UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Changing patterns of pottery production during the Longshan Period of northern China, ca. 2500-2000 B.C. 1990

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
UBC_1990_A1 U52.pdf
UBC_1990_A1 U52.pdf [ 12.24MB ]
Metadata
JSON: 1.0058415.json
JSON-LD: 1.0058415+ld.json
RDF/XML (Pretty): 1.0058415.xml
RDF/JSON: 1.0058415+rdf.json
Turtle: 1.0058415+rdf-turtle.txt
N-Triples: 1.0058415+rdf-ntriples.txt
Citation
1.0058415.ris

Full Text

CHANGING PATTERNS OF POTTERY PRODUCTION DURING THE LONGSHAN PERIOD OF NORTHERN CHINA, CA'. 2500-2000 B.C. by Anne P. U n d e r b i l l B.A., Duke U n i v e r s i t y , Durham, North C a r o l i n a , 1977 M.A., U n i v e r s i t y of B r i t i s h Columbia, 1983 A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES (Department of Anthropology and S o c i o l o g y ) We accept t h i s d i s s e r t a t i o n as conforming to the r e q u i r e d standard THE UNIVERSITY OF BRITISH COLUMBIA J u l y 1990 (c) Anne P. U n d e r h i l l In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of fW^^'kyj ^ ^ ^ " . ^ The University of British Columbia Vancouver, Canada Date TW^£ l"7| (ITO DE-6 (2/88) ABSTRACT This study investigates how systems of pottery production change in r e l a t i o n to increasing c u l t u r a l complexity. A revised version of the important model outlined by Rice (1981) i s presented and tested with ceramic data from the Longshan Period of northern China-. At the end of the period, at least one state evolved i n the Huanghe (Yellow River) v a l l e y region. The model describes s o c i a l factors that may cause ceramic change in chiefdoms. It describes three a l t e r n a t i v e strategies of producers: d i v e r s i f i c a t i o n , s i m p l i f i c a t i o n , and conservatism. Consumer demand f o r labor-intensive vessels used i n displays of status may also cause changes i n production. After Rice (1981), the model predicts that v a r i e t y of ceramic categories should increase and that vessels should become in c r e a s i n g l y standardized. Further, there should be a change i n mode of production as s o c i o p o l i t i c a l complexity increases. The model i s tested with ceramic data from three s i t e s i n Henan (Hougang, Baiying, Meishan) and one i n Shandong (Lujiakou). During a period of si x months i n 1987, I examined reconstructed vessels from these s i t e s i n museums and archaeological work stations located i n Henan and Shandong provinces. The following analyses are described: analysis of shape classes defined i n s i t e reports (Chapter 4) , d i v e r s i t y of shape classes, dimensional standardization, within-class standardization, and assessment of labor-intensive vessels per phase (Chapter 5). In i i addition, evidence f o r pottery production at s i t e s and techniques of pottery production are discussed (Chapter 6). Two chapters examine published data on d i f f e r e n t i a t i o n with respect to nonceramic goods at si t e s as well. Since sample size i s small for each analysis, the conclusions made here should be regarded as hypotheses that can guide future research. In b r i e f , the model i s p a r t i a l l y supported. A pattern of d i v e r s i f i c a t i o n r e s u l t s i n some phases and regions. However, there i s no i n d i c a t i o n of increasing standardization or change i n mode of production. Ceramic production i n west-central Henan as exemplified by the s i t e of Meishan may have been impacted by a developing bronze industry. i i i ACKNOWLEDGEMENTS I am g r a t e f u l to several i n d i v i d u a l s f or t h e i r encouragement and support. F i r s t , I thank my committee for guiding me so well as I sought to c l a r i f y my ideas on pottery production during the Longshan Period. I am indebted to my advisor Professor Richard Pearson f o r his valuable advice, enthusiasm, and expertise i n employing an anthropological approach i n conducting research on Chinese N e o l i t h i c data. I was very fortunate that Professor Prudence Rice, University of F l o r i d a , kindly consented to be a long distance committee member. Despite her busy schedule, she sent thorough comments on my d r a f t s , to China as well as B r i t i s h Columbia. Her important work on ceramic production provided i n s p i r a t i o n f o r t h i s study. Professor Michael Blake provided a great deal of advice about conducting research on several aspects of complex s o c i e t i e s as well as constant encouragement. This study could not have been completed without the generosity of several scholars i n China. I am e s p e c i a l l y indebted to Professor Yan Wenming of B e i j i n g U n i v e r s i t y f o r giving me a valuable introduction to Longshan pottery and for writing l e t t e r s of introduction on my behalf to i n s t i t u t i o n s i n Henan and Shandong. I am grateful to archaeologists at the Xiaotun (Anyang) work station, Yang Xizhang i n p a r t i c u l a r , for t h e i r permission to examine a number of vessels from Hougang. The members of the Puyang C i t y C u l t u r a l Research Bureau generously provided accommodations while I examined vessels from Baiying i n the Puyang C i t y i v Museum. I thank Sun Dexuan, Zhao Liansheng, Zhang Xiangmei, and Zhao Yan. i n Luoyang, Xu Jianyuan gave me permission to see the exhibit of vessels from Meishan and other s i t e s . Other scholars in Henan that kindly gave me information on the Longshan Period include An Jinhuai (Henan Province Cultural Research I n s t i t u t e , Zhengzhou) and at the Archaeological Research Station in Gaocheng (Dengfeng), Wang Zhiguo and Liu Shumin. I thank the following people i n Shandong f o r t h e i r help: Du Zaizhong (Weifang C i t y Museum), Zheng Xiaomei and Wang Yongbo at the Shandong Province Archaeological Research I n s t i t u t e , Jinan; Cai Fengshu of Shandong University, and Zhong Huanan. In addition, two other students at B e i j i n g University at the time, Janet Lucas and Chris Feng, gave me much p r a c t i c a l advice and encouragement. I thank The Social Sciences and Humanities Research Council of Canada f o r a Doctoral Fellowship, 1986-1987. I am g r a t e f u l to several others f o r t h e i r help. Barbara M i l l s was i n Vancouver conducting research before I l e f t f o r China. She gave me a great deal of advice about studying ceramic form and function. I thank Greg Schwann, Professor David Pokotylo, and Professor R.G. Matson f o r s t a t i s t i c a l advice. Other fellow graduate students at UBC and elsewhere - Sheila Greaves, Heather Pratt, V i c k i Feddema, Liz Furniss, and Dana Lepofsky provided humor and support. I am indebted to Joyce Johnson, Susan Matson, and K i t t y Bernick f o r drawing the f i g u r e s . Two s p e c i a l friends to thank for t h e i r constant support and p r a c t i c a l help are Pat and John Berringer. F i n a l l y , I am fortunate to v have a family that supports me in a l l my endeavors. I owe special thanks to my parents for a l l of t h e i r help i n the l a s t two years. v i TABLE OF CONTENTS ABSTRACT i i ACKNOWLEDGEMENTS i v CHAPTER 1. INTRODUCTION: RESEARCH PROBLEM AND ANALYTICAL APPROACH 1 Introduction 1 Analytical Approach 4 Procedures of Analysis 13 CHAPTER 2. THE LONGSHAN PERIOD IN CHINESE PREHISTORY 28 Introduction 28 Regional Diversity During the Longshan Period 28 Approaches to Cultural Complexity During the Longshan Period 34 Archaeological Evidence for Processes of Change 36 Conclusions 51 CHAPTER 3. MODEL OF CHANGE IN SYSTEMS OF CERAMIC PRODUCTION IN RELATION TO INCREASING CULTURAL COMPLEXITY 55 Operating Premises 55 Changes i n Strategies of Production 57 Changes i n Strategies of Consumers for Prestige Vessels 64 Change i n Mode of Production 78 Summary 85 vii CHAPTER 4. ANALYSIS OF SHAPE CLASSES AND HYPOTHESES ABOUT FUNCTIONAL TYPES 87 Introduction 87 Traditional Terms for Designating Shape Classes 88 Ki-nds of Ceramic Data i n Site Reports 95 Analysis of Shape Classes i n Site Reports 101 Hypotheses About Vessel Function 126 Conclusions 135 CHAPTER 5. TEST OF THE MODEL 137 Introduction 137 Analysis of Change i n Production: Strategies of Producers 139 Analysis of Change in Labor-intensive (Prestige) Vessels 168 CHAPTER 6. CHANGE IN MODE OF PRODUCTION AND ACCESS TO GOODS 185 Introduction 185 Evidence for Mode of Production 187 Patterns of Access to Goods 209 Conclusions 220 CHAPTER 7. CONCLUSIONS 223 REFERENCES 239 viii APPENDIX A. ANALYSIS OF SHAPE CLASSES IN ARCHAEOLOGICAL REPORTS 265 APPENDIX B. DETAILS ON ANALYSES FOR TESTING THE MODEL OF CHANGE IN SYSTEMS OF CERAMIC PRODUCTION IN RELATION TO INCREASING CULTURAL COMPLEXITY 303 APPENDIX C. DATA ON NONCERAMIC POTENTIAL PRESTIGE GOODS, UTILITARIAN ARTIFACTS, AND HOUSING 344 IX LIST OF FIGURES figure page 1. Types of Longshan Culture in Northern China. 30 2. Location of Important Sites from the Longshan Period. 31 3. Model of Change in Systems of Pottery Production i n Relation to Increasing Cultural Complexity. 58 4. Major Forms of Vessels i n Longshan Sites. 118 5. Guan Jar Size Classes, Hougang. 288 6. Size of Guan Jars, Class Eight, Baiying. 289 7. Pingdipen Basin Class, Hougang. 290 8. Pingdipen Basin Size Classes, Hougang. 291 9. Pingdipen Basin Class One, Baiying. 292 10. Sizes of Pingdipen Basins, Class One, Baiying. 293 11. Wan Bowl Class One, Hougang. 294 12. Sizes of Wan Bowls Class One, Hougang. 295 13. Wan Bowl Class One, Meishan. 296 14. Sizes of Wan Bowls, Class One, Meishan. 297 15. Wan Bowl Classes One and Two, Lujiakou. 298 16. Sizes of Wan Bowls, Class One, Lujiakou. 299 17. Ding Tripod Class Seven, Lujiakou. 300 18. Bei Cup Classes One and Two, Baiying. 301 19. Sizes of Gai Lids, Classes One to Eleven, Baiying. • 302 X 20. Variation in Dimensions by Period, Large Guan Jars, Hougang. 335 21. Variation in Dimensions by Period, Medium Size Guan Jars, Hougang. 337 22. Variation i n Dimension by Period, Medium Size Pingdipen Basins, Hougang. 339 23. Variation i n Dimension by Period, Medium Size Wan Bowls, Class One, Hougang. 340 24. Variation i n Dimensions by Period, Guan Jars, Meishan. 341 25. Variation i n Dimensions by Period, Ding Tripods, Meishan. 342 26. Variation i n Dimensions by Period, Ding Tripods, Lujiakou. 343 xi LIST OF TABLES table page 1. Dating of Phases at Hougang, Baiying, Meishan, and Lujiakou. 14 2. Samples of Whole and Reconstructed Vessels at Sites. 16 3. Cultural Features at Sites. 19 4. Description of Archaeological Sites. 20 5. Context of Deposition for Vessels i n Sample for Analysis. 21 6. Large, Walled Sites from the Longshan Period. 38 7. Evidence for Bronze Metallurgy During Longshan Period. 47 8. Display Behavior with Containers in Chiefdoms and Other Ranked Societies. 69 9. Display Behavior with Containers: The Chinese H i s t o r i c a l Context. 76 10. I d e n t i f i c a t i o n of Mode of Production: Ethnographic Data on Ceramic V a r i a b i l i t y . 83 11. Published Definitions for Traditional Terms Describing Shapes of Vessels Found i n Longshan Sites. 89 12. Formats of Describing Vessels i n Site Reports. 96 13. Schematic Representation of Levels in Interpreting Samples of Pottery Described i n Chinese Neolithic Site Reports. 100 14. Ratios and Individual Measurements with Formal- Functional Significance Used in Analysis of Shape Classes. 104 XII 15. Summary of Results from Analysis of Shape Classes at Hougang, Baiying, Meishan, and Lujiakou. 110 16. Proposed Definitions for Major Shape Classes of Vessels (Xingzhuang) in Longshan Sites. 113 17. Comparison of Total Number of Shape Classes Made During Each Phase at Hougang, Baiying, Meishan, and Lujiakou. 140 18. Change Over Time in Quantity of Shape Classes by Hypothesized Functional Category at Hougang, Baiying, Meishan, and Lujiakou. 143 19. Quantities of Forms by Hypothesized Functional Category at Hougang, Baiying, and Lujiakou. 147 20. Analysis of Dimensional Standardization. 152 21. Suggested Evidence for Relative Lack of Labor Input i n Vessel Forming for Wan Bowls, Class One, Hougang. 158 22. Suggested Evidence for Relative Lack of Labor Input i n Decoration and F i r i n g for Ding Tripods, Class Seven, Lujiakou. 158 23. Summary of Change in Strategy of Pottery Production at Hougang, Baiying, Meishan, and Lujiakou. 165 24. Labor-intensive Vessels That May Have Been Used for Display at Hougang, Baiying, Meishan, and Lujiakou. 170 25. Very Large Vessels at Other Sites From the Longshan Period. 176 26. Labor-intensive Vessels That May Have Been Used For Display at Other Sites From the Longshan Period, By Cultural Region. 178 27. Hypothesized Archaeological Indicators for Change i n Mode of Production. 188 28. Direct Evidence for Pottery Production at Hougang, Baiying, and Meishan by Excavation (T) Area. 195 xiii 29. Evidence for Ceramic Production at Other Sites From the Longshan Period. 201 30. Characteristics of Kilns From Other Longshan Period Sites. 206 31. Diversity and Quantity of Nonceramic Potential Prestige Goods at Hougang, Baiying, Meishan, and Lujiakou. 211 32. Diversity and Quantity of U t i l i t a r i a n Tools at Hougang, Baiying, Meishan, and Lujiakou. 213 33. Variation i n Average Size (Floor Area) of Houses By Different Types of Construction Material at Hougang and Baiying. 215 34. Characteristics of Burials with T a l l Stemmed "Eggshell" Cups or Gao Bing Bei at Chengzi. 217 35. Characteristics of Burials with T a l l Stemmed "Eggshell" Cups or Gao Bing Bei at Sanlihe. 218 36. Changes Made i n Shape Classes of Vessels Identified in the Hougang Report. 266 37. Original Shape Classes Accepted From the Hougang Report. 26 9 38. Summary of Results from Nonparametric Significance Tests for Hougang Vessels. 271 39. Changes Made i n Shape Classes of Vessels Ide n t i f i e d in the Baiying Report. 272 40. Original Shape Classes Accepted from the Baiying Report. 275 41. Changes Made in Shape Classes of Vessels Ide n t i f i e d i n the Meishan Report. 276 42. Original Shape Classes Accepted From the Meishan Report. 277 43. Changes Made i n Shape Classes of Vessels Ide n t i f i e d i n the Lujiakou Report. 279 xiv 44. Original Shape Classes Accepted From the Lujiakou Report. 281 45. Lids at Hougang for Covering Other Vessels or For Serving Food. 282 46. Lids at Baiying for Covering Other Vessels or For Serving Food. 284 47. Lids at Meishan for Serving Food. 286 48. Lids at Lujiakou for Covering Other Vessels or For Serving Food. 287 49. Distribution of Vessels i n Shape Classes Per Phase at Hougang. 304 50. Distribution of Vessels in Shape Classes Per Phase at Baiying. 307 51. Distribution of Vessels in Shape Classes Per Phase at Meishan. 310 52. Distribution of Vessels in Shape Classes Per Phase at Lujiakou. 312 53. Diversity of Shape Classes and Hypothesized Functional Types Per Period at Hougang. 314 54. Diversity of Shape Classes and Hypothesized Functional Types Per Period at Baiying. 316 55. Diversity of Shape Classes and Hypothesized Functional Types Per Period at Meishan. 318 56. Diversity of Shape Classes and Hypothesized Functional Types Per Period at Lujiakou. 320 57. Variation i n Shape of Leg for Xian/Yan Tripods, Hougang. 322 58. Variation in Handles for Xian/Yan Tripods, Hougang. 322 59. Variation in Shape of Leg for Ding Tripods, Hougang. 323 XV 60. Variation in Decorative Techniques for Large Guan Jars, Hougang. 324 61. Variation in Decorative Techniques for Medium- Sized Guan Jars, Hougang. 325 62. Variation in Decorative Techniques for Xian/Yan Tripods, Hougang. 326 63. Variation i n Type of Rim for Wan Bowls, Class One, Hougang. 327 64. Variation i n Handles for Ding Tripods, Class One, Baiying. 328 65. Variation in Decorative Techniques for Ding Tripods, Class One, Baiying. 329 66. Variation in Shape of Leg for Ding Tripods, Meishan. 330 67. Variation in Decorative Techniques for Guan Jars, Meishan. 331 68. Variation i n Decorative Techniques for Dou Stemmed Dishes, Meishan. 331 69. Variation in Decorative Techniques for Ding Tripods, Meishan. 332 70. Variation i n Shape of Leg for Ding Tripods, Class Seven, Lujiakou. 333 71. Variation i n Handles for Ding Tripods, Class Seven, Lujiakou. 333 72. Variation i n Decorative Techniques for Ding Tripods, Class Seven, Lujiakou. 334 73. Variation i n Decorative Techniques for Wan Bowls, Class One, Lujiakou. 334 74. Dist r i b u t i o n of Potential High Status A r t i f a c t s Among Different Types of Sites, with Location of Deposition by Excavation (T) Area. 345 75. Diversity of U t i l i t a r i a n Tools for Different Types of Sites. 348 XVI 76. Variation i n Construction Material for Houses (Fangzi) at Hougang, Including I d e n t i f i c a t i o n of Houses with Costly Materials. 349 77. Variation in Construction Material for Houses (Fangzi ) at Baiying, Including I d e n t i f i c a t i o n of Houses with Costly Materials. 350 xvii CHAPTER 1. INTRODUCTION: RESEARCH PROBLEM AND ANALYTICAL APPROACH INTRODUCTION An important issue i n anthropological archaeology i s the relationship between change i n production of craft goods and increasing c u l t u r a l complexity. This dissertation examines the relationship between change i n production and inferred use of one type of craft item, pottery, and increasing c u l t u r a l complexity in chiefdoms. I t tests a model outlining potential changes in systems of ceramic production with data from Longshan Period sit e s from the lower and middle reaches of the Huanghe or Yellow River region i n northern China. At the end of the Longshan Period, at least one state evolved i n th i s region. Several writers have proposed that specialized pottery production developed during the late Neolithic period (Song, L i and Du 1983:273; Keight- ley 1987; Feng et a l . 