Archaeological Investigation of Late Archaic Sites. Creamer, Ruiz y Haas

AnthropologyNEW SERIES, NO. 40

Archaeological Investigation of Late ArchaicSites (3000–1800 B.C.) in the Pativilca Valley,Peru

Winifred Creamer

Northern Illinois UniversityAnthropology DepartmentDeKalb, Illinois 60115U.S.A.

Alvaro Ruiz

Northern Illinois UniversityAnthropology DepartmentDeKalb, Illinois 60115U.S.A.

Jonathan Haas

Field MuseumDepartment of Anthropology1400 South Lake Shore DriveChicago, Illinois 60605-2496U.S.A.

Accepted April 9, 2007Published September 14, 2007Publication 1546

E 2007 Field Museum of Natural History

ISSN 0071-4739

PRINTED IN THE UNITED STATES OF AMERICA

Table of ContentsABSTRACT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

PREVIOUS RESEARCH IN THE NORTE CHICO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

THE DEVELOPMENT OF COMPLEX SOCIETY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

FIELD METHODS: PROYECTO ARQUEOLOGICO NORTE CHICO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

ANALYSIS OF MATERIALS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

FIELDWORK, 2002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Upaca (02PVGS-1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Sector A: Profile 1, West Platform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Sector A: Profile 2, Main Mound. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

Sector B: Profile 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Sector C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Sector D: Test Unit 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Sector D: Test Unit 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Sector E: Test Unit 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

Discussion: Upaca . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Punta y Suela (02PVGS-02) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Sector A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

Sector B: Sample Collection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

Sector C: Profile 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

Sector D: Test Units 1, 2, and 3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

Sector D: Test Unit 1 (DN1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Sector D: Test Unit 2 (DN2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

Sector D: Test Unit 3 (DS1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

Discussion: Punta y Suela . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

Pampa San Jose (O2PVGS-3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Sector A: Profile 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

South Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

North Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

West Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

East Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

Discussıon: Pampa San Jose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

Carreterıa (02PVGS-4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

Sector A: Profile 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

Discussion: Carreteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

Huayto (02PVGS-5). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

Sector A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

Sector B. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

Sector C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

Sector C: Profile 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

Wall A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

Wall B. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

Sector D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

Sector E. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

Sector F. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

Sector G . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

Sector H . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

Sector I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

Discussion: Huayto. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

Cerro R . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

Sector A: Units 1 and 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

Sector B. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

Sector C: Unit 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Discussion: Cerro R . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Cemetery P . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Sector A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

Sector B: Unit 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

Sector C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

Sector D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

Discussion: Cemetery P . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

Potao (02PVGS-8) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

Sector A: Profile 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

Sector B. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

Sector C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

Sector D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

Sector E. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

Discussion: Potao . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

Vinto Alto (02PVGS-9) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

Sector A: Profile 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

Sector C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

Sector C: Profile 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

Discussion: Vinto Alto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

DISCUSSION: PATIVILCA VALLEY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

LITERATURE CITED. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

List of Figures1. Map of the Norte Chico . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

2. Photo of a shicra bag . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

3. Map of Upaca . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

4. Depressions in walled compound in Sector D, Upaca. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

5. Upaca, Sector A, Profile 1, excavated units, carbon sample locations and construction phases . . . . 16

6. Upaca, Sector A, Profile 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

7. Photo of the main mound at Upaca.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

8. Upaca, Sector B, Profile 3, excavated units and carbon sample locations and construction phases . 21

9. Upaca, Sector D, Test Unit 1, excavated units. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

10. Upaca, Sector D, Test Unit 1, phases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

11. Upaca, Sector E, Test Unit 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

12. Map of Punta y Suela . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

13. Aerial view of Punta y Suela in 1970 and photo of Punta y Suela in 2002 . . . . . . . . . . . . . . . . . . 28

14. Photo of power pole at Punta y Suela. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

15. Punta y Suela, Sector C, Profile 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

16. Punta y Suela, Sector D, Test Unit 1 (DN1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

17. Punta y Suela, Sector D, Test Unit 2 (DN2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

18. Punta y Suela, Sector D, Test Unit 3 (DS1). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

19. Map of Pampa San Jose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

20. Photo of looter’s hole in Pampa San Jose main mound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

21. Pampa San Jose profile and profile schematic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

22. Pampa San Jose photo of dirt-filled woven bags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

23. Map of Carreterıa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

24. Plan view of area cleared at Carreterıa and Profile 1 at Carreterıa. . . . . . . . . . . . . . . . . . . . . . . . 47

25. Photo of Carreterıa in relation to Pampa San Jose. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

26. Air photo of Huayto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

27. Map of Huayto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

28. Huayto, Sector C, Profile 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

29. Incised figures on gourd from Cerro R . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

30. Photo of Cemetery P cleared area. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

31. Cemetery P, Sector B, profile of burial pit. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

32. Cemetery P incised figures on gourd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

33. Map of Potao. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

34. Photo of mound at Potao . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

35. Sector A, Profile 1, at Potao . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

36. Photo of twin peaks at Vinto Alto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

37. Map of Vinto Alto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

38. Huanca atop Mound A at Vinto Alto. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

39. Sector A, Profile 2, at Vinto Alto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

40. Photo of Sector C, Profile 1, at Vinto Alto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

41. Sector C, Profile 1, at Vinto Alto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

List of Tables1. Rivers and valleys of the Norte Chico. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

2. Characteristics of Norte Chico sites tested in 2002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

3. Architecture at Pativilca Valley sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

4. Dates from Upaca. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

5. Dates from Punta y Suela . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

6. Dates from Pampa San Jose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

7. Principal phases of occupation at Pampa San Jose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

8. Date from Carreterıa. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

9. Huayto samples and dates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

10. Date from Cemetery P. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

11. Date from Potao. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

12. Dates from Vinto Alto . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

13. Features of Late Archaic sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

14. Summary diagram of carbon 14 dates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

Archaeological Investigation of Late Archaic Sites(3000–1800 B.C.) in the Pativilca Valley, Peru

Winifred Creamer, Alvaro Ruiz, and Jonathan Haas

Abstract

This volume presents the results of archaeological testing at nine sites in the Pativilca Valleyon the coast of Peru. Included are the first systematically collected radiocarbon dates fromthis valley, part of the Norte Chico region. Descriptions and maps of the sites tested providethe basis for discussion of the Late Archaic period (3000–1800 B.C.) in the valley and withinthe surrounding Norte Chico region, where monumental architecture appears to have begunearlier than elsewhere along the coast. Forty-four radiocarbon samples were analyzed fromeight sites, demonstrating that seven of the eight were occupied between 2740 and 1870 CalB.C. during the Late Archaic period. One radiocarbon sample dating to 1480 Cal B.C. (ISGS-A-429) from the site of Potao showed that it was occupied in the subsequent Initial period.These findings are discussed in relation to the development of complex society in the region.

Este volumen presenta los resultados del programa de investigacion de nueve sitiosarqueologicos en el valle del Rio Pativilca, en la costa de Peru. Se presenta los primerosfechados proveniente de muestras recogidos sistematicamente del valle Pativilca en la regiondenominada el Norte Chico. Los mapas y descripciones forman una base de discutir elperiodo Arcaico Tardıo (3000–1800 aC) en este zona, donde se encuentra construccion dearquitectura monumental durante una epoca anterior del resto del Nuevo Mundo. Cuarenta ycuatro muestras de ocho sitios estaban fechados. Siete de los ocho sitios fecharon entre 2740 y1870 Cal aC, en el periodo Arcaico Tardıo. Un fechado de 1480 Cal aC (ISGS-A-429) del sitioPotao indica que este sitio estaba ocupado mas tardio, durante el Periodo Inicial. Estosresultados pueden estar discutido en relacion al desarrollo del sociedad complejo en el NorteChico.

Introduction

Andean Peru has long been identified as one ofthe six major world areas where ‘‘civilization’’developed under largely endogenous or ‘‘pris-tine’’ conditions (Fried, 1967). The last quarterof the 20th century saw a great deal of researchand writing focused on questions of how andwhy civilization first developed in the Andes(Moseley, 1975, 1992; Haas, 1982; Haas et al.,1987; S. Pozorski & T. Pozorski, 1987; Grieder

et al., 1988; Wilson, 1988; T. Pozorski & S.Pozorski, 1990; Shady, 1993, 1995; Billman,1999, 2001; Stanish, 2001). Continuing research

has steadily pushed back the very beginnings ofthe origins of what can be identified as emergentAndean civilization.

For many years, the Chavın cultural complex,with its distinctive art style and its ceremonialcapital at Chavın de Huantar in the centralhighlands, was considered the ‘‘Mother Culture’’of Andean civilization (see Lumbreras, 1970,1971, 1972, 1981, 1989; Kembel & Rick, 2004).More recent research, however, has shifted thefocus of attention away from the highland-basedculture of Chavın and moved it to a stretch ofthe central Peruvian coast. In this area, locatedroughly between the Lurın Valley on the south

FIELDIANA: ANTHROPOLOGY N.S., NO. 40, SEPTEMBER 14, 2007, PP. 1–79 1

and the Casma Valley on the north, archaeolog-ical research is revealing a pattern of largeceremonial centers with monumental architec-ture and elaborate art (Feldman, 1983, 1987,1992; Quilter, 1985, 1991; Burger, 1987; Burger& Salazar-Burger, 1991; Quilter et al., 1991; S.Pozorski & T. Pozorski, 1992; T. Pozorski & S.Pozorski, 2000; Haas & Creamer, 2004). Datingof the sites during these early coastal develop-ments is still being worked out, but radiocarbondates range from 4900 BP to 3200 BP, wellbefore the founding of Chavın de Huantar andthe Chavın expansion in the Early Horizon.

The remarkable assemblage of major LateArchaic centers in the Norte Chico region, allwith monumental and ceremonial architecture,presents an unparalleled opportunity for archae-ological research. The area offers the possibilityof opening a window into the prehistoricbeginnings of the earliest complex society toemerge in South America. This volume presentsthe first systematically collected radiocarbondates from one part of the coastal zone, fromnine Late Archaic sites in the Pativilca Valley.The Pativilca Valley is one of the four adjacentcoastal valleys where this pattern of earlymonument construction took place. In addi-tion, a detailed description of each of the sitessampled is provided, including the test unitsexcavated and the looter’s holes that werecleared and sampled. We include descriptionsof the sites tested, the samples collected and thecontext of each, and the resulting radiocarbondates. These are basic data on a series ofdistinctive sites and provide the basis fordiscussion of the Late Archaic period (3000–1800 B.C.) in the valley and within the surround-ing Norte Chico region, where monumentalarchitecture appears to have begun earlier thanelsewhere along the coast. This is followed bya discussion of the implications of our findings inrelation to the development of complex society inthe region.

The Norte Chico region includes the valleys (Sto N) of Huaura, Supe, Pativilca, and Fortaleza,part of the central Peruvian coast (Fig. 1). This isan area that has received intermittent attentionfrom archaeologists. Informal and limited sur-veys in the area (Williams & Merino, 1979; Vega-Centeno et al., 1998) have shown that settlementin the area was both long and dense. The NorteChico is at a biological, geographical andcultural crossroads between the north and southcoastal regions (Billman, 2001; Dillon et al.,

2003). Culturally, at least in late prehistorictimes, the Norte Chico was a fron-tier zone between the northern sphere of theChimu and the southern sphere of Chancay.Although there has been some research in theregion over the past 100 years, the archaeolog-ical record of the Norte Chico is comparativelyunknown.

Biologically, the southernmost occurrence ofequatorial biotic communities extends into thisarea, as do the northernmost examples of drierdesert communities. Billman (2001) has sug-gested that two environmental characteristicsmay have influenced the development of socialcomplexity in the Norte Chico. The coastal plainbetween the Andes and the Pacific Ocean iswidest in the Tumbes region in the far north ofPeru and narrows toward the south. The Andesapproach the shore by the Moquegua Valleynear the Chilean border where there is almost nocoastal plain at all. The Norte Chico is midwaybetween the large valleys of the north and thesmaller valleys of the south. Surplus agriculturalproduction might therefore be presumed to beproportional to the area of coastal plain present,decreasing from north to south along the coast.However, El Nino/ENSO events appear toimpact the coast differently from north to southin a way that impinges on agricultural pro-ductivity. The severity of rains and flooding fromthese events decreases from north to south; thus,the Norte Chico has felt less impact from ElNino/ENSO events than do the valleys of thenorth coast. The position of the region midwayin the gradient of coastal plain area and midwayin the scale of El Nino/ENSO effects may suggestfactors that contributed to the particular series ofcultural developments that occurred in thisregion during the third millennium B.C. (Bill-man, 2001).

Physical proximity and topographic featuresintegrate the four-valley system of the NorteChico from within. The three northern valleys,Fortaleza, Pativilca, and Supe, all empty into thePacific over a stretch of only 22 km. There are, inturn, inland passes linking the upper reaches ofall the valleys. In addition, the middle reaches ofHuaura are separated from Supe by only 15 kmthrough either of two inland passes. The NorteChico spans a total of only 50 km of coastline. Arecent analysis of least-cost pathways frominland sites to the coast suggests that the rivervalleys were not the best routes to the coast butthat travel directly across the hills from sites to

2 FIELDIANA: ANTHROPOLOGY

the coast was likely to have been used (Carlson &Craig, 2006).

Previous Research in the Norte Chico

Max Uhle (1925) was the first professionalarchaeologist to record sites and conduct ex-cavations in the Norte Chico region. Uhle wasspecifically after ceramic vessels that could be

used to help develop a broad ceramic chronologyfor the Peruvian coast. He excavated at sitesaround the mouth of the Supe Valley, includingChimu Capac (Middle Horizon), San Nicolas(Middle Horizon, Late Intermediate, and LateHorizon), around Aspero (Initial period/EarlyHorizon), and others (see also Kroeber, 1925).Subsequent to Uhle, Kosok and Schaedel(Kosok, 1965) visited the region in the 1940s

FIG. 1. Map of the Norte Chico region highlighting the Late Archaic sites in the Pativilca Valley.

CREAMER ET AL.: ARCHAEOLOGY OF PATIVILCA VALLEY, PERU 3

and used aerial photographs to locate and recorda number of sites in all four valleys. Kosok madesome grab sample collections while in the area,but other than a few drawings and photographs,none of this material has been published. Kosokalso published the first accounts of a complex ofsites that were seen at the time as enigmaticbecause they lacked surface ceramics. At thelargest of these sites, Chupa Cigarro Grande(now known as Caral), Kosok noted, ‘‘Thevirtual absence of potsherds at this site madeany accurate calculations as to its relative ageimpossible’’ (1965, p. 223, italics in original).

In the 1940s, Willey and Corbett (1954)conducted the first scientific excavations in theregion at several sites on the coast at the mouthof the Supe Valley. One of these is the largecommunity of Aspero, where excavations wereundertaken in a Late Archaic midden and a later,ceramic-bearing cemetery. At the time, the sitewas recognized as an early maritime settlement,but absolute dates were not available. Willeyrevisited Aspero in the 1970s with Moseley(Moseley & Willey, 1973; Moseley, 1975), whenthey recognized that a series of mounds at thesite (identified as ‘‘hillocks’’ in Willey andCorbett’s earlier publication) were in fact con-structed platform mounds. It was also recognizedthat the paucity of ceramics at the site was due tothe fact the site dated to a preceramic period.Further work was conducted at Aspero in the1970s by Feldman (1980, 1983, 1987, 1992), whoobtained radiocarbon dates ranging from 4900 6

160 BP to 3950 6 150 BP. The earliest of thesedates was rejected at the time as too old, and theearliest occupation of Aspero still needs to bedefined. Feldman tested a second Late Archaicsite in the area, As8, which yielded a singleradiocarbon date of 6914 6 190 BP. Salvageexcavations at the small maritime/coastal com-munity of Bandurria at the mouth of the HuauraValley have provided radiocarbon dates rangingfrom 4530 6 80 BP to 4300 6 90 BP (Fung,1988, 2004). Additional work was done in the1970s at other Late Archaic sites in the NorteChico (Fung, 1988) but no dates have beenreported. Silva (1975, 1978) excavated at the siteof Bermejo to the north of the Fortaleza Valley,which yielded Initial period and Early Horizondates. However, large parts of Bermejo lackceramics and have architecture similar to LateArchaic sites in the valleys immediately to thesouth. Shady and Ruiz (1979) tested a MiddleHorizon site in the Huaura Valley in 1978.

A survey of large sites with standing architec-ture in the Supe, Pativilca, and Fortaleza valleyswas undertaken by Williams, though only theSupe Valley results have been reported (Williams& Merino, 1979). Williams visited nearly 100 sitesin the Supe Valley and a similar number wereapparently visited in Fortaleza and Pativilca(Moseley, 2001). These sites run the temporalspectrum from Late Archaic through Late Hori-zon Incan sites. In the mid-1990s, Shady (Shady etal., 2003) visited sites recorded by Williams andMerino in an effort to distinguish Late Archaicoccupations from later Initial period and EarlyHorizon occupations. Systematic survey of thevalley bottom was conducted in the lower portionof the Fortaleza Valley in conjunction witha power line construction project (Vega-Centenoet al., 1998). More than 100 sites were recorded inthe course of this survey. Engel (1987) surveyedsites in the Pativilca Valley in the 1980s, but theyhave not been fully reported. A comprehensivesurvey of the Huaura Valley in 2004–2005recorded over 500 separate sites and over 2,000different site localities. Single sites with multipleoccupations were recorded separately as individ-ual localities (Nelson & Ruiz, 2005).

Zechenter (1988) tested a number of sites inthe Supe Valley in the 1980s in an effort toreconstruct subsistence strategies during the LateArchaic and Initial periods. In addition toextensive biological data, she reports individualradiocarbon dates from nine sites yielding datesfrom 7330 6 110 BP to 3110 6 80 RCYBP(approx. 6190–1370 Cal B.C.). Also in the SupeValley, Shady initiated extensive excavations atthe site of Caral (formerly ChupacigarroGrande) in the middle reaches of the Supe Valleyin the mid-1990s (Shady, 1997, 2003a,b,c,d,e,2004; Shady & Leyva, 2003). This work hasfocused on the description of architecturalfeatures of the site and retrieving ethnobotanicaland ethnozoological specimens for analysis.Shady’s research provides the first detailedaccount of a major inland Late Archaic urbancenter in the Andes. Recent publication ofradiocarbon dates from Caral range from 40906 90 BP to 3640 6 50 BP (Shady et al., 2001).

The Development of Complex Society

Moseley (1975) was one of the first torecognize the ‘‘precocious’’ nature of culturaldevelopment on the Peruvian coast. He pointedout that there were a number of large maritime-


based sites up and down the coast that dated tothe third millennium or even before. He also wasone of the first to highlight the fact that a numberof these sites, such as Aspero in the Supe Valleyand El Paraiso in the Chillon Valley, had largecommunal architecture yet lacked ceramics intheir cultural assemblages. This lack of ceramicswould indicate that the sites antedated theintroduction of ceramics in Peru at around 1800B.C. Subsequently as radiocarbon dates began tobe obtained for these Late Archaic sites they wereconfirmed to extend back into the third millenni-um B.C. Engel (1957) coined the term ‘‘CottonPreceramic Stage’’ to delineate coastal sites suchas these that had cotton but lacked ceramics andwere occupied between about 3000 and 1800 B.C.That term is applied here as a general marker toplace these sites in a wider comparative context inthe Peruvian Andes. Moseley (1975, 1985, 1992, inpress) then went on to argue that it was theorganization required for maritime-dependentsubsistence that was conducive to the emergenceof political centralization. Centralization, in turn,‘‘preadapted’’ these cultural groups to the sub-sequent development of even more complex,hierarchical and centralized forms of organizationbased on irrigation agriculture.

Moseley’s ‘‘maritime foundations of Andeancivilization’’ theory has been refined and debatednow for the past 30 years (see Osborn, 1977;Raymond, 1981; Wilson, 1981; Bonavia, 1982,1991, 1993–1995; Quilter & Stocker, 1983; S.Pozorski & T. Pozorski, 1990; Quilter, 1992). Itremains today a powerful explanation for howand why the Andean region started on the roadthat eventually led to much more complex statesand empires. More recent research on the coast,including that presented here, is raising questionsabout whether coastal maritime sites did indeeddevelop independently of an agriculturally basedsubsistence economy. Research in the SupeValley by Shady and her colleagues (Shady,1997, 1999a, 2000a,b,c; Shady et al., 2001; seealso Williams & Merino, 1979; Engel, 1987) hasshown that the maritime community of Asperoin the Supe Valley was not alone; rather, itappears to have been an integral part of a muchlarger cultural system that included large agri-culturally based centers.

Field Methods: Proyecto Arqueologico Norte Chico

Among the four valleys, the Pativilca Valleywas selected as the 2002 research area for the

Proyecto Arqueologico Norte Chico (PANC).The Pativilca River rises in the Andes and flowsthrough a long and narrow valley until it reachesthe coastal plain where it becomes wider,generating cultivable land. The lower reaches ofthe valley extend from the coast at the mouth ofthe river to the Quebrada Huanchay, where thevalley narrows, eliminating most cultivable land.The lower valley is the zone of river floodplainsand the lowest foothills of the mountains. Oneimportant characteristic of this river is that itmaintains a flow of water all year long, while theFortaleza and Supe rivers carry water duringonly some months of the year (Table 1), makingthe Pativilca the hydrological center of this area(Kosok, 1965, p. 218). Reconnaissance of thevalley by PANC members identified probableLate Archaic occupations at eight sites with largemounds—Upaca (formerly La Capitana), Puntay Suela, Pampa San Jose, Carreterıa, Huayto,Potao, Los Olmitos, and Vinto Alto—and twocemeteries, CR and CP. These two sites are onlyminimally identified here to avoid drawingattention to specific localities. Most of thesesites were mentioned by Engel (1987) andtentatively identified as belonging to either theCotton Preceramic or the Initial period, thoughhe does not appear to have excavated at any ofthem.

Before undertaking reconnaissance in the field,we systematically scrutinized aerial photos of thezone taken by the Servicio AerofotograficoNacional (SAN). Architectural remains of ar-chaeological sites such as mounds, platforms,sunken plazas, walls, and looted areas could beidentified in areas not covered by agriculturalfields, roads, or structures. Subsequently, inves-tigators visited the places identified as sites inthe aerial photos, sites identified by previousresearchers and by people living in the area. Thelocation of each identified site was initiallyrecorded with a handheld Global PositioningSystem (GPS), and the information was trans-ferred to maps at a 1:100,000 scale, obtainedfrom the Instituto Geografico Militar.

Boundaries for each site were established usingseveral criteria, including the area over whichtiny shell fragments could be observed on thesurfac, and the surrounding natural features,such as the walls of a quebrada. All of thePativilca Valley sites mentioned here have beenrecorded as polygons registered with the Insti-tuto Nacional de Cultura, Lima. This designa-tion is the first step toward providing legal


protection for each site. Each site was numberedwith the year the work was carried out (e.g., 02),the valley (e.g., PV 5 Pativilca Valley), type ofsurvey (e.g., GS 5 General Survey), anda number distinguishing each site. Names wereestablished for each site based primarily on localusage.

The inventory of archaeological sites recordedduring the project included nine elements: sitename, latitude and longitude, altitude, location,history, ceramics, period, state of conservation,and description:

1. Name: Names were assigned to a number ofthe sites by Engel (1987). When no namewas attached to the site, we assigned thename used by local people to refer to the siteor a geographical name from maps of theInstituto Geografico Militar of Peru.

2. Latitude and Longitude: These were de-termined by use of GPS or maps at the1:100,000 scale by the Instituto GeograficoMilitar of Peru.

3. UTM Coordinates: These were establishedwith a handheld GPS unit and/or maps atthe 1:100,000 scale of the Instituto Geogra-fico Militar of Peru.

4. Altitude: Determined by GPS unit or fromtopographic maps.

5. Location: This includes the location inrelation to settlements and towns nearby,geographic features, and political units ofdistrict and province.

6. Description: Site descriptions were preparedto include architectural components, theirform and size, along with archaeologicalmaterials encountered on the surface.

7. History: Previous investigations, if any, atthe sites visited.

8. Period: This includes the chronologicalposition of the site, determined by analysisof the cultural material on the surface,architectural features, and exposed profilesat each site.

9. State of Preservation: This is a preliminaryassessment of the conditions of deteriora-tion of the site and the level of destructionby human or natural forces.

To assess the actual chronology of the occupa-tion in the Pativilca Valley, sampling and testexcavations were implemented in 2002 at nine ofthe 10 potential Late Archaic sites. Los Olmitos,though included on the map, was not discovereduntil the middle of the 2002 field season, and itcould not be included in the permit for that year.With the exception of the two cemeteries and testpits at Upaca and Punta y Suela, excavationfocused primarily on retrieving radiocarbondates from within the monumental architecture(Table 2). At each site, areas of previousdisturbance were targeted for clearing. Theseholes were either excavated by looters or createdby various modern construction activities, suchas bulldozing a road or mining a site for gravel.The holes and profiles exposed were used as

TABLE 1. Rivers and valleys of the Norte Chico

River Year

Max volume* avgdaily discharge

(m3/sec) Month

Min volume* avgdaily discharge

(m3/sec) MonthIrrigated

area** (h)

Fortaleza 2002–2003 17.5 Mar 1.7 Sep, Oct 28082003–2004 9.2 Apr 1.1 Jul2004 1.1 Aug, Nov 1 Sep, Oct

Pativilca 2002–2003 88.4 Mar 13.7 Sep 271002003–2004 59.1 Feb 10.7 Jul2004 35.3 Nov 9.9 Aug

Supe 2002–2003 20.2 Mar 1.7 Jul 51622003–2004 8.7 Apr 1.3 Jun, Jul2004 2.5 Nov 1.1 Sep, Oct

Huaura 2002–2003 59.4 Mar 13.6 Sep 384972003–2004 78.5 Feb 12 July2004 14.3 Oct 10.7 Aug, Sep

* 2002–2003: http://www.minag.gob.pe/hidro_cau_lima2.shtml (PROM); 2003–2004: http://www.minag.gob.pe/hidro/hidro_cau_lima2004.shtml; 2004: http://www.minag.gob.pe/hidro/hidro_cau_lima.shtml (PROM). Dataavailable for Aug.–Nov. 2004 only.

** http://www.inrena.gob.pe/irh/irh_infointeres_atdr_barranca.htm (2001) or http://www.inrena.gob.pe/irh/irh_infointeres_atdr_huaura.htm (2005).


‘‘windows’’ into the interiors of platformmounds. Making use of existing damage allowedus to extract radiocarbon samples from interiorsurfaces and early construction phases in themounds that we could not reach with test pitsduring a single field season.

All but two of the mounds tested in thePativilca Valley were constructed at least in partwith shicra (Table 3), woven bags that were filledwith stone and used in mound construction,somewhat like a sandbag (Quilter, 1985). Thearid climate has preserved these bags (Fig. 2) andprovides an excellent source of annual plantremains that are suitable for radiocarbon dating(Shady et al., 2001). In each of the disturbed areasselected for sampling, all loose surface fill wasremoved to expose undisturbed, in situ construc-tion. Excavation beneath the disturbed surfacematerials was sufficient in every case to ensurethat undisturbed construction was exposed forsampling. The exposed surfaces were then photo-graphed and drawn. In addition to the remains ofshicra, plastered floors provided a good source oforganic material for radiocarbon dating, as theywere constructed with large amounts of fiber usedas tempering material mixed with clay. In a smallnumber of cases, mentioned below in the sitedescriptions, it was not possible to obtain in situannual plant fibers from the constructions. Inthese cases, pieces of charcoal from burned woodwere extracted from construction fill. Suchcharcoal, recovered in a dry environment suchas the Peruvian coast, introduces potentialproblems of ‘‘old wood’’ (Schiffer, 1986).

In arid environments, wood may survive for100 years or more after it is cut since decompo-sition is slowed by the inhospitable climate forbacteria growth. Many years after a tree dies,surviving branches may be incorporated intoconstruction or used for firewood. When thecharcoal produced in a hearth, for example,survives and is used as a sample for radiocarbondating, the date it yields can be highly inaccuratebecause the age of the wood is not a reflection ofthe age of the structure or feature of which it wasa part. For this reason, fibers of annual plantsare the preferred material for radiocarbon datingin dry regions since these materials live for onlya single season and do not yield the inaccurateresults that dry wood may provide. Despite thepotential problems of using wood charcoal forradiocarbon dating, it was judged worthwhile todate charcoal when annual plant fiber was notavailable. The resulting dates could be compared

TABLE2.

Ch

ara

cter

isti

cso

fN

ort

eC

hic

osi

tes

test

edin

20

02

.

