4. Distribution of sites with Ccoipa ceramics. it has a front face god motif (Drawing 3-32). The nose of the face has been raised approximately 1 cm from the surface of the vessel, while the eyes have been raised approximately 3 mm. A dark red pigment covers the exterior and interior surfaces of the vessel. Thick black lines divide the front of the drinking vessel into panels, which have been filled with nested orange and white rectan- Ceramics of the Middle Horizon 79 Fig. 3-9. Muyu Oreo ceramics. gles. The vessel probably stood on three or four small pedestal feet, although only one remains. A second interesting Muyu Oreo drinking ves- sel has a convex, horizontal band several centi- meters below its rim (Drawing 3-33). The band is outlined by a series of black, white, and orange lines, and the band itself holds a complex zigzag design. The rest of the vessel's exterior is dark red, while its interior is free of paint. The remains of a slightly concave, steep-sided bowl were also recovered from Muyu Oreo (Drawing 3-34). This bowl stands approximately 7 cm high. Along its base is a horizontal black line outlined by two adjacent white lines. The ves- sel is otherwise undecorated. Further excavations yielded an elaborately dec- orated annulated bowl with concave sides (Draw- ing 3-35). The exterior of the vessel contains a dark red background, while the interior remains unpainted. Thick black lines divide the exterior into panels, in one of which there is a white, tri- angular latticework and a series of three orange circles. A second panel is free of design. The ves- sel stood on a hollow pedestal base. The rim of the vessel originally displayed a series of scallops, of which only one remains. Several other scallops were recovered in surface collections and exca- vations in the Province of Paruro (Drawing 3- 30A-B), which suggests that they are a relatively common feature of Muyu Oreo ceramics. Test Excavation Results Test excavations at Muyu Oreo (536) focused on assessing the temporal relations of the many styles found on the surface of the site. The upper 30 cm of the excavations, representing the plow zone of the site, yielded an undifferentiated mix of Inca, Killke, Ccoipa, Qotakalli, Muyu Oreo, and Paruro Formative ceramics. Beneath the plow zone, a much darker and more compact stratum was encountered that yielded a mixture of Qota- kalli and Muyu Oreo and occasionally Paruro For- mative ceramics. The deeper excavations termi- nated with a thin level of Paruro Formative ce- ramics above the natural subsoil. The recovery of Muyu Oreo ceramics below the Inca- and Killke- 80 Chapter Three yielding plow zone and above the deposits of For- mative materials suggests that this style was pro- duced sometime between the end of the Early Ho- rizon and the beginning of the Late Intermediate period. Quantities of Muyu Oreo ceramics were also found at the site of Marcapata (624), a small ridge south of the Apurimac River near the village of Araypallpa. Surface collections at this site provid- ed examples of Inca, Colcha, Qotakalli, Ccoipa, Muyu Oreo, and Formative materials. Excava- tions at Marcapata yielded samples of Muyu Oreo ceramics below the Inca and Killke plow zone. Muyu Oreo ceramics were most frequently found mixed with Qotakalli shards, and to a lesser extent with Ccoipa shards. Dating Because no carbon remains suitable for dating were recovered, tentative dating of Muyu Oreo ceramics must be made from their stratigraphic position in excavations and from similarities with other ceramic styles. Excavations at the site of Muyu Oreo produced Muyu Oreo ceramics mixed with Qotakalli ceramics above a Formative level and below a plow zone containing Inca and Killke remains. Excavations at Marcapata produced Muyu Oreo ceramics mixed with Qotakalli and Ccoipa materials. It must be stressed, however, that the ceramic samples from both of these sites come from excavation units with no clear cultural context and thus have a limited value for dating this style. Nevertheless, the excavation results suggest that Muyu Oreo ceramics were made dur- ing the Middle Horizon. Three excavations in the Cuzco region have yielded what appear to be fragments of Muyu Oreo ceramics. Zapata has recovered Muyu Oreo vessels from burials at the site of Batan Oreo (pers. comm. 1994). Torres Poblete (1989, pp. 59-61, 93-94, Fig. 15) reports recovering a new ceramic style, and from his illustrations this style appears to be Muyu Oreo. Vessels found by Tor- res Poblete include a drinking vessel with a con- vex, horizontal band and a straight-sided side bowl with scallops. He classifies these fragments as "Araway — 5", or "Wari de Araway" (Wari of Arahuay), suggesting a relationship with the Wari occupation of the Cuzco Valley. Muyu Oreo ceramics, along with Qotakalli shards, may also have been recovered in 1978 at the site of Aqomoqo, located within the modern city of Cuzco, by Espinoza Martinez (1983). Es- pinoza Martinez, like Torres Poblete, equates this style with the Wari occupation of Cuzco, calling it "Wari de Aqomoqo" (Wari of Aqomoqo). Al- though additional research may prove Espinoza Martinez and Torres Poblete correct in their clas- sification of Muyu Oreo ceramics as Wari-related, in the conclusion of this chapter it will be proposed that Muyu Oreo is a Tiwanaku-related style. Distribution The distribution of the 14 sites containing Muyu Oreo ceramics in the research area is marked on Map 3-5. There is a Muyu Oreo site near the southern limits of the research region, near Pillpinto, as well as one on the northern boundary, close to Yaurisque. There are also clus- ters of Muyu Oreo sites near Colcha, Paruro, and Pacariqtambo. In other words, Muyu Oreo ceram- ics are not limited to or concentrated in any one zone of the research area, and research by other investigators has yielded similar vessels in the Cuzco Valley. Horizontally Incised Drinking Vessels Archaeological research in the Province of Pa- ruro provided numerous fragments of horizontally incised drinking vessels (Fig. 3-10). Examples of these vessels were recovered at sites with Ccoipa ceramics, including the Ccoipa village site and Rokeccasa. Horizontally incised drinking vessels were also found with fragments of Ocros-style ce- ramics at Muyu Roqo and with Arahuay ceramics in the lower excavation levels at Maukallaqta. Ad- ditional samples of horizontally incised drinking vessels have been recovered in burials with Wari- related vessels at Batan Oreo by Zapata (pers. comm. 1994, 1997). Although these findings sug- gest that horizontally incised drinking vessels may have been produced during the Middle Horizon, additional research is needed before their cultural and temporal affiliations can be stated with cer- tainty. Nevertheless, for the benefit of future re- search in the Cuzco region, I will provide a brief description and illustrations of these vessels. The exterior of the vessels is burnished and in- cised (Drawings 3-36 through 3-38). In rare cases, the remains of white pigment can been seen along the rim (Drawing 3-37A). The upper centimeters Ceramics of the Middle Horizon 81 0 10 20 i i i Contour lines every 400 m Map 3-5. Distribution of sites with Muyu Oreo ceramics. of the interior surface are burnished, while the lower portions are left unmodified. The sides of the vessels tend to be straight or slightly flaring; however, some lyrelike examples have been re- covered (Drawing 3-37D). The rims generally are slightly more flared than the vessel sides. The rims themselves are usually thickened, although some straight examples have been found. Char- 82 Chapter Three Fig. 3-10. Horizontally incised drinking vessels. acteristically there is a relatively deep horizontal band incised on the interior of the vessel 2-3 cm from the lip. The exterior of the vessel may carry a similar incised band, although not necessarily at the same level as the interior incision. Although the exterior body of the vessel may be left undecorated, it is more commonly covered with a series of horizontally incised lines. These incisions tend to be roughly parallel, and their quantity and quality vary from vessel to vessel. Some horizontally incised drinking vessels have in- cisions that are only slightly deeper than the bur- nishing, while others have incisions several milli- meters deep. The ware appears to be similar to Ccoipa wares. Examples of horizontally incised drinking vessels were recovered at 15 sites distrib- uted throughout the Province of Paruro (Map 3-6). Summary and Discussion Several different ceramic styles in the Cuzco region are believed to date to the Middle Horizon, and the complexity of their temporal relations is only now beginning to be addressed (see Glo- wacki, 1996). Middle Horizon ceramic styles re- covered in the Province of Paruro include Qotaka- lli and Ccoipa. Qotakalli ceramics are relatively well made. Finer Qotakalli pieces are recognized by their white slip and complex geometric deco- rations. Examples of this style have been found at numerous sites in the Cuzco region. The recovery of Qotakalli ceramics at the site of Pikillacta and in burials at Batan Oreo indicates that they were produced during the Wari occupation of the re- gion. Regional research in Paruro also identified a widespread ceramic style called Ccoipa. The dating of this style is currently problematic, al- though it may have been produced during the Middle Horizon and early Late Intermediate pe- riod. A limited number of Chakipampa and Ocros vessels were imported into the Cuzco region from the Wari heartland during the Middle Horizon. Wari influence is also reflected, however, in what appears to be the large-scale local production of Ceramics of the Middle Horizon 