1982:22; L i and Cheng 1984:14; An Jinhuai 1989:23). In the anthropological l i t e r a t u r e of the la s t twenty years, researchers have assumed that systems of craft production become increasingly more complex i n conjunction with other c u l t u r a l subsystems. The reasoning i s that part-time craft specialization i s prevalent i n 1 chiefdoms, and ful l - t i m e specialization i n states (Flannery 1972, Wright 1977, 1978). There are several models that describe how craft s p e c i a l i z a t i o n in general plays a role in the process of increasing c u l t u r a l complexity, s p e c i f i c a l l y with respect to the development of p o l i t i c a l central- i z a t i o n . Most of these models pertain to prestige goods rather than u t i l i t a r i a n or basic goods. They outline how increasing control over production and d i s t r i b u t i o n of prestige items enables e l i t e s to increase t h e i r p o l i t i c a l power (Brumfiel and Earle 1987). Models of th i s kind concerned with rank societies have been proposed by Frankenstein and Rowlands (1978) and Friedman and Rowlands (1977). Rice (1981) presents a model that outlines how pottery production changes in a context of increasing c u l t u r a l complexity. Her model shows how changes i n pottery production may be regarded as part of the process of increasing segregation, as defined by Flannery (1972). As she points out i n a l a t e r publication, increasing sp e c i a l i z a t i o n of pottery production i s part of the process by which social systems become increasingly d i f f e r e n t i a t e d (Rice 1984:256-7). Her model presents the hypothesis that as c u l t u r a l systems become more complex, there i s an increase i n va r i e t i e s of wares, p a r t i c u l a r l y e l i t e wares. Also, u t i l i t a r i a n wares become increasingly standardized (Rice 1981:222- 3,224). The model i s supported with data from a Maya s i t e i n Belize. The research question addressed i n th i s study i s : "How do systems of pottery production change during the Longshan Period i n r e l a t i o n to 2 increasing c u l t u r a l complexity?" The goal i s to present a revised version of the important model outlined by Rice (1981) and to test i t with data from another area of the world. The model offered here pertains to a subset of the model by Rice (1981). Ceramic change i n complex chiefdoms corresponds to the change from Step 3 i n her model, ranked societies, to Step 4, s t r a t i f i e d societies (Rice 1981:223). My model i s tested with data representing approximately 500 years of ceramic production, rather than 1000 years as i n the analysis by Rice (1981). An attempt i s made to describe in more d e t a i l how different components of ceramic production systems may change. On the basis of la t e r publications by Rice (1984, 1987), ethnographic data from several areas, and other archaeological studies, the model describes how production of prestige (labor-intensive) and non-prestige ( u t i l i t a r i a n ) vessels may change as chiefdoms evolve into states. An effo r t i s made to explain how human behavior, i . e . , strategies of producers and consumers i n chiefdoms, may cause different types of ceramic change. I attempt to investigate production and consumption as processes, by considering the goals and actions of people as causal factors affecting change in material goods (Gosden 1989). 3 ANALYTICAL APPROACH In my model the hypothesis i s made that systems of ceramic production become more complex i n conjunction with other c u l t u r a l subsystems, as i n the model by Rice (1981). Potters, especially s p e c i a l i s t s i n competition with one another, should adopt a strategy of d i v e r s i f i c a t i o n for production of one or more shape classes i n response to increasingly varied consumer demands for vessels. There should be an increase over time in vari e t i e s of ceramic categories such as decorative and shape classes. My model, l i k e that by Rice (1981), states that an increase i n status competition among consumers should create increased demand for vessels that symbolize prestige. Rice (1981) focuses on identifying change i n e l i t e wares. For reasons discussed below, I use the term "prestige vessels", or labor-intensive vessels for use i n social displays. My model proposes that two types of displays with labor- intensive pottery should be common i n prehistoric chiefdoms; in my terms, largesse and conspicuous consumption. Vessels for preparing, serving, and consuming food and alcohol could be displayed. Responding to consumer demand, potters should produce vessels exhibiting an increase i n degree of labor input, or produce a greater number of shape classes with labor-intensive techniques, and/or produce vessels with new labor-intensive techniques. 4 The model proposes two types of factors that may cause d i v e r s i f i c a t i o n i n production of non-prestige wares (not labor- intensive). They are change in diet and change in household r i t u a l s (Rice 1984). In response to consumer demand for new types of vessels, potters should produce new shapes and/or decorative techniques. Thus my model describes causal factors affecting u t i l i t a r i a n vessels that do not refer exclusively to sp e c i a l i s t s controlled by e l i t e s as i n the model by Rice (1981). Rice (1981:223) outlines how e l i t e s may force rur a l potters to produce surplus wares for tribute and trade purposes. My model also hypothesizes that potters, especially s p e c i a l i s t s i n competition with one another, should adopt a strategy of s i m p l i f i c a t i o n , or increased e f f i c i e n c y i n production, for one or more shape classes of prestige or non-prestige wares. Again, t h i s hypothesis i s made by Rice (1981). One potential causal factor for t h i s change i s increasing population size and density. After Rice (1981:223), I expect that there should be evidence for increasing standardization of wares. There may be an increase in dimensional standardization and/or within-class standardization i n terms of secondary shape features or decorative techniques. The model also predicts that there should be a change in organization of labor to produce pottery as s o c i o p o l i t i c a l complexity increases. There should be a change to a more complex mode of production, such as a household to a workshop mode. This i s another type of change that should take place as social systems become 5 i n c r e a s i n g l y d i f f e r e n t i a t e d . The model o u t l i n e d by R i c e (1981) i m p l i e s t h a t t h i s t y p e of change s h o u l d t a k e p l a c e , s i n c e i t e x p e c t s t h e development of mass p r o d u c t i o n i n s t r a t i f i e d s o c i e t i e s ( R i c e 1981:223). My model h y p o t h e s i z e s t h a t change i n mode of p r o d u c t i o n may be i d e n t i f i a b l e by change i n c e r a m i c a t t r i b u t e s ( i n c r e a s i n g s t a n d a r d i z a t i o n and i n c r e a s i n g d i v e r s i t y , a f t e r R i c e 1981, 1987, 1989), d i r e c t e v i d e n c e f o r p r o d u c t i o n ( s u c h as change from p r o d u c t i o n i n houses t o w o r k s h o p s ) , and t e c h n i q u e s of p r o d u c t i o n ( i n c r e a s i n g e f f i c i e n c y i n s h a p i n g and f i r i n g ) . The model d e s c r i b e s t h r e e d i f f e r e n t s t r a t e g i e s of p r o d u c e r s i n t o t a l : d i v e r s i f i c a t i o n , s i m p l i f i c a t i o n , and c o n s e r v a t i s m o r r e s i s t a n c e t o change. Thus i t p r o v i d e s e x p e c t a t i o n s t o t e s t t h e a l t e r n a t i v e h y p o t h e s i s t h a t some o r a l l components of c e r a m i c p r o d u c t i o n systems do not become more complex i n c o n j u n c t i o n w i t h o t h e r c u l t u r a l subsystems. L i k e R i c e (1984), I e x p e c t t h a t c e r a m i c p r o d u c t i o n systems a r e complex phenomena and t h a t d i f f e r e n t components may change i n d i f f e r e n t ways. G i v e n t h e a v a i l a b l e d a t a on c e r a m i c v a r i a b i l i t y f rom t h e Longshan P e r i o d , t h e model f o c u s e s on s o c i a l f a c t o r s t h a t can cause c e r a m i c change. I t cannot d e a l w i t h o t h e r c a u s a l f a c t o r s d e s c r i b e d by R i c e (1984) t h a t may have been i m p o r t a n t such as changes i n t e c h n o l o g y ( f o r any s t e p of p r o d u c t i o n such as f o r m i n g , d e c o r a t i n g , f i r i n g ) , changes i n s o u r c e s of raw m a t e r i a l s , o r changes i n exchange systems f o r raw m a t e r i a l s or v e s s e l s . 6 The unit of comparison for testing my model i s the shape class. I assess v a r i a b i l i t y in whole and reconstructed vessels from Longshan Period s i t e s . In contrast, the unit of analysis Rice (1981) uses i s the ware, defined from sherds by the type-variety system. Technological variation such as type of paste, temper, texture, etc. i s an important component of her model. However, very l i t t l e information on these attributes i s available for Longshan Period ceramics. My model makes a d i s t i n c t i o n between prestige and non-prestige vessels rather than e l i t e and u t i l i t a r i a n vessels for two reasons; 1) ethnographic data on display vessels used i n ranked societies, and 2) p r a c t i c a l i t y for the data set at hand. My survey of the ethnographic l i t e r a t u r e on pottery-producing societies suggests that e l i t e s , or persons occupying the uppermost positions of status such as chiefs, are not the only type of people who may use labor-intensive vessels for displays of status and prestige. For example, displays of conspicuous consumption at l i f e - c r i s i s ceremonies could be undertaken by a range of families. I expect that this situation characterizes prehistoric chiefdoms as well. Since pottery vessels were widely used, a number of households could acquire labor-intensive vessels, the quantity and degree of elaboration varying with resources of the households. Highly elaborated vessels should have been used by r e l a t i v e l y high status people, and less elaborated vessels by households lower on the social scale. Also, pottery vessels may not have been the most important part of the 7 c h i e f l y , versus domestic, economy (Goldman 1970:480-1) i n prehistoric chiefdoms. Other more re s t r i c t e d goods i n terms of raw materials and s k i l l s required for manufacture probably played a more important role. Rice (1981:223) and others define e l i t e wares on the basis of several attributes such as r e l a t i v e l y high labor input and high d i v e r s i t y i n terms of decoration, rare raw materials, and r e s t r i c t e d spatial d i s t r i b u t i o n on an i n t r a - s i t e as well as i n t e r - s i t e l e v e l . Feinman et a l . (1981) ide n t i f y vessels with r e l a t i v e l y great labor expenditure, p a r t i c u l a r l y i n decoration, by the production step index. A recently developed alternative, the production task index (Hagstrum 1988) evaluates a l l steps i n vessel formation. Analysis of paste composition allows researchers to define non-local wares that were presumably imported by e l i t e s (Rice 1977, Costin 1986, Berman 1986). Costin and Earle (1989) compare differences in labor input with context of recovery, examining wares found i n e l i t e and non-elite housing. Unfortunately, there i s l i t t l e relevant published information of t h i s nature for Longshan Period vessels with the exception of labor- intensive techniques for shaping and decorating. Focusing on iden t i f y i n g labor-intensive vessels i s worthwhile given t h e i r importance in the ethnographic l i t e r a t u r e on display a c t i v i t i e s . There are no s u f f i c i e n t independent sources of data available to confidently i d e n t i f y any vessels as used by e l i t e s . Very few vessels were found i n houses at the four sites examined in this study, as I show below. Also, there i s no information on spatial d i s t r i b u t i o n of pa r t i c u l a r wares within a 8 single settlement system. Therefore, this study emphasizes how prestige vessels could have been used rather than identifying s p e c i f i c types of consumers. Also, i t i s not always possible to iden t i f y vessels used by e l i t e s on the basis of variation i n labor input alone. I d e n t i f i c a t i o n of e l i t e vessels i s more r e l i a b l e i f differences i n labor input among vessels are extreme. In his survey of craft production i n ethnographic societies, Clark (1986:4-5) concludes that objects made for e l i t e s by attached s p e c i a l i s t s exhibit great investment i n labor, p a r t i c u l a r l y objects that are elaborated to the point where normal use i s not possible due to large size or f r a g i l i t y , etc. However, differences i n labor input among vessels may not always be d i s t i n c t . When there i s a low degree of d i f f e r e n t i a t i o n , i d e n t i f i c a t i o n of e l i t e wares on the basis of ceramic attributes alone cannot be made with confidence. There i s r e l a t i v e l y l i t t l e published information available on ceramic function during the Longshan Period, p a r t i c u l a r l y for individual assemblages. In th i s study, I in f e r rather than demonstrate that various shapes of labor-intensive vessels were used for display purposes, on the basis of ethnographic analogy. I hypothesize that labor-intensive vessels were a type of prestige good during the Longshan Period. Information on labor-intensive techniques i s limited as well. Unfortunately i t was not possible to quantify differences among vessels by means of the production step index (Feinman et a l . 1981) or the production task index (Hagstrum 1988). 9 Archaeological i d e n t i f i c a t i o n of prestige goods i n general during the Neolithic and early h i s t o r i c periods has not been the subject of extensive research. Two studies that systematically examine variation in goods (pottery, tools, ornaments) to make inferences about status d i f f e r e n t i a t i o n during the e a r l i e r Neolithic period are Pearson (1981) and Underhill (1983). In my description of the model, I summarize the l i t t l e information that i s published on use of containers (bronze, ceramic) for display purposes during the early h i s t o r i c period. No undisputed bronze vessels have been found from the Longshan Period. This i s the f i r s t published study of which I am aware that attempts to systematically examine pottery vessels and other a r t i f a c t s from Longshan Period sites as potential prestige items. However, i t has been commonly assumed that one type of ceramic ware, the "eggshell"-thin t a l l stemmed cups from sites i n Shandong, were prestige vessels. Before describing the an a l y t i c a l procedures undertaken i n th i s study, i t i s necessary to discuss d e f i n i t i o n s of terms that are used. Rice (1988, 1989) points out that the term "spe c i a l i z a t i o n " has several different connotations. This study i s concerned with s p e c i a l i z a t i o n of producers rather than specialization of production area or resources (see Rice 1989:110). My d e f i n i t i o n for specialization of producers i s derived from Costin (1986:328) and Kaiser (1984:184). I define s p e c i a l i z a t i o n as a type of organization of human labor i n which work units (individuals or groups) regularly produce a limited number of classes of goods rather than the f u l l range of goods available. These 10 work units regularly exchange t h e i r products for others that they do not produce themselves. Thus pottery vessels are commodities, or objects produced for exchange (Rice 1987:184). Among non-specialists, production takes place i n each household, and consumption takes place at the locus of production (Rice 1987:184). It i s possible to derive test implications for sp e c i a l i z a t i o n involving ceramic attributes from this d e f i n i t i o n , such as standard- i z a t i o n and d i v e r s i t y , as Rice (1981, 1987, 1989) has shown. My de f i n i t i o n does not incorporate other aspects of specialization that are especially d i f f i c u l t to operationalize such as time spent i n production ( i . e . , part- or full-time s p e c i a l i z a t i o n , see Benco 1988:68) and re l a t i v e s k i l l of individual potters (see Welch 1986:136-41). A mode of ceramic production represents a d i s t i n c t set of social relations between producers, and between producers and consumers. Modes d i f f e r in terms of scale of production, or quantities of labor and resources used, as well as quantities of vessels produced (Rice 1987:180-6). Therefore they d i f f e r in degree of i n t e n s i f i c a t i o n of production. I use the d e f i n i t i o n of i n t e n s i f i c a t i o n offered by Rice (1987:190, 1988:6): increased e f f i c i e n c y i n production for the purpose of increased y i e l d s . In r e a l i t y , organization of ceramic production i s represented by a continuous range of v a r i a b i l i t y (Rice 1987:180-6). Also, more than one mode of ceramic production may be represented i n any given society. 