Sit

en

o.

Na

me

Lo

cati

on

Dis

t.fr

om

coa

st(k

m)

Ele

vati

on

(ma

sl)

Are

ao

fA

rch

itec

-tu

re(h

)

Mo

un

ds

(no

.st

ill

visi

ble

)

Cir

cula

rp

laza

s(n

o.

stil

lvi

sib

le)

Ra

ng

eo

fd

ate

s(c

al

B.C

.)D

ate

dC

14

Sa

mp

les*

02

-PV

GS

-1U

pa

caE

ria

zo,

ba

seo

fC

erro

Mir

ad

or

82

09

35

32

(2)

27

40

–2

10

12

02

-PV

GS

-2P

un

tay

Su

ela

Riv

erte

rra

ce6

15

21

18

2(0

)9

17

0–

56

01

10

2-P

VG

S-3

Pa

mp

aS

an

Jose

Eri

azo

,b

ase

of

Cer

roM

irad

or

12

.52

35

58

(3)

1(1

)2

23

0–

18

70

60

2-P

VG

S-4

Ca

rret

erıa

Eri

azo

,b

ase

of

Cer

roL

om

as

de

Pa

tiv

ilca

12

.52

65

11

1(0

)2

18

01

02

-PV

GS

-5H

ua

yto

Hil

lto

p1

73

33

56

(3)

02

27

0–

22

40

20

2-P

VG

S-6

CR

Eri

azo

51

60

NM

00

n.a

.**

00

2-P

VG

S-7

CP

Eri

azo

51

52

NM

00

22

20

10

2-P

VG

S-8

Po

tao

Va

lley

flo

or

5.5

16

11

05

(3)

01

48

01

02

-PV

GS

-9V

into

Alt

oV

all

eym

arg

in1

32

50

18

31

(0)

25

80

–2

10

09

02

-PV

GS

-10

Lo

sO

lmit

os

Va

lley

ma

rgin

14

28

01

21

0n

.a.

0

*D

eta

ils

inH

aa

set

al.

(20

04

)

**

n.a

.5

no

sam

ple

sh

av

eb

een

pro

cess

edfr

om

this

site

.

NM

5N

ot

mea

sure

d.


TABLE3.

Arc

hit

ectu

rea

tP

ati

vil

caV

all

eysi

tes

Siz

e(m

)H

eig

ht

(m)

Vo

lum

e(m

3)

Ph

ase

s(f

rom

test

ing

)C

on

stru

ctio

nm

ate

ria

lD

ate

s(c

al

B.C

.)

Up

aca

A5

53

89

12

26

,67

3.3

41

An

gu

lar

rock

w/a

few

cob

ble

s,sh

icra

23

30

–2

19

0A

Des

tro

yed

44

An

gu

lar

rock

w/a

few

cob

ble

s,sh

icra

27

40

–2

31

0A

sso

cia

ted

CP

n.a

.*B

15

32

06

n.a

.4

11

10

,8

70

C1

03

20

5n

.a.

n.a

.N

on

eA

sso

cia

ted

CP

20

n.a

.D

-Tes

tU

nit

16

03

60

03

Are

ao

fp

its

du

gin

tosu

rfa

ce2

10

D-T

est

Un

it3

60

36

00

2A

rea

of

pit

sd

ug

into

surf

ace

No

ne

E-T

est

Un

it2

n.a

.0

2H

ill

slo

pe

11

60

Pu

nta

yS

uel

aA

40

35

01

21

5,8

97

.86

n.a

.N

on

eA

sso

cia

ted

CP

Des

tro

yed

n.a

.N

on

eB

12

43

26

43

35

.76

n.a

.2

42

0–

22

00

B2

n.a

.N

on

eB

3n

.a.

No

ne

Ass

oci

ate

dC

P1

5n

.a.

C3

73

74

63

78

.75

2C

ob

ble

an

da

ng

ula

rro

ck1

48

0D

N-1

Do

mes

tic

tra

sh3

97

50

–2

43

0D

N-2

Do

mes

tic

act

ivit

y3

18

40

–7

50

DS

-1D

om

esti

ctr

ash

57

41

0–

64

40

Pa

mp

aS

an

Jose

A7

33

10

12

05

2,0

42

.89

15

Co

bb

le,

an

gu

lar

rock

,sh

icra

,d

irt-

fill

edsh

icra

22

30

–1

87

0

Ass

oci

ate

dC

P3

9.6

n.a

.N

on

eB

38

34

27

3,1

97

.51

n.a

.N

on

eC

35

34

02

,86

8.0

4n

.a.

No

ne

D4

,09

7.7

1n

.a.

Ca

rret

erıa

A3

23

60

72

5,3

73

.59

5A

ng

ula

rg

ran

ite,

shic

ra2

18

0H

ua

yto

A6

83

75

71

4,3

61

.68

n.a

.C

ob

ble

sN

on

eB

55

31

27

3–

91

5,3

15

.52

n.a

.C

ob

ble

sN

on

eC

50

39

58

18

,80

3.1

53

Co

bb

les,

shic

ra2

27

0–

22

40

D5

53

65

1n

.a.

n.a

.C

ob

ble

sN

on

eE

No

mo

un

ds

n.a

.N

on

eF

Des

tro

yed

n.a

.N

on

eG

Des

tro

yed

n.a

.N

on

e


with those obtained from annual plants asa check on their consistency.

In addition to sampling in mounds, 1 3 2-mtest pits were excavated at Upaca and Punta ySuela. Both of the cemeteries, CP and CR, havebeen subjected to looting. We did not attempt toexcavate any undisturbed burials this first fieldseason but cleared 5 3 5-m areas in order toobtain a sample of human bone and associatedfunerary objects from the heavily looted surface.

Test excavations employed two strategiesdepending on the context and the condition ofthe area that was being tested. Test pits of 1 3

2 m were undertaken at sites in the PativilcaValley identified as Late Archaic or aceramiccenters. These test pits were located in areaswhere there appeared to be deposits of stratifiedtrash. Such deposits were identified by examina-tion of erosion channels, looter’s pits, or otherdisturbed areas and through the use of a tool toobtain samples (sediment tester). For each testpit, the contents were excavated in a combinationof natural layers and artificial levels, dependingon the context. When natural layers weredistinct, they were used as the principal horizon-tal control within the test unit. However, whennatural layers could not be detected or thenatural layers were more than 15 cm thick, thedeposits were divided into artificial levels of15 cm to maintain adequate control over pro-venience. All the excavated material wasscreened through quarter-inch mesh, and 10%of the sample from each unit was screenedthrough one-eighth-inch mesh. This was accom-plished by directing every tenth bucket ofmaterial to a separate screen. Two-liter soilsamples were taken for fine screening andflotation, and additional soil was taken forpollen analyses.

Linda Scott Cummings, director of the Paleo-research Institute, worked directly with the teamin the field to develop strategies for taking theflotation, fine-screen, and pollen samples. Sam-ples for fine screening and flotation weremeasured from material that had already passedthrough quarter-inch screen. Materials collectedwere placed in bags labeled with the date,provenience, excavators, and contents. Specialcare was made to recover radiocarbon sampleswith secure provenience. Samples for radiocar-bon dating were placed into aluminum foil in thefield and then into labeled bags. Each bag wasrecorded in a field catalog. A written descriptionof each site was accompanied by sketch maps

Siz

e(m

)H

eig

ht

(m)

Vo

lum

e(m

3)

Ph

ase

s(f

rom

test

ing

)C

on

stru

ctio

nm

ate

ria

lD

ate

s(c

al

B.C

.)

HN

om

ou

nd

sn

.a.

No

ne

ID

estr

oy

edn

.a.

No

ne

Po

tao

A6

93

73

17

n.a

.1

Co

bb

les

14

80

B3

23

85

7(3

)n

.a.

n.a

.C

ob

ble

sN

on

eC

91

31

59

7n

.a.

n.a

.C

ob

ble

sN

on

eD

27

33

81

.5n

.a.

n.a

.C

ob

ble

sN

on

eE

Des

tro

yed

n.a

.N

on

eV

into

Alt

oA

74

38

52

31

07

,80

0.3

1A

ng

ula

rro

ck,

shic

ra2

54

0–

24

80

B5

83

91

18

79

,37

9.2

8*

*n

.a.

No

ne

C8

63

10

63

51

,01

1.5

14

An

gu

lar

rock

,sh

icra

25

80

–2

10

0D

25

–3

53

65

n.a

.n

.a.

No

ne

*n

.a.

5n

osa

mp

les

ha

ve

bee

np

roce

ssed

fro

mth

issi

te.

**

Vin

toA

lto

:M

ou

nd

Bin

clu

des

bo

thM

ou

nd

sA

an

dB

.

TABLE3.

Co

nti

nu

ed.


and the field notes of the investigators. Theprimary focus of the 1 3 2-m test excavationswas to recover samples of tools and floral andfaunal remains in areas of stratified trash and toobtain radiocarbon samples from residentialcontexts. There was no attempt during theselimited excavations to identify and uncoverspecific architectural features of individualstructures or to excavate burials.

Analysis of Materials

The main objective of the research in 2002 wasto obtain radiometric dates for sites that werebelieved to date to the Late Archaic period(3000–1800 B.C.) based on layout, architecturalfeatures, and the absence of ceramics. Thisreport focuses on sampling methods, detailedcontext for each dated sample, and discussion ofthe dates obtained. Analysis of botanical andfaunal remains, pollen, textiles, lithics, and othermaterials will be presented elsewhere.

All materials from the 2002 field season havebeen processed (brushed or washed), labeled, andinventoried. Both screen and flotation samples ofmacrobotanical remains were analyzed. Plantstems, leaves, fruits, and seeds were identified onthe basis of comparison with reference materials(Alarcon, 2005). Two-liter samples were collect-ed for each process, though only one liter wasprocessed. Fine-screened materials were passedthrough 4-mm and 2-mm geological screens. Thematerials collected were then sorted into seeds,fish bone, and other identifiable groups, such asotoliths (see also http://www.paleoresearch.com/manuals/manual.html). Animal bone was identi-fied in macro-, fine-screen, and flotation sam-ples. Ceramics were collected from the surface ofsites during the 2002 field season. These appearto be surface-only deposits that postdate theoccupation of mounds at the sites tested (Bria,2004). Lithics were identified on the basis ofcomparison with materials from other Middleand Late Archaic sites (Dillehay et al., 1997;Rossen, 1998; Dunn & Heaton, 2005). Twined,looped, knotted, and other nonloom textiles were

FIG. 2. Excavated shicra bag illustrating simple twined construction (J. Haas).


identified by comparison with examples fromHuaca Prieta (Bird & Hyslop, 1985) and Asia(Engel, 1963) and examples at the MuseoNacional de Antropologia y Arqueologia andthe Huaca Prieta collection at the AmericanMuseum of Natural History (L. Haas, 2004,2005). Analysis is pending for the loom-woventextiles collected from the two heavily lootedcemetery sites.

Pollen samples were collected from eachlevel of excavated test pits. Twenty-threesamples have been processed at the TexasA&M pollen laboratory. One hundred grainidentifications have been made from eachsample. Procedures for analysis of pollen sam-ples in Peru are being tested at a new facility atCayetano Heredia University (Huaman et al.,2005). Collecting plants, processing samples tocreate a reference collection of pollen, andidentifying pollen from archaeological samplesis ongoing.

Each of the sites sampled in 2002 hasbeen identified in historic air photos (1943 and1969–1970). Initially, sites were recorded witha series of points recorded on handheld GPSunits. Subsequently, the geographical position ofeach site, structures, and features mentionedhave been recorded with high-resolution GPSreceivers (Trimble Pro-XR) corrected by post-processing with a GPS base station located in thetown of Barranca. All the Pativilca sites testedhave been intensively mapped with a Topcontotal station. Although the number of pointscollected varies for each site, a minimum of10,000 points per site was collected. Contourmaps and cross sections were produced withSurfer (ver. 8) and ARCGIS (ver. 9.1). AGeographic Information System is being de-veloped for the research area that facilitatesanalysis of site location and layout (Chun, 2003),architecture (Advincula, 2005; Perales & Haas,2005; Ruiz et al., 2005), and the application ofanalytical techniques, such as viewshed analysisand analysis of least-cost pathways (Carlson &Craig, 2005).

The materials recovered from surface collec-tion, test pits, and cleared profiles are temporar-ily stored at the field center of the ProyectoArqueologico Norte Chico during ongoinganalysis. The present work focuses on theidentification and dating of sites in the PativilcaValley, while analysis of botanical, lithic, andother materials will be published separately.Materials that have been analyzed are stored in

an air-conditioned, secure storage facility in thetown of Pativilca, Peru. We have begun co-ordinating with the Instituto Nacional de Cul-tura to determine the long-term storage appro-priate for project materials after analysis iscomplete.

Fieldwork, 2002

Each of the nine sites tested during 2002 wasbelieved to date to the Late Archaic period.While the results of radiocarbon dating suggestthat portions of one or more may date to theInitial period rather than the Late Archaic, theresults are impressive. Since none of these siteshas received systematic study in the past, the sitesthat we rely on for much of our discussion of thedevelopment of social complexity are describedthoroughly.

Upaca (02PVGS-1)

Upaca is located along the right margin of thePativilca River, Barranca province, in the districtof Pativilca, at the foot of the eastern slope ofthe Pativilca Hills (sheet 22h National mapseries). The site is located within a broad, opendry wash, or quebrada, surrounded by hills onthree sides and the valley floor on the fourth.

Upaca is a large site, measuring approximately120 hectares. It includes a series of mounds,circular plazas, residential areas, two largecompounds with what appear to be storagerooms, and a cemetery. There is a central largemound complex located in what is designatedSector A in the northeast corner of the site. Thiscomplex includes the largest platform mound atthe site, which measures 89 3 55 3 12 m tall.The mound was constructed primarily of quar-ried angular stones, though there are some rivercobbles incorporated into the fill. In contrast,mounds at other sites, such as Huayto andPotao, are constructed entirely of rounded rivercobbles. At the top of the main mound, brickand cement rubble indicate the location ofabandoned 20th-century construction. It is notclear whether these structures were ever com-pleted or occupied, but the pathway created toprovide access to the summit of the mound byheavy equipment left a notable scar.

A sunken circular plaza is associated withthe main mound. This circular plaza was placedinside a low rectangular platform extending outfrom the base of the main mound toward


the north. Toward the east and south side of themain mound are remnants of four smallerplatforms that have been seriously affectedby present-day occupation. To the west andsouth of Sector A are several smaller moundcomplexes and a second sunken circular plaza(Fig. 3).

A unique aspect of Upaca is the presence oftwo rectangular enclosures surrounded by low,wide stone walls. Located 300 m due west of themain mound in Sector D, the western quarter ofeach compound is filled with a large number ofsymmetrical room depressions arranged in rect-angular rows and columns (Fig. 4). The unifor-mity and arrangement of these rooms is reminis-cent of large blocks of storerooms found at muchlater, Late Intermediate or Late Horizon sites inPeru. There is, however, a complete absence ofany surface ceramics in the area of these rooms,and it appeared possible that these dated to theLate Archaic occupation of Upaca. The wallstoday are no more than 50 cm high but up to2 m wide. One compound measures 60 3 60 m,while the other, larger compound is 128 3

130 m. Remnants of a stone-walled road orpathway skirt the southern and western periph-ery of the site and have been followed to the westat least 3 km. The dating of this road wasimpossible to ascertain from surface observationalone.

Upaca has been subjected to considerabledisturbance and destruction in the past 50 years.The eastern margin has been turned intocultivated fields and modern housing. TheETECEN power station (Estacion Central Elec-trica del Norte, now called Red Peru) was placedin the middle of the site (see Cordova Herrera,2001). Aerial photos of the site before construc-tion of the power plant indicate that theconstruction removed what was probably resi-dential architecture but no large mounds orcommunal architecture. The nucleus of monu-ments at this site has been severely damaged byearthmoving activities during the past 10 years.Air photos taken in 1969, before construction ofthe power plant, show an extensive platformalong the west flank of the main mound. Thisplatform has been largely destroyed by heavymachinery that left only three fragments, slices ofthe platform that remained between passes of thebulldozer’s blade. The present surface, below theremoved platform mound, is covered withremains of shicra and angular rock. Portions ofrooms with plastered floors and stone walls that

have been mortared together and finished witha fine layer of plaster are among the features thatcan be seen in the remaining fragments of theplatform.

The local power company did not recognizethe presence of an archaeological site in thislocality and in 1987, according to local residents,decided to ‘‘improve’’ the power plant’s sur-roundings for the benefit of the local communityand to increase the terrain appropriate forhousing in the vicinity. A bulldozer was alsoused somewhat randomly on other parts of thesite, destroying portions of a number of otherfeatures, including a swath through the middle ofthe second circular plaza. Between January andMay 2001, an attempt was made to dig a well inthe floor of the main circular plaza at the base ofthe central platform mound. The well was leftunfinished when a depth of about 2.5 m did notreach water.

In addition to the modern and historicdisturbance, part of the Late Archaic componentof Upaca appears to have been disturbed bya later prehistoric occupation. A cemeteryadjacent to the main mound includes burialswith associated ceramics. A group of smallstructures were built within the circular plaza.These are associated with surface ceramics.

At Upaca, PANC teams cleared three profiles,each 3 m wide, and excavated two test pits: asmentioned, a remnant of an auxiliary platformattached to the west side of the principal moundin Sector A was largely destroyed by a bulldozer,but three 2-m-high strips of construction wereleft intact. Profile 1 was placed in one of theexposed, remnant sections of the west platform.A detailed drawing was made and samples ofshicra and other organic materials were collectedfor radiocarbon dating.

Profile 2 in the main mound of Sector A wascleared in an area that had been cut bya bulldozer along the western face of the mound.The bulldozer cut is almost 100 m long, anda portion in the center was cleared to examineand draw a complete profile from the top to thebase of the mound to collect shicra samples.

Profile 3 focused on a deep looter’s hole thathad been dug into the center of a mound inSector B. The structure is located 300 m south ofthe main mound in Sector A. In situ organicmaterial was clearly visible in the exposed profileand provided samples for radiocarbon dating.

Test unit DN-1 was excavated in the areaof these possible storage structures in an effort


to determine the date of these structures. Testunit 2 was excavated an area of undisturb-ed residential trash in Sector E. In each unit,we collected samples for radiocarbon dating as

well as samples for pollen analysis, macrobota-nical remains, fauna, and lithics. All test pitswere 1 3 2 m and were excavated to sterilesubsoil.

FIG. 3. Map of Upaca showing mounds and other features, sectors, canals, roads, and encroaching structures.


SECTOR A: PROFILE 1, WEST PLATFORM—Pro-file 1 is located in Sector A at Upaca, on theauxiliary platform west of the principal mound.Because of extensive modern damage, thisarea was identified as a platform mound onthe basis of historical air photos. The threeremnant ‘‘slices’’ of the mound are each about50 m long and 1–4 m high. For the presentinvestigation only a portion of the east face ofthe central remnant ‘‘slice’’ of the west platform,4 m long and 1.6 m high, was cleared andprofiled.

Screening the debris from Profile 1 yieldedmollusk shells, lithics, small textile fragments,coprolites, and plant remains, including shicrafragments, gourd fragments, carbonized seeds,and plant stems. Numerous fragments of mortarwere recovered that we associate with construc-tion, remodeling, and destruction of the plat-form. The mortar consists of clay mixed withplant fiber. Fragments often hold impressions ofcane stems and rocks or have a flat or plasteredsurface from a wall or floor.

The profile includes the following stratigraphicunits (Fig. 5):

1. Surface level of disturbed sediment contain-ing loose small to medium-sized rocks.

2. Layer of medium-grained, dark gray soilcompacted within the structure defined byWall 3 on the west. Cultural material wasalso present, including a few mollusk shells,plant remains, and loose rocks. ISGS-5395was obtained from annual plant fibercollected from this area, dating to 2650Cal B.C. (Table 4, 1F).

3. This deposit includes rocks and fragments ofmortar along with plant remains. A numberof thread or string fragments were found inthis layer, and a radiocarbon sample wascollected from the plant fibers present (GX-30117) of 2740 Cal B.C. (Table 4, 1E).

4. Floor of fine clay associated with Walls 2and 3. A basin 40 cm deep extended belowthis floor. This feature was covered witha layer of clay that was subsequently beenreplastered, suggesting that this may be theremains of a basin hearth.

5. Wall 2 was constructed of angular rock,mortared with a mix of clay and plant fibersand finished with a coat of fine plaster.

FIG. 4. Depressions in walled compound in Sector D, Upaca (J. Haas).


6. Fill of clumps of clayey soil or constructionfill (rock, mortar, and plant remains), fillingthe space between Floor 4 above and Floor14 below.

7. Construction fill of medium to large-sizedstones, hard fragments of constructionmortar, or clayey sediment mixed withplant fibers. This is the material into whichthe basin hearth (item 4 above) was built.

8. Rocky construction fill with numerousmedium-sized rocks and a few medium-sized hard chunks of clay mortar or clayeysediment.

9. Construction fill of shicra bags on bothsides of the profile was sampled and pro-vided two dates: ISGS-5291, 2310 Cal B.C.,and Beta 177319, 2350 Cal B.C. (Table 4,1A, 1B).

10. Wall 3 was built of angular rock heldtogether with mortar and covered withplaster, associated with Floors 4 and 14.Wall 3 also represents a remodeling orrebuilding of the room bounded by Wall 4.

11. Floor of fine clay associated with Wall 4that extends beneath Wall 3.

12. Wall 4 was constructed of mortared rockand covered with plaster.

13. Angular rock with compact mortar. Thisappears to be the base of Wall 2.

14. Fine clay floor associated with Wall 3.

15. Loose fill of small to medium-sized clumpsof clay mortar and rock.

16. Wall 1 was built of mortared rock andfinished with plaster. This appears to be theoldest element present in Profile 1.

Analysis identified four phases of occupationof the platform mound. Phase 1 corresponds tothe closure of the site and the postabandonmentperiod. This includes the two layers of fillimmediately above Floor 4. Layer 3 yieldeda date of (GX-30117) 2740 Cal B.C. and Layer 2(ISGS-5395) 2650 Cal B.C. These are the oldestdates of those obtained from this profile.

Phase 2 can be associated with the occupationof Floor 14, associated with the first use of Wall3. This layer also includes the use of Floor 4,associated with Wall 3, Wall 2, the possible pithearth, and the fill between the two floors (Layer6). Layer 13, the fill associated with Wall 2, mayalso date to this layer.

Phase 3 includes the occupation of Floor 11,which was constructed over construction fill inareas 8 and 9 dated to (ISGS-5291) 2310 Cal

B.C. and (Beta 177319), 2350 Cal B.C. Theplaster on Floor 11 did not appear to extend uponto the side of Wall 4. This suggests Floor 11was in use at a point following the building ofWall 4, not necessarily when the wall was firstconstructed.

Phase 4 corresponds to the oldest occupationthat can be recognized in the area cleared andincludes only Wall 1. In the profile, the floorassociated with this wall cannot be seen, as thewall continues below the bottom of the profile cut.

Conclusions—Four of the radiocarbon sam-ples from this profile were submitted for analysisand yielded dates from 41806110 to 3850670RCYBP (2740 to ,2310 Cal B.C.) (Table 4). Theradiocarbon results from Profile 1 stronglysuggest that the exposed portion of this platformwas constructed shortly after 2300 B.C. based onthe similar dates from two samples from Layer 2construction fill. The other two dates from thisprofile are substantially earlier than those fromthe shicra bags in the earlier construction fill. Apossible interpretation of this outcome is that thefill comprising Layers 2 and 3 includes disturbedmaterial. The earlier dates obtained from thesefill samples suggest that the profile displays oneof the later construction layers, and that thelower portion of the platform that remainsunexcavated dated between 2700 and 2600 B.C.Earthmoving in Sector A could have displacedearlier materials that were sampled and datedfrom the uppermost deposits in Profile 1 in a kindof reverse stratigraphy.

SECTOR A: PROFILE 2, MAIN MOUND—Profile 2is located in Sector A on the west side of themain mound. An area approximately 3 3 12 mwas cleared, extending to the top of the mound.

The entire west side of the main mound hasbeen cut by the activity of heavy machinery. Atrail made by a bulldozer crosses the west face ofthe mound. Although the trail has not been usedrecently, these activities disturbed the moundbeneath it to a depth of 2 m. This action left theside of the structure exposed and permitted us torecord the construction features on this side ofthe mound.

At the time Profile 2 was being cleared, thesurface of the mound was covered with stone,and no architectural remains were visible. Therock that covered the surface was angular,medium to large-sized fragments. Clearing theexposed profile revealed a substantial damagedarea. Cultural material recovered includedlithics, mollusks, fauna, textiles, coprolites,


a brilliant green feather, botanical material,seeds, corn, and shicra bags in secondarycontext. No distinctive whole artifacts wererecovered from this location, a pattern thatproved common in testing Late Archaic sites. Alarge quantity of the stone visible in the profile isloose rock without mortar. These loose rocksmay have been piled or shifted to facilitate thepassage of heavy machinery. The absence of

mortar suggests these rocks have been exposed tothe elements for an extended period of time.

The profile included the following stratigraph-ic units (Fig. 6):

1. Angular rock consisting primarily of frag-ments greater than 20 cm on a side

Floor 1: Fragments of a gray-colored clayfloor

FIG. 5. Upaca, Sector A, Profile 1, platform attached to main mound.


TABLE4.

Up

aca

rad

ioca

rbo

nd

ate

sfr

om

sam

ple

so

bta

ined

in2

00

2.

Pro

ven

ien

ce

Tex

ta

nd

illu

stra

tio

nlo

cati

on

Ma

teri

al

La

bn

o.

Ca

l.B

.C.

RC

YB

P12C

/13C

Sa

mp

le-

wei

gh

t(g

)C

ali

bra

ted

ag

era

ng

e1

sig

ma

(68

.3%

)B

.C.

Ca

lib

rate

da

ge

ran

ge

2si

gm

a(9

5.4

%)

B.C

.

Mo

un

dA

,P

rofi

le1

1B

Fib

erb

ag

Bet

a-1

77

31

92

35

03

88

06

60

–9

.51

52

46

1–

22

89

(10

0.0

)2

50

0–

25

41

(.7

),2

49

1–

21

96

(94

.7),

21

67

–2

14

4(1

.9)

Mo

un

dA

,P

rofi

le1

1F

Mix

edp

lan

tfi

ber

sIS

GS

-53

95

26

50

40

80

67

0–

11

.51

02

85

7–

28

13

(19

.0),

27

39

–2

72

5(4

.9),

26

97

–2

55

9(6

2.3

),2

53

5–

25

31

(1.4

),2

52

4–

24

96

(12

.5)

28

73

–2

79

9(1

8.1

),2

78

4–

24

70

(81

.9)

Mo

un

dA

,P

rofi

le1

1E

Mix

edp

lan

tfi

ber

sG

X-3

01

17

27

40

41

80

61

10

–2

4.3

6.6

28

85

–2

62

1(9

7.8

),2

90

7–

26

01

(2.2

)3

02

1–

24

68

Mo

un

dA

,P

rofi

le1

1A

Fib

erb

ag

ISG

S-5

29

12

31

03

85

06

70

–1

0.9

15

24

55

–2

44

1(5

.8),

24

35

–2

42

1(5

.5),

24

04

–2

36

0(2

0.2

),2

35

4–

22

69

(42

),2

26

0–

22

03

(26

.5)

25

46

–2

45

5(.

1),

24

88

–2

47

9(.

5),

24

74

–2

13

3(9

7.4

),2

08

0–

20

47

(2)

Mo

un

dA

,P

rofi

le2

2B

Mix

edp

lan

tfi

ber

sIS

GS

-52

80

21

90

37

70

67

0–

11

.21

52

29

4–

21

24

(80

.9),

20

96

–2

09

0(1

.9),

20

84

–2

04

0(1

7.2

)2

45

6–

24

21

(2.7

),2

40

4–

23

58

(4.9

)2

35

4–

20

14

(90

.7),

19

97

–1

97

9(1

.8)

Mo

un

dA

,P

rofi

le2

2E

Mix

edp

lan

tfi

ber

sIS

GS

-52

94

22

70

38

20

67

0–

10

.81

22

40

1–

23

77

(9.2

),2

35

0–

21

91

(76

.8),

21

78

–2

14

2(1

3.9

)2

46

6–

21

23

(94

.2),

20

96

–2

04

0(5

.8)

Mo

un

dA

,P

rofi

le2

2A

Mix

edp

lan

tfi

ber

sIS

GS

-52

95

22

70

38

20

67

0–

12

.81

1.8

24

01

–2

37

7(9

.2),

23

50

–2

19

1(7

6.8

),2

17

8–

21

42

(13

.9)

24

66

-21

23

(94

.2),

20

96

-20

40

(5.8

)

Mo

un

dA

,P

rofi

le2

2C

Mix

edp

lan

tfi

ber

sIS

GS

-52

96

23

30

38

60

67

0–

12

.28

.62

45

7–

24

17

(17

.7),

24

09

–2

27

8(6

5.5

),2

25

2–

22

30

(10

.9),

22

20

–2

20

1(5

.9)

25

54

–2

53

9(.