83 Ocros-style ceramics in the Cuzco region." The distribution of Ocros-style ceramics in the Prov- ince of Paruro indicates that Wari influence varied greatly across the Inca heartland. There are sev- eral sites with Ocros-style materials in the Lucre Basin, including the large site of Pikillacta. Sev- eral sites, east of Pikillacta in the Vilcanota River Valley, including Batan Oreo and Ccotocotuyoc, also contain quantities of Ocros-style materials. These findings suggest that Pikillacta controlled, or had considerable influence over, much of the east-west-running valley between the cities of Cuzco and Urcos. On the other hand, only a few sites examined in the research region had any Wari-style ceramics. The scarcity of Wari-style ceramics in the Province of Paruro is surprising, given its close proximity to the Lucre Basin. From these data, I conclude that Wari influence was substantially less to the south of Pikillacta than it was to the east and west. It is important to note that the Paruro Valley also holds the extraordinary site of Muyu Roqo. This small site has yielded a large quantity of bones and Wari-style vessels, many of which can be further classified as Ocros-style. Surface col- lections and test excavations at Muyu Roqo yield- ed an unusually high percentage of fine-quality bowls and drinking vessels and a low percentage of utilitarian vessels. These findings imply that rit- ual activities, including large-scale eating and drinking ceremonials, took place at the site. The exact nature of these activities and the purpose for which the fine ceramics were used and then aban- doned at the site remain to be investigated. The Wari presence in the Cuzco region influ- enced the development of new ceramic styles, the best example being Arahuay ceramics. The colors and vessel forms of this style reflect features of Wari ceramics, especially those of Ocros. Al- though little is known about Arahuay ceramics, I have proposed that their production began shortly after the establishment of Pikillacta and that they influenced, or were the direct antecedent of, Killke ceramics. Wari was not the only state to develop during the Middle Horizon. Tiwanaku, located close to the southwestern shore of Lake Titicaca, also rose to prominence during this period. It is generally 11 Ocros ceramics (especially straight-sided, flaring bowls) are found in large numbers at the site of Wari itself (Brewster- Wray, 1990), and Ocros-style vessels have been recovered at numerous sites across the Andes (Knobloch, 1991; Glowacki, 1996). believed that the Tiwanaku influence ended some- where near the pass of La Raya (the traditional boundary between the Quechua speakers of the Cuzco region and the Ayamara) and did not reach into the Inca homeland. Research in the Province of Paruro, however, recovered a ceramic style called Muyu Oreo, which I believe reflects Ti- wanaku influence in the Cuzco region. The classification of Muyu Oreo ceramics as a Tiwanaku-related style is based on the following observations. First, Muyu Oreo ceramics are paint- ed with bright black, white, and orange figures over a dark red background. Although these colors are found in Wari and Wari-related collections, they are a hallmark of Tiwanaku ceramics. Second, ves- sel attributes also indicate a relation between Muyu Oreo and Tiwanaku ceramics. For example, several scallops have been found among the Muyu Oreo fragments. Such scallops frequently appear on the rims of Tiwanaku bowls but are rarely found on Wari ceramics. Third, one of the Muyu Oreo ves- sels is a banded drinking vessel and another is a steep-sided bowl standing on a raised base. Both of these vessel forms are common components of Tiwanaku collections (although they do occur, less frequently, in Wari collections). Furthermore, I suggest that the colors and ves- sel forms of Muyu Oreo ceramics recovered thus far suggest a late Tiwanaku (Tiwanaku V) rather than an early Tiwanaku (Tiwanaku HI) influence in the Cuzco region. This suggestion is supported by the understanding that late Tiwanaku ceramics are associated with the expansion of the state from the Lake Titicaca region, whereas early-style Ti- wanaku is more closely associated with the period of state formation. Late Tiwanaku ceramic pro- duction is thought to have begun around a.d. 750 and to have ended around a.d. 1000. If Muyu Oreo is a Tiwanaku-related style, then these dates can be tentatively used to suggest a possible pro- duction period for Muyu Oreo ceramics, until more concrete data become available. It should be stressed that Muyu Oreo ceramics differ significantly in paste and design motifs from classic Tiwanaku V ceramics. These differ- ences suggest that Muyu Oreo ceramics were not imported into the Cuzco region from the Tiwa- naku heartland. The similarities that exist between Muyu Oreo ceramics and late Tiwanaku ceramics may indicate that Tiwanaku influence spread from the Lake Titicaca Basin into the Cuzco region. Although the Wari influence in the Cuzco re- gion is clearly established, the influence of Ti- wanaku on Cuzco is still tentative. Because ar- 84 Chapter Three 0 10 20 1 i i Contour lines every 400 m Map 3-6. Distribution of sites with horizontally incised drinking vessels. chaeological research began in the south central highlands, evidence for early Cuzco-Tiwanaku interactions has been suggested on several occa- sions but has been proved unreliable after further research. Most notably, fragments of Wari and Wari-style ceramics in Cuzco have been misiden- tified as Tiwanaku. I encourage additional re- search on Muyu Oreo ceramics to test the prop- osition that they reflect a Tiwanaku rather than a Wari influence in the region. Ceramics of the Middle Horizon 85 B A » | BLACK 11 RED/ORANGE □ WHITE INCISED 2 3 4 5CM Drawing 3-1. Incurving bowls from Muyu Roqo. 86 Chapter Three I YELLOW/ORANGE | BLACK M RED/ORANGE □ WHITE | GRAY 0 12 3 4 5CM Drawing 3-2. Straight-sided, flaring bowls from Muyu Roqo. Ceramics of the Middle Horizon 87 YELLOW/ORANGE BLACK RED/ORANGE Drawing 3-3. Straight-sided, flaring bowls from Muyu Roqo. ] WHITE HH GRAY 012345 CM 88 Chapter Three YELLOW/ORANGE J BLACK §§ RED/ORANGE Drawing 3-4. Straight-sided, flaring bowls from Muyu Roqo. ] WHITE HH GRAY 01234 5CM Ceramics of the Middle Horizon 89 B YELLOW/ORANGE BLACK HI RED/ORANGE ] WHITE |j|| GRAY 0 12 3 4 5CM Drawing 3-5. Orange-ware drinking vessels from Muyu Roqo. 90 Chapter Three B □ WHITE ; '! GRAY 01234 5CM YELLOW/ORANGE | BLACK P RED/ORANGE Drawing 3-6. Orange-ware drinking vessels from Muyu Roqo. Ceramics of the Middle Horizon 91 B L 1 YELLOW/ORANGE "I WHITE &M GRAY i BLACK |_ H! RED/ORANGE 0 1 2__3_4_SCM Drawing 3-7. Orange-ware drinking vessels from Muyu Roqo. 92 Chapter Three YELLOW/ORANGE BLACK PH RED/ORANGE ] WHITE \ ] GRAY 0 1 2 3 4 5CM Drawing 3-8. Orange-ware drinking vessels from Muyu Roqo. Ceramics of the Middle Horizon 93 A L B 0 12 3 4 5CM 3 INCISED □ WHITE Drawing 3-9. Black-ware incised drinking vessels from Muyu Roqo. 94 Chapter Three B 1 D i_ V 0r-L2345CM INCISED Drawing 3-10. Black- ware incised drinking vessels from Muyu Roqo. Ceramics of the Middle Horizon 95 B 0 12 3 4 5CM INCISED Drawing 3-11. Black- ware incised drinking vessels from Muyu Roqo. 96 Chapter Three 1 B / YELLOW/ORANGE BLACK RED/ORANGE 1 WHITE IM GRAY 0 12 3 4 5CM Drawing 3-12. Black-ware drinking vessels from Muyu Roqo. Ceramics of the Middle Horizon 97 BLACK RED 0 12 3 4 5CM Drawing 3-13. Arahuay bowl designs (courtesy of Nilo Torres Poblete and Luis Barreda Murillo). 98 Chapter Three B / 7 I BLACK m RED 0 12 3 4 5CM Drawing 3-14. Arahuay bowl designs. Ceramics of the Middle Horizon 99 B BLACK RED 0 12 3 4 5CM Drawing 3-15. Common Qotakalli designs. 100 Chapter Three B | BLACK H RED 0 12 3 4 5CM Drawing 3-16. Common Qotakalli designs. Ceramics of the Middle Horizon 101 B -» BLACK 0 12345CM ' V;" iiRED Drawing 3-17. Monochrome Qotakalli ceramics. 102 Chapter Three ^ B ? BLACK RED 0 12 3 4 5CM Drawing 3-18. Monochrome Qotakalli ceramics. Ceramics of the Middle Horizon 103 B ■tf BLACK q 1 2 3 4 5CM RED Drawing 3-19. Qotakalli bases from steep-sided bowls. 104 Chapter Three B (black 51 RED 0 12 3 4 5CM Drawing 3-20. Qotakalli rims from steep-sided bowls. Ceramics of the Middle Horizon 105 1 B L 7 BLACK pi 2 3 4 5 C M m red Drawing 3-21. Qotakalli rims from steep-sided bowls. 106 Chapter Three wm red BLACK □ WHITE 012345 CM Drawing 3-22. Ccoipa straight-sided drinking vessels. Ceramics of the Middle Horizon 107 A L B BLACK 0 12 3 4 5CM IBM H red Drawing 3-23. Ccoipa straight-sided drinking vessels. 108 Chapter Three B i J BLACK 0 12 3 4 5CM ■— Drawing 3-24. Ccoipa straight-sided drinking vessels. Ceramics of the Middle Horizon 109 ^^^ 1 B 7 BLACK 01234 5CM II RED Drawing 3-25. Ccoipa straight-sided drinking vessels. 110 Chapter Three / B BLACK 0 12 3 4 5CM Drawing 3-26. Ccoipa bowls. Ceramics of the Middle Horizon 111 A W$i& B I I J 012345 CM | BLACK II RED Drawing 3-27. Common Ccoipa designs. 112 Chapter Three B H Drawing 3-28. Common Ccoipa designs. Ceramics of the Middle Horizon 113 012345CM Drawing 3-29. Ccoipa bowl. 114 Chapter Three Q_1 2 3 A ftr.M HBLACK OR BROWN |||ORANGE Bred Q white Qnatural Drawing 3-30. Muyu Oreo ceramics. Ceramics of the Middle Horizon 1 15 IBLACK □ WHITE Hred I gray [orange 0 12 3 4 5cm (natural Drawing 3-31. Muyu Oreo ceramics. 1 1 6 Chapter Three Drawing 3-32. Muyu Oreo straight-sided drinking vessel with front face god motif. Drawing 3-33. Muyu Oreo straight-sided drinking vessel with a convex, horizontal band. Ceramics of the Middle Horizon 1 17 0 12 3 4 5CM ■ black I | WHITE PInatural Drawing 3-34. Muyu Oreo slightly concave, steep-sided bowl. Drawing 3-35. Muyu Oreo annulated bowl with concave sides. 