11 The theoretical approach of this study i s processualist. Ethnographic examples from several areas i l l u s t r a t i n g strategies of pottery production and consumption in chiefdoms are used to construct the model. It i s l i k e l y that there was substantial variation in economic, s o c i a l , and p o l i t i c a l organization among ranked societies in prehistory, given the variation noted in the ethnographic record (Feinman and Neitzel 1984, Earle 1987a, Carneiro 1979). However, l i k e Earle and Preucel (1987), I do not accept the symbolic-structuralist or structural-Marxist positions s t i p u l a t i n g that no generalizations about processes of c u l t u r a l change can be made. I t i s important to determine how changes i n craft production during the t r a n s i t i o n from chiefdom to state in China are similar to, as well as different from, other areas (see Keightley 1987:93 for a different opinion). This dissertation i s the f i r s t attempt at systematic analysis of Chinese Neolithic pottery for evidence of changes i n production and inferred use. Numerous researchers have made important contributions to other aspects of Chinese Neolithic pottery. Throughout t h i s study I refer to several a r t i c l e s written by Chinese archaeologists on technology. Valuable descriptions of technology have also been published i n English by Chinese and western researchers: Wu (1938), L i Chi (1956), Chang et a l . (1969), Medley (1976), L i and Cheng (1984), Chang (1986), and Vandiver (1988). Other topics of inquiry i n the English l i t e r a t u r e include ide n t i f y i n g r i t u a l sets of pottery i n cemeteries (Pearson 1988), making interpretations about mentality 12 (Keightley 1987), identifying post-marital residence patterns through s t y l i s t i c analysis (Li Kuang-chou 1981), and tracing the development of regional systems of technology (Huber 1983). PROCEDURES OF ANALYSIS The model of change i n systems of pottery production i n a context of increasing cu l t u r a l complexity i s tested on data derived from two sources: 1) visual observations of whole and reconstructed vessels i n archaeological work stations and museums in Henan and Shandong provinces in 1987 during a period of fieldwork six months in duration, and 2) descriptions of vessels i n Chinese archaeological reports. In addition, Professor Yan Wenming of Beijing University provided invaluable information and advice about studying Longshan ceramic production. Eight sites were o r i g i n a l l y selected for analysis on the basis of the following c r i t e r i a : r e l a t i v e l y detailed s i t e report, presence of more than one phase with adequate dating, and at least some vessels accessible for examination. It i s possible to use data from four of these site s i n the test of the model; three located in Henan and one i n Shandong (Table 1). They are: Hougang (northern Henan), Baiying (northern Henan), Meishan (central Henan), and Lujiakou (north-central Shandong). 13 Table 1. Dating of. Phases at Hougang, Baiying, Meishan, and Lujiakou. Hougang II Type: Hougang (near Anyang c i t y , Henan) 3 periods defined on the basis of several radiocarbon dates, stratigraphy, and ceramic seriation: Early ( t r a n s i t i o n a l to the Longshan Period), ca. 2700-2500 B.C.; Middle, ca. 2500-2300 B.C.; Late, ca. 2300- 2100 B.C. (Anyang Archaeological Team, IA, CASS 1985:82, Zhang and Zhang 1986:52) Baiying (Tangyin County, Henan) 3 periods defined on the basis of several radiocarbon dates, stratigraphy, and ceramic seriation: Early, ca. 2500-2300 B.C. (my estimate); Middle, ca. 2300-2200 B.C. (my estimate); Late, ca. 2200- 2000 B.C. (my estimate); (CPAM of Anyang D i s t r i c t , Henan Province 1983:40, Zhang and Zhang 1986:52,54); the Early Period i s roughly contemporary with the Middle Period at Hougang, and the Middle Period with the Late at Hougang (personal communication, Zhao Liansheng, 1987) Wangwan I I I Type: Meishan (Linru County, Henan) 2 periods defined on the basis of stratigraphy, ceramic seriation, and two radiocarbon dates: ZK 386, 2290+/160 B.C., ZK 349, 2005+/120 B.C., Early, ca. 2300-2100 B.C. (my estimate); Late, ca. 2100-1900 B.C. (my estimate) (Second Henan Archaeological Team, IA, CASS 1982:472, Zhang and Zhang 1986:53) (the Early Period corresponds to the late Henan Longshan Period, and the Late Period i s apparently t r a n s i t i o n a l to the E r l i t o u I Period, according to the Second Henan Archaeological Team, IA, CASS 1982:472) Liangcheng Type: Lujiakou (Weixian County, Shandong) 2 periods defined on the basis of stratigraphy, ceramic seriation, and two radiocarbon dates: ZK 317, 2340+/145 B.C., ZK 321, 2035+/115 B.C., Early, ca. 2350-2150 B.C. (my estimate); Late, ca. 2150-1950 B.C. (my estimate) (Shandong Archaeological Team, IA, CASS and the Art Museum of Weifang County, Shandong Province 1985:348) 14 At present, data are not a v a i l a b l e to investigate changes i n ceramic production within an i n d i v i d u a l settlement system. Patterns that r e s u l t from the analyses should be regarded as characterizing regions, rather than single s i t e s , since the extent to which vessels were exchanged between communities i s not known. Some of the shape classes at these s i t e s may have been produced at other locations and imported. The samples of vessels examined at these s i t e s are described in Table 2. The table indicates some problems that were encountered in a n a l y s i s . F i r s t , reports only describe a subset of vessels recovered from excavation. The quantity of vessels i n my sample i s not a random sample of the excavated vessels. Second, a r e l a t i v e l y low number of whole and reconstructed vessels was d i r e c t l y examined ( i . e . , outside of glass display cases) from each s i t e . I was not able to examine any vessels from Meishan i n t h i s manner. Third, sample sizes f o r i n d i v i d u a l analyses are very small. The s i t e of Hougang provides the best test of the model. I was given permission to d i r e c t l y examine a r e l a t i v e l y large number of vessels, and the report describes a r e l a t i v e l y high percentage of excavated vessels. For more than one analysis, i t i s possible to c i t e published data from other Longshan Period s i t e s to provide support for my i n t e r p r e t a t i o n s . The problem of missing information i n reports i s a l l e v i a t e d somewhat by the fact that reports tend to describe the f u l l range of ceramic v a r i a b i l i t y regarding shape, decorative technique, and type of 15 Table 2. Samples of Whole and Reconstructed Vessels at Sites. (note: t o t a l number of pots i n my sample means pots described i n reports and/or d i r e c t l y examined) s i t e Hougang (Henan) Baiying (Henan) Meishan (Henan) t o t a l # pots in si t e 454 188 no data, probably similar to Lu-. jiakou Lujiakou 134 (Shandong) # pots in my sample 221 114 103 103 # pots d i r e c t l y examined 125 73 0, 12 on display 60 location of pots examined storeroom, Archaeolog- i c a l Work Station, Xiaotun display room, Archaeog- i c a l Work Station, Puyang display room, pots in glass cases, Archaeolog- i c a l Work Station, Luoyang storeroom, Archaeolog- i c a l Work Station, Hanting 16 paste. Authors make an ef f o r t to describe d i v e r s i t y of techniques present at s i t e s , rather than to describe a large number of vessels of the same type. Emphasis i s placed on describing pots with unusual characteristics as well as pots that are r e l a t i v e l y well made. It i s not surprising that museum personnel choose vessels for display with these characteristics. The vessels that I examined d i r e c t l y from Baiying and the vessels I saw from Meishan f i t t h i s description. I examined a different set of vessels from Lujiakou and Hougang. Apparently, the "nicer looking" vessels recovered from Lujiakou were moved to another c i t y for storage. Since I examined a r e l a t i v e l y large number of vessels from Hougang kept i n a storage room, the sample that I saw should be an adequate representation of the kinds of vessels recovered i n excavation. The vessels that I examined d i r e c t l y from Hougang, Baiying, and Lujiakou do not constitute random samples of major shape classes (xingzhuang) i d e n t i f i e d i n the reports. In each case, permission to examine vessels was given under the condition that I complete my work within a short time period. Since the samples of vessels from Baiying and Lujiakou available for examination were not large, I examined every vessel present. At the Xiaotun Research Station, I concentrated on examining shape classes with r e l a t i v e l y large quantities of vessels. However, I was able to b r i e f l y examine every reconstructed vessel from Hougang i n the storeroom. 17 Another problem that affects the analyses i s a lack of information in reports on strategies of excavation. It i s not possible to ascertain how well samples of vessels represent the t o t a l archaeological context. The quality of interpretations one makes about ceramic v a r i a b i l i t y i s d i r e c t l y related to the quality of sampling methods used in excavation (Rice 1987:289-90). It appears that a judgmental sampling method was used to select areas for excavation at each s i t e . A number of c u l t u r a l features, especially houses and storage p i t s , were found at these sites (Table 3). It appears that the r e l a t i v e l y small number of vessels recovered at Lujiakou and Meishan i s largely a factor of the small areas excavated (Table 4). The small number of vessels recovered at Baiying in comparison to Hougang i s surprising, given the r e l a t i v e l y large area excavated and the quantity of c u l t u r a l features discovered. One reason for t h i s figure may be that Baiying contained a proportionally large number of vessels broken into small sherds. Therefore, r e l a t i v e l y few vessels were available for reconstruction. Or, reconstruction was given a lower p r i o r i t y at the Puyang research station than at the Xiaotun research station. The majority of vessels from these site s were found i n open test areas and storage p i t s (Table 5). Unfortunately, only a few vessels were recovered from houses, a context of deposition that can provide direct information about consumers. Most l i k e l y , emphasis was placed on 18 Table 3. C u l t u r a l Features at S i t e s . site houses pits burials other Hougang ** Early 2 12 1 Middle 14 35 17 Late 23 11 11 (to t a l ) (39)*** (58) (29) Baiying Early Middle Late ( t o t a l ) 9 8 46 (63) 16 21 50 (87) 0 3 9 (12) K : l , W:l (K:2 no date stated) Meishan Early Late ( t o t a l ) 17 16 (33) 2 42 (44) 12 3 •(15) K:5, W:2 0 (2,4) Lujiakou Early 6 19 Late 5 10 (to t a l ) (11) (29) ** b u r i a l s at s i t e s are jars containing the remains of chi l d r e n ; Meishan has a few b u r i a l s of t h i s type f o r adults and 3 p i t graves for adults with no grave goods *** one house at Hougang i s occupied during two phases K=kiln, W=well 1 9 Table 4. Description of Archaeological Sites. s i t e s i t e size portion dug Longshan phases h i s t o r i c phases date of excavation Hougang 100,000 m2 1809 m2 (1.81%) Early (layers 7,6) Middle (layer 5) Late (layers 4,3) Shang (thin layer) 1931-4, 1979 Baiying 33,600 m2 1830 m2 (5.45%) Early (layer 6) Middle (layers 5,4,3) Late (layer 2) Western Zhou (thin layer) 1976- 1978 Meishan not stated 547 m2 Early, E r l i t o u Late Periods (layers I - I I I , not 1 Qing c l e a r l y Dynasty stated) and 7 modern burials 1975 Lujiakou 40,000 m2 364 m2 (0.91%) Early (layers 5,4) Late (layers 3,2) Western Zhou (a few p i t s , b u r i a l s ) , Shang, Yueshi (a few pi t s ) 1973, 1974 20 Table 5. Context of Deposition at Sites for Vessels i n Sample for Analysis. context Hougang Baiying Meishan Lujiakou house 11 pots 18 pots 3 pots 2 pots (5.0%) (15.8%) (2.9%) (1.9%) p i t 82 41 31 35 (37.1%) (36.0%) (30.1%) (34.0%) test area 111 47 64 66 (50.2%) (41.2%) (62.1%) (64.1%) k i l n 0 4 0 0 (3.5%) well 0 1 0 0 (0.9%) bu r i a l 17 0 2 0 (7.7%) (1.9%) unknown 0 3 3 0 (2.6%) (2.9%) TOTAL NUMBER OF VESSELS 221 114 103 103 21 recovering and reporting reconstructable vessels from houses rather than sherds. The f i r s t step i n testing the model with these data i s description of archaeological evidence for chiefdoms and increasing c u l t u r a l complexity during the Longshan Period (Chapter 2 ) . At present there i s r e l a t i v e l y l i t t l e detailed published information on s i t e s . Also, interpreting data on a regional basis has not been a focus of research. The chapter discusses evidence for regional d i v e r s i t y i n the Huanghe (Yellow River) valley area and the available information on c u l t u r a l change i n different regions. Hougang and Baiying are located i n the Hougang II region, Meishan in the Wangwan I I I region, and Lujiakou i n the Liangcheng region. Hougang was probably a center of settlement, given i t s large size (Table 4) and evidence of a surrounding wall. At present i t i s necessary to assume that the Longshan Period i s characterized by increasing c u l t u r a l complexity. I t i s not possible to demonstrate, for example, an increase i n complexity of settlement hierarchies over time on the basis of published data. The model of change in ceramic production i n r e l a t i o n to increasing c u l t u r a l complexity i s presented i n Chapter 3. Chapter 4 describes i n more d e t a i l the kinds of information on ceramic v a r i a b i l i t y included i n Neolithic s i t e reports. It also evaluates the shape classes i d e n t i f i e d i n the reports for Hougang, Baiying, Meishan, and Lujiakou, a step essential to testing the model. I t assesses the procedures by which authors established shape classes. Variation among vessels in 22 q u a l i t a t i v e features and in values of ratios for major dimensions are examined. The chapter also includes a brief evaluation of hypotheses about vessel function during the Longshan Period. Although information on function i s extremely limi t e d , i t i s helpful in making inferences about changes i n production of different shape classes. Chapter 5 presents the test of the model and interpretation of resu l t s . Four analyses are undertaken: 1) variety of shape classes in each period, 2) dimensional standardization, 3) within-class standardization, and 4) i d e n t i f i c a t i o n of labor-intensive wares. In the f i r s t analysis, I simply count the quantity of shape and size classes at each s i t e , using classes established i n Chapter 4. The analysis of dimensional standardization assesses change over time i n the range of variation i n major dimensions for different shape classes. In the t h i r d analysis, I tabulate v a r i e t i e s of vessels i n terms of secondary shape features and decorative techniques for individual shape classes. The fourth analysis i d e n t i f i e s labor-intensive vessels, including those that are large i n size, elaborate i n shape, thin-walled, and with a number of decorative techniques. Two analyses are conducted i n Chapter 6. The f i r s t assesses whether there i s evidence for change i n mode of ceramic production during the Longshan Period. It examines ceramic v a r i a b i l i t y , direct evidence for production at sites such as k i l n s and tools, and the available information on shaping and f i r i n g techniques. In the second analysis, comparisons are made on an in t e r - and i n t r a - s i t e basis about 23 quantity and d i v e r s i t y of potential prestige goods and u t i l i t a r i a n items, both ceramic and nonceramic. Hougang i s compared to the other three sites since i t probably was a center of a settlement hierarchy. Inferences are made about changes over time i n access to goods. There is s u f f i c i e n t information to make inferences about changes i n social d i f f e r e n t i a t i o n on the basis of variation i n size and construction material for housing at Hougang and Baiying. In b r i e f , the model of change in systems of ceramic production i n a context of increasing c u l t u r a l complexity i s p a r t i a l l y supported. There i s a pattern of d i v e r s i f i c a t i o n for the two sites of Baiying (Early to Middle Period) and Lujiakou (Early to Late Period) with respect to variety of shape classes produced. The pattern of change i n within-class standardization for Lujiakou i s d i v e r s i f i c a t i o n as well. It appears that production of labor-intensive wares d i v e r s i f i e d over time at Hougang (Middle to Late Period), Baiying (Early to Middle Period), and at Lujiakou (Early to Late Period). There i s no evidence for any d i v e r s i f i c a t i o n of production at the westernmost s i t e , Meishan. The results indicate that systems of ceramic production did not change i n a homogeneous manner throughout the Huanghe (Yellow River) valley region. Also, as expected from Rice (1984), there i s variation within individual production systems as represented by single s i t e s . For example, although there appears to be a pattern of d i v e r s i f i c a t i o n for labor-intensive wares at Hougang from the Middle to Late Period, the pattern for dimensional standardization i s conservatism, and for within- 24 class standardization, s i m p l i f i c a t i o n . S i m i l a r l y , two analyses for the Early to Middle Period at Baiying indicate d i v e r s i f i c a t i o n (variety of shape classes and labor-intensive wares), and the analysis of