9),

24

83

–2

13

6(9

8.5

),2

07

8–

20

66

(.6

)

Mo

un

dB

,P

rofi

le3

3B

Fib

erb

ag

Bet

a-1

77

32

18

70

27

00

66

0–

25

.41

48

98

–8

08

99

6–

98

9(.

9),

97

4–

79

2(9

9.1

)

Mo

un

dB

,P

rofi

le3

3A

Mix

edp

lan

tfi

ber

sIS

GS

-52

81

11

10

29

10

67

0–

25

.11

21

25

5–

12

44

(4.3

),1

21

2–

11

98

(6),

11

93

–1

13

8(2

4.1

),1

13

3–

10

01

(65

.5)

13

67

–1

36

2(.

4),

13

14

–9

16

(99

.6)

Mo

un

dB

,P

rofi

le3

3D

Fib

erb

ag

GX

-30

11

81

11

02

91

06

80

–2

2.1

5.5

12

57

–1

23

7(7

.2),

12

15

–9

98

(92

.8)

13

71

–1

35

7(.

01

2),

13

17

–9

02

(97

.9),

13

52

–1

31

4(.

9)

Sec

tor

D,

Tes

tU

nit

1

T1

Ch

arc

oa

lIS

GS

-52

73

21

02

16

06

70

–2

6.3

63

56

–2

88

(34

.2),

25

7–

25

6(4

.9),

23

3–

10

7(6

0.9

)3

83

–4

4(1

00

.0)

Sec

tor

E,

Tes

tU

nit

2

T2

Ch

arc

oa

lIS

GS

-52

75

11

60

29

50

67

0–

26

.05

12

87

–1

28

3(1

),1

26

1–

10

46

(99

)1

38

0–

13

35

(5.4

),1

32

1–

97

3(9

3.5

),9

56

–9

42

(1.1

)


Floor 2: Fragments of floor, a mix of grayclay and small stones

Only one phase of ancient occupation could beclearly identified in this profile, as the areacleared was covered with a continuous layer ofplaster. The bulldozer disturbed the area at thefoot of the profile, but the profile itself proved tobe an undamaged exterior surface of the mound(Fig. 7). A later phase may be indicated by thepatch of Floor 3 remaining at the top of Wall 1.This small area was a paler color and a finertexture than the plaster on the rest of the profileand may indicate a late replastering that was notcarried out across the entire structure. Noexcavation into the mound itself was carriedout, and samples for radiocarbon dating wereobtained from the disturbed area at the base ofthe mound, where shicra bags were exposed, andin places at the top of each retaining wall wherethe upper plastered floor surface was damagedand shicra bags were visible. Samples were alsocollected from fiber present in wall plaster.

Conclusions—Four samples from Sector A,Profile 2, were submitted for radiocarbonanalysis. Floor 1 yielded a date of 2190 CalB.C. and Floor 2 a date of 2270 Cal B.C. Ashicra bag at the base of the excavated areayielded a date of 2330 Cal B.C., while fiber froma plaster on Wall 1 dated to 2270 Cal B.C. Thesedates suggest that the mound was completed and

covered with plaster by 2270 Cal B.C., thoughits construction took place earlier, by 2330Cal B.C., based on the earlier date from a shicrabag, clearly used in construction. These datescome from the uppermost layers of construc-tion and plastered surfaces. Interior portions ofthe mound and the unexcavated portion of themound below the present ground surface are allassumed to be older.

SECTOR B: PROFILE 3—Sector B is at theextreme southern end of the site. The centralfeature in this sector is a small mound of angularrock, though some rounded river cobbles arescattered around the upper surface of themound. The mound is rectangular, approximate-ly 15 3 20 m at the base and an average of 6 mhigh.

A hole roughly 2 3 2 m and 2.2 m deep isvisible at the top of the mound, which exposeda series of rooms. Since the unit was left open, itwas initially believed to be the work of looters,despite its rectangular shape, since no report ofarchaeological investigation of this description atUpaca has been identified. When a series of 1 3

2-m test pits were identified extending across thesite below a set of transmission lines, investiga-tion revealed that these were part of anevaluation associated with expansion of theelectrical power plant (Cordova Herrera 2001).The hole in the top of Mound B may have been

FIG. 6. Upaca, Sector A, Profile 2, west side of main mound.


excavated as part of this evaluation, though it isnot mentioned in the report.

The hole in Mound B appears likely to havebeen made at the time the power line wasinstalled nearby. The open hole revealed seg-ments of walls, some unfinished and some finelyplastered, along with layers of construction fill.Shicra bags used in the mound construction werealso visible. The surfaces exposed did not containceramics, and many mollusk valves and flakes ofstone were present both in the deposits and onthe surface. The mound has also been disturbedon its northeast corner, where an irregular-shaped, shallow pit revealed plastered wallsurfaces.

Profile 3 was positioned to take advantage ofthe faces of the hole on the top of the mound andits four exposed profiles. The east and westprofiles were cleared and drawn. Several super-imposed floors were identified, along with twowalls, which delimited a narrow passagewayblocked by a sloping wall. Screening the sedi-ment associated with this profile yielded a num-

ber of mollusk shells, lithics, small textilefragments, botanical remains including frag-ments of shicra bags, gourd fragments, carbon-ized seeds, plant stalks, mortar and wallfragments covered with plaster painted white,red and orange.

The west profile of the looted area in Mound Bincluded the following units (Fig. 8A):

1. Fill of loose, angular rock.

2. Fill of construction debris with a few rocks.Embedded plant fibers are visible across thenorth face of the profile. It is not clearwhether these are discarded fragments,fragments shed from construction fill, orparts of shicra bags. A sample of thesefibers dated to 870 Cal B.C. (Beta 177321).

3. Lens of clayey sediment of a light beigecolor that appears to have been depositedby wind or water.

4. Fine clay floor (Floor 4) associated withWall 4. Destruction of a wall at the northend of the profile is represented by theupturned edge of the floor that suggestsa wall was present when the floor was laidbut is no longer there. An even moreprominent turn in floor material is visiblein the lower section of Floor 4.

5. Fill having a fine, soft texture from thepresence of clay. Lumps of clay, possiblymortar fragments, are also present.

6. Fill of medium-sized rocks with medium-sized clods of clayey soil, shells, and plantremains. This appears to be a zone in whichwalls of the room fell in. From plant fibersin this area, a date of 1110 Cal B.C. wasobtained (ISGS-5281).

7. Floor of fine white clay.

8. Fill of small clumps of clay or mortar andsmall stones with numerous plant frag-ments.

9. Fill of medium-sized rocks and someclumps of clayey soil and lumps of mortar.

10. Wall 1, made of plaster-covered, mortaredrock finished with fine layer of white plaster.

11. Wall 2, made of plaster-covered, mortaredrock finished with fine layer of whiteplaster. A hole in the plaster covering Wall2 permitted us to collect plaster fragmentsand extract plant fiber from it for a radio-carbon date of 1110 Cal B.C. (GX-30118).

12. Wall 3, made of plaster-covered, mortaredrock finished with fine layer of whiteplaster.

FIG. 7. Photo of the northwest facing side of themain mound at Upaca in the area cleared as Profile 2(W. Creamer).


13. Wall 4, made of plaster-covered, mortaredrock finished with fine layer of whiteplaster.

14. Fill of rock and construction debris. Thisformed the floor of the hole and noadditional excavation was carried out.

Analysis of both the east and west profiles allowsus to identify four phases that can best be seen inthe west profile.

Phase 1 corresponds to the closure or aban-donment of the structure when the fill in Layer 2was added (Fig. 8B). Layer 1 is an intrusion intothis fill in the south corner of the west profile. Atthe interface between these two units, a ceramicsherd was found, part of a neckless jar.

Phase 2 includes the occupation of Floor 4,constructed over fill of Layer 6. What remains ofWall 4 stands 25 cm high and is topped bya patch of clay that may be a patch of floor orthe surface of a step.

Phase 3 includes occupation of Floor 7, whichis associated with the fill in Layers 8 and 9.Construction during this phase covered Wall 3but is associated with Wall 1.

Phase 4 corresponds to the occupation begin-ning with construction of Walls 1, 2, and 3. Wall2 is slightly inclined toward the interior of themound, and at its lower end is a step that goes inthe same direction. On the extreme east end ofWall 3, a few centimeters from the east profile, isone side of a niche filled during later construc-tion.

Conclusions—Based on the architecture andthe material observed in the exposed profiles, itappeared that the mound in Sector B was notconstructed during the Late Archaic. Threesamples submitted for analysis from this profileyielded dates of 870 Cal B.C. and two dates of1110 Cal B.C. (Table 4). Those first millenniumB.C. dates, during the Initial period, suggestalternative interpretations. The Sector B moundlies far enough from the main mound at Upacathat it could date to another period, an exampleof reuse of the Upaca locality. The only aspect ofProfile 3 that suggests an earlier date than thatobtained from radiocarbon dates is the fact thatonly one ceramic fragment was recovered fromthis locality. However, the dates come from clearcontexts, so they are likely to be correct. Sector Bcould date to a later period at Upaca, or it maybe a separate and later occupation.

SECTOR C—Sector C includes the extremesouthwest portion of Upaca. It includes a small

rectangular platform mound with a sunkencircular plaza. The approximate dimensions ofthe mound are 10 3 20 3 5 m, while the circularplaza is about 20 m in diameter. As in Sector B,the principal construction material is angularrock, though some river cobbles were observedon the surface of the mound.

The structures on the upper part of the moundin the central area are double-faced walls andsmall rectangular rooms approximately 2 3 3 m.The circular plaza has been damaged by heavymachinery. On the surface of this sector, molluskvalves, organic materials, and fragments offlaked stone are visible. This mound is thesecond with a sunken circular plaza within thesite. It is one of the smallest mounds that hasbeen reported associated with a circular plaza,but the architectural characteristics suggest themound dates to the Late Archaic period. Noresearch was conducted in this sector. Thebulldozer trench through the circular plazaappears to have been random destruction, asthere is no evidence of looting.

SECTOR D: TEST UNIT 1—Sector D is located inthe northeast portion of the site, 400 m west ofthe principal mound (Sector A). Sector Dincludes an extensive area of regularly spaceddepressions aligned in rows that suggests anintentional layout. The depressions do notdisplay any architectural features, but they aresimilar in size, approximately 4 3 5 m andregularly distributed across an area some200 3 200 m. These depressions appeared tohave been storage spaces, probably pits thatwere not roofed, but were covered with matting,cloth, or some similar material to protect thecontents. A stone wall at least 2 m thicksurrounds these features, which may have beena row of masonry structures that served asa walled border. Fragments of ceramics arevisible on the surface, along with flakes of stoneand mollusk valves.

There is some evidence of excavation bylooters in this area. There is also a series of opentest pits each approximately 1 3 2 m extendingacross Sector D. During surface examination ofthe site, ceramics, lithics (including a bifacialtool), and botanical remains were found in theexcavated fill of several of these pits, which washeaped at the side of each unit. Subsequentresearch suggests these excavations were carriedout by Cordova Herrera (2001), who conductedthe work for the power company at a time theysought to increase the number of transmission


FIG. 8. Upaca, Sector B, Profile 3, (A) Excavated units and carbon sample locations. (B) Construction phases.


poles crossing the site. Despite the presence ofthese artifacts and others visible in the strata ofthe open pits, the area tested was declared clearof any cultural remains, and additional postswere set in place in the site.

SECTOR D: TEST UNIT 1—This unit was 1 3

2 m in size, located within one of the rectangularstructures which cover this sector. A datum wasset 5 cm above the present ground surface, andarbitrary excavation units of approximately15 cm were employed. The first level beganat 40 cm below datum. On the surface wereangular stones, many small stones, and flakes.The sediment was dark gray, loose-textured,and composed of fine sand, which was foundbeneath a maroon-colored burned area. On theeast side of the unit was a hard patch ofsediment. None of the distinctive surface materi-als continued in the excavated layers. From thisunit lithics, mollusk shells and botanical remains,including flotation and pollen samples, wererecovered.

Seven levels were excavated before sterilesubsoil was reached. While recording and draw-ing the levels, it was possible to identify thedifferent natural and cultural levels present(Fig. 9).

Layer 1: Compact sand with small angulargravel less than 5 cm. This constitutes about 30%of the fill in this layer.

Layer 2: Groups of large rocks separated byloose sandy sediment containing botanical re-mains.

Layer 3: Sandy laminated sediment with bandsof small rocks less than 5 cm on a side.

Layer 4: Small angular rock, less than 3 cm ona side at the east side of the unit, graduallychanging to 5–7 cm on the west side. In this layer,there is also compact sandy sediment, a largequantity of charcoal, and mollusk fragments.

Layer 5: Compact lens of hardened andcalcified salt deposits (caliche).

Layer 6a: Small angular gravel compacted insandy sediment. Gravel dimensions less than 5 cm.A chunk of charcoal was collected in this level thatprovided a date of 210 Cal B.C. (ISGS-5273).

Layer 6b: Small angular gravel compacted insandy sediment. Gravel dimensions greater than5 cm.

Layer 7: Compact lens of hardened andcalcified salt deposits (caliche).

Analysis of the excavation suggests the datarepresent three phases of possible use andabandonment.

FIG. 9. Upaca, Sector D, Test Unit 1, excavated units.


Phase 1 is the uppermost deposit, Layer 1(Fig. 10). This is a substantial layer of post-abandonment fill, probably the result of windand to a lesser extent water from occasionalrain.

Phase 2 appears to be an episode of disposalof rock and trash. The rock in Layer 2 wasdeposited on a sloping layer of laminatedsediment banded with small angular gravel. Thelaminated material, Layer 3, appears to be a winddeposit, suggesting some time elapsed with thesurface exposed to the elements between de-position of Layers 2 and 4. Layer 4 includesmaterial deposited on a sloping surface, perhapsthe side of the depression visible on the surface.Abundant charcoal and mollusk remains suggestthat this was a trash deposit. Together, the Phase2 deposits had a leveling effect, the three episodesof deposition turning a sloping surface intoa horizontal one.

Phase 3 is the earliest visible and is found inexcavation Layers 5–7. These were probablyinitially a single deposit and the caliche in Layers5 and 7 formed in situ. The single radiocarbonsample from this unit came from Layer 6. The

date of 210 Cal B.C. (ISGS 5273) comes fromcharcoal and may date earlier than comparablesamples of annual plant fiber as a result of theold wood problem.

Conclusions—Only a single radiocarbon sam-ple was recovered from Sector D, Unit 1, 210 CalB.C. (ISGS 5273), from a charcoal fragment.This provides only minimal dating of the areabut places this compound during the EarlyHorizon. The date comes from below the clearestlevels of intentional deposition, Layers 2, 3, and4. The area of Upaca covered by regularly spacedpits, therefore, appears to be a feature postdatingthe construction of the main mound and otherstructures in Sector A.

The walls or outer margins of the depressioncould not be clearly identified during excavationor from a close examination of the profile of thisunit. These depressions are proposed to havebeen rooms or storage spaces, though thefunction and construction of these features isnot clear in the areas excavated. The radiocarbondate of 210 Cal B.C. suggests the circularfeatures in Sector D appear to have been in useduring the Early Horizon (900 B.C.–A.D. 200).

FIG. 10. Upaca, Sector D, Test Unit 1, phases.


SECTOR E: TEST UNIT 2—This sector is locatedin the southwest portion of the site and includesthe area that slopes toward the hills as well asa portion of the hillside. On the surface, a largenumber of rocks can be seen, and in spacesamong the rocks, some mollusk shell is alsovisible. Within this sector there is also part ofa prehispanic road that crosses the site followingthe middle of the hill slope. It is possible toidentify two terraces in this sector, one of whichis associated directly with the road, while thesecond terrace has stone flakes and the remainsof mollusks on its surface. Test Pit 2 was placedon this terrace at the extreme southern end ofSector E, on a slight slope that extends froma small terrace about 6 3 9 m (Fig. 11). This unitwas excavated in natural layers.

Layer A included many angular rocks ofmedium and small size distributed homoge-neously in fine loose soil. A few fragments ofmollusk shells were present along with a smallquantity of ceramics. There were also pupacasings of flies and small fragments of charcoal.This layer was formed by aeolian deposition overdomestic trash. The presence of fly pupaeindicates that this trash was not buried but wasleft exposed.

Layer B consisted of very fine, loose, light graysoil, darker than that in Layer A, with a largenumber of angular rocks of medium and smallsize distributed evenly through the level. The

layer included a large number of mollusk valvefragments, a small quantity of ceramics (distin-guished by an incised decorated fragment), somebotanical remains (plant stalks), and pieces ofcharcoal, all of which were distributed evenlyacross the unit. In addition, a coprolite anda fragment of quartz were found. This layer wasformed by the deposition of domestic trash.

Layer C was composed of a large quantity ofangular rock of all sizes, small to large, in a fine,beige sediment. Cultural material was presentonly in the upper part of this layer, the interfacewith Layer B, including minimal quantities ofmollusk shells and botanical remains. This wasa natural layer, part of the natural depositionand slope of the hill, that displayed culturalmaterial only at the interface with Layer 2.

Conclusions—The excavation data indicateone phase of occupation in this area. Each layer,A and B, indicates an episode of deposition. Thesurface material, Layer A, was windblownsediment deposited by natural forces afterabandonment of the locality and long after itsuse as a midden. Layer B includes the middendeposit. One sample of charcoal from Layer Cyielded a date of 1160 Cal B.C. (ISGS-5275).Since the bulk of Layer C did not containcultural material, this charcoal appears to datethe earliest use of this locality. Along with thepresence of ceramics in both of the layers bearingcultural materials, it appears that this midden

FIG. 11. Upaca, Sector E, Test Unit 2.


deposit dates to the Initial period. Further, thesample tested was charcoal and may possiblyhave been old wood; thus, the date may be olderthan the locality where it was collected.

DISCUSSION: UPACA—The three profiles andthree test pits excavated at Upaca yielded a broadrange of chronological data. Radiocarbon sam-ples yielded dates that confirmed a significantLate Archaic occupation in Sector A, includingthe main mound and associated circular plaza,with dates ranging from approximately 2700 to2200 Cal B.C. (Table 4). A second area, SectorC, with a smaller mound and sunken circularplaza, appears to be structurally similar to themound complex in Sector A and may also date tothe Late Archaic. This sector remains to betested.

All the radiocarbon samples collected fromprofiles were annual plant fiber, and most ofthose came from shicra bags. In the test units,however, only charcoal samples were recovered.The charcoal could be affected by the old woodproblem to yield dates skewed earlier than thoserecorded from annual plants. However, datedcharcoal samples appear to come from contextsthat fall during the Initial period or later.Whether or not the dates are absolutely accurate,they do not date to the Late Archaic period, andthey do not affect our interpretation of the LateArchaic dates collected at Upaca thus far.

Six of the eight dated samples from Sector Acome from shicra bags. One sample of plant fiberwas taken from in situ wall plaster on the SectorA mound, and one sample included plant fiberfrom under a hearth. Two of the dates fromProfile 1 are earlier than those from more deeplyburied contexts (2740 Cal B.C. and 2650 CalB.C.), suggesting that while there may be buriedstrata of this age, the dates obtained are not fromin situ material. Of the remaining two dates fromProfile 1, the destroyed platform appears to beslightly earlier than the main mound dated fromProfile 2. The low mound may have been builtearlier than the taller structure flanking it, and atthis point it is the earliest securely dated in situcontext at Upaca. Dates overlap betweenmounds at 2330–2310 Cal B.C., suggesting thatboth the main mound and the lower platformhad been built by that date. Fiber in the moundplaster dates to 2270 Cal B.C., possibly the lastplastering of this part of the main mound,though the latest date from a shicra bag higheron the main mound is 2190 Cal B.C. (Table 4).Therefore, the last Late Archaic use of the

mounds in Sector A may have come as early as2190 Cal B.C. While the range of occupationappears to extend from at least 2740–2190 CalB.C., the areas tested were from the latter part ofthe occupation when the mounds in Sector Awere in their final form, 2330–2190.

Both the Sector B mound and the test unitsyielded date more than 1,000 years later. InSector B, shicra bag fragments yielded dates of1110–870 Cal B.C., as did a small sample ofplant materials that appeared in the corner ofa plastered room floor in Mound B. Thepresence of ceramics in Profile 3, Sector B,and in the test pits in Sectors D and E furtherindicates that these structures postdate thethird millennium B.C. At some sites, intrusivefeatures disturbed Late Archaic features, but thisdoes not appear to be the case in Sectors B, D,and E.

The Late Archaic occupation at Upaca islikely to have covered an area greater than SectorA. There may be Late Archaic deposits in SectorC, for example, where the small mound andsunken circular plaza appear to be a similar ifsmaller version of Sector A, though this sectorremains to be tested. In addition, residentialconstruction probably extended over a portionof the site. The later structures in Sector B andthe test excavations in Sectors D and E did notreveal Late Archaic levels below those bearingceramics. Thus, additional testing would benecessary to define the dimensions of LateArchaic habitation at Upaca.

The radiocarbon dates, though few in numberfor the size of the site, suggest an extensiveoccupation during the Late Archaic period, whenthe Sector A mound and circular plaza wereconstructed, and a smaller occupation (Sector Bmound) during the Initial period. During theEarly Horizon, the extensive area of pits wascreated, though their function is still not clear.Upaca’s long occupation may have been discon-tinuous, but there was consistent use of the siteover at least 2,500 years.

Punta y Suela (02PVGS-02)

Punta y Suela is located on the right margin ofthe Pativilca River, 8 km from the coast and4.5 km east of the town of Pativilca. Coveringapproximately 100 hectares, the site is a complexwith four large mounds, six secondary moundsor low platforms, and an extensive residential


area. Four sectors have been delineated withinthe site (Fig. 12). Sector A includes the principalmound and the associated sunken circular plaza.Sector B is comprised of five smaller moundslocated north of Sector A, one of which also hasan associated sunken circular plaza. Sector C islocated to the east of Sector A and includesanother mound smaller than that in Sector A butlarger than the mounds in Sector B. Sector D isan extensive area on the northern part of the sitewhere there are numerous lithic flakes andmollusk shells on the surface.

Punta y Suela is situated at the southeast footof the Pativilca Hills (Instituto GeograficoNacional sheet 22h). Comparison of historic airphotos with the present landscape show that oneof the two circular plazas has been completelydestroyed by cultivators, and only a small cornerof the other now remains after recent roadwork.The part of the site that includes the principalmounds, Sectors A and B, is now surrounded byfields that have come under cultivation since1970, and an irrigation canal now circles most ofthe base of the principal mound in Sector A. Thearea immediately west of the main mound hasrecently been damaged by heavy equipment withthe result that many structures have beencompletely destroyed (Fig. 13A, B).

Some areas at a distance from the centralprecinct have survived, while others have alsosuffered serious damage. In the extensive resi-dential areas, the excavation of several canals hasdestroyed architecture. The occupants of theland around the site have also excavated a varietyof holes and pits to extract sand or to search forartifacts.

Destruction is the result of several factors, theexpansion of agricultural fields, roadwork, andthe construction of electric power transmissiontowers. Today, cultivated fields cover the sectorsof Punta y Suela with mound construction.Previously, this was eriazo, land that is neitherirrigated nor cultivated. A large part of theresidential portion of the site is still eriazo, butan irrigation canal is under construction thatappears to be intended to convert much of theremaining dry area of Punta y Suela intoagricultural fields.

At Punta y Suela, field crews cleared oneprofile 3 m wide in Sector C and three 1 3 2-mtest units in Sector D. Eleven radiocarbonsamples from those collected have been pro-cessed and yielded dates (Table 5).

Profile 1 was cleared along one face of thetrench that exposed the interior of the Sector Cmound. A detailed drawing was made, andsamples of organic remains were collected forradiocarbon dating.

Test Unit 1, or DN-1, was 1 3 2-m test unitplaced in the northern portion of Sector D atPunta y Suela in an area that had a scatter ofshell and chipped stone on the surface.

Test Unit 2, or DN-2, was placed in thenorthern portion of Sector D, where it was alsopossible that habitation associated with the sitewould have been located. There was someevidence of looting in this part of the site,though the materials discarded on the surfacesuggested these were later burials.

Test Unit 3, or DS-1, was excavated in thesouthern portion of Sector D. This 1 3 2-m unitwas placed in an undisturbed area north of theSector C mound. It was possible that the remainsof habitation associated with the mounds wouldbe identified and samples collected for radiocar-bon dating.

SECTOR A—Located on the southeast portionof the site, this sector includes a large, raisedplatform mound surrounded by cultivated fields,principally cornfields, irrigated by modern ca-nals. Construction material was primarily rivercobbles at Punta y Suela, though some angularrock could be seen on the surface. The mainmound measures 40 3 50 m and is approximate-ly 12 m high. On the summit of the mound isa fallen monolith, or hvanca, and a series ofstructures that are roughly delineated by thebases of walls, though the form of the rooms isnot clear.

The sunken circular plaza in Sector A is clearlyvisible in the 1970 SAN aerial photo. Today,however, the sunken circular plaza has beenleveled, and the field is cultivated, though a patchof gravel can been seen in the field roughly in thelocation of the former circular plaza when thecrops have been harvested and the field is clear.No excavations were carried out in this sector forlack of exposed areas to clear.

SECTOR B: SAMPLE COLLECTION—This sector islocated 50 m north of the main mound in SectorA. Sector B includes three rectangular mounds,each with dimensions smaller than the principalmound. One mound’s surface is covered withsmall and medium river cobbles. The largest ofthe mounds in Sector B, measuring 24 3 26 m 3

4 m high, has been heavily damaged by efforts tolevel it to extend agricultural fields. Several other


FIG. 12. Map of Punta y Suela showing mounds and other features, sectors, canals, roads, andencroaching structures.


FIG. 13. (A) Aerial view of Punta y Suela in 1970 (SAN). (B) Photo of Punta y Suela in 2002 (J. Haas).


TABLE5.

Pu

nta

yS

uel

ara

dio

carb

on

da

tes

fro

msa

mp

les

ob

tain

edin

20

02

.

Pro

ven

ien

ce

Tex

ta

nd

illu

stra

tio

nlo

cati

on

Ma

teri

al

La

bn

o.

Ca

l.B

.C.

RC

YB

P12C

/13C

Wei

gh

to

fsa

m-

ple

(g)

Ca

lib

rate

da

ge

ran

ge

1si

gm

a(6

8.3

%)

B.C

.C

ali

bra

ted

ag

era

ng

e2

sig

ma

(95

.4%

)B

.C.

Mo

un

dB

S/D

AC

ha

rco

al

ISG

S-A

-42

22

20

03

77

56

35

–1

0.7

12

27

8–

22

52

(23

.7),

22

30

–2

22

0(8

.4),

22

07

–2

14

0(6

8.1

)2

30

3–

21

25

(91

.4),

20

95

–2

09

2(.

3),

20

84

–2

04

1(8

.3)

Mo

un

dB

S/D

BM

ixed

pla

nt

fib

ers

ISG

S-A

-42

12

42

03

93

56

35

–1

4.9

12

47

2–

24

01

(73

.8),

23

79

–2

34

9(2

6.2

)2

55

8–

25

36

(4.6

),2

49

5–

23

02

(95

.4)

Mo

un

dC

,P

rofi

le1

PA

Wo

od

ISG

S-5

28

41

48

03

21

06

70

–2

7.1

18

15

99

–1

58

7(5

.6),

15

81

–1

57

0(4

),1

52

9–

14

07

(90

.4)

16

82

–1

66

8(1

.3),

16

61

–1

64

9(.

9),

16

40

–1

37

0(9

4.2

),1

35

8–

13

50

(.5

),1

34

2–

13

17

(3.1

)S

ecto

rD

-N1

,T

est

Pit

1T

AM

ixed

pla

nt

fib

ers

Bet

a-1

77

31

75

60

24

30

67

0–

25

.85

75

8–

68

4(3

0.5

),6

60

–6

44

(6),

58

6–

58

4(1

.1),

54

4–

40

4(6

2.4

)

76

5–

39

6(1

00

.0)

Sec

tor

D-N

1,

Tes

tP

it1

TB

Ch

arc

oa

lG

X-3

01

20

64

02

55

06

70

–2

6.1

12

80

3–

75

6(2

6.5

),6

99

–5

40

(73

.5)

82

3–

48

3(9

4.1

),4

66

–4

48

(2.4

),4

42

–4

13

(3.5

)S

ecto

rD

-N1

,T

est

Pit

1,

Lev

el1

0

TC

Mix

edp

lan

tfi

ber

sIS

GS

-53

93

91

70

97

50

61

10

–2

2.8

15

93

07

–9

11

7(7

1.9

),8

99

1–

89

07

(22

.5),

88

76

–8

85

7(4

),8

85

0–

88

43

(1.6

)

95

97

–9

56

0(1

.5),

94

54

–9

44

1(.