118 Chapter Three B 0 1 2 3 4 5CM Drawing 3-36. Horizontally incised drinking vessels. Ceramics of the Middle Horizon 119 012345 CM Drawing 3-37. Horizontally incised drinking vessels. 120 Chapter Three B 0 12 3 4 5CM Drawing 3-38. Horizontally incised drinking vessels. Ceramics of the Middle Horizon 121 Ceramics of the Early Intermediate Period and the Early Horizon (1000 b.c.-a.d. 550) The era before Wari expansion and after the beginnings of ceramic production in the Cuzco region (ca. 1000 B.C.) is traditionally di- vided into two time periods: the Early Interme- diate period (400 b.c.-a.d. 550) and the Early Ho- rizon (1000 b.c-400 B.C.). In the first section of this chapter, I review the ceramic styles of these two time periods for the Cuzco region: Huaru, Pucara, Chanapata, Derived Chanapata, and Mar- cavalle. In later sections, the survey and excava- tion findings for what are called "Paruro Forma- tive" ceramics are presented, and information is provided on the distribution of incised, early Ti- wanaku-related vessels. The Early Intermediate period ceramic styles are the least understood styles of the Cuzco re- gion. The period preceding Wari influence in the Cuzco region is associated, in the traditional ce- ramic sequence, with a style called "Huaru" (or Waru). This style was identified in 1952 by Cha- vez Ballon during his excavations at Batan Oreo. Huaru ceramics are thought to resemble a style of ceramics called "Carmenca," identified by Rowe (1944, pp. 19-20) in the early 1940s at the site of Chanapata (Rowe, 1956, p. 142). Huaru ma- terials remain virtually unknown, however, be- cause neither a description nor a single illustration of Huaru ceramics has been published (see Ap- pendix 1). During the Early Intermediate period, the site of Pucara, located approximately 200 km south- east of Cuzco in the Lake Titicaca Basin, reached its largest area of influence as the center for one of the earliest complex societies in the southern Andes (Kolata, 1993, pp. 70-78). Pucara ceram- ics, characterized by incised vessels with complex yellow and black figures painted on a dark red background (Kidder, 1943; Rowe & Brandel, 1971) have been found in the upper Vilcanota and Apurimac River drainages. Reconnaissance by Nunez del Prado Bejar (1972), S. Chavez (1988), and Lantaron Pfoccori (1988) recovered examples of Pucara ceramics throughout the Province of Chumbivilcas, some 75 km south of Cuzco. Most recently, Zapata's (pers. comm. 1994) excavations at Batan Oreo have provided clear examples of Pucara ceramics. The completion of his work will most certainly change our views on the relations between the Cuzco and Lake Titicaca regions dur- ing this early period of cultural development in the Andes. Chanapata and Derived Chanapata Ceramics The next ceramic styles in the standard Cuzco sequence, following that of Huaru, are called "Chanapata" and "Derived Chanapata." The di- vision between these two related styles is based on the relative frequency of fine black and red wares in the collections. The earlier Chanapata materials are marked by a predominance of fine Ceramics of the Early Intermediate Period and Early Horizon 123 black wares. This situation is reversed in the later Derived Chanapata collections, in which there is a higher percentage of fine red wares. Chanapata ceramics were first identified by Rowe (1943, 1944, pp. 10-23) during his early 1940s research in Cuzco. The style is named after the site of Chanapata, located just north of Cuzco in the suburb of Santa Ana, where Rowe exca- vated in 1942. During his initial Cuzco research, Rowe found examples of Chanapata ceramics at three other sites, two in the Cuzco Valley (Picchu and Limpillay [Wimpillay]) and one near the community of Maras (Pacallamocco), 30 aerial ki- lometers from Cuzco. Rowe's 1942 work at Chanapata was remark- able because it defined the first pre-Inca ceramic type of the Cuzco region. The excavations recov- ered a large sample of Chanapata materials. Rowe writes: The pottery characteristic of the pre-Inca levels at Chan- apata is so distinctive that even its plain sherds can be readily recognized when not too worn. The bulk of it is cooking ware, more or less globular ollas, plain or dec- orated with punctuate incision or pattern burnishing. The finely made pottery is mostly plate and bowl shapes with a polished surface, and decoration in incision or applied modeling (adornos). A small percentage has rather sim- ple painted decoration, either by itself or combined with incision. (Rowe, 1944, p. 15). The surface treatment of Chanapata ceramics includes burnishing and incising. The intensity of the burnishing can vary from vessels that have only a few narrow strokes on the exterior to items that have been carefully polished inside and out. The most common decoration found on Chana- pata vessels consists of punctations around the base of the neck or around the bulge of globular bodies. Fine incisions are also common designs and generally form simple straight-line geometric patterns, although some cases of curved lines have been found (Rowe, 1944, p. 16). The excavations at Chanapata yielded examples of both black and red wares. Using decoration and ware-type characteristics, Rowe divided his Chan- apata collection into several subtypes: plain, in- cised plain, punctated, pattern burnished, polished black, incised black, polished red, and incised red. Examples of painted ceramics were also recov- ered at Chanapata. However, because painted pieces were more numerous at the site of Pacalla- mocco, Rowe classed these fragments as Pacalla- mocco white on red and Pacallamocco red on buff. Chanapata ceramics were also found to con- tain a variety of modeled animal and human fig- ures as well as small figurines on their handles and other exterior surfaces. These aspects of Chanapata collections have been studied by Jorge Yabar Moreno (1972, 1982) and Dwyer (1972). As one of the earliest sites in the Cuzco Valley, Chanapata has received considerable archaeolog- ical attention. In 1960, Chavez Ballon excavated there with members of the Tokyo University Sci- entific Expedition to the Andes. Additional studies at Chanapata have been conducted by Jorge Yabar Moreno (1959, 1972, 1982), Frederick Engel, Barreda Murillo, and various other faculty and students of the Universidad San Antonio Abad del Cuzco. The site has been destroyed by Cuzco's expansion. Continued research by Rowe in the Cuzco re- gion in the mid-1950s identified several Chana- pata sites that contained a greater frequency of fine red ware than of black ware. These findings caused Rowe to review his data from Chanapata, and he found that there was a decrease in the fre- quency of polished black ware from the bottom to the top of the excavations. Rowe (1956, p. 143) writes, "This situation suggests that the sites with red fired ware only are later than the main occu- pation at Chanapata. We gave the name 'Derived Chanapata' to the newly identified red phase. . . ."' Radiocarbon dates from the sites of Chan- apata and Marcavalle support the proposition that there is a shift from black to red wares through time in the Chanapata sequence. Despite this shift, many of the vessel forms for Derived Chanapata appear similar to those found in Chanapata. One important exception to this statement has been found by Patterson (1967, p. 143), who notes that perhaps the most typical features of Derived Chanapata collections "are thickened rims on en- closed vessels and thick beveled or flat rims on open plates and bowls." Because Rowe's initial report on the archae- ology of Cuzco contained an extensive description and numerous drawings of Chanapata ceramics, this style is relatively well understood and has been found at various sites in the Cuzco region. Chanapata ceramics have been identified by Cha- vez Ballon (pers. comm. 1992), Barreda Murillo (pers. comm. 1992), and Zapata (pers. comm. 1992) in their separate excavations at Batan Urco, and by Dwyer (1972) at Minas Pata. Work at the site of Marcavalle by Chavez Ballon in 1953, Bar- 1 Derived Chanapata is at times called Pacallamocco (Patterson, 1967, p. 143). 124 Chapter Four reda Murillo and Lyon in 1963, and K. Chavez between 1966 and 1968 has also yielded numer- ous examples of Chanapata and Derived Chana- pata ceramics. These styles have also been found at Wimpillay (Valencia Zegarra, pers. comm. 1992), at Pisac (Gibaja Oviedo, pers. comm. 1992), on the Pampa de Anta, and at Pacallamoc- co (Zapata, pers. comm. 1992), as well as at var- ious sites in the Cusichaca region (Kendall, 1976, 1982; Hey, 1984; Lunt, 1984) and the Lucre Basin (McEwan, 1987, pp. 9-21). Dating Chanapata and Derived Chanapata Ceramics Two dates are available for Derived Chanapata materials from K. Chavez's (1980, p. 241) exca- vations at the site of Marcavalle (Lawn, 1971, p. 373). The first is 2131 ±55 b.p. ([P 1560] 181 ± 55 B.C.) and the second is 2096 ± 51 b.p. ([P 1561] 146 ± 51 B.C.). There are several radiocarbon dates from ex- cavation levels with Chanapata ceramics. Work done in 1960 by Chavez Ballon at Chanapata fur- nished two dates (Yamasaki et al., 1966, p. 337). The first sample dated to 2520 ± 150 b.p. ([N 89] 570 ± 150 B.C.). The second yielded the less use- ful date of 2360 ± 760 b.p. ([N 90] 410 ± 760 B.C.), owing to a small sample size. Another sam- ple (Gak ?) from the site is mentioned by Patter- son (1967, p. 143) as dating to 2600 ± 150 b.p. (650 ±150 B.C.). A sample submitted by Engel (Krueger & Weeks, 1966, p. 155) from Chanapata provided a date of 3330 ± 240 b.p. ([GX 203] 1380 ± 240 B.C.), which seems too early, in light of the other dates from the site. Kendall's work in the Cusichaca region has supplied another