within- class standardization indicates conservatism. There are two patterns for the Middle to Late Period at Baiying, conservatism for variety of shape classes and s i m p l i f i c a t i o n for within-class standardization. For Hougang, Meishan, and Lujiakou, the pattern that results from the analysis of dimensional standardization i s conservatism. I t was not possible to conduct this analysis for Baiying. A variety of patterns resulted from the analysis of within-class standardization for each s i t e . There i s no evidence for change in mode of ceramic production over time at any s i t e . The results p a r t i a l l y support the hypothesis of Rice (1981) that there should be an increase i n v a r i e t i e s of wares produced over time as s o c i o p o l i t i c a l systems become more complex. I hypothesize that d i v e r s i f i c a t i o n of production at Hougang, Baiying, and Lujiakou i s primarily i n the realm of prestige wares, a pattern predicted by Rice (1981). However, there i s no clear evidence for increasing standardization of non-prestige or u t i l i t a r i a n wares as her model predicts, from the analysis of dimensional standardization or within- class standardization. Also, the prediction about change i n mode of production i s not supported. The test should not be regarded as conclusive, since sample sizes are small for each analysis. In the future larger classes of vessels, 25 p a r t i c u l a r l y from Meishan and Lujiakou, should be examined to determine whether there i s support for the conclusions made here. Results for the assessment of change i n va r i e t y of shape classes are the most secure. Patterns of change regarding within-class standardization are not e n t i r e l y c l e a r due to small sample s i z e . The analysis of dimensional standardization includes a few shape classes that probably constituted vessels used for basic needs such as cooking and water storage. Sample siz e did not permit analysis of other functional types of vessels such as labor-intensive wares which could have yielded d i f f e r e n t r e s u l t s . Information on v a r i a t i o n i n assemblages with respect to labor-intensive techniques i s e s p e c i a l l y l i m i t e d . It i s only possible to i d e n t i f y some i n d i v i d u a l vessels that may have been used f o r s o c i a l displays. A more thorough assessment of v a r i a t i o n i n labor input within i n d i v i d u a l assemblages should be a p r i o r i t y i n future research. F i n a l l y , at present few r e l i a b l e t e s t implications are avai l a b l e f o r i d e n t i f y i n g change i n mode of ceramic production. Despite these problems, the a n a l y t i c a l r e s u l t s provide new hypotheses that can be used to guide future research and new a n a l y t i c a l approaches that can be useful i n understanding c u l t u r a l change during the Longshan Period. Also, they demonstrate the importance of t e s t i n g the hypotheses made by Rice (1981) with ceramic assemblages from d i f f e r e n t areas. Chapter 7 discusses p o t e n t i a l explanations f o r the differences i n r e s u l t s achieved i n t h i s study compared to the r e s u l t s described by Rice (1981) f o r B e l i z e . 26 One possible explanation i s scale of analysis, both temporal and s p a t i a l . The analysis by Rice (1981) indicates a f a i r l y homogeneous pattern of long-term changes i n systems of ceramic production for a single s i t e . This study examines changes that represent a shorter time span i n four assemblages representing a larger region. A variety of patterns resulted. However, considering the Longshan Period as a whole, one could conclude that the model formulated by Rice (1981) and the revised version offered here i s supported. Differences in c u l t u r a l h i s t o r i c a l processes between the Mayan and Longshan areas may be important as well. I suggest that the development of bronze production during the late Longshan Period had an important impact on ceramic production i n central Henan as represented at the Meishan s i t e . 27 CHAPTER 2. THE LONGSHAN PERIOD IN CHINESE PREHISTORY INTRODUCTION This chapter describes archaeological evidence for c u l t u r a l complexity during the Longshan Period. It i s l i k e l y that complex chiefdoms were present in a number of areas. Although published data on processes of cultu r a l change are limited, they suggest that chiefdoms began to evolve into states i n at least one area. The f i r s t section of the chapter describes regional d i v e r s i t y during the Longshan Period. The second section outlines Chinese and western views on cul t u r a l evolution during the Longshan Period. The t h i r d section describes archaeological evidence for processes of change. REGIONAL DIVERSITY DURING THE LONGSHAN PERIOD There i s substantial regional d i v e r s i t y represented by archaeo- l o g i c a l remains from the terminal Neolithic period i n northern China. It i s more appropriate to refer to a Longshan Period rather than a single Longshan Culture (Yan 1981). Hundreds of sites have been discovered i n regions along the lower reaches of the Huanghe or Yellow River including the modern provinces of Shaanxi, Shanxi, Hebei, Henan, 28 and Shandong. The range of dates for these sites i s approximately 2500- 2000 B.C. (Yan 1981). On the basis of s i m i l a r i t i e s and differences i n remains from t h i s wide area, archaeologists have i d e n t i f i e d seven "types" (leixing) of Longshan culture (Yan 1981, 1986, 1987a). From west to east, these types are: 1) Kexingzhuang II (southern Shaanxi), 2) Taosi (southern Shanxi), 3) Wangwan I I I (west-central Henan), 4) Hougang II (northern Henan, southern Hebei), 5) Wangyoufang (eastern Henan), 6) Chengziya (west-central Shandong), and 7) Liangcheng (eastern and south-central Shandong) (Figure 1). 1 The locations of important sites discussed in this study are depicted i n Figure 2. There i s debate as to whether additional types should be id e n t i f i e d i n southern and western Henan (Institute of Archaeology, CASS 1984:78) as well as southern Shandong (Cai Feng, personal communication 1987). Also, there i s debate over c u l t u r a l c l a s s i f i c a t i o n of the Miaodigou II Culture in western Henan, dated to ca. 2780 B.C. Some publications state that Miaodigou II corresponds to the early Longshan Period (Institute of Archaeology, CASS 1983:72, 1984:69). Following Yan (1981; personal communication, 1987b), I regard t h i s culture as tr a n s i t i o n a l to the Longshan Period. More than one physiographic zone i s represented by the seven branches of Longshan culture. Sites from the Kexingzhuang I I , Taosi, and Wangwan I I I types are located at the western part of the North China Central P l a i n . This area i s characterized by loess s o i l (see Tregear 29 —7 T Y P E S O F L O N G S H A N C U L T U R E 1 Kexingzhuang II 2 Taos i 3 Wangwan III 4 Hougang II 5 Wangyoufarvg 6 Chengz iya 7 L iangcheng KB/89 Figure I. Types of Longshan Culture in Northern China. 30 Figure 2 . Locations of important s i tes from the Longshan Per iod. 31 1970:32, Map 7). Sites from the Hougang I I , Chengziya, Wangyoufang, and Liangcheng types are located in areas with a l l u v i a l s o i l ( i b i d ) . Sites from the Hougang I I , Chengziya, and Wangyoufang types are situated i n the eastern part of the Central P l a i n . The Liangcheng type incorporates more than one physiographic region: floodplain, the central Shandong mountains, and mesa-like landforms i n southern Shandong (see Ren et a l . 1985:214). In the central Shandong mountains, elevations exceed 1000m above sea l e v e l ( i b i d ) . The North China Plain i s characterized by a temperate semi-humid climate and deciduous broadleaf forests (Ren et a l . 1985:208). The four westernmost cu l t u r a l types are p a r t i c u l a r l y important i n Chinese prehistory. Communities i n these regions were succeeded by state-level societies. The Wangwan I I I and Taosi types of Longshan culture were succeeded by the f i r s t dynasty in China, the Xia. The Hougang II type i s ancestral to the Shang dynasty, and the Kexing- zhuang II type i s ancestral to the Western Zhou dynasty (Yan 1986, 1987a). The Shang dynasty began ca. 1700 B.C., and the Western Zhou dynasty, ca. 1100 B.C. (Chang 1986:296-7). No written records from the Xia dynasty have been discovered, but textual data from l a t e r periods suggest that the Xia dynasty existed i n western Henan and southern Shanxi beginning i n ca. 2000 B.C. (Chang 1986:307). The Xia dynasty i s probably represented at least i n part by remains from the E r l i t o u Culture (Chang 1986:316, 1983a). Calibrated 32 radiocarbon dates indicate that Period I of the E r l i t o u Culture began ca. 1900 B.C. (Institute of Archaeology, CASS 1983:73), a date that roughly matches the period suggested by h i s t o r i c a l texts. 2 The archaeological remains from E r l i t o u Culture sites represent the f i r s t state-level society in China (personal communication, Yan 1987b, Zou 1987). Another view i s that the state level of organi- zational complexity evolved e a r l i e r , during the l a t e r Longshan Period i n western Henan and southern Shanxi, ca. 2300 B.C. (Zhang and Zhang 1986:55, An Jinhuai 1983a, 1987, 1989). The argument i s that the archaeological remains in question could only have been produced by a state-level society. Sim i l a r l y , Tian (1986) compares archaeological data to legends described i n h i s t o r i c texts and maintains that states developed i n the Wangwan, Taosi, and Wangyoufang regions during the Longshan Period. Archaeological data from the Longshan Period are compatible with those from complex chiefdoms (Wright 1984:42-3, Johnson and Earle 1987:211) i n other areas of the world. It i s l i k e l y that complex p o l i t i e s developed in more than one region, not just western Henan and southern Shanxi. After approximately 2000 B.C., states may have eventually developed in the Kexingzhuang II (proto-Zhou) and Hougang II (proto-Shang) regions that began to compete for power with the Xia state (see Chang 1983a, 1986:361). 33 APPROACHES TO CULTURAL CHANGE DURING THE LONGSHAN PERIOD In China as i n western countries, there i s debate over suitable theoretical approaches for investigating development of cu l t u r a l complexity. Some scholars such as An Jinhuai (1989) employ Marxist theory to explain how and why culture changed over time during the Longshan Period. They describe how primitive societies organized as p a t r i l i n e a l clans evolved into states and c i v i l i z a t i o n s with the i n s t i t u t i o n of slavery. Other scholars such as Tong (1989) consider how theories developed i n the West compare to ideas advanced by Chinese writers and how they may illuminate processes of change i n Neolithic societies. Yan (1986, 1987a) describes archaeological evidence for potential causal factors in state formation such as warfare and social s t r a t i f i c a t i o n . Archaeologists in the West have been concerned with comparing cu l t u r a l evolution in northern China to other areas. Chang (1986:243- 4,286-7; 1983a) considers evidence for processes of change that have been considered important i n the western archaeological l i t e r a t u r e such as development of increasingly wide spheres of interaction. In another work, Chang (1983b) uses textual data to describe s p e c i f i c features that characterize the early dynastic periods i n China and argues that these features developed from late Neolithic cultures. He maintains that certain kin groups amassed p o l i t i c a l power primarily by cont r o l l i n g access to production and use of wealth items with r i t u a l significance: 34 bronze vessels and other craft items with animal designs, etc., that were necessary for communicating with the gods and ancestors. Some generalizations about state formation in northern China have been made on the basis of a limited review of the English-language archaeological l i t e r a t u r e . Haas (1982, 1986) examines data from the Shang period and in contrast to Chang (1983b), argues that accumulation of economic and physical power was more important i n state formation than ideological power. He compares processes of change i n China with Mesoamerica. Employing a structural-Marxist approach to explaining c u l t u r a l evolution in northern China, Friedman and Rowlands (1977), and more recently, Maisels (1987) describe how the "A s i a t i c state" (said to be represented by the Shang and Zhou dynasties) developed from t r i b a l systems. They also argue that lineage organization and development of increasingly wide spheres of interaction played a key role i n c u l t u r a l change. Maisels (1987) compares state formation i n northern China to that i n Mesopotamia. It i s not possible at the present time to make broad conclusions about processes of state formation in northern China. Relatively l i t t l e research on state formation has been conducted (An Zhimin 1988a:759, Chang 1983b:129). Also, i t i s presently impossible to describe c u l t u r a l change over time in a detailed manner on the basis of published archaeological data. It i s p a r t i c u l a r l y d i f f i c u l t to reach conclusions on c u l t u r a l change for separate regions. Archaeological research i n 35 China has focused on the s i t e rather than the region as the cu l t u r a l unit of analysis. At t h i s stage, archaeological data from the Longshan Period should be examined on a r e l a t i v e l y fine scale. I t i s necessary to document accurately culture change at a si t e and regional l e v e l . This procedure requires careful examination of chronological data. Generalizations about processes of change should be demonstrable by archaeological remains. In the following section, I outline the limited evidence at hand for processes that have been regarded as important i n the evolution of cultur a l complexity in other areas of the world: changes i n settlement pattern, warfare, and social s t r a t i f i c a t i o n i n terms of d i f f e r e n t i a l access to goods. I attempt to draw conclusions about cu l t u r a l change i n northern China on a site and regional basis (see also Pearson and Underhill 1987; Underhill i n press a,b). ARCHAEOLOGICAL EVIDENCE FOR PROCESSES OF CHANGE Settlement Hierarchies Detailed data on settlement patterns for individual regions have not been published; however, i t i s possible to describe evidence for the presence of settlement hierarchies on the basis of s i t e size and presence of sig n i f i c a n t architectural features. Relatively large sit e s 36 with surrounding walls made of rammed earth (hangtu) have been found in four of the seven Longshan cult u r a l regions: Wangwan I I I (the Haojiatai and Wangchenggang s i t e s ) , Hougang II (Hougang), Wangyoufang (Pingliang- t a i ) , and Liangcheng (Bianxianwang) (Table 6). 