5),

93

92

–8

77

9(9

7.2

),8

77

0–

87

48

(.8

)S

ecto

rD

-N2

,T

est

Pit

2T

DM

ixed

pla

nt

fib

ers

ISG

S-5

29

27

50

26

00

67

0–

26

.21

88

33

–7

59

(56

.4),

68

3–

66

3(9

.3),

64

0–

58

8(2

0.9

),5

81

–5

45

(13

.4)

90

3–

51

6(9

8.1

),4

62

–4

51

(.8

),4

39

–4

29

(.7

),4

21

–4

14

(.4

)S

ecto

rD

-N2

,T

est

Pit

2T

EC

ha

rco

al

ISG

S-5

27

21

84

03

52

06

70

–2

5.2

15

19

35

–1

93

4(.

9),

19

21

–1

74

4(9

9.1

)2

03

2–

16

83

(99

.7),

16

67

–1

66

3(.

2),

16

46

–1

64

4(.

1)

Sec

tor

D-S

1,

Tes

tP

it1

,T

FC

ha

rco

al

Bet

a-1

77

31

85

41

06

44

06

70

–2

3.8

18

54

75

–5

36

1(9

7.2

),5

34

6–

53

44

(2.1

),5

23

6–

53

25

(.7

)

55

40

–5

29

0(1

00

.0)

Sec

tor

D-S

1,

Tes

tP

it1

TG

Ch

arc

oa

lG

X-3

01

19

54

10

64

50

69

0–

25

.71

85

47

9–

53

38

(94

.9),

53

32

–5

32

2(5

.1)

56

13

–5

58

6(2

.7),

55

60

–5

25

9(9

7),

52

36

–5

23

5(.

1),

52

17

–5

21

4(.

2)

Sec

tor

D-S

1,

Tes

tP

it1

TH

Ch

arc

oa

lIS

GS

-52

71

62

80

74

10

67

0–

25

.11

86

38

3–

62

18

(10

0.0

)6

41

5–

61

97

(82

.5),

61

94

–6

15

9(7

.2),

61

39

60

89

(10

.4)


Sector B mounds visible in the aerial photos areno longer visible. These smaller mounds havebeen either completely demolished or ploweddown around the margins.

The remains of a sunken circular plazaapproximately 15 m in diameter can be seenalong the south side of the largest mound inSector B, though little remains after roadworkand expansion of the cultivated fields over thesite. Two samples of charcoal were recoveredfrom a profile exposed along the road throughthis sector. One sample of plant fibers was AMSdated to 2420 Cal B.C. (ISGS-A421) (Table 5).The second sample was charcoal AMS dated at2200 Cal B.C. (ISGS-422). No other eroded orexposed areas were identified that appearedlikely to yield samples for radiocarbon dating,and no excavation was undertaken in this sector.

SECTOR C: PROFILE 1—This sector is located400 m southeast of Sector A. It includes a largemound 37 3 74 m at the base and 6 m high. Thesurface of the mound is covered with rivercobbles. The mound in Sector C was cut in halfwhen a power line was constructed and thecentral portion of the mound was removed by

heavy machinery in order to increase the distancebetween the power lines and the ground once thepower pole had been installed (Fig. 14). Sector Cwas further damaged by heavy machinery duringwidening of a jeep trail that runs between themound and an irrigation canal. Two profileswere exposed by the work of heavy machinery.One of these profiles was cleared and recorded.This profile offered an opportunity to determinethe construction methods of a mound from topto bottom without conducting a large-scaleexcavation. Carbon samples were collected fromthe material exposed in the profiled cuts,designated Profile 1.

Profile 1 is located in the southern part of thecut made by heavy machinery in the center of themound. A swath through the mound wasremoved, cut in a way that created a profilehaving three tiered levels, which indicate themovements of the bulldozer. PANC team mem-bers cleared an area 3 3 7 m.

Construction fill included river cobbles, a smallnumber of angular stones, clay, and some loosesediment. The construction did not incorporateshicra bags but employed a simpler method,

FIG. 14. Photo of power pole set in Mound C, Punta y Suela (J. Haas).


stacking rock and adding in earth and wet claymixed with a little bit of trash.

Fill from clearing Profile 1 included numerousriver cobbles, a small amount of pottery, molluskshells, botanical remains (gourd fragments,charcoal, seeds, and stems), possible lithics, anda few animal bone fragments.

The profile was cleared carefully, revealingseveral levels of material (Fig. 15).

1. The upper part of the fill revealed anirregular surface about 5 cm thick. Thissurface was distinguished by the very slightcompacting of the material.

2. Fill covering the summit of the moundincluded fine loose sediment and somecobbles. The entire layer represents moderndisturbance.

3. This was a layer of material produced bymodern disturbance along the slope of themound, including small lumps of clayey earthin fine sediment along with some cobbles.

4. An intrusion located in the center part ofthe cleared area proved to have been cut byheavy machinery. However, the fill belowthe floor surface, composed of river cobbles,was exposed in this area. A segment of treetrunk was recovered from a posthole withinthe intrusion. A fragment of this post wasdated to 1480 Cal B.C. (ISGS-5284).

5. A small concentration of reddened sediment(20 3 25 cm in area) was visible in thecenter of the cleared area.

6. A level area proved to be formed by heavymachinery and was neither a use surface,a floor, nor a terrace.

7. A second level area also proved to beformed by heavy machinery.

8. In the lowest area cleared in Sector C,removal of sediment revealed two layers ofclay. The lower edge of the cleared area wasthe result of earthmoving machinery andnot a use surface, floor, or terrace.

Analysis of this heavily disturbed profile identi-fied two phases, one of which postdates occupa-tion of Sector C at Punta y Suela.

Phase 1a included loose fill, small lumps ofsediment and river cobbles that have fallen oreroded along the slope of the bulldozer cut.Contemporary trash in this area suggests thatthis is relatively recent erosion of the slopefollowing earthmoving and placement of powerpoles on the mound. Phase 1b representsa slightly earlier deposit that was still disturbedby construction and dates to the recent past. Thisdeposit includes fine textured wind-depositedsediment and river cobbles on the upper surfaceof the profile.

Phase 2 includes the undisturbed remains ofthe mound. No individual walls or floors couldbe distinguished while clearing the profile. Whileit would take further investigation to determinethe details of mound construction, at leasta portion of the Sector C mound is stillundisturbed.

FIG. 15. Punta y Suela, Sector C, Profile 1.


Conclusions—The Sector C mound at Punta ySuela may be a solid construction of rivercobbles and clay. If shicra was used in con-struction of the mound, it lies more deeplyburied or is present in better-preserved parts ofthe mound. As a result, only a sample of thepost in the upper surface of the mound wasdated. The date obtained, 1480 Cal B.C. (ISGS-5284), falls long after the Late Archaic period ofoccupation at Punta y Suela, and, in addition,the post itself could be older than context inwhich it was used, a result of the excellentconditions for preservation in the Norte Chico.The two levels of clay identified in the clearedarea of Sector C suggest that there were multipleconstruction episodes. However, no shicra bagswere found in this area, and the mix of cobblesand layers of clay could be part of the un-derlying construction. Only further excavationof the undisturbed portion of the Sector Cmound will clarify our understanding of moundconstruction.

SECTOR D: TEST UNITS 1, 2, AND 3—Sector Dincludes an extensive area that appears to havebeen habitation. A large quantity of angular rockis visible on the surface along with remains ofmollusks, dark sediment, and stone flakes,though no walls or other architectural remainsare visible. This sector extends to the base of thehills and covers an area greater than 3003

500 m. An exposed profile includes a layer ofdense mollusk shells 60 cm thick.

In Sector D, three test units were excavated inareas that had been looted or where agriculturalactivities exposed trash deposits below thesurface. These 1 3 2-m test units were excavatedto sterile material. Carbon samples were re-covered from plant fibers in each unit. Theseunits were designated Test Unit 1 (DN1), TestUnit 2 (DN2), and Test Unit 3 (DS1).

SECTOR D: TEST UNIT 1 (DN1)—This 1 3 2-mtest unit was located in the northern part ofSector D at Punta y Suela and was excavated inarbitrary levels 7–20 cm thick (Fig. 16). The

FIG. 16. Punta y Suela, Sector D, Test Unit 1 (DN1).


surface shell and stone suggested this might be anarea of Late Archaic occupation.

Level 1 included gray sandy soil with gravelinclusions and more cultural material than wasvisible on the surface, including nondiagnosticceramics, lithics, and plant remains in addition toshell fragments.

Level 2 continued with remains suggestingtrash deposition similar to Level 1.

Level 3 included loose, gravelly soil thatcontained some shell fragments and ceramicfragments. No charcoal or plant material waspresent in this level.

Level 4 was a layer of caliche, soil cementedtogether by salt. The top of this level was muchharder than the previous levels, and breakingthrough the caliche layer and removing itestablished the base of this level. The materialat the base of this level was darker and appearedto be a trash deposit.

Level 5 consisted of fine textured soil witha large quantity of angular rock. Culturalmaterial in Level 5 included shell, charcoal, andsome fish bones, and a few ceramic fragmentsappear to be midden fill. Level 5 is at the samelevel as Level 4 but is a separately excavatedarea.

Level 6 was composed of fine soil mixed withsmall angular rock. The matrix in this level wasloose and uncompacted. Midden contents con-tinued to be recovered in this level, includingplant fiber, charcoal, and a small quantity ofmaize. One sample of annual plant remainsrecovered from this level was dated 560 CalB.C. (Beta 177317) (Fig. 16, TA), while a sampleof charcoal fragments from the same level datedto 640 Cal B.C. (GX 30120) (Fig. 16, TB).

Level 7 continued the deposit of trash, in-cluding shell, charcoal, lithics, some ceramicfragments, and plant remains, including a smallamount of maize.

Level 8 was also composed of loose fine soilmixed with angular rock and shell fragments.The quantity of midden decreased markedly,including less charcoal and plant material. Oneside of the unit was rocky, while the other wasprimarily fine soil.

Level 9 was the final level initiated in this unit.Cultural material recovered in this level includedlithics, shell, plant material, and one potsherd.

Level 10 includes only one sample collected forradiocarbon dating from the upper surface ofthis level in the northeast corner of the unit andconsisted of a twig that was radiocarbon dated to

9170 Cal B.C. (ISGS 5393). The remainder of thelevel was gravelly soil without any culturalmaterial. This level was only excavated in thenorthwest end of the unit. Excavation wasterminated when it no cultural remains wererecovered.

Three depositional units were distinguished inTest Unit 1.

Phase 1 was a level of surface material thatincluded gray sand and gravel deposited by windand water. There was also a scatter of artifactsincluded in this level.

Phase 2 included rocky fill with mollusks,botanical remains, lithics, charcoal, bone, andceramics. Within Layer 2 were a small lens ofbotanical material and a separate pocket ofsimilar material. These appeared to be depositedalong with the other materials in the level ratherthan having been specific features.

Phase 3 included fine gray sandy sedimentwith no cultural material.

Conclusions—Test Unit 1 (DN-1) revealed thepresence of a single level of deposition thatappears to be domestic trash. No features wereidentified in this unit, supporting its being a mid-den. The three radiocarbon dates from this unitbracket the occupation of this area. Two samplesdate to the first millennium B.C. One sample ofplant fiber yielded a date of 560 Cal B.C. (Beta177317), while a charcoal sample from the samelevel yielded a date of 640 Cal B.C. (GX-30120).These may date the deposit, and they appear tocome from undisturbed contexts, though the soilwas not highly compacted. Nearby are theremains of looted burials, which seem to beintrusive into the rocky fill of Sector D. Thus, theradiocarbon dates appear to pre-date the burials.However, there could easily be burials froma broad span of dates in this area, although nonehave been excavated. The dates do not indicatethere was use of this area during the Late Archaicperiod, and ceramics were present in everyexcavated level. The early sample from the baseof the unit pre-dates the Late Archaic period andappears to pre-date any occupation of the sitesince it was a twig not associated with culturalmaterial. The sample from Layer 3 may date anearlier land surface that was covered up by theaction of wind and water.

SECTOR D: TEST UNIT 2 (DN2)—This 1 3 2-munit was located in the north part of Sector D atPunta y Suela. The unit was placed adjacent toa looted burial. The profile of the looted areasuggested there might be a layer of trash that


included botanical material below the surface(Fig. 17). This test unit was excavated in naturallayers.

Layer A: Sandy silt with angular rock rangingfrom 0.5 to 7 cm. Shell fragments, two ceramicfragments, a textile fragment, and botanicalremains were present, including a cluster of plantremains in the southeast corner.

Layer B: Fine sandy sediment with smallangular rock approximately 5 cm in size. Theentire layer was cemented together by caliche. Acouple of small textile fragments were recoveredfrom this level.

Layer C: Gravelly sediment including fine rockfragments less than 5 cm in size. The material isa brownish gray color with a very small numberof shells and ceramic fragments. A river cobblewith a ground area on one side was recoveredfrom this layer. A lens of sand and gravellymaterial was identified at the top of Layer C.Another area of sandy and gravelly materialextended into Layer C but did not extend acrossthe entire test unit.

Layer D: This proved to be entirely withoutartifacts. The layer of silty sand was the lastexcavated in this unit.

Three phases of deposition can be identified inthis excavation.

Phase 1 postdates occupation of the site andincludes fine wind-deposited sediment in LayerA. The small quantity of cultural material foundin the surface level, one shell fragment and twoceramic fragments, probably does not representprimary deposition.

Phase 2 includes Layer B, including a gravellylens and a pocket of botanical remains. This isalso related to occupation of the site based on itscultural contents similar to the level below.

Phase 3 includes Layer C and the gravellylens that extends into Layer C. This representsdeposition of material during a period of site use,as indicated by the presence of cultural materialin the level, including shell, sherds, lithics, anda bone fragment.

Conclusions—Two radiocarbon dates wereobtained from samples in Test Unit 2 (DN2).One of these samples consisted of a fragment ofwoven mat from a burial that intruded into thesoutheast corner of the unit. The burial waspresent in the unexcavated area and was notremoved, though a fragment of the mat thatwrapped the burial was collected. The resulting

FIG. 17. Punta y Suela, Sector D, Test Unit 2 (DN2).


date of 750 Cal B.C. (ISGS-5292) indicates thatthe burial postdates the Late Archaic period. Thepresence of ceramics in this excavated unit inLayers B and C supports this view. Further, thisdate may indicate use of this area comparable tothat in Test Unit 3 (DS-1), where dates are notdissimilar. Although from a deeper location, thisburial was intrusive. Ceramics were found in alllevels of this unit.

A second radiocarbon date from this unit wasfrom a sample of charcoal collected from Layer Cand associated with peanut shell (Fig. 17, TE).This sample yielded a date of 1840 Cal B.C.(ISGS-5272). This date may indicate there wasa Late Archaic occupation present in this part ofSector D, below the ceramic layers. However, thiswould include only the lowest portion of Layer C.

SECTOR D: TEST UNIt 3 (DS1)—This pit islocated in the southwest part of the site in SectorD, a possible residential area. This test unit wasexcavated in 10-cm levels.

1. (Surface): Loose, silty soil mixed with gray,medium-sized gravel. Cultural material in-cluded seeds, mollusks, charcoal, bone,ceramics, and botanical remains. A numberof heat-fractured river cobbles were presentas well.

2. Loose sediment with medium-sized gravel,along with numerous river cobbles. Someriver cobbles were heat fractured. Charcoal,fish bone, otoliths, and ceramics (44 frag-ments) were present.

3. Loose sediment with medium-sized graveland river cobbles. A large quantity ofcharcoal, very few ceramics, remains ofburned bone, and a large number of heat-fractured rocks were present.

4. Fine sediment combined with gravel. Halfthe unit at this level still contained culturalmaterial, including burned rock, shell,charcoal, bone, lithics, quartz, and ceram-ics. One sample of charcoal collected fromthis level was radiocarbon dated to 5410 CalB.C. (Beta 177318).

5. This level was similar to Level 4, includingthe presence of burned river cobbles, shell,bone, charcoal, quartz, lithics, and oneceramic fragment. The south half of theunit was sterile gravelly fill. A charcoalsample from this level was dated to 5410 CalB.C. (GX 30119).

6. The final level in this unit included very finegray-yellow sediment with a small quantity

of bone, lithics shell, and a fragment ofhuman cranium. The lower half of the levelwas gravelly and without any culturalmaterial. A radiocarbon sample of charcoalfrom the upper portion of this level wasdated to 6280 Cal B.C. (ISGS-5271).

There was a distinctively large quantity ofheat-fractured rock and abundant organic mate-rial in Test Unit 3, which supports the area beinga residential zone, though the presence ofceramics in all excavated levels indicates thatthis occupation postdated the Late Archaicperiod. Analysis of the profile identified fivephases of occupation (Fig. 18).

Phase 1 included grayish clayey sediment withgravel and cultural material including ceramics.This phase appears to postdate occupation of thesite. For example, a patch of charcoal visible onthe surface did not continue into subsequentlayers.

Phase 2 was the final period of trashdeposition in this area. Patches of sedimentwith cobbles, shell, and fire-cracked rockalternate with patches of charcoal, suggestingcontinued deposition of materials, includinghearth contents (charcoal), botanical remains,lithics, and bone. A narrow layer, 1.5–4 cmthick, of fine, soft, yellow-gold colored soil wasdeposited at the end of this phase over the trashdeposit, though it does not cover the areacompletely.

Phase 3 included a period of time during whichthe area was used for deposition of burnedcobbles and trash. Yellowish-gray fill in thislayer included fragments of fire-reddened rivercobbles, other lithics, and shell along the slopingsurface of the cobbles below.

Phase 4 included a period during whicha mound of cobbles, both unaltered and firereddened, were deposited in the north end of theunit in a matrix of reddish-brown soil. A sampleof charcoal from Level 4 dates this Phase to 5410Cal B.C. (Beta 177318).

Phase 5 marked the initial use of this portionof Sector D. Material remains were sparse,though the variety suggests the start of a middenarea. The yellowish-gray sediment is composedof large-grained sand from the southwest cornerto the center of the unit. The two radiocarbondates were obtained from the deposits of thisphase—6280 Cal B.C. (ISGS-5271) and 5410 CalB.C. (GX 30119)—suggesting a Middle Archaicdate for this area, even when the fact that these


samples were charcoal is taken into consider-ation.

Conclusions—The radiocarbon dates fromTest Unit 3 (DS-1) suggest that there may bean earlier occupation at Punta y Suela than hasbeen identified at the other sites tested in thePativilca Valley. Test units of 1 3 2 m are sosmall compared to the overall size of the sitestested that there could easily be other MiddleArchaic occupations associated with Preceramicsites that were not identified during 2002. Sincethe early dates in Test Unit 3 come from charcoalsamples, it would be necessary to expand theexcavation in this area to see whether specificfeatures like the mound of cobbles could befeatures associated with a Middle Archaicoccupation. The preceramic deposits in TestUnit 3 appear to be associated with Phase 2,and, again, these would need to be confirmed byadditional excavation.

DISCUSSION: PUNTA Y SUELA—The three testunits, each 1 3 2 m in size, served to identify thisportion of Sector D as a zone of residential anddomestic uses, based on the large quantities ofmollusk fragments, heat-fractured rock, andhousehold trash. Radiocarbon dates suggest thatthis area was occupied over a very lengthyperiod, perhaps as early as the Middle Archaic:

N Four samples from Sector D, including TestUnit 1, Level 10, and Test Unit 3, Levels 4–6,date from 9170 to 5410 Cal B.C.

N Three samples date to the Late Archaicperiod, including two from Sector B and onefrom Sector D, Test Unit 2, Capa C.Together, these span the period 2420–1840Cal B.C.

N A single sample from the upper surface of theSector C mound dates to 1480 Cal B.C., theInitial period.

N Three samples, one from outside Sector D,Test Unit 2, associated with a burial and twofrom Sector D, Test Unit 1, Level 6, date to560–750 A.D., the Early Intermediate period.

The radiocarbon dates from Punta y Suelasuggest there may have been a Middle Archaicoccupation in the area of Test Unit 3 (DS-1). Theearly dates come from the lower levels (4–6) ofa test unit where ceramics were also recovered inLevels 4 and 5. In this part of Sector D, there isno evidence of Late Archaic occupation despitethe relative proximity of the mounds in Sectors Cand A.

Late Archaic dates were obtained frommounds in Sector B from the lowest layer ofSector D, Test Unit 2. However, the now-destroyed circular plaza and the mounds inSector A suggest that there was a substantialLate Archaic occupation at Punta y Suela. Themound in Sector C also appears likely to be LateArchaic in date since an Initial period date wasobtained from the uppermost level of the moundidentified in clearing a damaged area.

FIG. 18. Punta y Suela, Sector D, Test Unit 3 (DS1).


The date of 1840 Cal B.C. from Sector D, TestUnit 2, suggests that there may be Late Archaicresidential occupation in Sector D. Althoughresidential structures were not identified in the 13 2-m test units, trash associated with residentialactivities was recovered in these units, suggestingthat it may be possible to locate residentialstructures in Sector D with additional testing.

The Initial period date of 1480 Cal B.C. fromSector C is not accompanied by diagnosticceramics or other data that would allow us todiscuss possible Initial period use of the site.Later occupation, as indicated by Early In-termediate period dates from the test units inSector D, appear to be associated with intrusiveburials but possibly also with trash deposition inthis area.

It is more likely that Punta y Suela was used asa burial site than as a habitation site during theEarly Intermediate period. In 2002, no structureswith evidence of occupation after the LateArchaic were identified. In the Norte Chico,there is a precedent for Late Archaic sites to beused as burial sites in subsequent times, and theuse of Punta y Suela as a burial site seems themost likely possibility. Despite being heavilydamaged in some areas by a jeep trail and thedevelopment of agricultural fields on the site,there are still substantial architectural and mid-den remains at Punta y Suela that can beprofitably studied with further excavations.

Pampa San Jose (O2PVGS-3)

Pampa San Jose is located on the right bank ofthe Pativilca River, 500 m southwest of thehamlet of the same name, at the base of thewestern slope of Cerro Mirador. This large andimportant site has suffered a great deal ofdamage. The mounds at Pampa San Jose arecompletely surrounded by agricultural fields. It isunlikely that remains of residential structuresand their associated trash have been preserved.All that remains in this locality are a series ofmounds and a sunken circular plaza. Thecircular plaza itself is being used as a reservoirand is filled with water most of the time.

Although the circular plaza at Pampa San Josehas been heavily damaged by the expansion ofagricultural fields, the mounds are in a goodstate of preservation. The mounds are part of anarchitectural complex oriented toward the south-east (Fig. 19). In addition to the large platformmound with an associated sunken circular plaza,

two smaller platform mounds are located north-east of the main mound, while five subsidiarystructures are situated southwest of the mainmound. Directly across from the main mound isa low mound, now largely destroyed, that seemsto have been constructed in the center of the U-shaped site layout.

Sector A includes the large rectangular plat-form mound and associated sunken circularplaza. The main mound measures 73 3 101 mand is 20 m high. The circular plaza is 39.6 m indiameter. Construction material included rivercobbles but also some angular rock. On thesummit of the main platform mound, a large,central atrium feature can be observed thatincluded low walls on three sides of this openspace. Remnants of a broad staircase descendingto the foot of the mound on the open side ofthe atrium are visible.

Sector B at Pampa San Jose includes a rectan-gular platform mound located to the east of themain mound. Construction employed river cob-bles. Lithic and shell materials can be seen allover the area of the mound, though no ceramicswere visible. This mound is 38 3 42 m and 7 mhigh, and agricultural fields surround the entirestructure.

Like the other sectors, Sector C at Pampa SanJose includes a rectangular platform mound, inthis case located to the east of Sector B and alsooriented to the southeast. The mound is 35 3

40 m, constructed of river cobbles, with noceramics visible on the surface.

Although this site was mentioned by Williams(1985) and suggested to date to the Initial period,the architecture and the surface material atPampa San Jose suggest occupation during theLate Archaic period.

Work by PANC during 2002 at Pampa SanJose was limited to clearing the exposed walls ofa large hole that was excavated into the top ofthe principal mound in Sector A by looters.Numerous structures and floors are visible inthe walls of this hole and provided valuableinformation about the construction techniquesand the occupation sequence of the mound.Abundant plant material exposed in the profilespermitted us to obtain radiocarbon dates fordifferent construction phases.

SECTOR A: PROFILE 1—A large looter’s pit wasexcavated in the summit of the mound. Itmeasures 5 3 5 m and is more than 4.3 m deep(Fig. 20). The hole is on the southeast side of theatrium. In the sides of this pit, construction


employing a variety of different materials isvisible, including different types of shicra bagsand large river cobbles. On the site surface,remains of mollusks, lithic materials, and organicremains, including fragments of shicra, can beseen. The hole has a nearly rectangular form thatprovided four profiles. These were recorded andcarbon samples collected.

For the purposes of description, the walls ofthis unit were recorded separately (south, north,west, east). A letter indicating the face that wasbeing recorded makes it possible to identifysimilarities and differences among the profiles(Fig. 21A).

South Profile—SA: Disturbed area, no stratig-raphy visible.

SB: Fill of clay and stone.

SC: Multiple thin layers of clay, apparentlyrepresenting numerous replasterings of a floor.

SD: Fill of loose clay, gravel, and sandbetween floors.

SE: Clay floor 3 cm thick.

SF: Fill of rock and sand under the floor.

SG: Area of sand, gravel fill, and small rivercobbles.

SH: Wall of clay and river cobbles associatedwith Floor SL.

SI: Fill of large river cobbles and angularsandstone and quartzite chunks.

SJ: Clay floor cut by the excavation associatedwith SI.

SK: Fill of clay mortar and river cobblesbetween Floors SJ and SL.

SL: Clay floor associated with Wall SH.

SM: Trash fill between Layers SL and SN.Four carbon 14 samples were taken from thislevel. One of these, Sample 4, was processed,yielding a date of 2100 Cal B.C. (Table 6).

FIG. 19. Map of Pampa San Jose showing mounds and other features, sectors, canals, roads, andencroaching structures.


SN: Floor at the base of the layer of trash fillSM.

SO: Sandy fill with shicra bags, from whicha carbon 14 sample was obtained.

North Profile—NA: Wall 70 cm thick com-posed of river cobbles set in clay mortar.

NB: Fill of sand and gravel beside Wall NC.

NC: Wall of stone and mortar.

ND: Deposit of rocks of various sizes andsome shicra.

NE: Small deposit of sand, gravel, and fiberlocated on top of a collapsed roof and fallenmaterial NF.

NF: Collapsed roof material of interlacedcanes. Radiocarbon samples 5 and 6 come fromthis roof material. Sample 5 yielded a date of1950 Cal B.C.

NG: Fill from above Floor NG associated withwall ED. This fill consists of a mix of shicra bagsfilled with sand and gravel and others filled withrock. Carbon sample 8 comes from this level.

NH: A section of clay-plastered surface thatmay be a floor between Layers NG and NI.

NI: A layer of fill composed of carbon sample9 comes from this level, dated to 2100 Cal B.C.

NJ: Remnants wall fall from higher walls.

NK: Lenses of sand and gravel.NL: Floorassociated with Wall NC.

NM: Fill of loose river cobbles and clay, whichmay be associated with ND.

West Profile—WA: Fill river cobbles in claymortar, possibly a collapsed wall. No structures,floor, or features.

WB: Laminated layers of clay adhering to theplaster on Wall NA. These probably representnumerous replasterings. The top of this layer wasa floor, though it is impossible to determine howmany remodelings of this floor there may beincluded in this layer.

WC: Layer of fill with river cobbles, sediment,and some shicra bags. Carbon sample 1 wasobtained from this provenience.

WD: Loose sand and gravelly fill in the lowerpart of the north side are remains of shicra bagsused to contain and to carry the sand. Carbonsample 2 comes from this layer.

WE: A layer of loose river cobbles and grayclay extending beneath the base of wall NA.

WF: This is a layer of shicra bags filled withsand, delineated by bags visible in the profile.The base of this layer is not fully exposed in thelooter’s pit.

WG: Wall of river cobbles in clay mortar. Noplaster finish is visible on this side. The wallmay be a retention wall, part of moundconstruction.