sample with a date of 2380 ± 70 b.p. ([BM 1633] 414 ± 70 B.C.) from the site of Huillca Raccay (Burleigh et al., 1983). General time estimates for the production of Chanapata and Derived Chanapata ceramics in the Cuzco region can be offered from these limited dates. It seems that Chanapata production began around 700 or 800 B.C. Derived Chanapata may have started about 500 years later, around 300 B.C. The duration of Derived Chanapata is difficult to estimate, because there are no dates from deposits with Huaru ceramics, or any other Early Inter- mediate period ceramic style, from the Cuzco re- gion. Marcavalle Ceramics The site of Marcavalle is located between Cuz- co and San Sebasian on the west bank of the Cachimayu River. The prehistoric occupations at this site were first noted by Chavez Ballon and Yabar Moreno in 1949. Chavez Ballon and Rowe conducted surface collections there in 1954 (K. Chavez, 1980, p. 211). In his 1956 article, Rowe concluded correctly that the site contained Chan- apata as well as pre-Chanapata remains. The pre- Chanapata ceramics of the site are now called Marcavalle. The first excavations at Marcavalle were di- rected by Barreda Murillo and Lyon in 1963 and 1964. The preliminary results of that work are presented in Barreda Murillo (1973), along with the first detailed description of the ceramics from the site. K. Chavez (1980, p. 213) reports that various other surface collections and excavations were conducted at the site during the 1960s and 1970s by members of the Cuzco academic com- munity, including Leandro Zans Candia, Jorge Yabar Moreno, Maria Luisa Nunez del Prado de Guzman, and other professionals such as Edward P. Lanning, John Rowe, Tom Patterson, and Fred- eric Engel. Despite its unique position in the cul- tural history of the Cuzco region, the site is now nearly destroyed. Valencia Zegarra and Gibaja Ovieda (1991) have chronicled the destruction of Marcavalle and have pleaded for its preservation. The largest study at Marcavalle was supervised by K. Chavez from 1966 through 1968. Her re- search is described in a number of publications (K. Chavez, 1977, 1980, 1981a, 1981b, 1982). On the basis of extensive attribute study, K. Chavez defined 10 major vessel forms (ollas, bowls, square bowls, jars, shallow bowls, carinated bowls, spouted bottles, oval bowls, incurved bowls, and double bowls) and divided the Mar- cavalle collections into four phases (A through D). She also provided descriptions of 8 surface finishes, 16 paste-temper groups, and numerous decorative techniques (paint [cream on brown, black on cream, plum red on cream], specular he- matite coloring, punctations, grooves, incisions, fillets, pattern burnishing, and zoomorphic mo- tifs). As K. Chavez (1980, p. 204) notes, Marcavalle ceramics precede Chanapata ceramics, represent- ing not only the earliest pottery-using inhabitants of the Cuzco Valley but also, according to current research, the first occupants of the area. Unlike Chanapata, Marcavalle ceramics have not been Ceramics of the Early Intermediate Period and Early Horizon 125 foun i elsewhere in the Cuzco region and appear to be limited to the valley. As the earliest known ceramic style in the Cuzco region, considerable effort has been made to date Marcavalle. A car- bon sample (GX 0453) from Barreda Murillo's and Lyon's 1963 excavations yielded a date of 695 ± 115 B.C. (Patterson, 1967, p. 143; Lawn, 1971, p. 373). 2 K. Chavez furnished five dates for Marcavalle ceramics: 2571 ± 45 b.p. ([P 1562] 621 ± 45 B.C.), 2661 ± 46 b.p. ([P 1563] 711 ± 46 B.C.), 2685 ± 49 b.p. ([P 1564] 735 ± 49 B.C.), 2860 ± 47 b.p. ([P 1566] 910 ± 47 b.c), and 2916 ± 55 b.p. ([P 1567] 966 ± 55 B.C.). These dates indicate that the production of Marcavalle ceramics began around 1000 b.c. and continued until about 700 B.C. Paruro Formative Ceramics Marcavalle ceramics are recognized as the ear- liest form of pottery in the Cuzco region and have been dated from 1000 b.c. until 700 b.c. by K. Chavez (1982, p. 2). Chanapata ceramics, and a broad spectrum of related ceramics called Derived Chanapata, follow Marcavalle. The early ceramics of the Province of Paruro share closer similarities to Derived Chanapata ceramics, as described by Rowe (1944, 1956), than to Chanapata or Mar- cavalle ceramics. Nevertheless, it should be noted that the early Paruro ceramics are not identical to the Derived Chanapata ceramics found in the Cuzco Valley. Accordingly, they will be classified as "Paruro Formative" ceramics. Ware The formative ceramics from Paruro contain a coarse fabric with a substantial quantity of white nonplastic inclusions that vary greatly in size from 0.01-0.30 cm. There is a considerable pres- ence of gold and black mica as well as numerous matte, black inclusions. The ware is medium soft in hardness and fires to orange, red, or, more rare- ly, black. 2K. Chavez (1980:214) reports that two additional dates (1-3093 and 1-3094) were obtained by Engel at the site in 1966. Design Elements, Color, and Surface Treatment The formative ceramics of Paruro tend to be extensively burnished on their exteriors, yet the quality of the burnishing varies. Some pieces are casually burnished and present a loose matrix of vertical and horizontal lines, while others are ex- tensively burnished with their exteriors retaining a medium gloss. Even among the finest pieces, however, individual burnish strokes can be iden- tified. Some of the Formative fragments recovered in Paruro contain the remains of a dark red paint, and the possible remains of white pigment have been noted on a few additional pieces. Fine inci- sions and other carvings appear on the exteriors of vessels, and in a few cases, punctations were noted along the neck of the jars. Vessel Forms Because no whole or reconstructable formative vessel was recovered during research in the Prov- ince of Paruro, it is difficult to discuss specific vessel forms. From the available fragments, it ap- pears that the most common vessels were globular and were produced with a wide variety of shoul- der, neck, and rim forms (Drawing 4-1). Many of these globular forms contain short flaring rims with distinct points of inflection (Drawing 4-2). It is not uncommon for the interior lips of these short flaring rims to be flattened (Drawing 4-2A- C). Other vessels contained long, slightly flaring necks (Drawings 4-3 and 4-4A) or short abbre- viated necks (Drawing 4-4B-C). Small incurving bowls have been found (Draw- ing 4-4E), as have numerous examples of large, straight-sided flaring bowls. The lip of some of these flaring bowls has been pinched 1 or 2 cm from the rim (Drawing 4-5A-C), while others have a bulbous lip that has been flattened on the interior (Drawing 4-5D-F). More rarely, the entire surface of the interior lip may contain elaborately incised designs (Drawing 4-6). These straight-sid- ed, flaring bowls are similar to the Derived Chan- apata vessels described by Patterson (1967, p. 143) as open plates and bowls with beveled or flat rims. There is also a wide variety of bowls with ver- tical or straight flaring sides in the Paruro For- mative collections (Drawing 4-7). The interiors of some of these bowls are incised (Drawing 4-7B- 126 Chapter Four Fig. 4-1. Paruro Formative vessel with molded face. C) or carved (Drawing 4-7E), and others have raised bands (Drawing 4-7 A). The lips of several bowls are recovered with oval impressions (Draw- ing 4-7D), and some bowls have lug handles in- cised with short parallel lines (Drawing 4-8A-B). Flat slab figurines, long noted among Chana- pata collections in the Cuzco region (Dwyer, 1972; Yabar Moreno, 1982), have also been re- covered in the Paruro region (Drawing 4-8C). Furthermore, some vessel necks were found to be modeled with human faces that have coffee bean- like eyes and raised ears and mouth (Fig. 4-1, Drawings 4-9 and 4-10). be located on the upper slopes of the valleys, fre- quently on hilltops, knolls, promontories, and the ends of ridges. These findings are not inconsistent with those of other researchers working elsewhere in the Cuzco region. For example, the largest Chanapata-related site in the Cusichaca region, Huillca Raccay, is located on a high promontory (Hey, 1984), and the site of Chanapata is on a small knoll well off the floor of the Cuzco Valley. It should also be noted that there is considerable variation in the vessel forms and attributes within the Paruro Formative materials. This suggests that there may have been a number of local areas of ceramic production during this period. Distribution and Dating Although there are currently no radiocarbon dates for the Paruro Formative materials, the sim- ilarities they share with Derived Chanapata ce- ramics of the Cuzco Valley suggest that they were produced at approximately the same time. For- mative ceramics have been found at 31 sites in the research region (Map 4-1). These sites tend to Incised, Early Tiwanaku-Related Vessels Muyu Oreo ceramics are not the first Tiwa- naku-related ceramics to be identified in the Cuz- co region. K. Chavez (1985) has noted the wide- spread distribution of incised incensarios frag- Ceramics of the Early Intermediate Period and Early Horizon 127 574- 528-^? 203 0 10 20 Contour lines every 400 m Map 4-1. Distribution of sites with Paruro Formative ceramics. 