3 As survey, excavation, and publication proceed, the p o l i t i c a l , s o c i a l , economic, and r i t u a l functions of these sites w i l l become clearer. At present, four of the f i v e sites do not have complete excavation reports (Hougang i s the exception). Two of the sites were only recently discovered, Haojiatai and Bianxianwang. Three sites are similar in size with respect to t o t a l surface area: Pingliangtai, Haojiatai, and Bianxianwang. Hougang and Wangchenggang represent opposite ends of the spectrum, at 10 and 1 hectares, respectively. However, i t may be necessary to revise these figures after more excavation and publication are completed. For example, i n a recent a r t i c l e Sui (1988:47) maintains that Bianxianwang i s 100,000 m2 or 10 hectares i n size. These large, walled sites were probably either primary or secondary centers. Large sites with architectural features that require r e l a t i v e l y great amounts of labor input tend to be centers i n settlement hierarchies (Earle 1987a; Carneiro 1979, Peebles and Kus 1977, Stepon- a i t i s 1981). One well-known example of a walled settlement from a chiefdom i s the Moundville s i t e (Peebles and Kus 1977:444). Also, sites from the Deh Luran Plain i n Iran of a similar size range to those i n northern China are i d e n t i f i e d as centers. Wright (1984:63) maintains 37 Table 6. Large, Walled Sites from the Longshan Period. Wangwan I I I Type: Wangchenggang (Dengfeng County, Henan) surface area ca. 10,000 m2 (1 hectare) wall b u i l t during period I I , 2455+/145 B.C., ZK 581 (Henan Province Cultural Research Institute and the Archaeological Section of the Museum of Chinese History 1983:8,11- 13,16; Zhang and Zhang 1986:53) (5 periods defined at s i t e on the basis of three radiocarbon dates, stratigraphy, and ceramic s e r i a t i o n ; ca. 2500-1900 B.C.) Haojiatai (Yancheng County, Henan) surface area ca. 60,000 m2 (6 hectares), wall b u i l t ca. 2500 B.C. (Renmin Ribao 1986) Hougang II Type: Hougang (near Anyang c i t y , Henan) surface area ca. 100,000 m2 (10 hectares), wall b u i l t at end of Middle period according to Sui (1988:47), date for p i t H2 i s ZK 133, 2340+/140 B.C. (Zhang and Zhang 1986:52, Anyang Archaeological Team, IA, CASS 1985:33,82) (3 periods defined at s i t e on basis of several radiocarbon dates, stratigraphy, ceramic seriation; occupation ca. 2700- 2100 B.C.) Wangyoufang Type: Pingliangtai (Huaiyang County, Henan) surface area ca. 50,000 m2 (5 hectares) wall b u i l t before period I I I , probably during period I I , date for period I I I i s WB 81-2, 2405+/175 B.C. (Henan Province Cultural Research Institute and the Cultural Objects Division of the Zhoukou D i s t r i c t 1983:21,36; Zhang and Zhang 1986:53) (5 periods defined at s i t e from two radiocarbon dates, stratigraphy, ceramic seriation; occupation ca. 2500-2000 B.C.) Liangcheng Type: Bianxianwang (Shouguang County, Shandong) surface area ca. 44,000 m2 (4.4 hectares) wall b u i l t ca. 2050 B.C. (Renmin Ribao 1985) 38 that settlements ca. 10 hectares in size are major centers and that sites approximately 3 hectares in size are subsidiary centers. At least some high status individuals probably occupied the large walled communities, whether status i s defined on the basis of power, wealth, occupation, or other attributes. This expectation i s p a r t i c u l a r l y relevant to northern China, because textual data suggest that high ranking kin groups l i v e d i n walled communities during the early dynastic period (Chang 1986, 1983a). Also, there i s other archaeological evidence for the presence of high status individuals in these settlements such as d i f f e r e n t i a t i o n i n housing and consumption of prestige goods, as described shortly. The three walled sites from the Wangwan I I I (west-central Henan) and Wangyoufang (eastern Henan) types of Longshan culture were b u i l t at approximately the same time, i . e . , during the early Longshan Period, ca. 2500 B.C. 4 The wall at Wangchenggang was b u i l t during period II (Henan Province Cultural Research In s t i t u t e and the Archaeological Section of the Museum of Chinese History 1983:8,11,12), ca. 2455 B.C. (ZK 581, Zhang and Zhang 1986:53). The Haojiatai s i t e dates to ca. 2500 B.C. (Renmin Ribao 1986). The wall at Pingliangtai was b u i l t sometime before period I I I , or before ca. 2405 B.C. (Zhang and Zhang 1986:53, Henan Province Cultural Research Institute and the Cultural Objects Division of the Zhoukou D i s t r i c t 1983:21,36). The wall at Hougang was b u i l t s l i g h t l y l a t e r in time, approxi- mately 2300 B.C., according to Sui (1988:47). The easternmost walled 39 s i t e was established considerably l a t e r i n time than the sites i n Henan, ca. 2050 B.C. (Renmin Ribao 1985). Competition between social groups must have i n t e n s i f i e d at ah e a r l i e r period i n Henan than in Shandong. Furthermore, competition i n t e n s i f i e d f i r s t i n the Wangwan I I I and Wangyoufang regions, necessitating establishment of nucleated, defendable communities. Warfare Defense was probably one reason that the walled settlements were b u i l t . There are several lines of archaeological evidence for warfare during the Longshan Period: structural features at settlements i n addition to walls, skeletal remains indicative of violent death, and weapons. At present there i s no information on causal factors or the type of fi g h t i n g taking place. Warfare must have been an important process in state formation i n northern China (Yan 1986). Increasing d i f f e r e n t i a t i o n i n wealth may have caused an increase i n the frequency of warfare to acquire booty ( i b i d ) . Warfare may have erupted over competition for land after the death of a chief, a situation that occurred during the prehistoric period i n Hawaii (Johnson and Earle 1987:232). Of course, internal rather than inter-group c o n f l i c t may have been a factor as well. The s i t e of Pingliangtai has a number of additional structures that were probably b u i l t for defensive purposes: two small buildings 40 that appear to be guardhouses, two gates, and a wide ditch (Henan Province Cultural Research Institute and the Cultural Objects Division of the Zhoukou D i s t r i c t 1983, Chang 1986:265). According to Tian (1986), the ditch i s 30 meters wide and was used as a moat. The newspaper report for Haojiatai mentions the presence of a di t c h , too (Renmin Ribao 1986 ). Another function of the surrounding walls may have- been to protect habitation areas from floodwater (Du, personal communication 1987). The s i t e with the most convincing evidence for skeletal remains indicating violent death i s Jiangou (Handan County, southern Hebei) in the Hougang II region. The remains at Jiangou indicate a degree of violence not present during the pre-Longshan period (Underhill 1989). F i r s t , several skeletons representing both sexes and a range of ages were found i n a well. Some individuals were decapitated, and a few people may have been buried a l i v e (Chang 1986:270). Second, there are six skeletons from'Jiangou that have marks from hacking or scalping with a stone tool (Yan 1982a:38-9). Two are female youths, two are male adults, and two are unidentifiable. Only the two skulls i d e n t i f i e d as female show traces of scalping. Yan (1982a:39) suggests that the skulls of the dead were cut open and used by the victor s as drinking cups. He states that the skeletal remains from Jiangou date to ca. 2300 B.C. (Yan 1982a:39). It may not be coincidental that t h i s i s the estimated time period for the building of the wall at Hougang, as discussed i n the previous section. 41 Several s i t e s , from more than one region and time period, contain incomplete skeletons in p i t s . The significance of these skeletons i s unclear. Hao (1983:42) believes that these skeletons represent war captives who may have been put to death, or, persons who died from fig h t i n g . Skeletons of th i s kind were found at the Wangwan si t e (near Luoyang c i t y , western Henan) in the Wangwan III- region and at the Kexingzhuang s i t e (near Xian c i t y , eastern Shaanxi) in the Kexing- zhuang II region (Hao 1983:6, 39-40). Fang (1986:274) also argues that skeletons of th i s kind from the Wangchenggang s i t e represent war captives who were s a c r i f i c i a l victims. Both authors do not state that the skeletal remains are associated with a sp e c i f i c time period. Archaeological reports i d e n t i f y a number of stone and bone tool forms as weapons such as mao spearhead. Unfortunately, i t i s not possible to determine whether these forms were used as weapons or for subsistence tasks such as hunting. Axes (fu) and knives (dao) could have been used for many tasks. Yan (1986, 1987a) maintains that p r o j e c t i l e points (zu) increase dramatically i n number over time and that spearheads appear at the beginning of the Longshan Period. Also, improvements i n design make pr o j e c t i l e points more l e t h a l . In another paper, I attempted to tabulate the number of stone and bone potential weapons at f i v e Longshan site s (Underhill 1989:234-5). It was possible to examine trends over time at Baiying i n the Hougang II region and Meishan i n the Wangwan I I I region. At both s i t e s , there i s 42 an increase over time i n quantity of p r o j e c t i l e points, a possible indication of an increase i n the frequency and/or in t e n s i t y of warfare during the Longshan Period. Also, Chang (1986:250) points out that spearheads and p r o j e c t i l e points constitute a r e l a t i v e l y high proportion of stone tools at two sites from the Liangcheng region i n Shandong, Chengzi (Zhucheng County) and Yaoguanzhuang (near Weifang C i t y ) . Differentiation in Consumption of Goods A common feature of ranked societies i s display of social status with a range of material goods. Wealthy families may l i v e i n large houses b u i l t with r e l a t i v e l y costly construction materials (Blake 1988, Kamp 1987, Feinman and Neitzel 1984). Wealthy households tend to be large in size because they have adequate resources for supporting a large number of people (McC. Netting 1982, Hayden and Cannon 1984). Also, they often display t h e i r positions of status by using goods that are costly to acquire (Kamp 1987, Smith 1987). There i s evidence for d i f f e r e n t i a t i o n i n consumption of goods throughout the Longshan Period. Walled sites contain large houses b u i l t with r e l a t i v e l y costly materials. There i s marked variation in mortuary treatment at some cemeteries. Bronze, jade, and turquoise items were probably prestige goods. 43 Variation i n size and construction material for housing at two walled s i t e s , Pingliangtai and Wangchenggang, i s marked. At least 12 large adobe houses have been found at Pingliangtai. Adobe i s known as a r e l a t i v e l y costly construction material in other areas (Blake 1988:51). One of these houses, F l from Period IV, i s reported as 12.54 by 4.34 meters (Henan Province Cultural Research Institute and the Cultural Objects Division of the 2houkou D i s t r i c t 1983:30). There are houses in th i s size range from periods II and I I I as well (Sui 1988:51). At Wangchenggang, variation i n housing i s apparently most d i s t i n c t during Period II (ca. 2500 B.C.), the period in which the wall was b u i l t . Large hangtu or rammed-earth house foundations were found (Li Xiandeng 1983:11). Rammed-earth i s a costly construction material for buildings used by e l i t e s during the Shang Dynasty (Chang 1986:323). Another unique feature of the walled settlement at Pingliangtai i s the presence of pottery pipes for drainage of water (Henan Province Cultural Research Institute and the Cultural Objects Division of the Zhoukou D i s t r i c t 1983). Wells are another architectural feature which require substantial labor to construct. However, they are common on sites with and without walls (see Institute of Archaeology, CASS, 1984:83-4). There i s marked d i f f e r e n t i a t i o n in mortuary treatment at sites i n Shanxi and Shandong in type of grave, and quality and quantity of interred goods. The Taosi s i t e (Xiangfen County, Shanxi) i s unique i n terms of the large quantity of graves and wealth of objects recovered. 44 More than 1000 graves have been discovered to date. Three d i s t i n c t sizes of graves have been i d e n t i f i e d : large, medium, and small (for descriptions in English see Chang 1986:276 and Underhill, i n press, a). A l l of the nine large graves contain adult males with large quantities of f i n e l y made items, including a pottery pan dish with a painted dragon design, an a l l i g a t o r skin drum, qing chime stones, and jades. One of the large b u r i a l s , M3015, contains as many as 178 items (Shanxi Archaeological Team, IA, CASS and the Linfen D i s t r i c t Cultural Bureau 1983). As research and publication proceed, i t w i l l be possible to c l a r i f y how mortuary treatment changes during the 600 years that the cemetery at Taosi was used. Three periods have been defined on the basis of stratigraphy, ceramic s e r i a t i o n , and several radiocarbon dates: Early, ca. 2500-2300 B.C.; Middle, ca. 2300-2100 B.C.; and Late, ca. 2100-1900 B.C. (Zhang and Zhang 1986:51, Gao et a l . 1984:28). At least some of the large burials are dated to the Early and Middle Periods (Shanxi Archaeological Team, IA, CASS and the Linfen D i s t r i c t Cultural Bureau 1983:32-4). The large grave with the painted pan dish dates to the Early Period ( i b i d ) . One cemetery i n Shandong also exhibits d i f f e r e n t i a t i o n i n mortuary treatment, but to a lesser degree than Taosi. At Yinjiacheng (Sishui County), a large grave also contains the remains of an a l l i g a t o r drum (Institute of Archaeology, CASS 1984:20). There are two radiocarbon dates for Yinjiacheng, ca. 2550-2400 B.C. (Han 1989:144). 45 Information on production and use of non-ceramic prestige goods such as bronze items during the Longshan Period i s limited. Fragments of bronze have been found at only a few s i t e s , at both walled sites and sites without walls (Table 7). Four c u l t u r a l regions are represented: Wangwan I I I , Wangyoufang, Taosi, and Liangcheng. Chang (1983a:101) expects that p o l i t i c a l power during the early dynastic period was achieved by controlling production and use of bronze items, especially vessels. However, there i s no undisputed evidence for production of bronze vessels during the Longshan Period. Also, there i s no indication that e l i t e s successfully controlled production of bronze vessels or tools at t h i s time. E l i t e control may not have been achieved u n t i l a f t er the state form of s o c i o p o l i t i c a l organization had been established. Most of the evidence dates to the late Longshan Period. The e a r l i e s t remains are from Pingliangtai, traces of bronze dated to ca. 2400 B.C. (Henan Province Cultural Research Institute and the Cultural Objects Division of the Zhoukou D i s t r i c t 1983:31,36; Zhang and Zhang 1986:53). According to An Jinhuai (1983b:5), the remains show signs of smelting and casting. The significance of the metal fragment from Period IV at Wang- chenggang i s unclear. Some claim that the fragment i s from a cast vessel (Li Xiandeng 1983:10, An Jinhuai 1989:23). Others argue i t i s merely a piece of copper (Zou, personal communication 1987). Two sites without walls from the Wangwan I I I region contain fragments of metal dating to the late Longshan Period, Meishan (Second Henan Archaeological 46 Table 7. Evidence for Bronze Metallurgy During the Longshan Period. Wangwan I I I Type: Wangchenggang bronze fragment i n p i t H617 (6.5 cm wide and 5.7 cm long), possibly from a vessel; composed of lead, t i n , copper (concentrations not stated) p i t H617 dated to period IV, ZK 955, 1900+/165 B.C. (Henan Province Cultural Research Institute and the Archaeological Section of the Museum of Chinese History 1983:13, Zhang and Zhang 1986:53) Meishan traces of bronze on two ceramic crucible fragments from the Late period, in p i t s H40 and H28; composition of metal i n H40 i s 95% copper; Late period date ZK 349, 2005+/120 B.C. (Second Henan Archaeological Team, IA, CASS 1982:453-4, 472; Zhang and Zhang 1986:53) Niuzhai fragments of bronze, no date available (Yan 1984:38) Wangyoufang Type: Pingliangtai fragment of bronze 1.31 cm long in p i t H15, dated to period I I I , WB81-2, 2405+/175 B.C. (Zhang and Zhang 1986:53, Henan Province Cultural Research I n s t i t u t e and the Cultural Objects Division of the Zhoukou D i s t r i c t 1983:31,36) Taosi Type: Taosi small cast bronze b e l l (97% copper, traces of lead and zinc) i n burial M3296, ZK 1314 i s 1885+/130 B.C., but the excavators believe a more accurate date i s ca. 2085 B.C.; also a small bronze b e l l found on the surface (Shanxi Archaeological Team, IA, CASS and the Linfen D i s t r i c t Cultural Bureau 1984:1070); 2085 B.C. belongs to the either the Middle or Late period (Gao et a l . 1984:28, Zhang and Zhang 1986:53) Liangcheng Type: Sanlihe two pieces of an awl made of brass in a b u r i a l , according to Sun and Han 1985:275) 47 Team, IA, CASS 1982:453-4) and Niuzhai near Zhengzhou c i t y (Yan 1984:38). Fragments of metal have been found from more than one s i t e from the Liangcheng region i n Shandong (Yan 1984:38, Sun and Han 1985). The two pieces of an awl from Sanlihe (Jiaoxian County) have been described in the most d e t a i l . According to Sun and Han (1988:275), i t i s made of brass rather than bronze. One small bronze bell-shaped object was found in grave M3296 at Taosi, dating to ca. 2085 B.C. (Shanxi Archaeological Team, IA, CASS and the Linfen D i s t r i c t Cultural Bureau 1984:1069-1070). I t i s important to note that t h i s b u r i a l i s r e l a t i v e l y small and contains few grave goods. Bronze objects may not have been considered as highly important prestige items u n t i l the early dynastic period. Jade and turquoise were considered prestigious raw materials during the early dynastic period (Chang 1986:312, 331). Jade and turquoise objects from the Longshan Period have been found at walled and non-walled habitation s i t e s . Jade items have been reported at the walled s i t e of Haojiatai (Renmin Ribao 1986), and turquoise ornaments at Wangchenggang, Period II (Henan Province Cultural Research Institute and the Archaeological Section of the Museum of Chinese History 1983:11). Unfortunately, dates are not available for the long, thin jade axes with engraved designs recovered over f i f t y years ago at the s i t e of Liang- chengzhen, Rizhao County, Shandong, type s i t e of the Liangcheng c u l t u r a l branch (see Institute of Archaeology, CASS 1984:103, L i and Gao 1979). 