WH: Fill of loose sediment and river cobbles,which may be disturbed.

WI: Dense deposit of clay layers, without anyvisible floors.

WI2: Narrow layer of loose river cobbles.

WJ: Second retention wall built of rivercobbles and clay mortar, without plaster-finishedsides.

East Profile—EA: Fill of loose river cobbles,sediment and gravel.

EB: Loose fill, including some shicra bagsplaced behind the wall (to the east of the wall).The fill is a light orange color, and the base ofthis layer lies on SJ. Carbon sample 10 wascollected from the northern edge of this layer.

EC: Compacted river cobbles and clay mortar.

ED: Small deposit of loose gravel and smallriver cobbles.

EE: Floor of river cobbles and clay connectingwith SL.

FIG. 20. Clearing the profile of the deep looter’shole in Pampa San Jose, 2002 (J. Haas).


EF: Small deposit of clay and sand withoutany visible floor.

EG: Wall of river cobbles in clay mortar thatruns north to south crossing the entire looter’shole, though heavily damaged. Carbon sample 7was collected from this layer.

The base of the profile at Pampa San Jose didnot extend to sterile subsoil. A number ofdifferent occupation phases can be detected,though these do not extend to the base of thestructure.

Phase I includes postoccupational depositionafter use of the structures visible in this unitceased, including NB, ND, SB, EA, ED, and EF.NB, for example, includes fill of sand and gravelon both sides of Wall NC, and no further use ofthe room is visible.

On the surface, not visible in profile, wasa remnant of wall 70 cm wide. Only the broadbase of the wall was exposed on the surface ofthe mound. The wall was built of cobbles in claymortar and may be ancient or modern.

Phase II includes the period of use of Structure3, a room formed of an older wall, NA, with newWalls NC and NJ and a floor (NL) inside thestructure. The deposition of NK to level FloorNL is included in this phase.

Phase IIIa includes leveling that took placebefore construction of Structure 3. This appearsto be a single episode of deposition, ND, in-cluding shicra bags in a layer of fill that leveledthis area.

Phase IIIb material deposited after abandon-ment of Structure 2. This layer of windblownmaterial, cobbles, and sediment is probably theresult of the collapse of nearby walls and thespreading effects of the wind. Phase IIIb includesdeposits NE, EA, SB and WH.

EF and ED are included in Phase IIIb sincethey are surrounded by EA and because thePhase II deposit ND appears to cut into thesedeposits. WA is also included here, though it isa layer of disturbed material that pre-dates PhaseIIIb but postdates Phase IV.

FIG. 21. (A) Pampa San Jose profile. (B) Pampa San Jose profile schematic of construction and dates.


TABLE6.

Pa

mp

aS

an

Jose

rad

ioca

rbo

nd

ate

sfr

om

sam

ple

so

bta

ined

in2

00

2.

Pro

ven

ien

ce

Tex

ta

nd

illu

stra

tio

nlo

cati

on

Ma

teri

al

La

bn

o.

Ca

l.B

.C.

RC

YB

P12C

/13C

Wei

gh

to

fsa

m-

ple

(g)

Ca

lib

rate

da

ge

ran

ge

1si

gm

a(6

8.3

%)

B.C

.C

ali

bra

ted

ag

era

ng

e2

sig

ma

(95

.4%

)B

.C.

Mo

un

dA

loo

ter’

sp

it2

,W

DM

ixed

pla

nt

fib

ers

Bet

a-1

77

31

41

87

03

54

06

70

–1

4.9

15

19

49

–1

76

7(9

6.7

),1

75

9–

17

52

(3.3

)2

11

4–

20

99

(1.1

),2

03

4–

16

88

(98

.8)

Mo

un

dA

loo

ter’

sp

it1

,W

CM

ixed

pla

nt

fib

ers

GX

-30

12

51

95

03

60

06

40

–2

2.7

15

20

14

–1

99

8(1

5.4

),1

97

9–

18

90

(84

.6)

21

25

–2

09

6(3

.3),

20

90

–2

08

4(.

4),

20

40

–1

87

7(9

1.5

),1

84

1–

18

25

(2.9

),1

81

9–

18

15

(.3

),1

79

6–

17

80

(1.6

)M

ou

nd

Alo

ote

r’s

pit

5,

NF

Mix

edp

lan

tfi

ber

s

Bet

a-1

77

31

51

95

03

60

06

60

–2

5.4

15

20

33

–1

88

0(9

7),

18

38

–1

83

1(3

)2

13

7–

20

76

(10

.5),

20

71

–1

85

8(7

6.9

),1

84

5–

17

70

(12

.6)

Mo

un

dA

loo

ter’

sp

it4

,S

MM

ixed

pla

nt

fib

ers

ISG

S-5

27

62

10

03

71

06

70

–1

4.4

10

22

00

–2

01

8(9

2.9

),1

99

7–

19

80

(7.1

)2

29

7–

18

91

(10

0.0

)

Mo

un

dA

loo

ter’

sp

it9

,N

IM

ixed

pla

nt

fib

ers

ISG

S-5

28

22

10

03

71

06

70

–1

1.6

72

20

0-2

01

8(9

2.9

),1

99

7-1

98

0(7

.1)

22

97

–1

89

1(1

00

.0)

Mo

un

dA

loo

ter’

sp

it7

,E

GM

ixed

pla

nt

fib

ers

GX

-30

12

22

23

03

79

06

60

–1

3.3

15

23

30

–2

32

2(2

.5),

23

09

–2

13

6(9

4.2

),2

07

8–

20

67

(3.3

)2

45

6–

24

37

(1.7

),2

42

4–

24

21

(1.3

),2

40

4–

23

58

(6.2

),2

35

4–

21

10

(79

.5),

21

02

–2

03

5(1

1.4

)


Phase IV includes the construction, use andremodeling of Structure 2, a complex includingWalls NA, WG, and WJ and Floor SE with itsassociated leveling and remodeling layers SC,SD, and SF. Floor SF appears to have beenconstructed first, possibly a patio area, while thewalls were placed next, and apparently theinterior surface was leveled with WD. WD wasa layer of sandy fill with shicra bags toward thenorthern end at the base of the unit, filled withsand, not a common construction technique.Part of a shicra bag in WD (#2), the surface onwhich wall WG was built, was dated to 1870 CalB.C. (Beta-177314).

Remodeling of Structure 2 includes depositsWD, WC, WB, WI, and WI2. WC was a layer offill or remodeling that included a sample of plantfibers dated to 1950 Cal B.C. (GX-30125).

Phase V includes the deposits that filled inStructure 1 and leveled the area for Structure 2,EB, SG, SI, WE, WF, NE, NG, NF, and NM.WF included an unusual shicra variant, basketrybags filled with dirt (Fig. 22), rather than themore frequently used shicra bags filled with rock.Adjacent to this was NF, a collapsed roof of

interlaced canes. According to local informants,the discovery of this roof material by the lootersmotivated them to continue digging, even thoughthe unit had reached more than 2 m in depth,as they believed it was the top of a tomb. Theydid not find any corresponding tomb. Twosamples of roof material were collected forradiocarbon dating (#5 and #6). Sample #6was processed, yielding a date of 1950 Cal B.C.(ISGS-5276).

Phase VI includes the construction, use, andremodeling of Structure 1, including Floors EE,SL, SJ and NH. Walls constructed at the start ofPhase VI include SH and EG. Sample #7 wasplant fiber from the mortar over Wall EG. Ityielded a date of 2230 Cal B.C. (GX-30122).

Phase VII includes all deposits below the floorof Structure 1 formed by Walls EG and SH withFloor EE. This includes a layer of trash andconstruction fill, apparently intended to level thesurface over Floor SN. This includes fill identi-fied as NI and SM. Two samples from Phase VIIwere dated. Both sample NI (#9) (ISGS-5282)and sample SM (#4) (ISGS-5276) were dated to2100 Cal B.C.

FIG. 22. Pampa San Jose photo of dirt-filled woven bags, shicra (J. Haas).


Phase VIII, the earliest present in this unit,includes only a small fragment of floor (SN) andthe fill below it (SO). This deposit is located atthe base of the south profile and is a tinyremnant of the earliest construction featurevisible in this unit. One radiocarbon samplewas obtained from this Phase (SO #11), thoughit has not been processed.

DISCUSSION: PAMPA SAN JOSE—Williams (1985)suggested that Pampa San Jose dates to theInitial period. However, the architecture and thesurface material recorded by PANC at PampaSan Jose in 2002 suggest occupation during theLate Archaic period. All six dates, ranging from1870 to 2230 Cal B.C., each from separateproveniences in the profile, date to the LateArchaic. As specific dates, the samples fromPampa San Jose (Table 6) do not fall in precisetemporal order from the lowest to the highestlevels. Among the earliest dates that obtainedfrom fiber on Wall EG, 2230 Cal B.C. is earlierthan the two dates from the fill below Wall EG,both 2100 Cal B.C. From this we can concludethat Wall EG was constructed after 2100 CalB.C. However, radiocarbon dates are bestexpressed as ranges of probability, which in thiscase eliminates the overlap. Although the resultsappear less precise, they are more accurate thanthe individual dates. In comparison with othersites, Pampa San Jose has a relatively tight daterange, indicating primary occupation toward theend of the Late Archaic period. Radiocarbondates show the site was under construction frombefore 2230 Cal B.C. until after 1870 Cal B.C.Since the looters did not reach the base of themound, the profile from which samples wereobtained at Pampa San Jose is not complete.Construction probably began before 2230 Cal

B.C., and significantly earlier dates might beobtained by sampling lower levels of the mainmound or in other mounds.

Three principal periods of use are visible(Fig. 21B, Table 7). The first is represented bythe structure formed by Walls EG and SH withFloors EE, ED, EC, SL, and SJ. Next, thecomplex of floors and structures created byWalls NA, WJ, and WG display a period of useand remodeling, while the room defined by WallsNC and NJ demonstrates a third locality on themound where there was construction and use ofa room. These three different structures or areasof construction were identified in different partsof the looter’s pit profile. When we consider thatthe looter’s hole takes up only a fraction of thetop of the mound, 5% or less, there seems to havebeen a wide variety of structures on the surfaceof the mound during each phase.

The numerous construction and occupationlayers visible in this profile suggest active use andreuse of the mound at Pampa San Jose, whichrose ever higher with each remodeling. Themound was not a structure that was continuallyrebuilt to a single pattern through repeatedreplastering of a structure or group of structures.It was a place where different structures indifferent positions were built, used, covered upor leveled, and eventually replaced by otherstructures.

Carreterıa (02PVGS-4)

Carreterıa is located on the right bank of thePativilca River about 12.5 km from the Pacificcoast, north of the former hacienda Carreterıa. Acanal passes along the northern edge of the

TABLE 7. Principal phases of occupation and associated radiocarbon dates at Pampa San Jose.

Elements Dates Date range (1 d)* Strata

14 postdates 1870 cal B.C. Structures defined by Walls NA, NC, and NJ9–12 Structures created by Walls NA, WJ, and WG

1950 cal B.C. (1GX-30125) 2020–1880 WC: a layer of fill or remodeling of structure WG-NA1870 cal B.C. (2Beta-177314) 1950–1740 WD: fill or leveling of structure WJ-NA; the surface

on which Wall WG was built1950 cal B.C. (5ISGS-5276) 2040–1820 NF: fiber from collapsed roof at side of structure EG-

SH4–8 Structure formed by Walls EG and SH

2230 cal B.C. (7GX-30122) 2340–2060 EG: fiber from plaster on Wall EG2100 cal B.C. (9ISGS-5282) 2200–1880 NI: plant fibers from fill below Wall EG2100 cal B.C. (4ISGS-5276) 2200–1880 SM: plant fibers from fill below Wall SH

* Detailed probability ranges can be found in Table 6.


mound (Fig. 23). Like Pampa San Jose, littleremains at Carreterıa apart from the mainmound and a small portion of the sunkencircular plaza.

SECTOR A: PROFILE 1—Carreterıa’s centralrectangular platform mound measures 32 3

60 m and is 6–7 m high. Cultural material visibleon the surface includes mollusk valves and lithicfragments. Cement footings remain from modernstructures that were atop the mound but havebeen demolished. From the top of the mound

facing northeast, it is possible to observe theremains of a sunken circular plaza approximately27 m in diameter that was enclosed by a lowrectangular platform and wall. This area mea-sured 51 3 56 m and was destroyed in large partduring February 2002. Local cultivators havedestroyed or covered over the residential archi-tecture to create fields; every possible area ofresidential construction associated with themound has been heavily damaged or completelydestroyed.

FIG. 23. Map of Carreterıa showing mounds and other features, sectors, canals, roads, and encroachingstructures.


Profile 1 in Sector A was the only area clearedat Carreterıa during the PANC investigation.The profile was located on the central part of themound on its north side, looking out over thesunken circular plaza. The irrigation canal thatcuts across the base of the main mound along itsnorth side left an exposed profile, removinga significant deposit from the base of the mound.The Profile 1 trench was cleared from the base todisplay levels of mound construction and toextract radiocarbon samples.

The mound was constructed of angular rock,primarily irregular chunks of speckled granite,with a light scatter of burned river cobbles,obscuring poorly preserved walls and floors. Anextensive series of stratigraphic units were identi-fied during the clearing of Profile 1 at Carreterıa.The cleared area proved to include the edge ofa stairway that appears to have run from the topof the mound to its base on the side facing thecircular plaza. The northwest side of the profilerevealed steplike tiers of construction, consistingof masonry retaining walls filled with rock andmud. The southeast side of the profile includedportions of at least two layers of stairs thatclimbed the structure. The uppermost set of stepswas just below the loose rock and windblown soilcovering the mound. A second set of stairs wasidentified in the upper third of the area cleared,though these seem to have been an earlier set ofstairs that ran under the upper set.

A. a–f. Six shallow steps built of angular rockand clay mortar. These were partially erodedand irregular.

B. Retaining wall of angular rock in claymortar extending step f. Additional stepsin the sequence that began with a–f did notcontinue, possibly being eroded away.

C. Retaining wall of angular rock and cobblesin clay mortar. B and C are the upper andlower portions, respectively, of a retainingwall. However, a distinct layer of fill,possibly a former floor, is sandwichedbetween the two segments of wall. It appearsthat wall C may have been a step at anearlier stage and was later covered with clayas part of a remodeling. At a later time, thestaircase was reconstructed by raising theheight of each step.

D. Floor of packed and smoothed clay. Theradius fragment was found above thissurface. This surface was wider than thespace between steps higher on the structure.

It is possible that shallow terraces like floorD divided short flights of steps.

E. Retaining wall at leading edge of Floor D.Wall E did not extend across the areacleared. This may be the result of erosion.

F. This floor of packed and smoothed clayformed another large step or shallow terraceextending from the base of Wall E to the topof Wall G.

G. Retaining wall at leading edge of floor F.

H. A broken area of Floor D revealed angularrock and clay mortar construction below thefloor. A small portion of a buried step wasrevealed as well. The surface was carefullyplastered, but any other stairs were coveredby Floor D. Step H appears to be a portionof a staircase below the set of steps identifiedas A.

I. A layer of fill between Wall B and Wall C,a result of remodeling or rebuilding.

J. Fragment of clay floor at the base of WallG. The slope of the mound was moregradual in this part of the profile thanhigher on the mound. There was not a clearseries of constructed steps in this area butrather two sections of clay floor that slopeddifferently, suggesting that they were differ-ent constructions or the result of collapse ofthe rock and clay mortar below, resulting inan uneven surface.

K. This fragment of clay floor sloped ata sharper angle than floor fragment J, asmentioned previously. Neither fragment Jnor fragment K extended across the clearedarea, possibly a result of erosion or collapse.Carreteria has been mined for rock and thesurface could have been cleared in thisspace.

L. A low, raised ridge of adobe was built acrossthis part of the cleared area, forming a semi-circular enclosure. It appears to be a poste-rior construction. There was no evidence ofits age or function, though one of the humanbone fragments came from this area.

M. A low, raised ridge of adobe was built acrossthis part of the cleared area, forming a semi-circular enclosure. It appears to have beenbuilt after partial collapse of the structuresince it crosses part of a terrace and whatappears to be a collapsed portion of Wall O.There was no evidence of its age or function,though one of the clusters of vertebrae camefrom this area.


N. Possible wall at leading edge of Floors J andK. This possible wall was eroded andshallow.

O. Wall at leading edge of Terrace VI. Unlikethe other retaining walls, Wall O extendsacross most of the cleared area. This mayhave been the base of both the staircase andthe large steps adjacent to the staircase ora place where the staircase changed dimen-sions.

P. A broad terrace fragment extended from thebase of Wall O downslope. It appeared to bea terrace that had slumped because oferosion and canal construction below it onthe mound.

Q. A damaged portion of the terrace surfacewas cleared and revealed an earlier wallsegment buried by Terrace P. The construc-tion material was angular rock and claymortar with no evidence of shicra.

R. Step fragment, possibly a continuation ofstairs (a–g) or part of an earlier staircase,such as (h).

S. From the lowest identifiable margin ofTerrace P, clearing revealed a rubble slopeabove the canal. This area was steeplyinclined and was not further cleared.

This profile revealed different features in theeastern and western halves of the unit cleared(Fig. 24). On the southeastern side, the edge ofa stairway that extended from the summit to thebase of the mound was uncovered, along withevidence of remodeling. On the northwesternside of the unit, five large steps suggest themound was constructed of several superimposedterraces, as indicated by a series of large stoneblocks that formed the corners of retaining wallsI–VIII. Each terrace was constructed of angularrock and clay mortar. No evidence of shicra usedin construction was revealed during clearing.This may be a result of remodeling efforts or theuse of shicra only for extensive areas ofconstruction, like platform fill.

Unlike the other sites examined, fragments ofhuman bone were found dispersed across theentire unit at Carreterıa, though no articulatedelements were recovered. These included a singleskull fragment, a humerus fragment, and a radiusfragment along with a metacarpal and a carpalbone. Vertebrae were found in two places(Fig. 24).

Analysis of this profile identified fiveoccupation phases at Carreterıa, though it is

likely that additional layers would be revealedby further excavation, as the trench wasintended only to clear a section that had beendamaged, not to excavate the surface of themound that was preserved below the looserock and soil.

Phase 1: At the end of or following theoccupation of the mound at Carreterıa, a seriesof low, curved walls were constructed over thestaircase to create small enclosures.

Phase 2: The upper surface visible in Profile 1 atCarreterıa was the remodeling of a staircase (a–f).

Phase 3: Below the Layer 1 staircase (a–g) wasa previous stairway (h and possibly r).

Phase 4: Construction of the platform moundincluded a series of stepped terraces (I–VII).

Phase 5: An early mound surface or an early,broad staircase is represented by Wall Q, whichwas covered by construction of Terrace VII.

This does not represent a complete constructionsequence but summarizes the construction visiblein the area cleared. These construction layerswere revealed by erosion of the mound surface.A more complete understanding of the construc-tion sequence would require additional excava-tion.

DISCUSSION: CARRETERıA—Carreterıa is adja-cent to the Pativilca River floodplain, wherethere is an abundance of rounded river cobbles.Although cobbles were readily available, therewas deliberate selection of angular quarriedstone for the mound construction, and atCarreterıa, the angular rock is quite uniform,a distinctive pale granite with dark speckles.

By clearing Profile 1 in this area, it waspossible to make a detailed record of themodifications made in the access from thecircular plaza to the atrium of the principalmound. The mound in the area of Profile 1 waslargely built of rock set in clay mortar with novisible shicra. Clearing alone did not penetratethe structure enough to reveal shicra bags in theconstruction material of the main mound in thiscentral and heavily remodeled area. When noshicra was identified during the clearing ofProfile 1, the mound was examined closely foreroded places where construction might bevisible. Remains of shicra bags were found ina small profile exposed by the irrigation canal onthe southwest side of the mound. Apparently,shicra was used in initial mound construction butnot in later remodeling. On a flanking structure,shicra bags were visible in eroded areas above theirrigation canal. This area was apparently not


FIG. 24. Profile 1 at Carreterıa, including location of human remains.


subject to remodeling after its initial constructionsince the shicra was both in situ and exposed onthe side of the mound and not an isolatedoccurrence. The sample collected from thislocation was dated 2180 Cal B.C. (ISGS-5286)(Fig. 23, Table 8). This single date suggeststhat the Carreterıa mound was constructedduring the period 2290–2030 Cal B.C. and thatit was subsequently remodeled on multipleoccasions.

While a single date is not definitive, it doescorrelate with the absence of ceramics in all fillmaterial and the overall pattern of a largeplatform mound associated with a sunken circu-lar plaza, the Late Archaic pattern. It should alsobe noted that the one date from Carreterıa fallswell within the available suite of dates for PampaSan Jose. The main mound complexes for thesetwo sites are less than 2 km apart, and the twomain mounds with associated sunken circularplazas face one another (Fig. 25). While anypossible relationship between these two sitescannot be determined at present, it is likely thatthey were aware of each other and had some kindof connection, whether it was temporal, compet-itive, cooperative, or political.

Huayto (02PVGS-5)

Huayto is located on the right bank of thePativilca River on the summit of an elevatednatural feature immediately north of the villageof the same name. The site occupied the north/northwest portion of a large landform, a dissect-ed plateau or foothill spur extending out fromthe hills to the west (Fig. 26). The Huayto siteconsists of several mounds constructed aroundan open plaza (Fig. 27). Three mounds arepreserved and form Sectors A, B, and C. SectorD was a small mound located at the extremenorth end of the site. Sectors F, G, and I areother mounds that are partially or completelydestroyed. Sector H, southwest of Sector D, isnot a mound but an area with marine shell andstone fragments on the surface.

Unlike the other sites tested, there does notappear to have been looting at Huayto; thedestruction is the result of local housing con-struction. Air photos dating to 1969 show thetop of this landform only lightly occupied. Sincethat date, the expansion of construction hasaffected all but Sectors A, B, and C. Structureswere built around the ceremonial nucleus of thesite that destroyed or largely covered an area that

TABLE8.

Ca

rret

erıa

rad

ioca

rbo

nd

ate

sfr

om

sam

ple

so

bta

ined

in2

00

2.

Pro

ven

ien

ceM

ate

ria

lL

ab

no

.C

al.

B.C

.R

CY

BP

12C

/13C

Wei

gh

to

fsa

mp

le(g

)C

ali

bra

ted

ag

era

ng

e1

sig

ma

(68

.3%

)B

.C.

Ca

lib

rate

da

ge

ran

ge

2si

gm

a(9

5.4

%)

B.C

.

Ma

inm

ou

nd

fill

Fib

erb

ag

ISG

S-5

28

62

18

03

76

06

70

–1

2.2

15

22

87

–2

24

6(1

8.5

),2

23

9–

21

12

(57

.3),

20

97

–2

03

9(2

4.3

)2

45

6–

24

47

(.5

),2

43

2–

24

22

(.5

),2

40

4–

23

61

(3.7

),2

35

4–

20

07

(91

.6),

20

03

–1

97

5(3

),1

97

2–

19

61

(.7

)


presumably had been occupied by residentialarchitecture. Today five houses form the south-ern boundary of the ancient settlement. The localcommunity recently constructed a soccer field inwhat was probably the ceremonial center of thesite, completely destroying the Sector F moundand part of the others. From the architecture andconstruction materials, Huayto appeared to dateto the Late Archaic period, and surface ceramicsindicate that the site was later reoccupied.

One profile was cleared and recorded atHuayto, on the N/NE slope of Sector C. Atrench 3 3 7 m was cleared in an area thatproved to have been heavily disturbed. Theprofile exposed some original construction andprovided information about building techniquesand the possible sequence of occupation. Theorganic material included shicra and other plantfiber suitable for radiocarbon dating.

SECTOR A—Sector A includes a rectangularplatform mound 68 3 75 m at the base and 7 mhigh. The surface of the mound is covered withriver cobbles of different sizes and a very smallquantity of ceramics. There has been somecollapse on the rear of the mound. In the

exposed profile, river cobbles in mud mortarare mixed with mollusk valves. No ceramics werefound in the exposed portion of the Sector Amound. A shallow depression indicating thepresence of an atrium can be seen on the summitof the mound on the plaza side of the structure.

SECTOR B—Sector B includes an elongatedrectangular mound measuring 55 3 127 m. Thenorthwest half of the mound is 9 m high, endingin a level terrace. The other half of the Sector Bmound is lower, approximately 3 m high, and isthe location of modern corrals. The surface of thisstructure is covered with river cobbles with veryfew ceramics. The margin of Sector B that facesaway from the plaza (Sector E) slopes steeply toan escarpment that appears to be natural but hasbeen made even steeper by excavation of a canaljust below the rim of the escarpment.

SECTOR C—Sector C includes an elongatedrectangular mound measuring 50 3 95 m. SectorC forms the right arm of the overall rectangularor U-shaped arrangement of structures on thenorth face of the platform. The Sector C moundincludes two rectangular platforms that dividethe summit in two. Both platforms are similar insize and are about 8 m high. The platformmound that comprises Sector C has been de-graded by the action of heavy machinery on thelower slopes of the mound and by the construc-tion of a corral, now abandoned, on the summit.The surface is covered with medium and largecobbles, a small quantity of ceramics, mollusks,possible lithics, a few animal bones, and botan-ical remains, including gourd fragments, char-coal, seeds, and stems.

SECTOR C: PROFILE 1—A trench 3 3 5 m, wascleared and then extended to 3 3 7 m. Thedeposits varied from the upper to the lower endof the unit.

Layer 1 was a thin layer of partially com-pacted orangey sand that contained no cultural

FIG. 25. Air photo (SAN 1970) showing the relationship between Pampa San Jose (left) and Carreteria (right).

FIG. 26. Air photo of Huayto, located atopa triangular landform dissected from nearby hills.


material. It appeared to have been depositedrelatively recently at the highest point cleared.

Wall A was built of cobbles approximately10 cm in diameter set in clay mortar. Theconstruction was irregular in appearance, likelyto have been part of a corral or a temporarystructure. Beneath the wall of river cobbles wereremains of sugarcane ash and fragments ofmortar.

Wall A—Modern wall of medium cobblesapproximately 10 cm in diameter. Beneath thewall of river cobbles were remains of sugarcaneash and fragments of mortar. The wall was ofpoor but consistent construction.

Layer 2 was a layer of loose cobbles in gray/beige sand with some gravel. Very little culturalmaterial was present, including a few molluskshells mixed with clods of clay that may comefrom collapsed walls.

Layer 3 was an accumulation of loose cobblesand pieces of mortar. Very little cultural materialand very little soil was present in this rocky area.The rock may come from collapsed structures orwalls.

Wall B—This ancient wall was built of largecobbles with mortar filling the spaces betweencobbles. The cobbles were placed in uniformrows. Wall B was identified when the loosematerial had been cleared from this area. Wall Bwas built of large cobbles with mortar filling thespaces between cobbles. The cobbles were placedin uniform rows.

Layer 4 consisted of gray soil and mediumsized cobbles at the south end of the cleared area.

A shicra bag fragment was uncovered at theedge of Wall B (Fig. 27A). This sample wasdated 2240 Cal B.C. (ISGS-5285). A sample fromanother shicra bag found at the south end of theunit (Fig. 27B) was dated 2270 Cal B.C. (Beta-117312) (Table 9).

Analysis of this profile identified three phasesof occupation in the area that was cleared.

Phase 1 includes the relatively recent deposi-tion of an orangy layer, probably contemporaryconstruction debris.

Phase 2 leveled the summit of the mound.Deposition of a thick layer of sandy materialappears to have occurred after some wall collapseas indicated by the lumps of mortar present. WallA separates Layers 2 and 3. The wall may havebeen built to contain the material deposited inPhase 2 or may have been reused as a retainingwall.

Phase 3 showed deposition of gray soil at thesouth end of Profile 1. Only a small corner of thisdeposit was identified, but it appears to be underdeposition related to the collapse of structures.

Phase 4 may be the same date or later thanLayer 2. From the thickness of the layer and thescanty cultural remains this appears to haveaccumulated over a period of time from collapseof structures and the gradual erosion of themound summit and sides.

Phase 5 includes Wall B, which appears to bepart of the initial construction of the mound.

Conclusions—From the information revealedduring excavation of Profile 1 at Huayto, someof the mound remains are preserved, as indicatedby the presence of Wall B. However, thatsurviving portion of Mound B is covered byalmost a meter of collapse and other subsequentdeposits. Substantial additional excavationwould be required in order to reveal the extentof the structure that is preserved in this sector.The presence of shicra bags in the cleared profilesupports the suggestion that Layer 4 accumulat-ed from the collapse of structures after the sitewas abandoned. If the area were abandoned foran extended period, many shicra bags may havedisintegrated, leaving only the few that werecovered when walls collapsed. The two datedsamples were obtained from Layer 4, suggestingthat the mound had been constructed andpossibly was abandoned and in the process ofcollapse by 2270–2240 Cal B.C. The consistencyof these two dates is reinforced by the fact thateach was analyzed at a different laboratory.