128 Chapter Four 0 10 20 i i i Contour lines every 400 m Map 4-2. Distribution of sites with incised, early Tiwanaku-related ceramics. Ceramics of the Early Intermediate Period and Early Horizon 129 . I* I 1 pi , Fig. 4-2. Incised, early Tiwanaku-related ceramics. ments, or ceremonial burners, at sites between the Cuzco Valley and Lake Titicaca.3 Chavez's work documents what appears to be early Tiwanaku in- fluence in the Cuzco region. Systematic surface collections and excavations in the Province of Paruro recovered fragments of incised, early Tiwanaku-related ceramics (Draw- ing 4-11) from five sites (Map 4-2)4. Of these sites, Muyu Oreo (536), which is situated less than 15 aerial kilometers from the city of Cuzco, yielded the largest number of fragments. This fact is particularly interesting because this site also provided the largest collection of Muyu Oreo ce- ramics, which are believed to reflect late Tiwa- naku influence in the region. The designs on the incised, early Tiwanaku-re- lated vessels tend to be geometric (Fig. 4-2). Cir- cles are made with a hollow tool, which leaves a raised, solid area in their centers (K. Chavez, 1985, p. 138). The lines are made by dragging a 3 For an extensive description of these vessels see K. Chavez (1985). 4 K. Chavez (1985) reports finding two fragments of incised ceremonial burners at Tejahuaci (496). U-shaped tool across the vessel's surface. Their exteriors are roughly burnished, whereas the in- teriors have simply been smoothed. Two of the fragments appear to be rim scallops (Drawing 4- 1 1F-G), one of which is decorated with a series of circles (Drawing 4-1 1G), while another con- tains a remnant of a Tiwanaku puma motif (Draw- ing 4-1 IF). A third is a rim fragment with a series of small circles (Drawing 4-11 A). The other ex- amples display complex series of circles, lines, and dots (Drawing 4-11B-E). K. Chavez (1985) suggests that incised incen- sarios were produced during the period of early Tiwanaku development (or Tiwanaku III [a.d. 100-400]; Kolata, 1993, p. 78). She states that the fragments she recovered — which are similar to those found during the survey and excavation work in the Province of Paruro — are Tiwanaku related. Nevertheless, based on her study of their wares, vessel shapes, and motifs, she concludes that these incised incensarios do not represent im- ports from the Tiwanaku heartland but instead re- flect stylistic influence on ceramic production that traveled northward from the Lake Titicaca region. 130 Chapter Four V r 012345CM Drawing 4-1. Paruro Formative vessel rims. Ceramics of the Early Intermediate Period and Early Horizon 1 3 1 01234 5CM r Drawing 4-2. Paruro Formative vessels with short flaring rims. 132 Chapter Four ^^iu^- Drawing 4-3. Paruro Formative vessels with long, slightly flaring necks. Ceramics of the Early Intermediate Period and Early Horizon 133 ^ 012345 CM Drawing 4-4. Paruro Formative vessel rims. 134 Chapter Four 0 12 3 4 5CM f f t Drawing 4-5. Paruro Formative flaring bowls. Ceramics of the Early Intermediate Period and Early Horizon 135 0 12 3 4 5CM Drawing 4-6. Paruro Formative flaring bowl. 136 Chapter Four I i Drawing 4-7. Paruro Formative flaring bowls. Ceramics of the Early Intermediate Period and Early Horizon 137 f 012345 CM Drawing 4-8. Paruro Formative bowls and figurine. 138 Chapter Four 012345 CM Drawing 4-9. Paruro Formative vessels with modeled faces. Ceramics of the Early Intermediate Period and Early Horizon 139 012345CM Drawing 4-10. Paruro Formative vessel with modeled face. 140 Chapter Four A L I 1 1 12 3 4 5CM Drawing 4-11. Incised, early Tiwanaku-related ceramics. Ceramics of the Early Intermediate Period and Early Horizon 141 A Revised Ceramic Sequence for the Cuzco Region Ceramic chronologies are critical tools that ar- chaeologists use in examining prehistoric and historic cultural developments. Relatively lit- tle archaeological research can be conducted with- out them. The first ceramic sequence for the Cuz- co region was proposed by Rowe in 1956, and this general sequence divided the pre-Hispanic pe- riod of the region into seven broad temporal-ce- ramic classifications. Although much of the cur- rent archaeological research in the Cuzco region has focused on describing and analyzing its pre- Inca cultures, few new ceramic styles have been formally proposed. The lack of descriptions and illustrations for many of the Cuzco styles limits archaeological research in the region. Because few published sources exist for Cuzco ceramics, each researcher must begin anew in identifying, defining, and describing the styles of the region. In this study, a ceramic sequence for the Prov- ince of Paruro is developed and combined with the results of research conducted elsewhere in the Cuzco region (Fig. 5-1). The resulting revised ce- ramic sequence for the Cuzco region spans more than 2,000 years. It begins with the rustic wares and vessels of the Early Horizon and Early Inter- mediate period and ends with the spectacular ce- ramics of the Inca Empire. The earliest ceramic style so far identified is Marcavalle, which is found at only one site in the Cuzco Valley. Radiocarbon dates indicate that Marcavalle ceramics were produced between 1000 and 700 B.C. Further work in the Cuzco Val- ley may push the origins of Marcavalle ceramics further back in time or discover a precursor to them. The Chanapata, Derived Chanapata and related wares, such as the Formative ceramics from Pa- ruro, which follow Marcavalle in the sequence, are spread widely across the Cuzco region. Most likely there were many small production centers for these ceramic styles throughout the region, each producing vaguely similar products. Exten- sive ware analyses are needed to better understand the distribution of these early styles. The ceramics of the Early Intermediate period are poorly understood. The production of some Derived Chanapata styles may have continued throughout this period or may have been replaced by the production of currently unidentified styles. The ceramics of the Cuzco region may have been influenced by outside polities during the Early In- termediate period. It has been suggested that in- cised, early Tiwanaku-related wares were brought into or produced in the region during this period. Furthermore, it is likely that Pucara ceramics en- tered the Cuzco region by trade at this time. During the Middle Horizon, two, perhaps com- peting, polities controlled the central and south central Andean highlands. The Wari state expand- ed from the Ayacucho region of Peru between a.d. 550 and 900. The Tiwanaku polity, centered in the Lake Titicaca region, began to expand around a.d. 500 and collapsed about the same time as Wari. During this period of state devel- opment in the Andes, the Cuzco region witnessed a florescence of imported and locally produced ceramics. Cuzco lies between the highland centers of Wari and Tiwanaku. The Cuzco region was oc- cupied by the Wari from around a.d. 600 to the A Revised Ceramic Sequence for the Cuzco Region 143 CD "B E a> CD CO .32 ■o ■o CO LU o3 LU AD 1600 1500 1400 1300 1200 1100 1000 900 800 700 600 500 400 300 200 100 r 0« to * l> «-■ =1 100 200 300 400 500 600 700 800 900 ^2| !!» Fl -^> LU I Spanish Invasion Wari Abandonment of Pikillacta Wari Founding of Pikillacta 'LH = 1000 BC Late Horizon Fig. 5-1. Revised ceramic sequence for the Cuzco region. collapse of the empire. Pikillacta, one of the larg- est sites in the south central highlands, was built by the Wari in the Lucre Basin, some 35 km southeast of Cuzco. The construction of this site represented an enormous investment of time and labor for the Wari. It is thought to have been the Wari administrative center for the southern region of their domain. Nevertheless, the effects that the centuries-long Wari occupation had on the local Cuzco-based populations, and the nature and ex- tent of Wari control in the region, remain to be studied. Researchers have reported finding small num- bers of Chakipampa and Ocros vessels in the Cuz- co region. These vessels represent imports from the Ayacucho region. Wari-style ceramics were, 144 Chapter Five however, also produced in the Cuzco region dur- ing the Middle Horizon. Wari-style ceramics closely imitate ceramics from the Wari homeland but are locally produced. Ocros-style bowls have, for example, been found at several sites in the Cuzco region and are frequently used to identify sites occupied during the period of Wari control. There are, however, relatively few sites with these ceramic styles in the Paruro research region. The few sites that do contain Wari and Wari-style ce- ramics are relatively close to the Lucre Basin in the Paruro River Valley. These findings suggest that Wari influence was concentrated in the Lucre Basin and did not extend a significant distance to the south. It has been long recognized that there were a number of ceramic styles produced in the Cuzco region during the Middle Horizon. One of the most important, newly identified styles used in the Cuzco region during the Middle Horizon is Ara- huay. This style is classified as Wari-related, be- cause it was influenced by Wari ceramic traditions, but it does not directly imitate ceramics from the Ayacucho region. It is characterized by the use of broad red bands outlined with narrow black lines. Arahuay pottery seems to have been produced during the Wari presence in the Cuzco region, since fragments have been recovered in Pikillacta and with Wari-related styles at Batan Oreo. Qotakalli is a distinct style with a cream slip and elaborate geometric designs. First identified in the Cuzco Valley in the early 1960s, it is now known to be relatively widespread. No radiocar- bon dates have been run for Qotakalli ceramics; however, excavations at Pikillacta and Batan Oreo have recovered Qotakalli in association with Wari-style fragments, indicating that Qotakalli was used during the period of Wari control. The Ccoipa ceramics style, a relatively rustic style fea- turing simple geometric designs in alternating red and black paint, seems to have been produced in the Paruro region during this same time period or slightly