48 A much greater quantity and variety of jade items have been recovered from mortuary contexts i n Shanxi, Shaanxi, and Shandong. There are several forms of jades i n the graves at Taosi from more than one period including ornaments such as bracelets (bihuan) and yue axes (Shanxi Archaeological Team, IA, CASS and the Linfen D i s t r i c t Cultural Bureau 1983). A recent a r t i c l e reports a remarkable quantity and variety of jade objects from the Shimao cemetery s i t e i n northern Shanxi (Shenmu County), a s i t e that belongs to a cu l t u r a l branch related to Kexingzhuang II (Dai 1988). Jade and turquoise items were found i n a few graves from more than one period at the Sanlihe cemetery si t e i n the Liangcheng region of Shandong (Institute of Archaeology, CASS 1988). Although the dimensions of social status expressed i n mortuary r i t u a l are unclear, burying prestige items with the dead would have been an effective method of displaying rank and wealth. Chang (1983b, 1986) proposes that a number of a r t i f a c t s from the Longshan Period have r i t u a l significance, including jade and ceramic objects with certain designs, ceramic p h a l l i , and divination bones. He (1983b:101) argues that e l i t e s obtained p o l i t i c a l power during the early dynastic period by cont r o l l i n g access to a variety of a r t i f a c t s used for r i t u a l s i n communicating with the gods and ancestors, i n addition to bronze vessels. He implies that t h i s process began during the Longshan Period (Chang 1986:287, 366). Claim to direct communication with the gods i s a common method of leg i t i m i z i n g power and authority in chiefdoms (Peebles and Kus 1977, 49 Wright 1984). However, the evidence for th i s process during the Longshan Period i s limited. Also, some types of r i t u a l items such as ceramic p h a l l i and divination bones (bu) probably were used by a l l segments of society, not just high status individuals competing for power. Common at late Neolithic s i t e s , they were probably used for ancestor worship (Chang 1983b, 1986). Two types of designs on jade and ceramic objects from late Neolithic sites may have had r i t u a l significance: animal motifs and the thundercloud (leiwen) pattern. A r t i f a c t s with animal designs have been found i n the Taosi and Liangcheng c u l t u r a l regions. The Taosi s i t e provides some support for Chang's hypothesis: the ceramic vessel with painted dragon design was found in a large grave with abundant grave goods. One of the large jade axes from the Liangchengzhen s i t e i n Shandong has an engraved design of the mythical taotie animal. Axes with t h i s design may have been used i n r i t u a l s by e l i t e s , because the design i s present on bronze vessels from the early dynastic period (Li and Gao 1979:61-2) . Sherds of black ware with the thundercloud (leiwen) pattern have been found at two sites i n Shandong, Liangchengzhen and Shangzhuang i n the Chengziya cul t u r a l region (ChipiTig County). This design i s present on Shang and Zhou bronze vessels, too (Institute of Archaeology, CASS 1984:103). Unfortunately, there i s no information on context of recovery at either s i t e . The sherd at Shangzhuang was found on the surface (Institute of Archaeology, Shandong Province 1985:494). 50 Other a r t i f a c t s may have been used by high status individuals for r i t u a l purposes. Tian (1986 ) maintains that a wooden "storehouse - shaped" vessel from a large grave at Taosi i s a symbol of the sky and was used by rulers to make sa c r i f i c e s to heaven (see Shanxi Archaeo- l o g i c a l Team, IA, CASS and the Linfen D i s t r i c t Cultural Bureau 1983:38). Chang (1986:276-7) suggests that the musical instruments from large graves at Taosi (qing chime stone, a l l i g a t o r skin drum, etc.) symbolize special status of the deceased, since h i s t o r i c texts mention that items of t h i s sort were used by royalty. CONCLUSIONS Sites from the Longshan Period are located i n a wide area along the lower reaches of the Huanghe or Yellow River. Seven d i s t i n c t c u l t u r a l types have been i d e n t i f i e d , each of which probably contained complex chiefdoms. There i s evidence for similar processes of change in each region: establishment of settlement hierarchies, warfare, and display of socia l d i f f e r e n t i a t i o n with prestige goods. Cultural complexity probably increased over time i n more than one region. At the end of the Longshan Period, the state evolved in at least one region, Wangwan I I I i n west-central Henan and southern Shanxi. As research and publication continue, i t w i l l be possible to test hypotheses about c u l t u r a l change i n a more detailed manner. One pattern that emerges from t h i s assessment i s that settlement hierarchies with 51 walled s i t e s as centers were established i n the early Longshan Period, ca. 2500 B.C. i n the Wangwan I I I (west-central) and Wangyoufang (eastern) c u l t u r a l areas of Henan. Settlement hierarchies were probably established at a lat e r time in the other areas, p a r t i c u l a r l y i n Shandong. There i s evidence for warfare and display of social d i f f e r e n t i a t i o n at walled sites and at sit e s without walls throughout the Longshan Period. Variation i n housing and prestige goods i s r e l a t i v e l y marked at two walled s i t e s , Wangchenggang and Pingliangtai. This pattern was established during the early periods at the s i t e s . S i m i l a r l y , status d i f f e r e n t i a t i o n i s expressed i n mortuary r i t u a l during the early period at Taosi. Several forms of a r t i f a c t s may have had r i t u a l significance. More data are required to test the hypothesis raised by Chang (1983b, 1986) that high status individuals began to seize power by controlling production and use of r i t u a l a r t i f a c t s . S i m i l a r l y , more data are necessary to examine the r e l a t i v e importance of competition for economic, s o c i a l , and ideological power by e l i t e s , the issue raised by Haas (1982, 1986). There i s l i t t l e evidence to suggest that material factors such as technological change were as important i n cu l t u r a l change as social factors (Chang 1983b:124), although some changes i n a g r i c u l t u r a l production may have taken place. The r e l a t i v e l y large surface area and thickness of c u l t u r a l layers at late Neolithic sites suggest that 52 improvements i n a g r i c u l t u r a l production had been made (Yan 1989), allowing population density to increase. For example, there may have been improvements i n tool forms (An Jinhuai 1989:22). At present there i s l i t t l e information on foods produced during the Longshan Period. F o x t a i l m i l l e t (su), broomcorn m i l l e t (shu), and ri c e (dao) (Oryza sativa) were common crops in the late Neolithic period (An Zhimin 1988b, Yan 1989, 1982b). 5 There may have been an increase in species of domesticated animals by the Longshan Period: goat, sheep, c a t t l e , chicken, pig, dog, and possibly the horse (An Zhimin 1988b:377). notes: 1) For descriptions of distinguishing features (mainly ceramic) for c u l t u r a l types i n the western provinces, see Zhang and Zhang (1986); for the c u l t u r a l types in Shandong, see Han (1989) and Wu and Du (1984). 2) According to Pankenier (1985), astronomical records support the conclusion that the Xia Dynasty began ca. 1900 B.C. In approximately 1953 B.C., there was a rare clustering of fi v e planets, an event that may have influenced development of the Mandate of Heaven concept. 3) The well-known walled s i t e of Chengziyai in Shandong has been included by Chang (1986) and Sui (1988) with the group of sites l i s t e d above. Others have argued that Chengziyai probably dates to the post- Longshan Period Yueshi Culture (personal communication, Yan 1987b, Zou 1987). For t h i s reason, I do not discuss the si t e further. Also, 53 Chang (1986:250) states that remains of a wall were found around the site of Yaoguanzhuang in eastern Shandong. However, there i s no mention of a wall i n the si t e report (see Institute of Archaeology, Shandong Province et. a l 1981). 4) Radiocarbon dates cited i n th i s thesis are based on a h a l f - l i f e of 5570 and calibrated by means of dendro-chronology, as described by Institute of Archaeology, CASS (1983:1-6). Some sources c i t e the spec i f i c c a l i b r a t i o n curve used ( i . e . , Institute of Archaeology, CASS 1983), but most do not (such as Zhang and Zhang 1986). 5) F o x t a i l m i l l e t and broomcorn m i l l e t have been found in e a r l i e r Neolithic s i t e s from the Huanghe (Yellow River) valley area (An Zhimin 1988b:372-5). I t i s l i k e l y that farmers continued to grow these crops in most regions during the Longshan Period. Yan (1989) states that f o x t a i l m i l l e t was the most common variety of m i l l e t during the late Neolithic period i n the lower Huanghe region. I t has been recovered at two Longshan site s i n eastern Shandong and at one si t e i n Shaanxi. The sit e i n Shaanxi has also yielded broomcorn m i l l e t . Rice has been found at one Longshan s i t e i n eastern Shandong (An Zhimin 1988b:375) as well as one s i t e i n Shaanxi (Yan 1989). According to Yan (1982b:29, Figure 12), r i c e probably became a common crop in the lower Huanghe valley during the t h i r d millennium B.C. 54 CHAPTER 3. MODEL OF CHANGE IN SYSTEMS OF CERAMIC PRODUCTION IN RELATION TO INCREASING CULTURAL COMPLEXITY This chapter has four parts. The f i r s t three sections consist of: 1) discussion of operating premises, 2) description of changes in strategies of production and test implications for ceramic v a r i a b i l i t y resulting from standardization and di v e r s i t y , and 3) description of changes i n strategies of consumers for prestige (labor-intensive) wares with test implications for d i f f e r e n t i a t i o n i n labor input. The fourth section i s a discussion of test implications for identifying change i n mode of production and consideration of expectations by researchers for specialization of production during the Longshan Period. OPERATING PREMISES As stated i n Chapter 1, the research question i s : How do systems of pottery production change during the Longshan Period of northern China i n r e l a t i o n to increasing c u l t u r a l complexity? The goal i s to test a revised version of the model by Rice (1981) which hypothesizes that there should be an increase i n vari e t i e s of wares, p a r t i c u l a r l y prestige vessels used for displays of status. Also, there should be evidence for increasing standardization of wares, especially u t i l i t a r i a n 55 (non-prestige) vessels. The model also hypothesizes that there should be a change to a more complex mode of production. In t h i s study I assume that changes in decisions about strategies of production by potters and strategies of vessel use by consumers may cause changes i n ceramic attributes. Production and consumption are related i n a c i r c u l a r manner (Douglas and Isherwood 1979:145, Gregory 1982:13). In a chiefdom, potters may i n i t i a t e changes i n production of vessels that, i n turn, cause changes i n strategies of consumers. At the same time, changes i n consumer demand for certain classes of vessels may cause changes i n strategies of production. Causal factors for change or s t a b i l i t y i n systems of production may be external or internal (Rice 1984). In my model one causal factor internal to social systems i s considered as p a r t i c u l a r l y i n f l u e n t i a l : competition, both among producers and among consumers. Another study dealing with competition between potters, but in state-level societies, i s described by Feinman et a l . (1984). They describe how competition i s affected by degree of administrative control over economy and degree of regional p o l i t i c a l consolidation. Any model i s a s i m p l i f i c a t i o n of r e a l i t y . Ceramic change i s often caused by a complex set of interacting variables, and after a time lag (Rice 1987:274,276). Another l i m i t i n g factor i s archaeological recognition of important causal variables. It can only be hoped that, as research on the Longshan Period proceeds, more detailed models of causes of ceramic change w i l l be constructed and tested. 56 CHANGES IN STRATEGIES OF PRODUCTION As Chapter 1 mentions, two general strategies of production are d i v e r s i f i c a t i o n , or e f f o r t s to produce more v a r i e t i e s of vessels, and s i m p l i f i c a t i o n , or e f f o r t s to reduce v a r i e t y f o r purposes of increasing e f f i c i e n c y . Potters may choose to vary production of paste, form, and/or decorative classes. A t h i r d strategy i s conservatism, or resistance to change in production. A given assemblage of vessels from a community may exhibit evidence f o r one or more of these strategies. As depicted i n Figure 3, the model employed here hypothesizes that potters should adopt a strategy of d i v e r s i f i c a t i o n f o r the production of one or more shape classes as s o c i o p o l i t i c a l complexity increases. The top half of the figure outlines the model i n general. The lower half describes s p e c i f i c changes that may take place with respect to u t i l i t a r i a n (non-prestige) and prestige (labor-intensive) wares. Potters may respond to incr e a s i n g l y d i v e r s i f i e d demands by consumers f o r non-prestige vessels used i n preparing, cooking, and serving food or drink, and for household r i t u a l s that take place i n private contexts. At the same time, potters may respond to increasingly d i v e r s i f i e d demands by consumers f or prestige or labor-intensive vessels used i n displays of status. Increases i n status competition would be l i k e l y to have an impact on consumer demand for labor-intensive vessels. Potters should also adopt a strategy of s i m p l i f i c a t i o n or increasing e f f i c i e n c y i n production of some shape classes, p a r t i c u l a r l y non- 57 Figure 3. Model of Change in Systems of Pottery Production In Relation to Increasing Cultural Complexity increase in socio- p o l i t i c a l differen- t i a t i o n , and in popula- tion size and density more diversified demands by consumers for vessels potters in competition respond by adopting a strategy of d i v e r s i f i - cation for one or more shape classes increase over time in varieties of shape classes, sizes, pastes, and/or decorative techniques; decrease i n dimensional and within-class standardization potters in competition adopt a strategy of simplification or increased efficiency for one or more classes change in mode of production increase in dimensional and within-class standardization; change in location of evidence for production (tools, k i l n s ) ; use of more e f f i c i e n t f i r i n g shaping techniques non-prestige wares: increased inter- settlement contacts through exchange relationships changes in agricultural practices change in diet change in house- hold r i t u a l (private contexts) increased demand by consumers for new types of pots for preparing, cooking, serv- ing food and/or drink increase in number of shape classes, decorative techniques D r e s t i a e (labor-intensive^ wares: increase in degree of labor input for techniques used previously, increase in number of shape classes with labor-intensive techniques, and/or introduction of new labor-intensive techniques increase in status competition \ increase in demand by consumers for vessels to use in displays of largesse or conspicuous consumption 7 7 1 58 prestige wares. There should be a change i n organization of labor over time as well. Below, I discuss alternative strategies of producers and consumers i n more d e t a i l . D i v e r s i f i c a t i o n Potters are l i k e l y to adopt a strategy of d i v e r s i f i c a t i o n when there i s an increase in s o c i o p o l i t i c a l d i f f e r e n t i a t i o n with an increase i n size and density of the consuming population. Two important aspects of c u l t u r a l evolution are increasing population size and density, and increasing so c i a l d i f f e r e n t i a t i o n (Flannery 1972, Wright 1978). Increases i n population size and density coupled with increases i n social d i f f e r e n t i a t i o n are l i k e l y to cause more d i v e r s i f i e d demands for vessels by consumers (Rice 1984:256-7). Consumers may demand greater v a r i e t i e s of non-prestige and/or prestige wares. An increase i n vari e t i e s of vessels desired by consumers of different social statuses i s l i k e l y (Rice 1987:301). Other factors may cause d i v e r s i f i c a t i o n i n production as well. Decreases i n available arable land, especially in conjunction with increases in population size and density, may force some households to make a l i v i n g by other means such as pottery making (Arnold 1985:168, 196). This may cause a change from a household mode of production to one involving s p e c i a l i s t s . Or, there may be a change from one sp e c i a l i s t mode to another. 