SECTOR D—This includes a small irregular-shaped mound measuring approximately 55 3

65 m located northwest of Mound B and about1 m high. The surface is covered with rivercobbles. A few ceramic fragments are present inthe areas where river cobbles are not present, andthe loose sediment includes some mollusk valves.

SECTOR E—This includes an open area sur-rounded by Sectors A, B, C, and F. The area ispartially covered with loose river cobbles.However, there are some places where the stoneshave been cleared, such as the soccer field, whereloose soil, a deep brown color, containingmollusk valves and fragments of ceramics, canbe seen. Toward the corner formed by themounds in Sectors A and B, a slight depressioncan be seen in the surface; this probably wasa sunken circular plaza that has been filled inwith cobbles from the surrounding area.


FIG. 27. Map of Huayto showing mounds and other features, sectors, canals, roads, and encroachingstructures.


SECTOR F—This sector is located east of SectorC and includes a destroyed mound on the surfaceof which are numerous loose river cobbles,mollusk valves, and fragments of ceramics. Nearthe mound in Sector C is a high spot. Aerialphotos of Huayto show that Sector F wasoriginally rectangular with dimensions similarto those of the mound in Sector C.

SECTOR G—Located to the east of Sector F,this includes a damaged mound. On the surfaceof Sector G, only a few cobbles, flakes, andmollusk shells are visible. No architecturalelements are visible, and ceramics were notobserved. The area of material culture coversabout 10 3 15 m.

SECTOR H—This includes a level area west ofthe mound in Sector D. On the surface, molluskshells and a few fragments of ceramics arepresent. In a few shallow holes in the surface,lenses of ash can be observed. It is likely that thiswas a domestic area. No walls or architecturalfeatures are present, and at present this sectorhas been divided into lots for the construction ofhouses.

SECTOR I—This is located southeast of SectorG. It includes a small mound, approximately 5 3

9 m and 1.5 m high. Only rounded cobbles arevisible on the surface, which is presently beingused as a corral. Between Sectors G and I area group of contemporary houses that may havedestroyed and covered archaeological remains.On the surface surrounding these houses aremollusk shells and ceramic fragments.

DISCUSSION: HUAYTO—Unlike other sites in thePativilca Valley, Huayto is situated in a defensivelocation, atop a steep sided terrace. While LateArchaic sites in the region often have a U-shaped

layout (Haas et al. 2004), Huayto formed a closedrectangle comprised of Sectors A–C and F. Thetwo radiocarbon dates from Huayto are closelyclustered, suggesting that the Sector C moundwas built with at least some shicra bag construc-tion by 2270–2240 Cal B.C. The symmetricalarrangement of structures in Sectors B and C,each apparently topped by two platforms, andthe position of the Sector A mound suggesta coordinated construction plan. The Sector Amound with the atrium facing into the plaza(Sector E) probably was connected to a sunkencircular plaza by a staircase. The circular plaza isno longer visible, though it could be buriedrather than destroyed.

Huayto is a site that is in imminent danger ofbeing destroyed by the expansion of modernsettlement. However, it also appears to have hada substantial ancient history. The mounds stillvisible appear to date to the Late Archaic period.The excavations revealed even more disturbancethan is visible on the surface. Although ceramicsare visible on the surface, these appear to bepostdate use of the site. Local residents describedextensive ceramic-bearing cemeteries in the areaaround Huayto away from the mounds that wereassociated with later occupations located closerto the river or agricultural fields.

Cerro R

Cerro R is located some 2.5 km northeast ofthe city of Barranca. The site is a funerary zone,located on the slopes of Cerro R on the leftmargin of the Pativilca River. There are at leastfour distinct looted cemeteries spread across

FIG. 28. Profile 1 at Huayto.


this extensive area of sandy and rocky hills.Cerro R is characterized by sand surroundingrocky outcrops whose slopes have been damag-ed by looting. The site is divided into threedifferent sectors by the presence or absence ofarchaeological material on the surface. Noarchitectural features were observed in any partof the site.

In almost all sectors of the site, human boneand lithic and mollusk fragments are abundantaround areas of looting. This is clearly a cemeterythat was used during different time periods. Asignificant number of burials have been lootedacross this site, though there are also areas thatare undisturbed. Many of the areas haveassociated ceramics in the back dirt, but somedo not. There are some fragments of twinedtextiles on the surface that may date to the LateArchaic period, though a majority of the textilefragments visible are flat-weave or other loomedvarieties from later, ceramic periods.

Areas of the cemetery that are likely to bepreserved have reduced numbers of mollusk,lithic, ceramic, gourd, and cotton textile frag-ments on the surface. Large areas of the site arebeing actively destroyed by new house construc-tion and associated residential infrastructure,however, so areas appropriate for testing maydisappear in the coming years.

The small quantity of ceramics observed andthe large number of mollusk fragments distrib-uted across Cerro R suggest that this could havebeen a funerary area of the Late Archaic periodwith a subsequent period of reuse. PANCmembers cleared three areas on the surface ofCerro R that had been looted to examine thesurface below the loose sand for evidence ofburial pits or other features.

In Sector A, two areas, Units 1 and 2, werecleared, each 5 3 5 m in area.

Sector B was an area similar to Sector A, withextensive evidence of looting.

Sector C was an area that appeared to haveexperienced less looting than the other portionsof Cerro R. One 5 3 5-m unit was excavated inthis area.

SECTOR A: UNITS 1 AND 2—Sector A is locatedon the southern slope of Cerro R, which slopestoward the basin at the center of the hill.Measuring 100 3 50 m, the surface of this sectoris covered with mollusk fragments, lithics, gourdfragments, and a small number of human bones.Sector A is a heavily disturbed area. The entirezone is being used as a sanitary landfill for the

TABLE9.

Hu

ay

tora

dio

carb

on

da

tes

fro

msa

mp

les

ob

tain

edin

20

02

.

Pro

ven

ien

ce

Tex

ta

nd

illu

stra

tio

nlo

cati

on

Ma

teri

al

La

bn

o.

Ca

l.B

.C.

RC

YB

P12C

/13C

Wei

gh

to

fsa

mp

le(g

)C

ali

bra

ted

ag

era

ng

e1

sig

ma

(68

.3%

)B

.C.

Ca

lib

rate

da

ge

ran

ge

2si

gm

a(9

5.4

%)

B.C

.

Mo

un

dC

,A

(Ba

g1

7)

Fib

erb

ag

ISG

S-

22

40

38

00

67

0–

12

.01

2.6

23

97

–2

38

4(4

.6),

23

45

–2

13

82

46

0–

21

11

(89

.8),

21

01

–2

03

6P

rofi

le1

52

85

(95

.4)

(10

.2)

Mo

un

dC

,B

(Ba

g1

8)

Fib

erb

ag

Bet

a-

22

70

38

20

67

0–

9.7

11

24

01

–2

37

7(9

.3),

23

50

–2

19

12

46

6–

21

23

(94

.2),

20

96

–2

04

0P

rofi

le1

17

73

12

(76

.8),

21

78

–2

14

2(1

3.9

)(5

.8)


inhabitants of the nearby new settlement. Thereis also active looting of the cemetery in this area.As a result, surface materials included textiles,hair, and human bone. A fragment of gourdinscribed with an upright figure was found on thesurface of Cerro R. The motif pyroengraved onthe surface of the gourd is similar to that foundat the neighboring site of Cemetery P (Fig. 29, cf.Fig. 32).

In Unit 1 of Sector A, an area of 5 3 5 m wascleared. The goal of the work was to identify theoutline of looter’s pits and to identify the size andconstruction of burial pits damaged by looting. Allsurface material was screened, and all artifacts andbone were collected. Carbon samples were ob-tained from fiber exposed while clearing, thoughno intact burials were identified. The carbonsamples from Cerro R have not been processed.

In Sector A, Unit 2, two layers were identifiedduring clearing.

Layer A included damp, unconsolidated sandysediment with fragments of shell, textiles, beads,gourd, some human bone, and plant materials,including seeds and gourd fragments. Layer Bwas the product of disturbance by lootersexcavating burials. Patches of sand of differentshades of brown showed overlapping piles ofdebris, including human bone, teeth, textiles,shell, gourd, botanical remains, and a fewceramic fragments. A piece of newspaper witha date of 1984 also came from this layer.

Conclusions—At this site, clearing of surfacematerial did not yield the same kind of results

that can be obtained by clearing architectural

remains. Thus, though radiocarbon samples werecollected, these were not in situ. The outlines of

burial pits that were anticipated in this area werenot preserved below the disturbed surface

material.

SECTOR B—Sector B is located in the nextsheltered slope of the hill to the north of Sector

A, to the southwest of the new settlement ‘‘LosArenales.’’ This sector is another funerary area,

and on the surface a great deal of human bone,textile fragments, and a few ceramic fragments

can be seen. Few mollusk shells were observed inthis sector, which has been looted over its entire

area.

SECTOR C: UNIT 1—Sector C is located 500 mnortheast of Sector B on the margin of the

central basin in the hill. Similar to the othersectors of Cerro R, Sector C covers an area that

slopes down to cultivated fields. This appears tobe another funerary area based on the mollusk

shells, textile fragments, and a few human boneson the surface. Only one ceramic fragment was

recorded here. This zone has only been slightlyaffected by looting.

Unit 1 is part of a looted cemetery in Sector

C. There are portions of this cemetery whereceramics are not visible on the surface or where

the quantity is very small. A nearby settlement isexpanding, covering part of the cemetery.

Further expansion of housing could cover moreof the cemetery, while gray water and sewage

FIG. 29. Incised gourd figure from Cerro R.


could eventually permeate the area, destroying

the archaeological materials below the surface.

Unit 1 was excavated in levels since the areaconsisted of an extensive deposit of sand withvery little visible differentiation across the areacleared or from top to bottom within the deposit.During the clearing, three levels were established,each approximately 10 cm thick. Bone, textiles,lithics, mollusks, ceramics, remains of paintedgourd, scraps of wood, and feathers wererecovered from the levels. It was possible todefine sectors of the unit where no ceramics at allwere present. Clearing of this previously dis-turbed area was intended to identify the densityof cultural material and to obtain samples oforganic remains related to the cultural depositssuitable for radiocarbon dating (Fig. 30).

Level 1 included ceramic fragments, shelltextiles, and human bone. Textile fragmentsfrom looted burials were so numerous that onlya sample was retained.

Level 2 was similar to the previous level,including sandy sediment with ceramics, shell,textile fragments, and human bone.

Level 3 was a continuation of the previouslevel, with a small, blackened area in thenortheast corner of the unit. Materials recoveredincluded bone, shell, textile fragments, andceramics.

Conclusions—The goal in this unit was to clearloose, surface material. Although no featurescould be distinguished, the volume of materialdemonstrates extensive looting in this area.Further excavation might yield the base of lootedburial pits. The profile view of the unit did notshow any stratigraphy in the sand.

Numerous textile fragments and other remainsassociated with funerary activities were recov-

ered from Unit 1. However, a distinction amongthe levels cannot be made because no stratigra-phy was visible, nor were any remains of burialfeatures present. Thus, the material cleared fromthe surface at Cerro R can be considered only asa single phase of cemetery use.

DISCUSSION: CERRO R—When looters leavetheir excavated pits, sand drifts in, resulting inan undulating surface across the entire area ofCerro R. However, below the surface, thecontours of areas excavated by looters were notpreserved as we had expected, nor was thereevidence of burial pits. A broad area similarto those cleared in Sectors A and C wouldneed to be excavated, not just cleared of loosesurface sediment, in order to obtain in situsamples for radiocarbon dating and to recordpit outlines.

The immediate goals of clearing areas at CerroR were to obtain samples for radiocarbon datingand to examine burial pits remaining afterlooting. Although some samples of plant fiberwere collected, these were not from materialsthat could be clearly distinguished as in situ. Thesandy surface material was loose and shifting,and the samples were obtained from screenedsediment. The objective of obtaining radiocar-bon samples can be considered only minimallysuccessful as a result. No dates were processedfrom Cerro R, though the fragment of incisedgourd fragment from this site is similar to thefragment from the Cemetery P site that is dated2220 Cal B.C. (see below).

Cemetery P

Cemetery P is located on the left margin of thePativilca River, about 2 km northeast of the cityof Barranca on the slopes of a small hill of thesame name. Cemetery P is a long, sandy areawith a north–south axis and is clearly a funeraryzone.

This site consists of a single extensive cemeteryarea that was heavily looted in the past anda small area of current looting. On the slopeopposite the hill south of the looted burials isanother area where mollusk fragments and lithicscan be seen on the surface that probablyrepresents an undisturbed Late Archaic ceme-tery. The surface of the looted area is coveredwith human bone, mollusk fragments, lithics, anabundance of gourd bowls, and cotton textiles.

Cemetery P can be divided into four sectorsdifferentiated by the presence or absence of

FIG. 30. Photo of Cemetery P cleared area(W. Creamer).


archaeological material on the surface. It isprobable that these funerary areas include LateArchaic layers, which would explain the verysmall quantity of ceramics in Sector A. Portionsof Sectors B, C, and D have been heavily looted,yet only a small quantity of ceramics is apparenton the surface (cf. L. Haas, 2004). In spite of this,it is still possible to distinguish intact areas inthese sectors.

All sectors have numerous lithic fragments andmollusk shells on the surface. Large blocks ofscoria from smelting metals cover part of SectorB. The sediment around the scoria is reddened,suggesting that this was a metalworking area atsome point, though no architectural elementswere identified.

SECTOR A—Sector A is located on the south-west slope of the hill and includes an areaapproximately 100 3 100 m. On the surface,which is only slightly disturbed, were molluskvalves, a few fragments of ceramics, gourd vesselfragments, a few corncobs, and a fragment ofspondylus shell. A small number of humanremains were spread across the surface alongwith remains of textiles and basketry. Sector Aappears to be an early cemetery.

SECTOR B: Unit 1—Sector B is located on thenortheast slope of the hill, including an area 1003 200 m. Like Sector A, there was no architec-ture present, and the surface was heavilydisturbed by looting. There were numerousremains of textiles, some decorated, principallywith parallel brown lines. Fragments of basketry,ceramics, and beads were also found. There wasalso cotton, possibly fragments of funerarybundles, and corncobs. The mollusk valves werespread across the entire surface, though theonly distinctive mollusk was the fragment ofspondylus. In the southern portion of Sector B,an area approximately 40 3 60 m, were somelarge blocks of scoria from metal smelting, in anarea where the sediment had been burned red. Inthis area, a small piece of sheet copper wasrecovered. In this sector, at least two differentoccupation layers are present, one having a fu-nerary function and the other a metallurgyworkshop.

In Sector B, a 5 3 5-m disturbed area on theeast side of the cemetery was cleared in an effortto identify the dimensions of constructed burialpits. All surface material was screened, and allthe artifacts and bones were collected. The recentlooter’s holes were cleared in order to drawprofiles. Radiocarbon samples were collected

from the plant fibers exposed during the clearingof profiles. No burials were excavated.

Five layers were identified in Unit 1.Layer A consisted of aeolian sand. Remains of

textiles, ceramics, lithic artifacts, wood, andbasketry were widespread since the cemeteryhas been heavily looted (Fig. 28).

Layer B was composed of loose sand that wasrecently deposited. Cultural material was presentas a product of the looting of burials. Burialswere placed in sand mixed with a few fragmentsof angular rock. The layer included disturbedhuman remains, though no features were visible,and the sand was uniform in texture andcontents. Artifacts recovered included ceramics,botanical remains (especially squash seeds andsome black-colored grains), fragments of cotton,and a variety of other textiles, human and animalbone, lithics, beads, feathers, and worked wood.None of the human remains were articulated.

Layer C deposits were composed of loose sandwith some rock inclusions. Disarticulated animaland human bone, textiles, botanical remains,beads, metal (small fragments of copper), andother materials were present, though no featurescould be identified, and the layer was uniform intexture and color. The base of Layer C wasmarked by a distinctive color and a change oftexture.

Layer D was a compact layer of dark reddish-brown clayey material with caliche dispersedthrough the layer. No artifacts were recovered.

Layer E’s compact gray sandy soil wasexcavated to be sure that Layer D was not a lensof sterile sediment that was covering culturaldeposits. Layer E included no cultural material.

Conclusions—Although it would have beenpossible to continue excavation, the objective wasto clear loose surface material, and for this reason,excavation did not continue. It would be pro-ductive to excavate an entire unit at Cemetery P tosterile subsoil or bedrock to establish the date ofuse of each sector and the depth of the deposits.Unit 1 did yielded minimal stratigraphy with colorchange in the sand in Layers D and E. However,the bulk of the artifacts were recovered fromLayers A, B, and C, where no stratigraphy wasvisible, yet there were numerous artifacts andhuman bone fragments present from looted graves.

SECTOR C—Sector C is located on the north-east slope of the hill. On the surface are molluskvalves, textile fragments, pieces of gourd vessel,and human bone dispersed across an area 50 3

100 m. Looting has damaged this sector.


SECTOR D—Sector D is located north of SectorB and is relatively small, 30 3 40 m. This sectorhas been heavily looted, and numerous humanremains are present, along with shell, textiles,and some ceramics.

DISCUSSION: CEMETERY P—Excavation at Cem-etery P was an effort to identify areas ofundisturbed burials. It was difficult to determinethis given the extensive looting that had beendone in this sector. No clear evidence of intactburials was found. Observation of a recentlylooted burial suggested one reason why it isdifficult to identify individual burial pits atCemetery P. The looter had discarded the bonesof the individual and woven wrappings in the pit,but one side of the burial pit was still preserved.The pit had been excavated into the sand of thedune that covers the area and was not lined withadobes, rock, or any other material. Apparently,the burial bundle was placed in a pit in the sandand covered. The only indication of the burial pitwas a thin brown crust of cemented sand grainsthat extended over the burial, observed in theprofile of what remained of the pit (Fig. 31). Thisbrown crust may have formed from moistureoriginally in the burial bundle, but whatever itssource, the burial pit seems to be something likean eggshell and just as fragile, likely to be groundback into sand very shortly after it is brokenopen.

A fragment of gourd engraved with a standingfigure was recovered from the surface ofCemetery P, and a small portion of the gourdwas AMS dated to 2220 Cal B.C. (Fig. 32,

Table 10) (Haas et al., 2003). The forward-facingpose of the individual, headdress, extended armsending in fingers or claws, the fanglike teeth, andlarge eyes are all elements found in latericonography, especially that associated with theStaff God (Demarest, 1981, pp. 43–69).

The presence of Late Archaic textiles (L. Haas,2005) in the material recovered from the test unitat Cemetery CP, plus the radiocarbon date fromthe pyroengraved gourd fragment on the surface,points to the use of this cemetery during the LateArchaic. However, the extent or intensity of theLate Archaic use cannot be assessed on the basisof the work done to date.

Potao (02PVGS-8)

Potao is located on the left margin of thePativilca River, adjacent to the northwest edgeof the community of Potao. The location ofPotao is different from other Late Archaic site inthis and neighboring valleys in that it is locatedon the floodplain, while the other large LateArchaic sites are on terraces or hills above thevalley and always adjacent to active irrigationcanals. Construction of houses and agriculturalfields of the local community have heavilydamaged Potao in recent years. Aerial photossuggest that there was originally a single mainmound in the northwest portion of the site, withtwo long arms extending outward on each sidein the form of a U (Fig. 33). Sectors A, B, andE appear to have been articulated. The mound

FIG. 31. Cemetery P, Sector B, profile of burial pit recently cleared by looters.


in Sector C is not physically connected to anyother, while the mound in Sector D is a naturalfeature, though the surface demonstrates evi-dence of early occupation. The southern armhas been covered with houses, while the plazaarea between the two arms of the main complexis today a cultivated field. Local residentsindicate that they have found artifacts duringhouse construction along with charcoal and

mollusk fragments, though no ceramics havebeen reported. In aerial photos, it is possibleto identify two additional mounds in Sector Ethat are no longer visible. According to localresidents, one of the mounds was apparentlyan archaeological construction. During theleveling of these mounds, local residentsreport seeing numerous river cobbles, and somebone.

FIG. 32. Cemetery P incised gourd fragment.


Sector A includes a platform mound 69 3 733 17 m completely constructed of rounded rivercobbles The mound faces northeast, with twolevels of terracing around the sides. From thetop of this mound, the rest of the site is easilyvisible, and on this upper surface, the tops ofseveral walls made of angular stone are visible.The east side of the mound has collapsedslightly, revealing a small cut or profile in whicha large number of mollusk shells can be seenin the construction fill of river cobbles and clay.On the surface of the mound are numerous rivercobbles and mollusk shells, while a few ceramicfragments can be seen along the sides.

The surface is badly disturbed by the con-struction of a chapel on the peak of the hill(Fig. 34). A cement platform for altars was builthalfway up the hill, and a path for processionswinds up one side of the mound. Houses andcorrals along the north, east, and southeastoccupy the lower margin of the mound.

In Sector A, a profile was cleared in an area onthe upper part of the western slope of the mainmound This locality was the best available forarchaeological work, considering the other sec-tors were occupied by houses or corrals or in thecase of Mound C, a later Ceramic periodoccupation.

SECTOR A: PROFILE 1—Profile 1 was located inSector A, the main mound of the U-shapedstructure, at the rear of the U. In this area, heavymachinery had removed a great deal of the upperportion of the mound. According to members ofthe community, this destruction was carriedout more than 30 years ago, when the haciendaowner tried to build a house on the summit ofthe mound. Today, only a few pieces offoundation rock and clay in each of the cornersof the mound mark the base of the structure.Heavy machinery reshaped the upper half of themound to create a truncated pyramid with twolevels. Screening the material from clearingProfile 1 yielded only a few mollusk valves,botanical remains, and a few fragments ofceramics. The profile fill also included moderncultural material, such as rectangular adobebricks.

Profile 1 revealed extensive construction fillcomposed of medium- and large-sized cobblesand mortar that was relatively unstable, pro-ducing frequent collapse of the sides of theprofile during clearing.

The following stratigraphic units were revealedduring clearing of the profile at Potao (Fig. 35).

TABLE10.

Cem

eter

yP

rad

ioca

rbo

nd

ate

sfr

om

sam

ple

so

bta

ined

in2

00

2.

Pro

ven

ien

ceM

ate

ria

lL

ab

no

.C

al.

B.C

.R

CY

BP

12C

/13C

Wei

gh

to

fsa

mp

le(g

)

Ca

lib

rate

da

ge

ran

ge

1si

gm

a(6

8.3

%)

B.C

.C

ali

bra

ted

ag

era

ng

e2

sig

ma

(95

.4%

)B

.C.

Su

rfa

ceG

ou

rdfr

ag

men

tIS

GS

-A-0

33

22

22

03

78

96

48

–2

8.1

.22

28

9–

21

86

(71

.3),

21

84

–2

14

1(2

8.7

)2

40

2–

23

76

(3.2

),2

35

2–

21

13

(87

.8),

21

00

–2

03

7(9

.0)


Layer 1 consisted of loose beige soil withabundant large and small cobbles. This layer washighly disturbed.

Layer 2 included unconsolidated beige soilwith numerous large and small cobbles. Analignment of cobbles was visible in this layer.

Layer 3 continued with chunks of adobe ina loose mix of soil with medium and smallcobbles. Two pockets of loose cobbles wereidentified in this layer, as was a pocket of looseangular rock. Each of these pockets couldrepresent disturbance rather than a featureassociated with mound construction, however.

Layer 4 included cobbles and clay mortar.

Conclusions—Analysis of these layers did notreveal undisturbed evidence of mound construc-tion. The mix of materials best shows the mid- tolate 20th-century attempt to remodel the moundinto a site appropriate for house construction.

The alignment of cobbles in Layer 2 may be theremains of ancient construction, but furtherexcavation would be necessary to be certain.Plant fibers and other botanical materials fromthe fill were recovered to obtain radiocarbondates. A small sample of charcoal was collectedfor radiocarbon dating, yielding a date of 1480Cal B.C. (ISGS-A429).

SECTOR B—Sector B is located to the north ofSector A and includes a long platform moundwith a NE/SW axis that measures 32 3 85 3

3 m. The surface of this mound has been heavilydamaged by modern construction, and corralsand animal pens largely cover it. No architec-tural remains such as walls are visible today.Little cultural material is present, only a fewfragments of ceramics and numerous loose rivercobbles that were used in the ancient construc-tion.

FIG. 33. Map of Potao showing mounds and other features, sectors, canals, roads and encroaching structures.


Longtime residents note that more than 30years ago, the mound was modified with heavymachinery to level the upper surface. Theoriginal height of the mound was similar to thatof the mound in Sector C, approximately 7 mhigh.

SECTOR C—Sector C is located to the northeastof Sector B and is separated from it byapproximately 4 m of cultivated fields. Rectan-gular shaped with a NE/SW axis, it measures 913 159 3 7 m on the southeast, while thenortheast corner is 4 m higher than the sur-rounding area. On the extreme southern cornerof the site, the surface is covered with angularrock, while in the central part of this sector, thereare numerous river cobbles. At the north end ofthe sector, there is a greater presence of sedimentwith river cobbles, and it is in this latter areawhere the greatest evidence of looting can beseen. At this end of the site, ceramics are visibleon the surface, and adobe bricks can be observedin the disturbed areas. In the central area of thissector, large rectangular hollows are present inthe mound surface. These are now largely filledin with loose river cobbles and were probablyrooms oriented NE/SW. The west side of themound has been altered by construction ofstorage areas for animal fodder, such as corn-stalks that are dried after harvesting. Looting has

damaged the northern margin of the sector, andthe southern edge of the sector is covered withmodern corrals.

SECTOR D—Sector D is located 15 m north ofSector C, on the north-central part of a naturalmound of irregular form. The surface is coveredwith river cobbles, and although there are noarchitectural features present, the cobbles coveran area 27 3 38 3 1.5 m, suggesting that therewas remodeling of the mound to create theappearance of a structure. From the proximity ofthe mounds in Sectors C and D, occupation ofboth probably began during the Late Archaic,with a later reoccupation.

SECTOR E—Sector E is located to the east ofSector A. Today it has been destroyed by thesettlement of Potao. The main mound structurecan be identified from aerial photos taken in1970 and by the height of the center of thesettlement compared to its surroundings. Themound in Sector E is likely to have had similardimensions and orientation to that in Sector B.In the few exposed vertical surfaces, river cobblesset in clay are visible along with a few molluskshells. No fragments of ceramics were visible,and the exposed profiles in the mound suggestthe mound was Late Archaic.

DISCUSSION: POTAO—Clearing Profile 1 atPotao revealed that the Sector A mound is an

FIG. 34. Photo of mound at Potao (J. Haas).


artificial structure built of cobbles set in claymortar. The area cleared was so heavily dis-turbed that it was uncertain whether any in situconstruction was present. The single piece of insitu charcoal recovered from the profile in themound of Sector A yielded an AMS radiocarbondate of 3215 6 35 RCYBP (,1480 Cal B.C.)(Table 11). This would appear to place theconstruction of this mound in the Initial periodrather than the Late Archaic. An Initial period

date is consistent with the presence of at leastsome ceramics and the unusual location of thesite on the floor of the valley rather than ona raised terrace or hill.

One date is not conclusive, and with theextensive disturbance at Potao, more excavationwould be necessary to determine when the moundwas initially constructed. The form, cobbleconstruction, and disturbance suggest that therecould be an earlier component to the site.

FIG. 35. Sector A, Profile 1 at Potao.


The cobble construction of the Sector Cmound also suggests an early occupation duringthe Late Archaic with an intensive reoccupationat a later date based on the presence of adobebricks and burials with ceramics.

Vinto Alto (02PVGS-9)

Vinto Alto is located on the left margin of thePativilca River, part of the town of the samename. The present community of Vinto Alto hasexpanded to cover much of the site and hascaused serious damage. A contemporary ceme-tery is located in the area, which may havecontained a sunken circular plaza, and houseshave been constructed around the base of eachmound. The site is composed of four sectors.Sectors A and B are two large mounds that arenearly side by side. Sector C, however, isa natural hill having two peaks (Fig. 36). Oneside of the hill was modified to construct a largeopen ‘‘atrium’’ area in the space between the twopeaks. Sector D is a platform near Sector C(Fig. 37).

Because of their close proximity, Sectors Aand B will be described together.

Sectors A and B each include platformmounds covered with both river cobbles andangular rock. Although they appear to constitutea single structure, they are in fact two distinctmounds. The base of several rooms can be seenon the top of each mound. Mound A is 74 3 853 23 m and Mound B 58 3 91 3 18 m in size.