later. Despite the long period of Wari presence in the Cuzco region, Inca mythology suggests that the Inca associated themselves more closely with Ti- wanaku than with Wari. The ruined city of Ti- wanaku is mentioned specifically as a location of primeval importance in Inca myths, and Lake Ti- ticaca is identified as the origin place of the sun and the moon. This apparent "Tiwanaku affilia- tion" has been something of a quandary because there has been little evidence — unlike that for Wari — that this Lake Titicaca polity influenced the Cuzco region. However, it is becoming appar- ent that some contact may have existed between the Cuzco and Tiwanaku regions during the Mid- dle Horizon. Materials thought to be influenced by Tiwanaku III traditions have been found be- tween Tiwanaku and Cuzco. Research in the Province of Paruro recovered such fragments less than 15 aerial kilometers from the city of Cuzco. Furthermore, a new ceramic style called Muyu Oreo has been identified in the Province of Pa- ruro, and similar materials have been found in the Cuzco Valley. This style is characterized by bright black, white, and orange figures painted over a dark red background; these colors are traditionally associated with late Tiwanaku (Tiwanaku V) ce- ramics. Various vessel forms and vessel attributes found in the Muyu Oreo collections also suggest ties with Tiwanaku rather than with Wari. With the recovery of Muyu Oreo ceramics in the Cuzco region, it can be proposed that influence from the Titicaca area may have continued up to, and per- haps during, the Wari occupation. The fall of Wari and the concomitant abandon- ment of its administrative center in Pikillacta, in the Lucre Basin, perhaps as late as a.d. 900 or 1000 marked the beginning of autochthonous state development in the Cuzco region. By about a.d. 1400, the Inca had united the Cuzco region under their rule, and the city of Cuzco had emerged as its capital. Evidence from systematic regional sur- vey work in the Province of Paruro suggests that Killke ceramics were produced in the Cuzco Val- ley during this period of state formation and that they were traded in a region with a 60 km radius. Other ceramic styles were being produced in the Cuzco region during the Late Intermediate pe- riod. Colcha ceramics, manufactured south of the Apurimac River in the area of Araypallpa, repre- sent one such style. Colcha ceramics have been classified as Killke-related because they share a number of attributes, including geometric design elements and vessel forms, with Killke ceramics. As research continues in the Cuzco region, it is expected that additional Killke-related styles will be identified. It should be noted that the proposed chronolog- ical relations between many of the ceramic styles presented here are speculative and require future research. Some of the styles, such as Chanapata, Marcavalle, and Ocros-style, have been dated with several radiocarbon samples. The dates of others, such as Killke, Colcha, Ccoipa, and De- rived Chanapata, rest on three or fewer samples. A Revised Ceramic Sequence for the Cuzco Region 145 Qotakalli and Arahuay are dated stratigraphically in the excavations at the Wari sites of Pikillacta and Batan Oreo. The dates of still other ceramic materials, such as Muyu Oreo, are still poorly un- derstood, although their ages can be inferred from their style. Future research may require the redefinition of styles described here. Excavations at stratified sites and the recovery of additional radiocarbon dates are especially important in testing the pro- posed temporal affiliations. Because the pace of archaeological research in the Cuzco region is steadily increasing, there is little doubt that new observations will be made and new ceramic styles will be identified over the course of the next de- cade. These advances are necessary if we are to begin to understand the cultural history of the Cuzco region and the processes that led to the development of one of the greatest empires of the New World. 146 Chapter Five Literature Cited Albarracin-Jordan, J., and J. E. Mathews. 1990. Asentamientos Prehispanicos del Valle de Tiwanaku. Vol. 1. Producciones CIMA, La Pazi. Alcina Franch, J., M. Rivera, J. Galvan, C. Garcia Palacios, M. Guinea, B. Martinez-Caviro, L. J. Ra- mos, and T. Varela. 1976. Memorias de la Mision Cientffica Espanola en Hispanoamerica. Vol. 3. Ar- queologia de Chinchero; Ceramica y Otros Materiales. Ministerio de Asuntos Exteriores, Madrid. Archivo General de la Nacion, Archivo Agrario, and Miscelanea: Hacienda Larapa. 1596. Petition y petition aceptada de Catalina de Guevara Vda. de la Parroquia de San Sebastian. 1591-1596: folio 348- 349. Arnold, D. E. 1980. Localized exchange: An ethnoar- chaeological perspective, pp. 147-150. In Fry, R. E., ed., Models and Methods in Regional Exchange. SAA Papers 1. Society for American Archaeology, Wash- ington, D.C. Baca Jenaro F 1974. Motivos de Ornamentation de la Ceramica Inca Cusco, vol. 1. Libreria Studium S. A., Camana, Peru. . 1989. Motivos de Ornamentation de la Cer- amica Inca Cusco, vol. 2. Libreria Studium S. A., Ca- mana, Peru. Barreda Murillo, L. 1973. Las culturas Inka y pre- Inka de Cuzco. Thesis, Department of Archaeology, University of San Antonio Abad, Cuzco. . 1982. Asentamiento humano de los Qotakalli del Cuzco, pp. 13-21. In I. Oberti Rodriguez, comp. Arqueologia de Cuzco. Instituto Nacional de Cultura, Cuzco. . 1991. Historia y arqueologia del Qosqo pre- Inka. Revista Municipal del Qosqo, 1(2): 20-36. Bauer, B. S. 1989. Muyu Orqo y Ccoipa: Dos nuevos tipos de ceramica para la region del Cusco. Revista Andina, 7(2): 537-542. . 1990. State development in the Cusco region: Archaeological research on the Incas in the Province of Paruro. Ph.D. diss., Department of Anthropology, University of Chicago. . 1991. Pacariqtambo and the mythical origins of the Inca. Latin American Antiquity, 2(1): 7-26. -. 1992a. The Development of the Inca State. Uni- versity of Texas Press, Austin. 1992b. Investigaciones arqueologicas recientes en los asientos de Maukallaqta y Puma Orqo, depar- tamento del Cusco, Peru; pp. 67-108. In Bauer, B., ed., Avances en Arqueologia Andina. El Centra de Estudios Rurales Andinos "Bartolome de las Casas," Cuzco. Bauer, B. S., and C. Stanish. 1990. Killke and Killke- related pottery from Cuzco, Peru, in the Field Muse- um of Natural History. Fieldiana: Anthropology, n.s., 15. Bingham, H. 1913. In the wonderlands of Peru: The work accomplished by the Peruvian Expedition of 1912. National Geographic, 24: 387-573. . 1915. Types of Machu Picchu pottery. Ameri- can Anthropologist, 17(2): 257-271. . 1922. Inca Land: Explorations in the Highlands of Peru. Riverside Press, Cambridge, Massachusetts. Brewster-Wray, C. 1990. Moraduchayuq: An Admin- istrative Compound at the Site of Huari, Peru. Uni- versity Microfilms, Ann Arbor. Brundage, B. C. 1963. Empire of the Inca. University of Oklahoma Press, Norman. Burleigh, R., J. Ambers, and K. Matthews. 1983. Brit- ish Museum natural radiocarbon measurements XVI. Radiocarbon 25(1): 39-58. Cabello Balboa, M. Miscelanea Antartica, una Historia del Peru Antiguo. Edited by L. E. Valcarcel. Univer- sidad Nacional Mayor de San Marcos, Instituto de Et- nologfa, Lima, Peru. Calderari, M. 1991. Estilos ceramicos Incaicos de La Paya. pp. 151-164. In Actas del XX Congreso Na- cional de Arqueologia Chilena, vol. 2. Nacional de Historia Natural, Sociedad Chilena de Arqueologia, Santiago de Chile. Calderari, M. and V Williams. 1991. Re-evaluacion de los estilos ceramicos Incaicos en el Noroeste Ar- gentino. Revista Comechingonia, 9: 73-95. Chavez, K. L. M. 1977. Marcavalle: The ceramics from an Early Horizon Period site in the Valley of Cuzco, Peru, and implications for South Highland socio-eco- nomic interaction. Ph.D. diss., Department of Anthro- pology, University of Pennsylvania. . 1980. The archaeology of Marcavalle, an Early Horizon site in the Valley of Cuzco, Peru: Part I. Baessler-Archiv, n.E, 28(2): 203-329. -. 1981a. The archaeology of Marcavalle, an Early Horizon site in the Valley of Cuzco, Peru: Part II. Baessler-Archiv, n.F, 29(1): 107-205. 1981b. The archaeology of Marcavalle, an Ear- ly Horizon site in the Valley of Cuzco, Peru: Part III. Baessler-Archiv, n.F, 29(1): 241-386. -. 1982. Resumen de los trabajos en Marcavalle. pp. 1-8. In Rodriguez, I. O., ed., Arqueologia de Cuz- co. Instituto Nacional de Cultura, Cuzco. -. 1985. Early Tiahuanaco-related ceremonial burners from Cuzco, Peru, pp. 137-178. In Dialogo andino, no. 4. Departamento de Historia y Geograffa Universidad de Tarapaca, Arica, Chile. Chavez, S. J. 1985. Ofrendas funerarias dentro de los limites meridionales del territorio Huari en el Depar- Literature Cited 147 lamento del Cuzco, pp. 179-202. In Dialogo Andino, no. 4. Departamento de Historia y Geograffa Univer- sidad de Tarapaca, Arica, Chile. 1987. Funerary offerings from a Middle Hori- zon context in Pomacanchi, Cuzco. Nawpa Pacha, 22- 23(1984-95): 1-48. -. 1988. Archaeological reconnaissance in the province of Chumbivilcas, South Highland of Peru. Expedition 30(3): 27-38. Cieza de Leon, P. 1976. The Incas of Pedro Cieza de Leon [Part 1, 1553, and Part II, 1554]. H. de Onis, trans., and V. W. von Hagen, ed., University of Okla- homa Press, Norman. El Comercio. 1952a. Hasta 5 kilos de oro ha dado el riachuelo de Huaroy-mayo que tiene intrigado al Cuz- co. 6 August, Lima, [newspaper article] . 1952b. Motivo de investigacion son los hallaz- gos de Batan-Orco. 8 August, Lima. -. 1952c. Huaqueros de Batan-Orco fueron pues- tos en libertad. 10 August, Lima. 1952d. No hay oro en Batan-Orco. Tumba pre- hispanica profanada contenia objetos valiosos . . . In- forme del visitador general de monumentos historicos. 