59 Potters are l i k e l y to make changes i n strategy of production i f an element of competition i s involved. As I discuss below, potters may r e s i s t change. Specialist potters in competition with one another for a growing and increasingly d i v e r s i f i e d consuming population should d i v e r s i f y production of prestige and non-prestige wares. Demand for a variety of goods gives potters incentive to compete (Foster 1965:55). Birmingham (1975) observes that competing potters i n the Kathmandu Valley make a wide variety of vessels i n order to please consumers. Competition between potters to d i v e r s i f y production may involve the harboring of secret technologies (Nicklin 1971:33). Competition among producers i s i n t e n s i f i e d i f there i s an increase i n number of work units sp e c i a l i z i n g i n pottery production. There are three categories of archaeological evidence for d i v e r s i f i c a t i o n i n production. The f i r s t i s an increase over time i n variety of paste, shape, and/or decorative classes. The second i s a decrease i n dimensional standardization. The t h i r d i s a decrease i n within-class ( i . e . , shape class) standardization with regard to secondary shape features and techniques of decoration. There may be evidence for one or more of these ceramic categories. Two external factors may cause d i v e r s i f i c a t i o n of production of non-prestige ( u t i l i t a r i a n ) wares. One i s changes i n diet necessitating adoption of new methods for preparing, cooking, and/or serving food or drink (Rice 1984:246). For example, declining a v a i l a b i l i t y of arable land could stimulate farmers to grow new crops that are better adapted 60 to poorer s o i l . Potters could respond by producing new forms and/or pastes for vessels used to prepare these new foods. A second causal factor i s changes in r i t u a l practices (Rice 1984:246-7). Change i n household r i t u a l practices could result from contact with another cult u r a l group. Potters would respond by producing new form and/or decorative classes of vessels. However, high status families may adopt changes i n diet and in r i t u a l practices and also create a demand for changes in vessels (in form, paste, and/or decoration). Vessels used in public contexts would exhibit labor-intensive techniques. S i m p l i f i c a t i o n The strategy of s i m p l i f i c a t i o n of production involves a decision to produce fewer varieties of ceramic classes defined i n terms of paste, shape, and/or decorative techniques. The strategy of s i m p l i f i c a t i o n may be caused by at least three types of external factors. The f i r s t and more common situation involves two of the same causal factors that may stimulate d i v e r s i f i c a t i o n : increasing population size and density, and decline i n a v a i l a b i l i t y of arable land. In t h i s case potters are l i k e l y to compete for the increasing size of the consuming population by producing non-prestige wares as e f f i c i e n t l y as possible. There are p o t e n t i a l l y two types of archaeological evidence for s i m p l i f i c a t i o n i n this instance, i n addition to decrease i n variety of 61 ceramic categories. The f i r s t i s increase i n dimensional standardi- zation. The second i s increase i n within-class standardization, or v a r i e t i e s of size classes, decorative techniques, and secondary shape features within each shape c l a s s . Evidence f o r r e l a t i v e lack of labor input i n production techniques such as decoration, forming, and f i r i n g may indicate e f f i c i e n c y i n production as well. There i s no incentive for potters to invest time and labor on classes of vessels with r e l a t i v e l y low value to consumers. In the Kathmandu Valley, vessels regarded as "cheap" and "disposable" by consumers are made with a r e l a t i v e lack of labor input by s p e c i a l i s t potters (Birmingham 1975:386). A s i m i l a r s i t u a t i o n has been observed i n Veracruz, Mexico (Krotser 1974:133-4). The t h i r d external causal factor i s increasing s o c i a l d i f f e r e n - t i a t i o n . Potters may also decide to produce some prestige wares more e f f i c i e n t l y , to meet consumer demand. There should be evidence of an increase i n dimensional standardization and/or within-class standard- i z a t i o n of labor-intensive vessels. Prestige vessels w i l l not exhibit evidence of lack of labor input. As defined i n t h i s study, they are produced by labor-intensive techniques. If i n t e n s i f i c a t i o n of production, or e f f o r t s to produce greater quantities of vessels more e f f i c i e n t l y , i s successful there can be a feedback s i t u a t i o n i n which i t i s advantageous f or potters to increase t h e i r e f f o r t s at i n t e n s i f i c a t i o n . For example, i f the size of a d i s t r i b u t i o n area increases, potters may respond by increasing t h e i r 62 efforts at e f f i c i e n t production. Davis and Lewis (1985) conclude that when size of d i s t r i b u t i o n area increased during the late Bronze Age on Crete, there was an increase i n e f f i c i e n t l y applied decorative techniques and an increase i n standardization of shape. Cultural stress from factors such as increase i n warfare, famine, disease, or environmental degradation can force potters to simplify production to a great degree. These factors may cause the size of the population using and making pottery to decline dramatically. Also, potters would have less time, energy, and/or resources to spend on production. There should be a r e l a t i v e l y large decrease i n var i e t i e s of vessel classes (paste, form, and/or decoration) produced. Production of prestige wares should decline dramatically or cease. Relative lack of labor input i n production should increase. Decoration i s most l i k e l y to change (Rice 1987:464). There are two cases from the period of European contact i n the New World that i l l u s t r a t e t his process (Rice 1987:268- 71). During the sixteenth and seventeenth centuries i n central Mexico, decoration on Aztec pottery became simp l i f i e d after population levels declined. S i m i l a r l y , contact with Europeans caused disease among the Arikara of the North American Plains and population decreased. Subsequently, quality of surface f i n i s h on Arikara pottery declined. 63 Conservatism Conservatism in production i s l i k e l y , especially with respect to vessels that serve basic needs such as vessels used for household r i t u a l s (in private contexts), water ja r s , and cooking pots (Rice 1984:245-6). Potters are reluctant to change production methods when consumers r e s i s t change. Also, potters do not want to r i s k changing methods that have proven successful i n producing vessels that serve basic needs. Aspects of production that have functional significance, i . e . , paste and form, are least l i k e l y to change (Rice 1984:241). CHANGES IN STRATEGIES OF CONSUMERS FOR PRESTIGE VESSELS Researchers often ignore the main purpose of using pottery vessels in prehistoric societies: preparation, storage, and serving of food and drink. Because these a c t i v i t i e s have universal significance, attributes of pottery vessels can symbolize social r e l a t i o n s . Two other ceramic studies that address t h i s topic are Sinopoli (1986) and Die t l e r (1988). Sinopoli (1986) uses h i s t o r i c data to investigate how social relations are symbolized during a c t i v i t i e s involving food vessels i n a medieval Hindu society. Dietler (1988) describes the social importance of drinking alchohol and the effect of importing foreign wine and drinking vessels on p o l i t i c a l economy during the French Iron Age. 64 Strategies for social manipulation have an impact on consumer demand for goods (Appadurai 1986:29). In a chiefdom, competition for maintaining and enhancing positions of status i s common behavior. Status may be defined in terms of wealth, sex, age, p o l i t i c a l position, r i t u a l position, genealogical relationship, or other variables. Status competition creates demand for craft items that have prestige value. This part of the model describes how people use pottery vessels that exhibit r e l a t i v e l y great amounts of labor input during displays of status. Pottery vessels are on display as well as the food and alchohol that they contain. Social value of goods varies with context of use (Douglas and Isherwood 1979). Pottery vessels for food and alchohol may be used i n two kinds of soc i a l contexts: household consumption and consumption at inter-household social events. If display behavior with pottery i s present i n chiefdoms, i t should take place most often i n public contexts, such as inter-household social events. Social messages tend to be sent when they w i l l be v i s i b l e to r e l a t i v e l y large, diverse groups of people (Wobst 1977:330). Ethnographic data on use of pottery vessels support t h i s prediction. In the Yucatan, vessels with the greatest amount of labor input are those used i n the most public areas (Lischka 1978:231). In a chiefdom there are two types of behavior i n which people may use inter-household social events involving food and/or alchohol to maintain or enhance positions of status. They are: 1) display of 65 largesse, or generosity i n giving food and/or alchohol to other people, and 2) conspicuous consumption, or display of personal consumption of food and/or alchohol. For these a c t i v i t i e s consumers demand containers exhibiting r e l a t i v e l y great amounts of labor input i n order to symbolize status. Displays of largesse and conspicuous consumption with labor- intensive vessels must have been common in prehistoric chiefdoms, given the extent they are mentioned i n the ethnographic l i t e r a t u r e (discussed below). During the t r a n s i t i o n from chiefdom to state there should be fluctuations i n competition for positions of socia l status. Competition may increase, decrease, or remain stable. When competition increases, competing individuals or groups create an increase i n demand for craft goods that symbolize status. Increasing size and density of population may be a causal factor. As societies increase i n size and complexity, the need for people to communicate positions of status with material items increases (Rice 1984:257). An increase i n status competition should cause an increase in demand for pottery vessels to use in displays of largesse and/or conspicuous consumption, resulting i n increased production of vessels d i f f e r e n t i a t e d i n terms of labor input. Increased production from one phase to another may involve: making greater quantities of elaborated vessels with no change in method, using a greater variety of techniques for elaboration, and/or increasing the l e v e l of labor expenditure i n making vessels. 66 A decrease i n status competition causes a decrease i n demand for pottery vessels used in displays of largesse and/or conspicuous consumption. This results i n a decrease i n production of elaborated vessels. One causal factor may be rejection of peaceful means of status competition by people in favor of increased warfare. However, a decrease i n production of elaborated vessels could indicate that competitors decided to replace pottery vessels with a new type of craft good to symbolize status, rather than an actual decline in status competition. Changes i n types of craft goods that serve as prestige items are l i k e l y as chiefdoms become more complex. Increases i n production of status items tend to cause further increases. Competitors demand increasingly elaborated and different items for demonstrating superior status. This process cannot continue i n d e f i n i t e l y ; inevitably the system of production must s t a b i l i z e or collapse. On the basis of ethnographic data about use of containers (pottery and wooden vessels, baskets) during displays of largesse and conspicuous consumption, I expect to f i n d a pattern showing a continuum of elabor- ation in. labor input. At one end are vessels exhibiting r e l a t i v e l y marked elaboration, and at the other, vessels that are r e l a t i v e l y non- elaborated. The greater the social position of the consumers, the greater the elaboration. Vessels exhibiting a low degree of elaboration may represent attempts by non-elite consumers to acquire vessels that emulate those used by e l i t e s . 67 Labor-intensive techniques have been i d e n t i f i e d from ethnographic observation and experience of western potters. For example, i t has been observed that large vessels require more labor input than smaller vessels (Reina and H i l l 1978:246; DeBoer and Lathrap 1979:120). Also, thin-walled vessels produced by scraping require r e l a t i v e l y great labor input (DeBoer and Lathrap 1979:120, Hagstrum 1986:16). D i s p l a y s of l a rgesse On the basis of ethnographic data summarized i n Table 8, I expect that there may be evidence for up to four types of displays of largesse in p r e h i s t o r i c chiefdoms. Three types pertain to e l i t e s : 1) chiefs giving food and alchohol to p o l i t i c a l supporters, 2) other e l i t e s holding occasional feasts, and 3) r i v a l r i e s of feast-giving by e l i t e s . The fourth type applies to e l i t e s and non-elites: providing food for participants at l i f e - c r i s i s ceremonies such as weddings and funerals. The goal of e l i t e s in any display of largesse i s to maintain or increase positions of status. Displays of largesse are displays of generosity in giving food and/or drink to r e l a t i v e l y large numbers of " people. This generosity i s a manipulative t a c t i c . A generous reputation i s an important asset for s o c i a l , especially p o l i t i c a l , status. Food and alchohol are g i f t s of great value. By receiving these g i f t s , people become indebted to the donor(s). In the f i r s t two types 68 Table 8. Display Behavior with Containers i n Chiefdoms and Other Ranked So c i e t i e s . type of display containers area reference largesse, chiefs give food to p o l i t i c a l supporters huge wooden troughs f o r serving taro Polynesia ( F i r t h 1965: 222) large quantities of serving baskets, over 100 i n one feast Polynesia ( F i r t h 1965: 227) huge pottery serving vessels f o r porridge A f r i c a (Goody 1982: 91) huge cooking pots f o r porridge, very large serving baskets A f r i c a (Richards 1939:148) high status people hold feasts large quantities of pots f o r corn beer Mexico (Pastron 1974:108) large cooking pots Guatemala (Nelson 1981:122) large quantities of big cooking pots Mexico and Guatemala (Hayden and Cannon 1984:175) 6 9 r i v a l r i e s of large, Melanesia generosity by elaborately (Oliver high status decorated 1955:367) people wooden baking frame for puddings large cooking pots for taro pudding huge wooden serving vessels Melanesia (Malinowksi 1922:171) Melanesia (Davenport 1986:96, 98,99) large quantities of cooking pots Melanesia (Oliver 1955:297) large quantities of beer apparently served i n many pots Af r i c a (Washburne 1961) largesse at l i f e - c r i s i s ceremonies large cooking pots Guatemala (Nelson 1981:113) large cooking pots Philippines (Longacre 1985:344 ) large, decorated pots for serving food such as basins Peru (Tschopik 1950:206) large quantities of cooking pots Guatemala (Nelson 1981:111) 70 porcelain food dishes, some with elaborate shapes, porcelain r i c e wine jars conspicuous large consumption quantities of pots large quantities of pots d i v e r s i t y of pot forms di v e r s i t y of pot forms chiefs use elaborately decorated wooden spit bowls elaborately decorated pots for drinking corn beer size of calabash for drink- ing beer correlates with status Philippines (Solheim 1965:256, 258,281) Peru (Tschopik 1950:215-6) India ( M i l l e r 1985:74) Peru (Tschopik 1950:215-6) India ( M i l l e r 1985:73-4) Hawaii (Earle 1987b:69) Peru (DeBoer 1984:551-4) Af r i c a (Washburne 1961) 71 of displays by e l i t e s , recipients are obliged to acknowledge the superior status of the donor(s). In the t h i r d type of e l i t e display, recipients attempt to surpass previous displays of generosity. The purpose of displays of largesse at l i f e - c r i s i s ceremonies i s to symbolize status of the hosting families. The ethnographic data summarized i n Table 8 indicate that there i s a tendency to use par t i c u l a r functional types of vessels for different types of displays. Displays of largesse tend to involve vessels for cooking and serving food, and for preparing and serving alchohol. In displays of largesse by chiefs for p o l i t i c a l supporters, two types of elaborated containers are common: extremely large containers for cooking and serving food, and great quantities of containers for serving food. Among the Gonja of Ghana, women i n the house of the chief cook and serve porridge i n huge pots to people during the most important community f e s t i v a l . This act of generosity on the part of the chief ensures loyalty of supporters (Goody 1982:91). Among the Bemba of Rhodesia, chiefs give people who provide labor at t h e i r households large amounts of food. Porridge i s cooked in huge pots and served i n baskets that are about eight times larger than those used i n dai l y l i f e (Richards 1939: 148). The extremely large size of containers and great quantities used symbolize the unsurpassable largesse of the chiefs. Feasts held by e l i t e s other than chiefs tend to involve vessels that are large in size, too. Among the Maya of Guatemala, lineage heads tend to own r e l a t i v e l y large quantities of big f i e s t a pots (Hayden and 72 Cannon 1984:175). Also, the wealthier the landowner, the larger the size of f i e s t a pots owned (Nelson 1981:122). E l i t e s other than chiefs attempt to surpass previous displays of generosity i n order to improve t h e i r positions of status. In Melanesia, men host feasts in order to humiliate p a r t i c u l a r r i v a l s (Oliver 1955: 365-90; Davenport 1986:97-8). Again, very large containers as well as large numbers of them are used to prepare and serve food. The larger the size of vessel, the more expense was undertaken by the host (Davenport 1986:96,98). Among the Thonga of southern A f r i c a , e l i t e males compete with each other for supporters by giving away beer (Washburne 1961). Apparently, donors have large quantities of pots for brewing and serving. When families provide food for participants of l i f e - c r i s i s ceremonies, they use vessels that are r e l a t i v e l y less elaborated and show more varied methods of elaboration. If families do not have adequate resources, they borrow large pots (Nelson 1981:113) or food and labor (Longacre 1985:344). Or, they may use larger quantities of normal-sized pots (Nelson 1981:111). Vessels for cooking and serving may be large in size, elaborately decorated, have elaborate forms, and have fine paste. 