On the summit of Mound A, there is a polishedstela, or huanca, 2.6 m long, 85 cm wide at oneend, 75 cm wide in the center, and 54 cm wide atthe other end, a form reminiscent of the‘‘lanzon’’ of Chavın de Huantar (Fig. 38). Thestela is approximately 20 cm thick. At the base ofthe stone, there are marks that may be whatremains of iconography, though these are muchdeteriorated. There are also circular indentationsworked into the surface of the stone.

Similar to Carreterıa, the Sector A mound atVinto Alto has been cut by a canal around thebase of the mound on the east side. Houses alsocover a large area on the east side of the mound,and no prehispanic structures or features arevisible. On the west side of the two mounds is anaccumulation of aeolian sand, while on thesurface of the two mound, a few flake fragmentscan be observed.

A room constructed of cement block thatfunctioned as a small chapel in the past was

TABLE11.

Po

tao

rad

ioca

rbo

nd

ate

sfr

om

sam

ple

so

bta

ined

in2

00

2.

Pro

ven

ien

ceM

ate

ria

lL

ab

no

.C

al.

B.C

.R

CY

BP

12C

/13C

Wei

gh

to

fsa

mp

le(g

)C

ali

bra

ted

ag

era

ng

e1

sig

ma

(68

.3%

)B

.C.

Ca

lib

rate

da

ge

ran

ge

2si

gm

a(9

5.4

%)

B.C

.

Mo

un

dA

,P

rofi

le1

Ch

arc

oa

lIS

GS

-A-4

29

14

80

32

15

63

5–

10

.3,

51

51

7–

14

87

(42

.3),

14

83

–1

44

4(5

7.7

)1

59

8–

15

88

(1.5

),1

58

1–

15

68

(1.8

),1

52

8–

14

10

(96

.7)


constructed on the summit of the mound inSector B. This construction destroyed whateverarchitecture might originally have existed on thesummit.

In Sector A, a trench 3 m wide and 6 m longwas cleared to remove disturbed material at thebase of the principal mound where it had beencut by heavy machinery. This area was designat-ed Profile 2. From this trench, plant fibers werecollected for radiocarbon dating.

In Sector C, a 3 3 5-m trench in the existinglooter’s holes in the atrium area was cleared sothat Profile 1 could be drawn.

SECTOR A: PROFILE 2—This profile was locatedon the south side of the principal mound inSector A, near the canal that forms the boundarybetween the site and the local community. This3 3 6-m unit extends toward the summit ofMound A from a point where Mounds A and Bmeet. It does not reach to the top or the base ofthe mound and is situated approximately in thecenter of one side of the mound. A large quantityof loose rock was removed to clear an in situprofile and to look for walls and other architec-tural elements. A variety of artifacts wererecovered, including charcoal, botanical materi-als (a burned wood fragment and a woodfragment), mollusks, and lithics.

Layer 1 included loose material on the surface,fine dry beige silt, and sand mixed with cobbles.Two samples of shicra in secondary context werecollected from this layer. One of these was dated2480 Cal B.C. (GX 30121).

Layer 2 was more compact than Layer 1 andconsisted of yellowish-gray sediment. Cobblesand angular rock were present in this layer.

Layer 3 included shicra in primary contextalong with cobbles and angular rock. Two shicrasamples from this context were dated to 2480 CalB.C. (ISGS-5287) and 2540 Cal B.C. (ISGS-5392) (Fig. 39A, B, C).

Conclusions—There were subtle signs of con-struction in the cleared profile, such as facedstone, angular rock in an alignment, an align-ment of cobbles, and shicra at the base of thewall identified near the base of the profile.However, these were not easily related to eachother because of the disturbance in this area. Thearea is best viewed as a single phase of depositionthat has been disturbed.

The frequency of lithics in material cleared inProfile 2 was high compared to other localities;excavators identified hammerstones, cores, andtools (retouched pieces). Fragments of burned rockwere also present in this area. Combined with theabsence of ceramics, these data suggest that Mound

FIG. 36. Photo of twin peaks at Vinto Alto (J. Haas).


A was a Late Archaic occupation. Radiocarbonanalysis of three samples of annual plant fibersupports this conclusion. Dates from Mound A of3970 6 70, 3970 6 70, and 4010 6 70 RCYBP(,2480, 2480, and 2540 Cal B.C.) are highlyconsistent. Two different laboratories conductedthese analyses, supporting the cluster of dates asmore than an artifact of analysis (Table 12).

SECTOR C—Located approximately 200 m eastof the present-day settlement of Vinto Alto,Sector C includes a large natural mound that hasbeen modified and built upon. The hill is 86 3

106 m at the base, which has been transformedinto a mound through construction of a series ofretention walls of angular rock though some

river cobbles are also present in the constructionfill. On the summit are rooms and a centralatrium that measures 15 3 17 m. The atrium isstill visible and faces east and west, eachdirection overlooking a lateral platform. All theconstruction within the atrium is oriented towardthe north, however.

In the atrium, the remains of plastered wallscan be seen, while in the exposed profiles, theconstruction fill can be identified (Fig. 40). Theatrium has been seriously affected by the collapseof the walls around its margins. The use of shicraas construction material and the absence ofceramics on the surface is part of the LateArchaic pattern.

FIG. 37. Map of Vinto Alto showing mounds and other features, sectors, canals, roads andencroaching structures.


Most of the remains of the principal moundwere in good condition, with undisturbedpatches on the summit and the slopes. In onearea, looters excavated a large hole and left wallsand construction fill exposed, while a houseoccupied one of the platform mounds adjacent tothe main mound and burro and pig corrals.

SECTOR C: PROFILE 1—In Sector C, existinglooter’s holes were cleared so that a profile couldbe drawn. This profile, located on the edge of theatrium, was designated Profile 1, an area 3 3

5 m. The plant fibers visible in the profile werecollected to obtain radiocarbon dates. The placewhere Profile 1 was cleared had been identifiedduring earlier examination of the site as an areawhere erosion and looting had been exposed atthe top of the Sector C mound. A floor super-imposed over another could be seen, cut throughby later activity. The area had been heavilydamaged by a large looter’s hole reaching to thebedrock below mound construction. Later thislarge excavated area was used as a goat corral,and the surface was liberally covered with

caprine feces. However, remains of shicra bagscould be observed, and some of these providedmaterial for radiocarbon dating. Clearing thisexposed profile permitted us to record severaldifferent layers of construction fill and architec-tural elements.

The following layers were recognized (Fig. 41):

Layer 1: A gray-brown layer composed ofloose sediment with abundant rocks 5–10 cmin size.Layer 2: A layer of loose, gray-brownsediment with numerous small rocks andpebbles less than 5 cm in size.Layer 3: Loose gray-brown sediment mixedwith gravel.

The four deposits distinguished here are part ofthe construction fill and show continual modifi-cation and remodeling of the area. Four occupa-tion phases were revealed in the cleared profile.

Phase 1, the earliest visible in Profile 1,includes Floor 6 and two subsequent replaster-ings, Floors 5 and 4.

Phase 2 included the fill over Floor 4, whichwas topped with Floor 3. A base of rock may havebeen placed over the fill and under the floor.

Phase 3 includes a shallow layer of fill overFloor 3 and Floors 1 and 2. Floor 1 isa replastering of Floor 2.

Phase 4 includes the materials that accumu-lated over the floors at a later time. Much of thismaterial covered the floors after looting, as thecleared profile shows the six exposed floorsunder layers of loose fill.

Conclusions—Although more than seven dif-ferent moments in time are represented by thefloors and layers of fill in Profile 1, they can bestbe grouped into four distinct phases of occupa-tion. However, additional excavation mightshow that Floor 3 lies directly below Floors 1and 2, making them part of Layer 3 rather thana separate layer. The presence of at least sixfloors in a single profile points to extensiveremodeling and reuse of this structure. Thesuccession of floors in the same area is alsoindicative that there was continuity in the formand function of the structure from one rebuildingto the next. This pattern is in marked contrast tothe remodeling done at Pampa San Jose, forexample, where the different phases of construc-tion were much more discontinuous and involvedstructural reorganization of the building.

Three samples of shicra bags from differentproveniences in Profile 1 were submitted for

FIG. 38. Huanca atop Mound A at Vinto Alto(J. Hass).


radiocarbon analysis and yielded dates of 2580 to2100 Cal B.C. (Table 12). Either this wide rangeof dates may indicate a long period of building,rebuilding, and reoccupations of this structure,or the date of 2100 with a wide margin of errormay be in error, skewing the overall distribution.Three additional samples of shicra bags werecollected on the summit of Sector C. Theseappeared to be in situ but were exposed on thesurface outside the area of the cleared profile(Table 12G, H, I). The dates obtained weretightly clustered at 2400, 2410, and 2420 CalB.C., suggesting there was a distinct episode ofconstruction in Sector C at this time, as well asboth earlier and later, based on the dates fromProfile 1.

Sector D is located 200 m west of Sector C. Itincludes a large platform some 65 m long,ranging from 25 to 35 m wide. The surface iscovered with river cobbles and angular rock.From the architecture and the absence ofceramics on the surface, this sector dates tothe Late Archaic. The platform has been affectedby construction of animal pens around the baseand by construction of a small chapel on thesummit.

DISCUSSION: VINTO ALTO—Testing at VintoAlto revealed long-term use of the site asindicated by numerous remodelings of thesurface of the atrium shown in Profile 1. Theexposed profiles yielded a consistent set of datesranging from 2580 to 2100 Cal B.C., and seven ofthe nine dates ranged from 2580 to 2410 Cal B.C.This is the earliest cluster of dates from amongthe sites tested and suggests that Sectors A, B,and C were all being modified by 2500 Cal B.C.and that activities continued at the site until atleast 2400–2100 Cal B.C.

Discussion: Pativilca Valley

Seven of the sites tested by PANC in 2002included monumental architecture. Of these,four were associated with sunken circular plazas(Table 13): Upaca (02PVGS-1), Punta y Suela(02PVGS-2), Pampa San Jose (02PVGS-3), andCarreterıa (02PVGS-4). These four sites arelocated on the right bank of the Pativilca River.On the right margin of the valley, we alsoidentified sites with large residential areas anda substantial architectural complex. Upaca and

FIG. 39. Sector A, Profile 2 at Vinto Alto.


TABLE12.

Vin

toA

lto

rad

ioca

rbo

nd

ate

sfr

om

sam

ple

so

bta

ined

in2

00

2.

Pro

ven

ien

ce

Tex

ta

nd

illu

stra

tio

nlo

cati

on

Ma

teri

al

La

bn

o.

Ca

l.B

.C.

RC

YB

P12C

/13C

Wei

gh

to

fsa

mp

le(g

)

Ca

lib

rate

da

ge

ran

ge

1si

gm

a(6

8.3

%)

B.C

.

Ca

lib

rate

da

ge

ran

ge

2si

gm

a(9

5.4

%)

B.C

.

Mo

un

dA

,P

rofi

le2

AM

ixed

pla

nt

fib

ers

GX

-30

12

12

48

03

97

06

70

–1

2.8

18

25

75

–2

50

8(3

5.9

),2

50

4–

24

01

(52

.2),

23

78

–2

35

0(1

1.9

)2

83

9–

28

17

(1.4

),2

66

5–

26

46

(1.3

),2

64

2–

22

77

(95

.2),

22

52

–2

22

9(1

.4),

22

21

–2

20

6(.

7)

Mo

un

dA

,P

rofi

le2

BM

ixed

pla

nt

fib

ers

ISG

S-5

28

72

48

03

97

06

70

–1

6.8

10

25

75

–2

50

8(3

5.9

),2

50

4–

24

01

(52

.2),

23

78

–2

35

0(1

1.9

)2

83

9–

28

17

(1.4

),2

66

5–

26

46

(1.3

),2

64

2–

22

77

(95

.2),

22

52

–2

22

9(1

.4),

22

21

–2

20

6(.

7)

Mo

un

dA

,P

rofi

le2

CF

iber

ba

gIS

GS

-53

92

25

40

40

10

67

0–

14

.11

32

83

1–

28

21

(2.6

),2

66

1–

26

50

(3.0

),2

62

4–

24

56

(91

.6),

26

24

–2

45

6(2

.9),

24

18

–2

40

6(2

.9)

28

60

–2

81

0(6

.0),

26

53

–2

72

2(2

.2),

27

01

–2

32

8(9

0.7

),2

32

4–

23

07

(1.1

)

Mo

un

dC

,P

rofi

le1

DF

iber

ba

gB

eta

-17

73

23

21

00

37

00

61

10

–1

3.4

15

22

78

–2

25

2(6

.6),

22

30

–2

22

0(2

.4),

22

07

–1

93

7(8

9.7

),1

93

0–

19

23

(1.4

)

24

57

–2

41

8(2

),2

40

7–

18

69

(94

.3),

18

43

–1

81

0(2

.2),

18

01

–1

77

6(1

.5)

Mo

un

dC

Pro

file

1E

Fib

erb

ag

GX

-30

12

42

33

03

86

06

60

–1

5.7

18

24

56

–2

42

0(1

7.4

),2

40

5–

22

80

(67

.4),

22

51

–2

23

1(1

0.4

),2

21

9–

22

09

(4.9

)

24

71

–2

18

9(9

5.3

),2

18

0–

21

42

(4.7

)

Mo

un

dC

,P

rofi

le1

FF

iber

ba

gIS

GS

-52

77

25

80

40

40

67

0–

12

.71

82

83

6–

28

18

(6.5

),2

66

4–

26

46

(6),

26

37

–2

46

7(8

7.5

)2

86

7–

28

04

(11

.1),

27

82

–2

76

7(1

.0),

27

63

–2

71

6(5

.6),

27

12

–2

42

8(7

9.3

),2

42

4–

24

03

(1.6

),2

37

5–

23

53

(1.4

)M

ou

nd

C,

sum

mit

NE

GF

iber

ba

gIS

GS

-53

94

24

10

39

30

67

0–

12

.71

82

25

1–

25

41

(4.4

),2

49

1–

23

04

(95

.6)

26

17

–2

61

2(.

4),

25

81

–2

20

1(9

9.6

)M

ou

nd

C,

sum

mit

NE

HF

iber

ba

gB

eta

-17

73

24

24

00

39

30

66

0–

11

.41

82

58

4–

25

43

(1.7

),2

48

9–

24

78

(5),

24

74

–2

33

1(8

7.4

),2

32

2–

23

09

(5.8

)

25

75

–2

50

8(1

2.6

),2

50

4–

22

76

(82

.6),

22

53

–2

22

9(2

.9),

22

21

–2

20

6(1

.6)

Mo

un

dC

,su

mm

itN

W

IF

iber

ba

gG

X-3

01

23

24

20

39

40

67

0–

13

.01

82

55

9–

25

36

(10

.4),

25

30

–2

52

4(2

.3),

24

96

–2

33

4(8

3.3

),2

32

0–

23

10

(3.9

)

26

19

–2

61

0(.

8),

25

97

–2

59

0(.

5),

25

83

–2

26

6(9

2.6

),2

26

3–

22

03

(6.1

)


FIG. 40. Adobe and clay-plastered floors visible in looted area of Sector C, Vinto Alto.

FIG. 41. Sector C, Profile 1 at Vinto Alto.


Punta y Suela both include two sunken circularplazas associated with large platform mounds.

On the left margin, however, we found thelargest structure in the Pativilca Valley, VintoAlto (02PVGS-9), Sector A, the principalmound, which measured 23 m high. Also onthe left margin of the river are Cemetery P(02PVGS-7) and Cerro R (02PVGS-6), bothlarge cemeteries that have been heavily looted,yet these extensive funerary areas include veryfew ceramic fragments. It was during investiga-tion at these sites that two fragments ofpyroengraved gourd were recovered bearingimages that may represent an ancient Andeandivinity. Carbon-14 analysis of one of thesefragments yielded a calibrated date of 2220 B.C.(Haas et al., 2003).

We can confirm that the most important siteand one that is clearly representative of the LateArchaic period in the Pativilca Valley is the siteof Upaca. Its great antiquity, large extension, thevolume of its mounds, and the presence of twosunken circular plazas indicate that it wasa center of great importance within the valley.Other sites of primary importance include VintoAlto and Punta y Suela. The site of Vinto Altoalso yielded an early date from well-preservedplatform mounds. Punta y Suela includedextensive residential areas with abundant botan-ical remains, which will assist us in reconstruct-ing the diet and the range of plant foodsconsumed. The radiocarbon dates from this sitesuggest it is also of considerable antiquity(Table 14).

Pampa San Jose has provided valuable data onconstruction of its mounds. By recording indetail the very extensive profiles exposed bya large looter’s pit in the summit of the mound,we took advantage of a unique opportunity toobserve construction techniques and the contin-

ual remodeling, construction, and labor invest-ment in these structures.

The recent investigations in the Pativilca Valleydescribed here have revealed a pattern of large siteswith monumental architecture that were occupiedduring the third millennium B.C. Sites of this typeare concentrated in three valleys in an area of only1,200 km2. Further, these sites are directly associ-ated with the introduction of irrigation in thearea and rapid transition from a hunting-and-collecting way of life to an agricultural one. Inwhat appears to be a situation of truly ‘‘pristine’’cultural development, this complex of sites pro-vides a window into understanding how leaderscame to exercise power over their respectivepopulations. For this, we need to understand therole of agriculture, warfare, and religion in theorigin and development of power relations duringthis extraordinary period of early political change.

These initial data from the Norte Chicodemonstrate that it was a center of intensivecultural development during the period from3000 to 1800 B.C. Kosok (1965), Burger (1992),Moseley (2001), and others have recognized andemphasized the extraordinary density of the LateArchaic and aceramic centers of the Norte Chicowith monumental architecture. The radiocarbondates span a much longer period but clusterbetween 2700 and 1800 B.C. (Table 14).

A rectangular, stepped pyramid shape charac-terized the majority of the large mounds in theregion. A broad stairway down the front of thepyramid conducted people from an open atriumon an upper terrace to a sunken circular plaza.Additional stairs and passageways led into otherrooms and structures on the highest level of themound, a pattern that can be reconstructed fromthe information available from excavations atCaral (Shady & Levya, 2003). Indications on thesurface correspond in general to the system of

TABLE 13. Features of Late Archaic sites in the Pativilca Valley.

Site Monumental architecture Circular plaza U-shaped layout Bank of river

Huayto Y ? Y RPampa San Jose Y Y N RPunta y Suela Y Y N RUpaca Y Y N RCarreterıa Y Y N RVinto Alto Y ? N RLos Olmitos Y ? ? LPotao Y N Y LCemetery P N N N LCerro R N N N L


stairways and passageways described by TomPozorski (1982) at Caballo Muerto. Generallythere is an open area at the lowest level, thesunken circular plaza, followed by a suggestionof stairs and passages that progressively narrowaccess to the higher sections of the platformmounds. Pozorski has interpreted this pattern asan indication of increasingly restricted access toa succession of religious rituals. A large numberof people observed or participated in activitiesheld in the open circular plazas, but a smallernumber went up the pyramid to the atrium, andonly a few people entered the sacred interiorspaces at the top of the mound.

Among the 30 large sites recorded in the NorteChico region, there is some variability in thegeneral pattern. Some sites, like Aspero, do not

have circular plazas, some, like Lurihuasi, do nothave several large mounds but a single tightly-packed mass of mounds. Some sites have onlya single mound, while some mounds do not haverecognizable features, such as an ‘‘atrium.’’Nevertheless, the consistent pattern is one inwhich each class of religious or ceremonialactivity took place in areas that would be visibleto a large audience. A small segment of thepopulation had access to a second layer of theceremony, while an even smaller group wouldhave participated in a third layer of ceremony.This pattern of open presentation of religiousactivities followed by restricted access to re-ligious works in interior spaces is a manifestationof the ideological base of power in the LateArchaic Norte Chico region.

TABLE 14. Summary of dates from Late Archaic sites in the Pativilca Valley.

Sites

UPC PYS POT PSJ VTA CAR HYT CP

CAL B.C.100200 x300400500600 x700 x800 x90010001100 xxx120013001400 x x1500160017001800 x x1900 xx20002100 x xx x x2200 xx x x xx x2300 xxx x2400 x xxxxx2500 xxx2600 x2700 x28002900300040005000 xx6000 x700080009000 x


During the investigations carried out during June,July, August, and September 2002 by the PANC, 10prehistoric sites were identified as possibly belong-ing to the Late Archaic period, and nine of thesewere targeted for testing. The search for sites in 2002was very limited in scope and consisted primarily ofreviewing air photos and driving the valley roadslooking for the telltale remains of stone platformmounds. Surface evidence, such as the plan ofmounds and plazas, construction material, anda general absence of ceramics on the surface and inexposed profiles of disturbed areas, demonstratedthat these sites dated to the Late Archaic orpossessed an occupation from this period. Acomplete inventory of possible Late Archaic sitesmust await the results of a comprehensive survey ofthe valley (Perales, Pativilca V survey report, 2006).

Six of the seven major centers with large scaleceremonial architecture yielded radiocarbondates between 2700 and 1800 Cal B.C. An eighthsite, Los Olmitos, with similar characteristics anda lack of ceramics in a number of exposedprofiles, could not be tested but is likely to havea Late Archaic component. One of the twocemeteries identified yielded a Late Archaic datefor a surface artifact—a pyroengraved gourd—and similar artifacts were found on the surface ofthe second cemetery.

Altogether, the very preliminary survey andtesting firmly established a significant LateArchaic occupation in the Pativilca Valley.Kosok (1965) was the first to point out thepotential importance of sites in Pativilca, andWilliams (1972, 1978–1980, 1985) pointed outthe presence of the pattern of platform moundsand sunken plazas. It was Engel (1987) who wasthe first to propose that there were a number ofsites in this valley that dated to what he calledthe ‘‘Cotton Preceramic,’’ the equivalent of theterm Late Archaic as used in this monograph.The work of the PANC expands on thisantecedent work and effectively confirms thatthe Pativilca Valley was a major center of largeceremonial centers with terraced platformmounds and associated sunken circular plazason the coast of Peru during the Late Archaic.The presence of this group of centers in Pativilcafits a similar pattern found in the FortalezaValley to the north (Vega-Centeno et al., 1998;Haas et al., 2005; Vega-Centeno, 2005) and theSupe Valley (Williams & Merino, 1979; Shady etal., 2003) to the south. Recent survey has shownthe presence of yet another group of Late

Archaic sites in the Huaura Valley, just southof Supe (Nelson & Ruiz, 2005).

As is the case in all of these valleys of theNorte Chico region, the Pativilca Late Archaicsites are close together and often visible fromone to the other, such as Pampa San Jose toCarreterıa. They are all concentrated in an areaof only about 70 km2. The close proximity ofthese sites to each other has interesting implica-tions in terms of possible intersite relations aswell as raising pressing questions in terms ofchronology. The physical placement of these sitesas well as their proximity points to a lack of anykind of direct conflict or warfare (Haas et al.,2005). The sites are positioned on open terracesnext to the river floodplain. Sites such as Upacaand Punta y Suela are immediately adjacent tolines of foothills that would have provided idealassault positions for attacking enemies (Keeley,1996; Haas, 2001), and the other sites are open toattack on all sides. These sites also lack any kindof defensive walls, lookouts, redoubts or hilltopfortifications. There are no signs of valleywidedefensive features, such as strategic fortificationsat points where the valley narrows as it comesdown from the mountains. Other access routesinto the valley from side quebradas also lackdefensive walls or fortifications. Overall, in thePativilca Valley, there are no hallmarks ofconflict or warfare at any level. The people inthis area were simply not concerned about attackfrom either their neighbors or outsiders.

If the different Late Archaic centers in thePativilca Valley were not at war with each other,what was the nature of intersite relationships?Although there are many possibilities, there arethree broad alternatives.

First, Shady (2003a,b,c, 2004) has proposedthat all the Late Archaic sites in the Norte Chicowere under the direct hegemony of Caral in theSupe Valley, which she maintains was the capitalof a Late Archaic state.

Second, there is the possibility that the LateArchaic sites in the Pativilca Valley and perhapseven the Norte Chico as a whole were in somekind of peaceful alliance or confederacy.

Third is the possibility that each of the LateArchaic sites in the Pativilca Valley was anindependent minipolity members of which werein competition with each other for members orparticipants by means of cyclical ceremonialactivities.

At this stage of regional research, there is nobody of empirical evidence to either support or


refute any of these alternatives. There is noevidence at all that the Pativilca sites weresomehow under the hegemony of Caral or anyother site and no indication that Caral was anymore or less important regionally than any numberof other large sites, such as Punta y Suela, PampaSan Jose, or Vinto Alto. Caral may be in bettercondition than these sites, but it is not larger, nordoes it have larger monumental architecture. Atthe same time, there is also no evidence that thepeople of these sites were in formal or informalalliances or confederacies with each other. Neitheris there indirect evidence for pressures that mightforge such alliances or confederacies, such asregional warfare or large-scale, valleywide irriga-tion systems. As with the other two possibilities,there is scant support for the third alternative ofcompetition between ‘‘peer polities’’ (Renfrew &Cherry, 1986; Creamer et al., 2006). The onlyindirect indication of possible competition is theconstruction of so many large mounds at sites soclose to one another. It is also interesting to notethat the mounds are constructed to maximize theirapparent size. Each presents a taller ‘‘front’’ to theside facing out into the valley. Thus, at Pampa SanJose, for example, the side of the main moundfacing the valley is several meters taller than theside facing in to the circular plaza. There does seemto be some indication that the size of the moundswas important, possibly for intersite politicalposturing. Elsewhere, the authors of this manu-script have made a more expanded case for theindependent polity model (Haas & Creamer, 2006).

The biggest problem with systematically testingany and all hypotheses about intersite relation-ships is establishing a solid valleywide andregional chronology. PANC has submitted 29samples for radiocarbon dating from eight sites inthe Pativilca Valley. While this has effectivelyestablished that these sites date to the 1,200-yearperiod between 3000 and 1800 B.C., they do notestablish with certainty whether these sites werecontemporaneous with each other. For example,most of the sites have radiocarbon dates from the100-year period between 2200 and 2100 Cal B.C.(3775 and 3700 RCYBP). Were these sites alloccupied contemporaneously during this 100-yearperiod? Because of the statistical variabilityinherent in radiocarbon dating, it is not possibleto answer this question. On the one hand, it ispossible that all the sites were occupied for theentire 100 years; on the other hand, it is equallypossible that sites were occupied for only a portionof this century or in fact that none of the sites were

occupied at all during this century (because of thebroad range of potential variability at a 95%confidence interval). All these scenarios havemajor implications in terms of potential intersiterelationships. Was there one site occupied at anygiven time or five sites all occupied at the sametime? The dates available at this time are simplyinadequate to definitively address this problem.

During the Late Archaic in the Norte Chico,in the absence of chronologically sensitiveceramics or any other type of artifact (with thepossible exception of textiles; L. Haas, 2005),firm dating of the occupation represents botha methodological and a theoretical challenge.Addressing the broad and extremely interestingtheoretical questions about the origins of the firstcomplex polities to arise in the Andes. A solutionis going to require extraction and analysis ofradiocarbon samples on an order of magnitudethat is currently unprecedented in the region.Recent research at Chavın de Huantar, wherehundreds of radiocarbon samples were collectedto determine the construction sequence of thissingle site (Kembel, 2001; Kembel & Rick, 2004),demonstrates the need for this kind of dating andthe value that can be derived from such analysis.

The Late Archaic archaeological record of thePativilca Valley makes up a critical componentof the emergence of an identifiably Andeancivilization in the Norte Chico region of thePeruvian coast in the third millennium B.C.Unfortunately, these important archaeologicalresources have suffered greatly through modernagriculture, construction, and vandalism in thepast 50 years. At least 50% of each and every sitehas been completely destroyed since the 1940s,and all of them continue to be immediatelythreatened by the expansion of fields, housing,and roads. Community education programs,a local museum in the town of Pativilca, andlimited tourism development plans are currentlybeing initiated to allow the people of the valley toplay a more active role in the long-termpreservation of these important parts of Peru’sancient cultural patrimony.

AcknowledgmentsSupport for the project has come from

a number of sources, including the NationalScience Foundation (Collaborative ResearchGrant BCS-05442160), Northern Illinois Univer-sity, The Field Museum, James H. Duncan Sr.,


James and Susan Hannigan, Jeri Ailes and DavidHanley, Donald and Marianne Tadish-Marzec,and others.