19 August, Lima. . 1952e. El Patronato arqueologico del Cuzco ex- cavara en Batan-Orco. 21 August, Lima. . 1952f. A pesar de todo, creese que Batan-Orco era un lavadero de oro. 26 August, Lima, [newspaper article] 1952g. Tumba Tiahuanacu se hallo en Batan- Orco. 16 September, Lima, [newspaper article] . 1952h. Informe de la Comision del Patronato de Arqueologfa que investigo en Batan-Orco. 19 Sep- tember, Lima, [newspaper article] -. 1952i. Conchas marinas y turquesas hallaron en las ultimas excavaciones de Batan-Orco. 21 Septem- ber, Lima, [newspaper article] 1952j. Laminas totemicas de oro se hallaron en Batan-Orco. 2 October, Lima, [newspaper article] . 1952k. El mamelon de Batan-Orco. 6 October, Lima, [newspaper article] Cook, A. G. 1992. The stone ancestors: Idioms of im- perial attire and rank among Huari figurines. Latin American Antiquity, 3(4): 341-364. D'Altroy, T. N. 1992. Provincial Power in the Inka Empire. Smithsonian Institution Press, Washington, D.C. D'Altroy, T. N., and R. L. Bishop. 1990. The provin- cial organization of Inka ceramic production. Ameri- can Antiquity, 55(1): 120-138. Dwyer, E. B. 1971. The early Inca occupation of the Valley of Cuzco, Peru. Ph.D. diss., Department of An- thropology, University of California, Berkeley. . 1972. A Chanapata figurine from Cuzco, Peru. Nawpa Pacha, 9: 33-40. Eaton, G. F. 1916. The collection of osteological ma- terial from Machu Picchu. Memoirs of the Connecti- cut Academy of Arts and Sciences, 5: 3-96. Engel, F. 1966. Geograffa humana prehistorica y agri- culture precolombina de la Quebrada de Chilca, Vol. I. Universidad Agraria, Lima. Espinoza Martinez, H. 1983. Evidencia cultural del Ho- rizonte Medio (Wari) Aqomoqo-Cuzco, pp. 16-22. In Gibaja Oviedo, A., ed., Arqueologfa Andina. Edi- ciones Instituto Nacional de Cultura, Cuzco. Fejos, P. 1944. Archaeological explorations in the Cor- dillera Vilcabamba. Viking Fund Publications in An- thropology, New York. Gade D. W., and M. Escobar Moscoso. 1982. Village settlement and the colonial legacy in southern Peru. Geographical Review, 72(4): 430-449. Gibaja Oviedo, A. M. 1973. Arqueologfa de Choque- pugio. Thesis, Department of Archaeology, University of San Antonio Abad, Cusco. -. 1983. Arqueologfa de Choquepugio, pp. 29-44. In Gibaja, A., ed., Arqueologfa Andina. Ediciones In- stituto Nacional de Cultura, Cuzco. Glowacki, M. 1996. The Wari occupation of the Southern Highlands of Peru: A ceramic perspective from the site of Pikillacta. University Microfilms, Ann Arbor. Gonzalez Corrales, J. A. 1984a. La arquitectura y cer- amica Killke del Cusco, pp. 189-204. In Kendall, A., ed., Current Archaeological Projects in the Central Andes. BAR International Series 210. British Archae- ological Reports, Oxford. . 1984b. Arquitectura y ceramica Killke del Cus- co. Revista de Museo e Instituto Arqueologico, 23: 37-46. Guaman Poma de Ayala, F. 1980. El Primer Nueva Coronica y Buen Gobierno [1584-1615]. Murra, J. V, and R. Adorno, eds., Urioste, J. I. trans. Siglo Vein- tiuno, Mexico City. Harth-Terre, E. 1959. Pikillacta — Ciudad de positos y bastimentos del imperio Incaico. Revista de Museo e Instituto Arqueologico, 18: 3-19. Heffernan, K. J. 1989. Limatambo in late prehistory: Landscape archaeology and documentary images of Inca presence in the periphery of Cusco. Ph.D. diss., Department of Prehistory and Anthropology, Austra- lian National University, Canberra. Hey, G. 1984. Early occupation on the Huillca Raccay promontory site, Cusichaca: The archaeological evi- dence, pp. 291-304. In Kendall, A., ed., Current Ar- chaeological Projects in the Central Andes. BAR In- ternational Series 210. British Archaeological Reports, Oxford. Hodder, I. 1974. Regression analysis of some trade and marketing patterns. World Archaeology, 6(2): 172- 189. . 1980. Trade and exchange: Definitions, identi- fication and function, pp. 151-156. In Fry, R. E., ed„ Models and Methods in Regional Exchange. SAA Pa- pers 1. Society for American Archaeology, Washing- ton, D.C. Hodder, I. and C. G. Orton. 1976. Spatial Analysis in Archaeology. Cambridge University Press, Cam- bridge. Hollowell, J. L. 1987. Precision cutting and fitting of stone in prehistoric Andean walls. Unpublished Re- search Report #2832-84, the National Geographic So- ciety, Washington, D.C. Jijon y Caamano, J. 1934. Los orfgenes del Cuzco. Im- prenta de la Universidad Central, Quito. 148 Literature Cited Jijon y Caamano, J., and C. Larrea. 1918. Un cemen- terio Incaico en Quito y notas acerca de los Incas en el Ecuador. Revista de la Sociedad Juridico-Literatria, 20: 159-260. Kendall, E. A. 1974. Aspects of Inca architecture. Ph.D. diss., Institute of Archaeology, University of London. -. 1976. Preliminary report on Ceramic data and the pre-Inca architectural remains of the (Lower) Uru- bamba Valley, Cuzco. Baessler Archiv, 24: 41-159. 1984. Archaeological investigations of Late In- termediate Period and Late Horizon period at Cusi- chaca, Peru, pp. 247-290, In Kendall, A., ed., Current Archaeological Projects in the Central Andes. BAR International Series 210. British Archaeological Re- ports, Oxford. -. 1985. Aspects of Inca architecture: Description, function and chronology, parts 1 and 2. BAR Inter- national Series 242, British Archaeological Reports, Oxford. Kidder, A. II. 1943. Some early sites in the Northern Lake Titicaca Basin. Papers of the Peabody Museum of American Archaeology and Ethnology. Vol. 27, no. 1. Knobloch, Patricia J. 1991. Stylistic date of ceramics from the Huari centers, pp. 247-258. In Huari Ad- ministrative Structure: Prehistoric Monumental Archi- tecture and State Government. Dumbarton Oaks, Washington D.C. Kolata, A. 1993. The Tiwanaku: Portrait of an Andean civilization. Blackwell, Oxford. Krueger, H. W., and F. C. Weeks. 1966. Geochron Lab- oratories, Inc., radiocarbon measurements II. Radio- carbon 8: 142-160. Lantaron Pfoccori, L. F. 1988. Prospeccion arqueolo- gica de la Provincia de Chumbivilcas. Thesis, Uni- versity of San Antonio Abad, Cuzco. Lawn, B. 1971. University of Pennsylvania radiocarbon dates XIV. Radiocarbon 13(2): 363-377. Liu, C. L., K. M. Riley, and D. D. Coleman. 1986. Illinois State Geological Survey radiocarbon dates VIII. Radiocarbon 28(1): 78-109. Lunt, S. W 1983. An introduction to the pottery from the excavations at Cusichaca, Department of Cuzco. Paper presented at the 44th Meeting of the Interna- tional Congress of Americanists, Manchester, En- gland. . 1984. An introduction to the pottery from the excavations at Cusichaca, Department of Cuzco, Peru, pp. 307-322. In Kendall, A., ed., Current Archaeolog- ical Projects in the Central Andes. BAR International Series 210. British Archaeological Reports, Oxford. . 1987. Inca and pre-Inca pottery: Pottery from Cusichaca, Department of Cuzco, Peru. Ph.D. diss., Institute of Archaeology, University of London. . 1988. The manufacture of the Inca aryballus, pp. 489-511. In Saunders, N. J., and O. de Mont- mollin, eds. Recent Studies in Pre-Columbian Archae- ology. BAR International Series 421 (ii). British Ar- chaeological Reports, Oxford. -, Lynch, T 1973. Current research, Andean . 1974. Current research, Andean South America. American Antiquity 39(2): 383-386. 1976. Current Research, Andean South Amer- ica. American Antiquity 41(2): 229-232. McEwan, G. E 1983. The Middle Horizon in the Valley of Cuzco, Peru: The impact of the Wari occupation of Pikillaqta in the Lucre Basin. Ph.D diss., Department of Anthropology, University of Texas at Austin. . 1984. Investigaciones en la cuenca del Lucre, Cusco. Gaceta Arqueologica Andina 9: 12-15. 1987. The Middle Horizon in the Valley of Cuz- co, Peru: The impact of the Wari occupation of the Lucre Basin. BAR International Series 372. British Archaeological Reports, Oxford. -. 1989. The Wari empire in the southern Peruvian highlands: A view from the provinces, pp. 53-71. In R. M. Czwarno, E M. Meddens, and A. Morgan, eds., The Nature of Wari: A Reappraisal of the Middle Ho- rizon Period in Peru. BAR International Series 525. British Archaeological Reports, Oxford. 1990. Architectural investigations at the Wari Site of Pikillacta, Cuzco, Peru. Paper presented at the 18th Annual Midwest Conference on Andean and Amazonian Archaeology and Ethnohistory, Chicago. -. 1991. Investigations at the Pikillacta Site: A provincial Huari center in the Valley of Cuzco, pp. 93-119. In Isbell, W H., and G. F McEwan, eds., Huari Administrative Structure: Prehistoric Monu- mental Architecture and State Government. Dumbar- ton Oaks, Washington, D.C. Meddens, F M. 1989. Implications of camelid manage- ment and textile production for Huari, pp. 146-165. In R. M. Czwarno, F M. Meddens, and A. Morgan, eds., The Nature of Wari: A Reappraisal of the Middle Horizon Period in Peru. BAR International Series 525. British Archaeological Reports, Oxford. Meyers, A. 1975. Algunos problemas en la clasificaci6n del estilo Incaico. Pumapunku, 8: 7-25. Muelle, J. C. 1945. Pacarectambo: Apuntes de viaje. Revista del Museo Nacional, 14: 153-160. Munsell. 1990. Munsell Soil Color Charts: 1990 Edi- tion Revised. Macbeth, New York. Nunez del Prado Bejar, J. V. 1972. Dos nuevas esta- tuas de estilo Pucara halladas en Chumbivilcas, Peru. Nawpa Pacha, 9(1971): 23-32. Orton, C. G. 1980. Mathematics in Archaeology. Cam- bridge University Press, Cambridge. Pardo, L. A. 1938. Hacia una nueva clasificacion de la ceramica cuzquena del antiguo imperio de los Incas, Peru. Revista del Instituto Arqueologico del Cusco, 3(4-5): 1-22. . 1939. Arte peruano: Clasificaci6n de la cer- amica Cuzquena (epoca Incaica). Revista de la Sec- tion Arqueol6gica de la Universidad Nacional del Cuzco, 4(6-7): 3-27. 