Our neighbors in the city of Barranca, especiallyGaim Byrne and his family, have facilitated ourstay in the area and have been good friends. Ourgratitude goes out to the leaders of local commu-nities and the civil authorities in Potao, Huayto,Upaca, Vinto Alto, Pampa San Jose, and Punta ySuela and especially to the mayor of the munic-ipality of Pativilca at the time the work was carriedout, Sr. Hilario Cruz Alvarado, and to the generalsecretary of the municipality, Sr. Eduardo MartelOliva, who have facilitated our fieldwork andassisted in making the First Workshop a reality.

The Proyecto Arqueologico Norte Chicoincludes archaeologists and supporters of ar-chaeology who together have changed ourknowledge of ancient Peru. We recognize allthe members of the project who have createda team that demonstrates their commitment tothe objectives of the project. The effort andenergy that every person has invested in theproject has forged a group that is up to any task.

Literature Cited

ADVINCULA, MARIO. 2005. Analisis de ArquitecturaMonumental Temprana en el Norte Chico: ElPreceramico Tardıo en los valles de Pativilca yFortaleza. Paper presented at the 70th annualmeeting of the Society for American Archaeology,Salt Lake City, UT.

ALARCON , CARMELA. 2005. Evidencias BotanicasDurante el Preceramico Tardio en el Norte Chicode Peru. Paper presented at the 70th annual meetingof the Society for American Archaeology, Salt LakeCity, UT.

BILLMAN, B. 1999. Reconstructing prehistoric politicaleconomies and cycles of political power in theMoche Valley, Peru, pp. 131–159. In Billman, B.and G. M. Feinman, eds., Settlement Pattern Studiesin the Americas: Fifty Years since Viru. SmithsonianInstitution Press, Washington, DC.

———. 2001. Understanding the timing and tempo ofthe evolution of political centralization on thecentral Andean coastline and beyond, pp. 177–204.In Haas, J., P. ed., From Leaders to Rulers. KluwerAcademic/Plenum Publishers, New York.

BIRD, J., AND J. HYSLOP. 1985. The PreceramicExcavations at the Huaca Prieta, Chicama Valley,Peru. Anthropological Papers of the AmericanMuseum of Natural History 62(1): Pt. 1.

BONAVIA, D. 1982. Preceramico Peruano, Los Gavi-lanes: Mar, Desierto y Oasis en la Historia delHombre. COFIDE and Instituto Arqueologico Ale-man, Lima, Peru.

———. 1991. Peru, hombre e historia: De los orıgenesal Siglo XV. Edubanco, Lima, Peru.

———. 1993–1995. La domesticacion de las plantas ylos orıgenes de la agricultura en los Andes Centrales.Revista Historica 38: 77–107.

BRIA, REBECCA. 2004. A Persistent Frontier: ExternalInfluences and Local Traditions in the PativilcaValley Peru as Reflected in Ceramic Style. Un-published M.A. thesis, Anthropology Department,Northern Illinois University, De Kalb, IL.

BURGER, R. 1987. The U-Shaped pyramid complex,Cardal, Peru. National Geographic Research 3(3):363–375.

———. 1992. Chavin and the Origins of AndeanCivilization. Thames and Hudson, London.

BURGER, R., AND L. SALAZAR-BURGER. 1991. Thesecond season of investigations at the Initial Periodcenter of Cardal, Peru. Journal of Field Archaeology18(3): 275–296.

CARLSON, K., AND N. CRAIG. 2005. Late PreceramicMound Site Locational Pattering in the Norte Chicoof Coastal Peru: A GIS Approach. Paper presentedat the 70th annual meeting of the Society forAmerican Archaeology, Salt Lake City, UT.

———. 2006. Late Preceramic and Initial PeriodSettlement Dynamics in the Huaura Valley, NorthCentral Coast, Peru: A GIS-Based LocationalAnalysis. Paper presented at the 71st annual meetingof the Society for American Archaeology, San Juan,Puerto Rico.

CHUN, TERESA. 2003. Preceramic Circular Plazas in theNorte Chico, Peru. Unpublished M.A. thesis,Anthropology Department, Northern Illinois Uni-versity, DeKalb, IL.

CREAMER, W., A. RUIZ, AND J. HAAS. 2006. LateArchaic Regional Organization in the Norte Chico.Paper presented at the 71th annual meeting of theSociety for American Archaeology, San Juan,Puerto Rico.

Field Crew for 2002 Field Season:Mario Advincula

Micaela AlvarezCarmela AlarconJesus HolguınCarlos EscobarVeronica OrtizFelipe Libora

James Duncan Sr.Kit Nelson

Rebecca OsbornDavid Meyer

Dan CorkillTeresa Chun

Lillian HaasRebecca Bria

Vanessa DoranEric Hubbard

Eugenio GalvezTana Ratana

Maximino BardalesGuillermo GalvezSantos Bardales

Lyra Haas

Barranca Crew for 2002 Field Season:Gaim Byrne Esperanza Natividad CarpioMiguel Loo LiliFrancisco (Pancho) Fiorela Byrne


CORDOVA HERRERA, G. M. 2001. Informe FinalProyecto de evaluacion arqueologica linea detransmision 66KV Paramonga—Huarmey. Copyon file, Instituto Nacional de Cultura, Lima, Peru.

DEMAREST, A. 1981. Viracocha: The Nature andAntiquity of the Andean High God. PeabodyMuseum of Archaeology and Ethnology, Cam-bridge, MA.

DILLEHAY, T., J. ROSSEN, AND P. J. NETHERLY. 1997.The Nanchoc tradition: The beginnings of Andeancivilization. American Scientist, January–February:46–55.

DILLON, M. O., M. NAKAWAKA, AND S. LEIVA

GONZALES. 2003. The lomas formations of coastalPeru: Composition and biogeographic history, pp.1–9. In Haas, J. and M. O. Dillon, eds., El Nino inPeru: Biology and Culture over 10,000 years.Fieldiana Botany, n.s. 43, Publ. 1524.

DUNN, S., AND A. HEATON. 2005. Hard to FindTools: Preceramic Lithics of the Norte ChicoRegion, Peru. Presentation at the 70th annualmeeting of the Society for American Archaeology,Salt Lake City, UT.

ENGEL, F. 1957. Sites et etablissements sans ceramiquede la cote Peruvienne. Journal de la Societe desAmericanistes 46: 67–155.

———. 1963. A Preceramic Settlement on the CentralCoast of Peru: Asia Unit 1. Transactions of theAmerican Philosophical Society, n.s. 53, Pt. 3.

———. 1987. De las begonias al maız: Vida yproduccion en le Peru antiguo. Ediagraria, Uni-versidad Agraria La Molina, Lima, Peru.

FELDMAN, R. 1980. Aspero, Peru: Architecture, Sub-sistence Economy and Other Artifacts of a Precera-mic Maritime Chiefdom. Ph.D. dissertation, De-partment of Anthropology, Harvard University,Cambridge, MA.

———. 1983. From maritime chiefdom to agriculturalstate in Formative coastal Peru, pp. 289–310. InLeventhal, R. and A. Kolata, eds., Civilizations inthe Ancient Americas. University of New MexicoPress, Albuquerque, NM.

———. 1987. Architectural evidence for the develop-ment of nonegalitarian social systems in coastalPeru, pp. 9–14. In Haas, J., S. Pozorski and T.Pozorski, eds., The Origins and Development of theAndean State. Cambridge University Press, Cam-bridge, UK.

———. 1992. Preceramic architectural and subsistencetraditions. Andean Past 3: 67–86.

FRIED, M. 1967. The Evolution of Political Society: AnEssay in Political Anthropology. Random House,New York.

FUNG, R. P. 1988. The Late Preceramic and InitialPeriod, pp. 67–96. In Keatinge, R., P. ed., PeruvianPrehistory. Cambridge University Press, Cambridge,UK.

———. 2004. Resena del salvamento arqueologicoen Bandurria: Perıodo Preceramico Tardıo, pp. 325–336. In Quehaceres de la Arqueologıa Peruana—Rosa Fung: Compilacion de escritos. Ediciondel Museo de Arqueologıa y Antropologıa, Lima,Peru.

GREIDER, T., ET AL. 1988. La Galgada: A PreceramicCulture in Transition. University of Texas Press,Austin, TX.

HAAS, J. 1982. The Evolution of the Prehistoric State.Columbia University Press, New York.

———. 2001. Warfare and the evolution of culture,pp. 329–350. In Feinman, G. and T. Price, eds.,Archaeology at the Millennium. Kluwer Academic/Plenum Publishers, New York.

HAAS, J., AND W. CREAMER. 2004. Cultural transfor-mations in the Central Andean Late Archaic, pp.35–50. In Silverman, H., P. ed., Andean Archaeol-ogy. Blackwell Publishing, Oxford, UK.

———. 2006. Crucible of Andean Civilization, thePeruvian Coast from 3000 to 1800 BC. CurrentAnthropology 47: 745–775.

HAAS, J., W. CREAMER, AND A. RUIZ. 2003. GourdLord. Archaeology 56(3): 19.

———. 2004. Dating the Late Archaic occupation ofthe Norte Chico region in Peru. Nature 432:1020–1023.

———. 2005. Power and the emergence of complexsocieties in the Peruvian Preceramic, pp. 37–52. InVaughn, K., D. Ogburn and C. Conlee, eds.,Foundations of Power in the Ancient Andes.Archaeological Papers of the American Anthropo-logical Association, No. 14.

HAAS, J., S. POZORSKI and T. POZORSKI, EDS. 1987. TheOrigins and Development of the State in theAndes. Cambridge University Press, Cambridge,UK.

HAAS, L. 2004. Using textiles to date sites in the NorteChico. Manuscript in possession of the author.

———. 2005. Using textiles to date sites in the NorteChico. Poster Presentation at the 70th annualmeeting of the Society for American Archaeology,Salt Lake City, UT.

HUAMAN, L., W. CREAMER, AND J. HAAS. 2005. Po-llen from Preceramic sites in the Norte ChicoPeru. Paper presented at the annual meeting of theSociety for American Archaeology, Salt Lake City,UT.

KEELEY, L. H. 1996. War before Civilization. OxfordUniversity Press, New York.

KEMBEL, S. R. 2001. Architectural Sequence andChronology at Chavın de Huantar, Peru. Ph.D.Dissertation, Department of AnthropologicalSciences, Stanford University, Palo Alto, CA.

KEMBEL, S. R., AND J. W. RICK. 2004. Buildingauthority at Chavın de Huantar: Models of socialorganization and development in the Initial Periodand Early Horizon, pp. 51–76. In Silverman, H., P.ed., Andean Archaeology. Blackwell Publishing,Oxford, UK.

KOSOK, P. 1965. Land, Life and Water in AncientPeru. Long Island University Press, New York.

KROEBER, A. L. 1925. The Uhle Pottery Collectionsfrom Supe. University of California Publications inAmerican Archaeology and Ethnology 21(6).

LUMBRERAS, L. 1970. Los templos de Chavın. Cor-poracion Peruana de Santa, Lima, Peru.


———. 1971. Towards a re-evaluation of Chavın, pp.1–28. In Benson, E., P. ed., Dumbarton OaksConference on Chavın. Dumbarton Oaks ResearchLibrary and Collection, Washington, DC.

———. 1972. De los orıgenes del estado en el Peru:Nueva cronica sobre el viejo Peru. Milla Batres,Lima, Peru.

———. 1981. Los orıgenes de la civilizacion en el Peru.Lima, 5th edition. Milla Batres, Lima, Peru.

———. 1989. Chavın de Huantar en el nacimiento dela civilizacion andina. INDEA, Lima, Peru.

MOSELEY, M. 1975. Maritime Foundations of AndeanCivilization. Cummings, Menlo Park, CA.

———. 1985. The exploration and explanation ofearly monumental architecture in the Andes, pp. 29–58. In Donnan, C., P. ed., Early CeremonialArchitecture in the Andes. Dumbarton Oaks Re-search Library and Collection, Washington, DC.

———. 1992. Maritime foundations and multilinearevolution: Retrospect and prospect. Andean Past 3:5–42.

———. 2001. The Incas and Their Ancestors: TheArchaeology of Peru, 2nd edition. Thames andHudson, New York.

———. In press. The maritime foundation of Andeancivilization: An evolving hypothesis. In Trillo, P., P.ed., Peru y el mar: 12000 anos de la historıa de lapescarıa. Sociedad Nacional de Pesquerıa, Lima,Peru.

MOSELEY, M., AND G. WILLEY. 1973. Aspero, Peru: Areexamination of the site and its implications.American Antiquity 38: 452–468.

NELSON, K., AND M. PERALES MUNGUıA. 2006. Socio-Economic System of the Northern Chancay: LateIntermediate Period Settlement Patterns in theHuaura Valley, Peru. Paper presented at the 71stannual meeting of the Society for AmericanArchaeology, San Juan, Puerto Rico.

NELSON, K., AND A. RUIZ. 2005. Archaeological Surveyof the Huaura Valley: New Methods and ExcitingFinds. Paper presented at the 70th annual meeting ofthe Society for American Archaeology, Salt LakeCity, UT.

OSBORN, A. 1977. Strandloopers, mermaids and otherfairy tales: Ecological determinants of marine re-source utilization—the Peruvian case, pp. 157–205.In Binford, L., P. ed., For Theory Building inArchaeology. Academic Press, San Diego, CA.

PERALES MUNGUıA, M. 2006. Proyecto de Investgacion:Reconocimiento Arqueologico en el Valle Bajo dePativilca, Lima—Peru, Informe Final Report sub-mitted to the Instituto Nacional de Cultura, Lima,Peru.

PERALES, MUNGUıA, AND J. HAAS. 2005. Toward andUnderstanding of Late Archaic Architectural Com-plexes in the Norte Chico Region of Peru: Excava-tions at Caballete, Fortaleza Valley. Paper presentedat the 70th Annual Meeting of the Society forAmerican Archaeology, Salt Lake City, UT.

POZORSKI, S., AND T. POZORSKI. 1987. Early Settlementand Subsistence in the Casma Valley, Peru. Univer-sity of Iowa Press, Iowa City, IA.

———. 1990. Reexamining the critical Preceramic/Ceramic Period transition: New data from coastalPeru. American Anthropologist 92(2): 481–491.

———. 1992. Early civilization in the Casma Valley,Peru. Antiquity 66(253): 845–870.

POZORSKI, T. 1982. Early social statification andsubsistence systems: The Caballo Muerto complex,pp. 225–253. In Moseley, M. E. and K. C. Day, eds.,Chan Chan: Andean Desert City. School of Amer-ican Research, University of New Mexico Press,Albuquerque, NM.

POZORSKI, T., AND S. POZORSKI. 1990. Huaynuna, a LateCotton Preceramic site on the North coast of Peru.Journal of Field Archaeology 17: 17–26.

———. 2000. Una reevaluacion del desarrollo dela sociedad compleja durante el Preceramico Tardıoen base a los fechados rediocarbonicos y a lasinvestigaciones arqueologicas en el valle deCasma, pp. 171–186. In Kaulicke, P. ed., El periodoarcaico en el Peru: Hacia una definicion de losorıgenes. Boletın de Arqueologıa PUCP, Vol. 4.Departmento de Humanidades, Expecialidad deArqueologıa, Pontificia Universidad Catolica delPeru, Lima, Peru.

QUILTER, J. 1985. Architecture and chronology atEl Paraıso. Journal of Field Anthropology 12:279–297.

———. 1991. Late Archaic Peru. Journal of WorldPrehistory 5(4): 387–438.

———. 1992. To fish in the afternoon: Beyondsubsistence economies in the study of early AndeanCivilization. Andean Past 3: 111–125.

QUILTER, J., B. OJEDA, D. PEARSALL, D. SANDWEISS, J.JONES, AND E. WING. 1991. Subsistence economy ofEl Paraıso, an early Peruvian site. Science 251:277–283.

QUILTER, J., AND T. STOCKER. 1983. Subsistenceeconomies and the origins of Andean complexsocieties. American Anthropologist 85: 545–562.

RAYMOND, J. S. 1981. The maritime foundations ofAndean civilization: A reconsideration of theevidence. American Antiquity 46(4): 806–821.

RENFREW, C., AND J. CHERRY. 1986. Peer PolityInteraction and Socio-Political Change. CambridgeUniversity Press, Cambridge, UK.

ROSSEN, JACK. 1998. Unifaces in early Andean culturehistory: The Nanchoc lithic tradition of northernPeru. Andean Past 5: 241–299.

RUIZ, A., G. ASCENSIOS, K. CARLSON, N. CRAIG, W.CREAMER, AND J. HAAS. 2005. Mapping on DifferentScales and GIS in the Norte Chico: Enhancementsto Traditional Archaeological Excavation. Paperpresented at the 33rd annual Midwest Conferenceon Andean and Amazonian Archaeology andEthnohistory, University of Missouri, Columbia,MO.

SCHIFFER, M. 1986. Radiocarbon dating and the ‘‘OldWood’’ problem: The case of the Hohokam chro-nology. Journal of Archaeological Science 13: 13–30.

SHADY, R. 1993. Del Arcaico al Formativo en losAndes centrales. Revista Andina 21: 103–132.


———. 1995. La neolitizacion en los Andes Centralesy los orıgenes del sedentarismo, la domesticacion yla distincion social. Saguntum 28: 49–55.

———. 1997. La cuidad sagrada de Caral-Supe en losalbores de la civilizacion en el Peru. UniversidadNacional Mayor de San Marcos, Lima, Peru.

———. 1999a. Flautas de Caral: El Conjunto MusicalMas Antiguo de America. Boletın del Museo deArqueologıa y Antropologıa 2(10): 4–5.

———. 1999b. El Sustento Economico del Surgi-miento de la Civilizacion en el Peru. Boletın delMuseo de Arqueologıa y Antropologıa 2(11): 2–4.

———. 1999c. Los Orıgenes de la Civilizacion y laFormacion del Estado en el Peru: Las EvidenciasArqueologicas de Caral-Supe, Primera Parte. Bole-tın del Museo de Arqueologıa y Antropologıa 2(11):2–4.

———. 2000a. Ritual de enterramiento de un recintoen el Sector Residencial A en Caral-Supe, pp. 187–212. In Kaulicke, K., P. ed., El Periodo Arcaico en elPeru: Hacia una Definicion de los Orıgenes. Boletınde Arqueologıa PUCP, Pontificia UniversidadCatolica del Peru, Lima, Peru.

———. 2000b. Los Orıgenes de la Civilizacion y laFormacion del Estado en el Peru: Las EvidenciasArqueologicas de Caral-Supe, Segunda Parte. Bole-tın del Museo de Arqueologıa y Antropologıa 3(2):2–7.

———. 2000c. Practica Mortuoria de la Sociedad deCaral-Supe Durante el Arcaico Tardıo. Boletın delMuseo de Arqueologıa y Antropologıa 3(3): 2–15.

———. 2003a. Los orıgenes de la civilizacion y laformacion del estado en el Peru: Las evidenciasarqueologicas de Caral-Supe, pp. 93–100. In Shady,R. and C. Leyva, eds., La ciudad sagrada del Caral-Supe: Los orıgenes de la civilizacion andina y laformacion del estado prıstino en el antiguo Peru.Instituto Nacional de Cultura, Lima, Peru, Origi-nally published in Bolet ın del Museo deArqueologıa y Antropologıa 2(12): 2–4, 1999, and3(2): 2–7, 2000.

———. 2003b. El sustento economico del surgimientode la civilizacion en el Peru, pp. 101–106. In Shady,R. and C. Leyva, eds., La ciudad sagrada delCaral-Supe: Los orıgenes de la civilizacion andinay la formacion del estado prıstino en el antiguoPeru. Instituto Nacional de Cultura, Lima, Peru,Originally published in Boletın del Museode Arqueologıa y Antropologıa 2(11): 2–4,1999.

———. 2003c. Sustento socioeconomico del estadoprıstino de Supe-Peru: Las evidencias de Caral-Supe,pp. 107–122. In Shady, R. and C. Leyva, eds., Laciudad sagrada del Caral-Supe: Los orıgenes dela civilizacion andina y la formacion del estadoprıstino en el antiguo Peru. Instituto Nacional deCultura, Lima, Peru, Originally published in Ar-queologıa y Sociedad 13: 49–66, 2000.

———. 2003d. Caral-Supe: La civilizacion mas anti-gua del Peru y America, pp. 335–240. In Shady, R.and C. Leyva, eds., La ciudad sagrada del Caral-Supe: Los orıgenes de la civilizacion andina y laformacion del estado prıstino en el antiguo Peru.

Instituto Nacional de Cultura, Lima, Peru. Origi-nally published in Multiple: Cultura Peruana 3: 60–68, 2002.

———. 2003e. Caral-Supe y la costa norcentral delPeru: La cuna de la civilizacion y la formacion delestado prıstino, pp. 139–146. In Shady, R. and C.Leyva, eds., La ciudad sagrada del Caral-Supe: Losorıgenes de la civilizacion andina y a formacion delestado prıstino en el antiguo Peru. Instituto Nacionalde Cultura, Lima, Peru. Originally published inLohmann, G., R. Burger, Y. Onuki, and R. Shady,eds., 2000, Historia de la cultural peruana I, pp. 45–87.Fondo Editorial del Congreso del Peru, Lima, Peru.

———. 2004. Caral. Centura SAB, Lima.

SHADY, R., C. DOLORIER, F. MONTESINOS, AND L.CASAS. 2003. Los orıgenes de la civilizacion en elPeru: El area norcentral y el Valley de Supe duranteel Arcaico Tardıo, pp. 51–92. In Shady, R. and C.Leyva, eds., La ciudad sagrada del Caral-Supe: Losorıgenes de la civilizacion andina y la formacion delestado prıstino en el antiguo Peru. Instituto Nacio-nal de Cultura, Lima, Peru, Originally published inArqueologıa y Sociedad 13: 13–48, 2000.

SHADY, R., J. HAAS, AND W. CREAMER. 2001. DatingCaral, a Preceramic urban center in the Supe Valleyon the central coast of Peru. Science 292: 723–726.

SHADY, R. and C. LEVYA, EDS. 2003. La cuidad sagradade Caral-Supe: Los origınes de la CivilizacionAndina y la formacion del estado pristino en elantiguo Peru. Instituto Nacional de Cultura, Lima,Peru.

SHADY, R., AND A. RUIZ. 1979. Evidence for In-terregional Relationships during the Middle Hori-zon on the north-central coast of Peru. AmericanAntiquity 44(4): 676–684.

SILVA SIFUENTES, J. 1975. Excavaciones en Bermejo,Ancash, 1972. Bachelor’s Thesis, Departamento deCiencias Historico Sociales, Programas Academicosde Ciencia Social, Universidad Nacional Mayor deSan Marcos, Lima, Peru.

———. 1978. Acercamiento al estudio historico deBermejo, Anchas, pp. 310–324. In Matos, R., P. ed.,Actas y trabajos, II Congreso Peruano el hombre yla cultura andina, Vol. II. Editorial Lasontay, Lima,Peru.

STANISH, C. 2001. The origin of state societies in SouthAmerica. Annual Review of Anthropology 30:40–64.

UHLE, M. 1925. Report on Explorations at Supe, pp.257–264. Appendix to The Uhle Pottery Collectionsfrom Supe, by Kroeber, A. L. University ofCalifornia Publications in American Archaeologyand Ethnology 21 (6).

VEGA-CENTENO, R. 2005. Ritual Architecture ina Context of Emergent Complexity: A Perspectivefrom Cerro Lampay, A Late Archaic Site in theCentral Andes. Ph.D. Dissertation, University ofArizona, Tucson, AZ.

VEGA-CENTENO, R., L. F. VILLACORTA, L. E. CACERES,AND G. MARCONE. 1998. Arquitectura monumentaltemprana en el valle medio de Fortaleza. BoletınArqueologico 2: 219–238.


WILLEY, G., AND J. CORBETT. 1954. Early Ancon andEarly Supe Culture: Chavın Horizon Sites of theCentral Peruvian Coast. Columbia Studies inArchaeology and Ethnology, Vol 3. ColumbiaUniversity Press, New York.

WILLIAMS, C. 1972. La difusion de los pozos ceremo-niales en la costa Peruana. Apuntes 2: 1–9.

———. 1978–1980. Complejos de piramides conplanta en U: Patron arquitectonico de la CostaCentral. Revista del Museo Nacional 44: 95–110.

———. 1985. A scheme for the early monumentalarchitecture of the central coast of Peru, pp. 227–240. In Donnan, C., P. ed., Early CeremonialArchitecture in the Andes. Dumbarton OaksResearch Library and Collection, Washington,DC.

WILLIAMS, C., AND M. MERINO. 1979. Inventario,catastro y delimitacion del patrimonio arqueologicodel Valley de Supe. Report submitted to theInstituto Nacional de Cultura, Lima, Peru.

WILSON, D. 1981. Of maize and men: A critique of themaritime hypothesis of state origins on the coast ofPeru. American Anthropologist 83: 93–120.

———. 1988. Prehispanic Settlement Patterns in theLower Santa Valley, Peru: A Regional Perspective on theOrigins and Development of Complex North CoastSociety.SmithsonianInstitutionPress,Washington,DC.

ZECHENTER, E. 1988. Subsistence Strategies in the SupeValley of the Peruvian Central Coast during theComplex Preceramic and Initial Periods. Ph.D.dissertation, Department of Anthropology, Univer-sity of California, Los Angeles, CA.


Subject Index

Agriculture 5, 70, 73Alliance 72, 73Asia (site) 11Aspero 3, 4, 5, 71

Bandurria 4

Caral, See Chupa Cigarro Grande 4, 70, 72, 73Carreterıa 5, 43, 44, 45, 46, 48, 63, 67, 72Casma 2Chancay 2Chavın 1, 2, 63Chavın de Huantar, See Chavın, Mother Cul-

ture 1, 2, 63, 73Chimu 2, 3Chimu Capac 3Chupa Cigarro Grande, See Caral 4Circular Plaza 11, 12, 20, 25, 26, 30, 36, 37, 44,

45, 46, 48, 50, 52, 63, 67, 70, 71, 72, 73Confederacy 72Cotton 5, 53, 55, 56, 72Cotton Preceramic 5, 72

Defensive Walls 72

Early Horizon 2, 3, 4, 23, 25El Nino 2

Flotation 9, 10, 22Fortaleza 2, 4, 5, 72Fortification 72

GIS 11GPS 5, 6, 11

Huaca Prieta 11Huanca 63Huaura 2, 4, 72Huayto 5, 11, 48, 49, 49, 50, 52, 74

Initial Period 1, 3, 4, 5, 11, 20, 25, 36, 37, 43, 62

La Capitana, See Upaca 5

Late Horizon 3, 4, 12Late Intermediate PeriodLos Olmitos 5, 6, 72Lurın 1

Maize 33Middle Horizon 3, 4Moquegua Valley 2Mother Culture, See Chavın 1

Pampa San Jose 5, 37, 40, 43, 44, 48, 66, 67, 70,72, 73, 74

Pativilca 1, 2, 4, 5, 6, 7, 9, 11, 25, 26, 36, 37, 43,46, 48, 52, 55, 57, 63, 67, 70, 72, 73, 74

Peanut 35Peer Polities 73Polity 72, 73Pollen 9, 10, 11, 13, 22Potao 1, 5, 11, 57, 59, 61, 62, 74Punta y Suela 5, 6, 9, 25, 26, 31, 32, 33, 36, 37,

67, 70, 72, 73, 74Pyroengraving 54, 57, 70, 72

Quebrada Huanchay 5

San Nicolas 3Servicio Aerofotografico Nacional 5Shicra 7, 12, 14, 15, 16, 18, 19, 25, 30, 32, 38,

39, 40, 42, 46, 48, 49, 50, 52, 64, 65, 66, 67Squash 56State 5, 6, 37, 72Supe 2, 3, 4, 5, 19, 46, 66, 72

Textiles 10, 11, 15, 53, 54, 55, 56, 57, 73Tumbes 2

Uhle 3Upaca 5, 6, 9, 11, 12, 14, 18, 20, 23, 25, 67, 70, 72

Vinto Alto 5, 63, 65, 67, 70, 73, 74

Warfare 70, 72, 73

FIELDIANA: ANTHROPOLOGY N.S., NO. 40, SEPTEMBER 14, 2007, PP. 79 79

Field Museum of Natural History1400 South Lake Shore DriveChicago, Illinois 60605-2496Telephone: (312) 665-7055

FIE

LD

IAN

A: A

NT

HR

OP

OL

OG

Y, N

EW

SE

RIE

S, N

O. 40

AnthropologyNEW SERIES, NO. 40

September 14, 2007Publication 1546

Archaeological Investigation of Late ArchaicSites (3000–1800 B.C.) in the Pativilca Valley,Peru

Winifred CreamerAlvaro RuizJonathan Haas

Documents

Archaeological Investigation of Late Archaic Sites. Creamer, Ruiz y Haas