1957. Historia y Arqueologia del Cuzco. Im- South America. American Antiquity 38(2): 239-242. prenta del Colegio Militar Leonico Pardo, Callao, Peru. Parsons, J. R., and C. M. Hastings. 1977. Prehispanic settlement patterns in the Upper Mantaro, Peru: A progress report for the 1976 field season. Unpublished Literature Cited 149 report. Instituto Nacional de Cultura, Lima, and the National Science Foundation, Washington, D.C. Patterson, T. C. 1967. Current Research, Highland South America. American Antiquity, 32(1): 143-144. Poole, D. 1984. Ritual Economic Calendars in Paruro: The Structure of Representation in Andean Ethnog- raphy. University Microfilms, Ann Arbor. Reichlen, H. 1954. Decouverte de tombes Tiahuanaco dans la region du Cuzco. Journal de la Societe des Americanistes, 47: 221-223. Renfew, C. 1975. Trade as action at a distance, pp. 3- 59. In Sabloff, J. A., and C. C. Lamberg-Karlovsky, eds.. Ancient Civilization and Trade. University of New Mexico Press, Albuquerque. Rice, P. M. 1987. Pottery Analysis: A Source Book. Uni- versity of Chicago Press, Chicago. Rivera Dorado, M. 1971a. La ceramica Killke y la ar- queologfa de Cuzco. Revista Espanola de Antropolo- gia Americana, 6: 85-123. . 1971b. Disenos decorativos en la ceramica Killke. Revista del Museo Nacional, 37: 106-1 15. . 1972. La ceramica de Cancha-Cancha, Cuzco, Peru. Revista dominicana de arqueologia y antropol- ogia, 2(2-3): 36-49. 1973. Aspectos tipologicos de la ceramica Cuz- quena del Periodo Intermedio Tardio. Atti del 40 Con- gresso Internazionale degli Americanisti, 1: 353-362. Rowe, J. H. 1943. Chanapata: La cultura pre-Incaica del Cuzco. Tupac Amaru, May, (2/3): 41-43. . 1944. An introduction to the archaeology of Cuzco. In Papers of the Peabody Museum of Ameri- can Archaeology and Ethnology, Vol. 27, no. 2. Har- vard University, Cambridge, Massachusetts. 1946. Inca culture at the time of the Spanish conquest. In Handbook of South American Indians, Vol 2. The Andean Civilizations, pp. 183-330. J. Steward, ed. Bulletin of the Bureau of American Eth- nology, no. 143. USGPO, Washington, D.C. 1956. Archaeological explorations in southern Peru, 1954-1955. American Antiquity, 22(2): 135- 150. . 1967. An interpretation of radiocarbon mea- surements on archaeological samples from Peru, pp. 16-30. In Rowe, J. H., and D. Menzel, eds., Peruvian Archaeology: Selected Readings. Peek Publications, Palo Alto, California. Rowe, J. H., and C. T Brandel. 1971. Pucara style pottery designs. Nawpa Pacha, 7-8(1969-70): 1-16. Ryden, S. 1957. Andean Excavations, Publication no. 4. The Ethnographical Museum of Sweden, Stockholm. Sanders, W. T. 1973. The significance of Pikillacta in Andean culture history. Occasional Papers in Anthro- pology, 8: 380-428. San Roman Luna, W 1979. Arqueologia de Pomacan- chi: Una introduction a su estudio. Licendiado en Ar- queologia Tesis, Facultad de Ciencias Sociales, Univ- ersidad Nacional San Antonio Abad, Cusco. . 1983. Datos ethnohistoricos de Pomacanchi, pp. 66-75. In Gibaja Oviedo, A., ed., Arqueologia Andi- na. Ediciones Instituto Nacional de Cultura, Cuzco. Seler, E. 1893. Peruanische Alterthiimer, insbesondere altperuanische Gefass. E. Mertens, Berlin. Squier, G. E. 1877. Peru: Incidents of Travel and Ex- ploration in the Land of Incas. Harper and Brothers, New York. Stuiver M., and P. J. Reimer. 1986. Programs Calib and Display. Radiocarbon, 28: 1022-1030. Torres Poblete, N. 1989. Sondeo arqueologico de Ar- away. Licenciado en Arqueologia Tesis, Facultad de Ciencias Sociales, Universidad Nacional San Antonio Abad, Cusco. Tschopik, M. H. 1946. Some notes on the archaeology of the Department of Puno, Peru. Papers of the Pea- body Museum of American Archaeology and ethnol- ogy, Vol. 27, No. 3. Harvard University, Cambridge, Massachusetts. Uhle, M. 1912. Los origenes de los Incas, pp. 302-352. In Actas del XVII Congreso International de Ameri- canistas, Buenos Aires. Valcarcel Vizuerra, L. E. 1934. Los trabajos arqueo- logicos del Cusco: Sajsawaman redescubierto II. Re- vista Del Museo Nacional, 3: 3-36, 211-233. . 1935. Los trabajos arqueologicos en el Depart- mento del Cusco: Sajsawaman Redescubierto III— IV. Revista Del Museo Nacional, 4: 1-24, 161-203. Valcarcel Vizguerra, L. E. 1946. Cuzco archaeology. In Handbook of South American Indians, Vol. 2. The Andean Civilizations, pp. 179-182. J. Steward, ed. Bulletin of the Bureau of American Ethnology, no. 143. USGPO, Washington, D.C. Valencia Zegarra, A. 1979. Coleccion arqueologica Cusco de Max Uhle. Instituto Nacional de Cultura- Cusco, Centro de Investigation y Restauracion de Bi- enes Monumentales, Cuzco. . 1984. Arqueologia de Qolqampata. Revista del Museo e Instituto de Arqueologia, 23: 47-62. Valencia Zegarra, A., and A. Gibaja Oviedo. 1991. Marcavalle: El rostro oculto del Cusco. Instituto Re- gional de Cultura de la Region Inka, Cuzco. Yabar Moreno, J. 1959. La cultura pre-Incaica de Chanapata. Revista del Museo e Instituto Arqueolo- gia, 18: 93-100. . 1972. Epoca pre-Inca de Chanapata. Revista Saqsaywaman, 2: 211-233. 1982. Figurillas de la cultura pre-Inka del Cuz- co, pp. 9-12. In Arqueologia de Cuzco. I, Oberti Rod- riguez, comp. Instituto Nacional de Cultura, Cuzco. Yamasaki, E, T. Hamada, and C. Fujiyaama. 1966. Ri- ken natural radiocarbon measurements II. Radiocar- bon, 8: 324-339. Zapata Rodriguez, Julinho. 1997. Arquitectura y con- textos funerarios Wari en Batan Urqu, Cusco. //; Bo- letin de Arqueologia PUCP, Vol. 1. La Muerta en el Antiguo Peru: Contextos y Conceptos Funerarios. Pe- ter Kaulicke, ed. Pontificia Universidad Catolica del Peru, Lima. 150 Literature Cited Appendix 1: Shallow Bowls During his 1940s excavations at the site of Chanapata, Rowe (1944, pp. 19-20) recorded finding a small quantity of shallow bowl frag- ments covered with a white slip and painted with simple red designs. He called this style "Carmen- ca" and provided several drawings of it (Rowe, 1944, fig. 16, pp. 3-8). I recovered various fragments of shallow bowls in the lower excavation levels of Marcapata (624) as well as on the surface of four other sites that appear to be similar to Rowe's Carmenca ceram- ics. (Map A-l, Fig. A-l, Drawings A-l and A-2). These vessels are painted with pigments that vary from dark red to brown or black. The decorative designs, drawn with thin pigments, have been ab- sorbed into the surface of the vessels. A few of the fragments have a white slip, although most are unslipped. There is low tonal contrast between the pigments and surfaces of the unslipped fragments. Although the depth at which most of the frag- ments were recovered suggests that they are of considerable antiquity, no secure date can cur- rently be offered. Fig. A-l. Shallow bowls. Shallow Bowls 151 t ) ; 0 12 3 4 5CM BLACK OR BROWN Ired Drawing A- 1. Shallow bowls. 152 Appendix One A L BL 7 1 0 12 3 4 5CM BLACK OR BROWN Drawing A-2. Shallow bowls. Shallow Bowls 153 0 10 20 i i i Contour lines every 400 m Map A-l. Distribution of sites with shallow bowls. 154 Appendix One 3 0 = W5 o Ot V u © ■o N 11 "O 9 3 e£ U ■5 w - o U - w a e .IS o - Cm o> r£ s es V e — o 83 Q a 0 e = S3 O OS - M - u o •— -9 o e 03 >> £ la © "3 E o XI c4 Sj -J X! — - Z - "S - u Oh B O < 0 •a S - a a r — - r — i — r^ i/~i r^ r- r- ON ON r<1 fN !~r i/i >o r- r- o ■* ■* — r- r- o. n. o. o. d. c_ . l/"i l/~J li"l l/~i l/~i I/*'* f OO 00 00 00 00 ON ON ON ON On O. Q. O. O. 3^ 9 in in 00 oc ON ON on on "> ^^cr- 00 _ _ 00 00 ON n£ On . O ON " vi in 00 00 ON ON . . ^ a. o.r r- r- -t- -r On On on!G!Conononon!G!Gon r- r~- r- O0 00 OO On On On r- r- r- fN (N 14 X ^ — e E r 03 03 1 E E __- _r o o o -a -o o o s c e = — o u 4J0J3,y=Jtj ££EE u **& *& '*C **& '& nO r~-r^r-r^r- oo ooNt--r- p~ nOnOnOnOnO r«~, "~> =*S Oonooooo ^r-~<^.ON O NO ^ °° NO no ir-, _ n ^oOiOr- On On On On iA — ' i,' — NJ^^OOX ^i^iL'ON ND^rJ-ONTj- *'**'0 *****J2„* .b-^ o- t>- ^Z acJSMSHSB* c3oB3c3csB5 ooooooorBlf Oo~oo — * a> * * » « =1 * 'i "i 3 vt" vr vr a. a. a. D a. wawayqo jamba leza Oil O O a- a- >. >% C3 C3 C3 S3 >, acta Casallaqt leza Oil smocco raccay leza Oil marca smarca A C3 « § i 3 -E C3 O 22 rB o C3 C3 O CJ « 03 E E 03 03 F F casmarca B oquepuquio mamarca keccasa * O 1 8 " O B C M SSCBECEO 4) O O 3 C C3 o Ut 3 3 O r3 e jz a o DOO- £ Qu, H U u. u < i= ^ a. a. < U Ou a< u ,E 03 x: E.a, 3 O £ H ?! n 03 03 03 o O O J2 _2 J j2 'J2 U '£ £ Radiocarbon Dates from the Cuzco Region 155 = ■♦-» = o U Q Z - f J ■a"3 U - o a 8. 7, 1987 7, 1987 7, 1987 r — — r-- C SO tN M IT) — 00 odd <<< — — — i (N o — ' r- ir> in -«t <"- f- t- Co n n n H H H 0- (J • U • U U • • • • (J ^U^U^^UsJUU^ o0QoCDo.nCQeQCQCQo so O >o vi 1/1 — ' sOOnmo ■* + 1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 0-*0~OV-)-hioOsOO — '— ■ (^ IT) lO +1 +1 +1 +1 + 1 O O O SO in m m os ■>* ■* o en Oh Oh Oh Oh Oh CQOhCQOhCDCQOhOhOhOhQQ "oa"Qa7:~;cQaaDaoH_ O O O vt © voommw-sooM^io^ + 1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 OOO — '©>/-! — iri O SO O sOOOtNt^O^fsOOOsO — en n m in in so \D >C sOr oo Os rn oi ?n csf (N csf ri csf rsf ri csf en so (N ^r<")Tj-sOr^o — SO^. OsOsOsOsOfN ^S00-^ X ZOhZOhOOOhO-OhOhO u # # ca ^a ^a ^a ^a ja ea "S-E o.a.a.a.>o,>>>>>a, ccccpcpppppc c3 c3 c3 c3 ed cd cd cd CO cd 2 2 £3 r3 zr OP ar > - 33 r3 73 U _*: _^ -^ rrj 5 £ £ £ 3 3 2 J= 2 = = S S Ctf C^ A 73 ?J Cl Cl CU cCfjcococs3Cs3cdcdcd c^cocoppp cc «s ?3 Is ^ b b b b r-- 3 ^- CO J— ra ra ra M _]J 3 71 « > c r3 r3 O C r3 — -1- u •a «S c/5 C 4) 1) TD 00 C OO a os X (D SO 1) ^ E «5 ^ c 3 y i b « 1 ^ >, g w * ^5 o Q ~z — E p r3 a 73 r^ O •— c/3 U || O * * 22 r5 r3 -3 ■2 U — -£ C _c 3 c o D r3 > O u -5 Q C -g «5 X c u u u II o.> 8 156 Appendix Two UNIVERSITY OF ILLINOIS-URBANA 3 0112 042936812 Field Museum of Natural History Roosevelt Road at Lake Shore Drive Chicago, Illinois 60605-2496 Telephone: (312) 922-9410