73 Displays of conspicuous consumption Displays of conspicuous consumption by e l i t e s and non-elites may take place at individual households and at inter-household events. People may use vessels designed for use by more than one person and/or individual-use vessels such as drinking cups. Ownership of a large quantity of pots and a great d i v e r s i t y of pots are two kinds of symbols of wealth and status. Another i s use of elaborated cups for drinking alchohol. In A f r i c a , for example, size of calabash (for drinking) correlates with status of consumer (Washburne 1961). Methods of elaboration are varied. Display of personal consumption with elaborated drinking cups should be common given archaeological as well as ethnographic evidence. In Denmark, elaborately shaped drinking cups of metal and containers for alchohol (metal, ceramic) are associated with adult male graves from prehistoric chiefdoms (Kristiansen 1984:86-93). Display of personal consumption i s evident, whether the vessels were made for consumption by the deceased i n the a f t e r - l i f e or by mourners at the funeral r i t e s . In Germany, thin-walled drinking cups of pottery are also associated with adult male graves (Wells 1985:11-3). 74 The Chinese historical context H i s t o r i c a l records mention that bronze and some types of pottery vessels were used i n displays of status during the Shang and Zhou dynasties. Apparently, displays took place i n several soc i a l contexts. Bronze vessels were exhibited i n bu r i a l s , ancestral h a l l s , and during feasts (Li Xueqin 1980:9). Table 9 describes types of displays mentioned i n the l i t e r a t u r e . It appears that size , weight, shape, and decoration were important c r i t e r i a , as well as quantity of vessels i n use at one time. These examples provide additional support for my expectation that there should be evidence for display behavior, largesse and conspicuous consumption, with pottery during the Longshan Period. H i s t o r i c a l records suggest that one type of bronze vessel i n par t i c u l a r , the ding cauldron, was often used i n displays of status. It i s possible that at least some large ding were used in displays of largesse. For example, two bronze ding that have been found are large enough i n volume to cook an ox (Li Xueqin 1980:10). Very large ding may have symbolized power and authority (Chang 1983b:95). Unfortunately, details about the social contexts i n which large ding were used are lacking. Ding cauldrons may have been used during the types of events in which displays of largesse are common i n chiefdoms: leaders feeding p o l i t i c a l supporters, other e l i t e s holding feasts, r i v a l r i e s of generosity by e l i t e s , and families feeding participants at l i f e - c r i s i s ceremonies. 75 Table 9. Display Behavior With Containers: the Chinese H i s t o r i c a l Context • type of display symbols of rank characteristics of vessels largesse' symbols of rank demonstration of power ceremony to maintain positions of rank by e l i t e males, Shang and Zhou Dynasties possibly a hierarchy of bronzes according to size, weight, shape, decoration, i n s c r i p t i o n s , Shang Dynasty (Chang 1980:207) huge bronze ding cauldrons: one at 875 kg, 133 cm i n height, 110 cm in length; two vessels 1/2 t h i s size; each has the capacity to cook an ox, Shang Dynasty (Li Xueqin 1980:10) different sizes and quantities of bronze ding cauldrons correlate with rank, Zhou Dynasty (Li Xueqin 1980:10) large bronze ding used to b o i l a l i v e a foreign leader, Zhou Dynasty (Li Xueqin 1980:10) apparently bronze and pottery eating and drinking vessels, quantity of food vessels correlates with age (Cooper 1982:107-14) 76 ceremony of thanks- giving by e l i t e males, and other drinking ceremonies, Shang and Zhou Dynasties apparently bronze and pottery drinking vessels (Cooper 1982:107-14) general, Shang and Zhou Dynasties apparently quantity of bronze and pottery food serving dishes correlates with rank (Chang 1978:131) 77 I t appears that there were displays of conspicuous consumption with containers during the Shang and Zhou dynasties, too. One text mentions that quantity of vessels for serving food correlates with age of consumer. Also, drinking vessels are associated with e l i t e males (Cooper 1982:108-14). Quantity of vessels for serving food may correlate with rank as well (Chang 1978:131). CHANGE IN MODE OF PRODUCTION The f i r s t topic discussed i n this section i s typologies for modes of production. Second, hypotheses in the Chinese and western archaeo- l o g i c a l l i t e r a t u r e for mode of production during the late Neolithic period are discussed. Third, expectations for change i n mode of production on the basis of ceramic v a r i a b i l i t y are described. There are few test implications that have been derived from ethnographic data. There are several typologies for mode of ceramic production in the archaeological l i t e r a t u r e . Four modes of production have been defined on the basis of ethnographic data by van der Leeuw (1977, 1984) and Peacock (1981, 1982:8-10): 1) household production (non-specialist), 2) household industry, 3) individual workshop industry, and 4) nucleated workshop industry. Other researchers make a further d i s t i n c t i o n between attached and independent specialization (Earle 1981:230, Brumfiel and Earle 1987:5). Independent s p e c i a l i s t s produce goods for a general 78 population of consumers. Attached s p e c i a l i s t s , sponsored by e l i t e s , produce prestige items exclusively for e l i t e consumption. Santley et a l . (1989:108) use the term "tethered sp e c i a l i z a t i o n " to refer to one type of attached spe c i a l i z a t i o n . Two recent studies employ more complex typologies. Costin (1986) focuses on distinguishing between modes involving independent and attached s p e c i a l i s t s . Sinopoli (1988) defines modes that pertain s p e c i f i c a l l y to state-level societies. This study employs the simpler typology of modes defined by van der Leeuw (1977, 1984) and Peacock (1981, 1982). Given a chiefdom level of c u l t u r a l complexity, the f i r s t three modes are pot e n t i a l l y applicable: household, household industry, and individual workshop industry. Nucleated workshop industry i s associated with urbanism and f u l l y developed market economies (Rice 1987:184). The study emphasizes i d e n t i f i c a t i o n of prestige (labor-intensive) versus non-prestige wares rather than i d e n t i f i c a t i o n of attached versus independent s p e c i a l i - zation . Rice (1987:187) suggests that the household industry mode encompasses a great deal of v a r i a b i l i t y . I t i s l i k e l y that modern ethnographic societies with household industry do not represent the range of variation that probably existed among prehistoric chiefdoms. In t h i s study I propose to define two types of household industry, "simple" and "complex". "Simple household industry" refers to the standard d e f i n i t i o n of household industry involving female producers using a simple technology. 79 Due to factors such as poor a g r i c u l t u r a l land, families attempt to supplement t h e i r incomes by making pots (Rice 1987:184). The Fulani of Cameroon are often regarded as exemplifying the household mode of production. I t i s not l i k e l y that t h i s group represents the only kind of household industry i n prehistory. The Fulani regard potting as a low status, undesirable a c t i v i t y . Containers made of modern materials such as p l a s t i c or metal are preferred (David and Hennig 1972:4,17). "Complex household industry" i s similar to the individual workshop industry mode as defined by van der Leeuw (1977, 1984) and Peacock (1981, 1982). In both modes, s p e c i a l i s t producers are men. Potting i s the major source of income. The former mode involves production i n houses, and the l a t t e r , in workshops. The individual workshop mode i s characterized by greater i n t e n s i f i c a t i o n of production. The ethno- graphic case described by M i l l e r (1985) for a v i l l a g e i n central India i s a useful i l l u s t r a t i o n of "complex household industry". Six houses, organized by men, produce vessels for others (1985:36,209). Vessels are either sold i n houses or in regional markets, and payment i s made with grain or cash (1985:86). Potters on the island of Thassos in Greece exemplify individual workshop industry (Peacock 1981:189-190). Two male potters, investing a considerable amount of resources i n technology, produce vessels for a l l other residents. Some Chinese and western researchers expect that there i s increasing e f f i c i e n c y i n ceramic production during the late Neolithic period. They imply that a workshop industry mode existed. 80 Song, L i , and Du (1983:273) state that there were gradual improvements i n techniques of production and that scale of production increased s i g n i f i c a n t l y . They envision a system i n which families specialized i n pottery making. L i and Cheng (1984:14) agree with this conclusion. Their i l l u s t r a t i o n (Li and Cheng 1984:7) of two men working on a wheel (one turning, one forming) seems equivalent to individual workshop industry because large output i s implied. The term that Keightley (1987) applies to late Neolithic pottery in China, "prescriptive production", seems to refer to a workshop mode of production. The term originates with Franklin (1983), designed to describe organization of bronze production during the Shang dynasty. I t indicates a highly organized d i v i s i o n of labor that regularly produces a large volume of output. According to Franklin (1983:96), with pre- s c r i p t i v e production there i s "an essential p r e d i c t a b i l i t y " , and "no room for surprise". Also, there i s standardization of form and material (Franklin 1983:97). By expecting the presence of overseers who plan a l l steps of production, Keightley implies that pottery vessels should exhibit a high degree of standardization (1987:107). There are serious problems in identifying modes of production on the basis of ceramic data. Few test implications for modes have been developed, partl y due to the limited ethnographic sources providing relevant information (Rice 1987:204-5). Studies involving state-level societies have had the most success i n iden t i f y i n g mode of production. There i s often more than one source of nonceramic data available such as 81 direct evidence of workshops and textual data describing organization of production (see Sinopoli 1986, 1988; Davis and Lewis 1985; Benco 1987, 1988; and Beaudry 1984). It i s d i f f i c u l t to determine whether changes in degree of standardization or di v e r s i t y indicate change from one mode to another, or simply change within a mode. The different conclusions reached in two recent studies i l l u s t r a t e t h i s dilemma. Milanich et a l . (1984:132- 6) conclude that a certain level of standardization and d i v e r s i t y represent specialized production i n the Weeden Island Culture of northern F l o r i d a . However, Kaiser (1984:292-5) concludes that increases i n d i v e r s i t y indicate i n t e n s i f i c a t i o n of production within a non- sp e c i a l i s t or household mode. There are a limited number of observations of pottery production i n t r a d i t i o n a l societies that provide support for the hypothesis that increases i n standardization and d i v e r s i t y indicate development of or increase i n spec i a l i z a t i o n (Rice 1981, 1987:202). There i s some indication that a progression from one mode to another i s characterized by an increasing degree of standardization and d i v e r s i t y (Table 10). Unfortunately, most observations of t r a d i t i o n a l societies pertain to the extreme ends of the continuum, the household and nucleated workshop industry modes. Only one ethnoarchaeological study with the e x p l i c i t purpose of examining v a r i a b i l i t y i n vessels i n re l a t i o n to modes of production has been published to date (Longacre et a l . 1988). I t supports the 82 Table 10. Id e n t i f i c a t i o n of Mode of Production: Ethnographic Data on Ceramic V a r i a b i l i t y . ceramic v a r i a b i l i t y ethnographic observation dimensional standardization Longacre et a l . (1988) show that pots made i n a nucleated workshop industry mode are much more standardized in shape and size than pots made in a household mode Balfet (1965) notes that pots made i n a nucleated workshop industry mode are highly standardized i n shape and size compared to pots made by semi- sp e c i a l i s t s within-class standardization none di v e r s i t y Longacre et a l . (1988) show that there i s a greater variety of size classes of pots made i n a nucleated workshop industry mode than i n a household mode van der Leeuw (1984) suggests that d i v e r s i t y of forms increases with complexity of mode 83 hypothesis that a high degree of standardization represents a complex s p e c i a l i s t mode of production. The authors conclude that vessels made among ful l - t i m e s p e c i a l i s t s (representing a nucleated workshop industry) in the Philippines, the Paradijon, are much more standardized i n shape and size than vessels made i n a household mode of production by the Kalinga (1988:111). One other study demonstrates with quantitative data that potters in a nucleated workshop industry mode, in Egypt, produce vessels that are highly standardized i n form (Lacovara 1985:56). Balfet (1965:166) makes a comment about dimensional standardization and mode of production that i s l a t e r supported by the results of Longacre et a l . (1988). She notes that vessels made by fu l l - t i m e s p e c i a l i s t s in north African workshops are highly standardized i n shape and size. Also, vessels made by semi-specialists (mode not specified) are less standardized (1965:170). Longacre et a l . (1988:111) also observe that a greater d i v e r s i t y of size classes of cooking pots i s produced i n the nucleated workshop mode than in the household mode. This finding supports the suggestion by Rice (1981, 1987:202) that d i v e r s i t y of ceramic categories increases with complexity of productive mode. In this vein, van der Leeuw (1984:757-60) suggests that d i v e r s i t y of vessel forms increases i n the progression from household production to each type of s p e c i a l i s t mode. If there i s archaeological evidence for e f f i c i e n c y i n production, i t i s l i k e l y that s p e c i a l i s t s are represented (Brumfiel and Earle 84 1987:5). Factors such as increasing population density can cause independent s p e c i a l i s t s to i n t e n s i f y production (Earle 1981:230). However, increasing e f f i c i e n c y or s i m p l i f i c a t i o n i n production over time may not necessar i l y e n t a i l a change i n mode of production. SUMMARY The model tested i n t h i s study outlines p o t e n t i a l changes i n strategies of ceramic production i n r e l a t i o n to increasing c u l t u r a l complexity i n chiefdoms. There are three strategies of production: d i v e r s i f i c a t i o n , s i m p l i f i c a t i o n , and conservatism. There may be evidence f o r more than one of these strategies within an assemblage. After Rice (1981), i t i s hypothesized that potters should adopt a strategy of d i v e r s i f i c a t i o n f o r one or more shape classes as s o c i o p o l i t i c a l complexity increases. Potters should also adopt a strategy of s i m p l i f i c a t i o n or increased e f f i c i e n c y f o r some shape. classes, and there should be ceramic evidence f o r a change i n mode of production. The model describes some external causal factors that can have an impact on production strategy such as increasing population size and density. One important causal factor i n t e r n a l to the system of production i s competition among potters i n response to consumer demand fo r an increasing v a r i e t y of prestige (labor-intensive) and non-prestige vessels. 85 The model also describes how people may use pottery vessels i n order to display status. E l i t e s and non-elites may undertake displays of largesse and/or conspicuous consumption. These displays are common in ethnographically-known chiefdoms. The model predicts that an increase in status competition creates demand f o r a greater v a r i e t y of prestige vessels. Also, some types of display tend to involve c e r t a i n functional classes of vessels. For example, displays of largesse tend to involve very large containers. F i n a l l y , h i s t o r i c a l texts from the Shang and Zhou dynasties in China suggest that bronze and pottery vessels may have been used f o r displays of largesse and conspicuous consumption. Therefore, there i s even more reason to expect that these displays were taking place during the Longshan Period. 86 CHAPTER 4. ANALYSIS OF SHAPE CLASSES AND HYPOTHESES ABOUT FUNCTIONAL TYPES INTRODUCTION This study r e l i e s on descriptions of shape classes i n archaeo- l o g i c a l reports for testing the model of ceramic change. The test requires e x p l i c i t l y defined and d i s t i n c t shape classes for each s i t e . Therefore, i t i s necessary to evaluate the individual shape classes defined i n reports. When examining vessels i n the f i e l d , i t became apparent that some shape classes might require r e d e f i n i t i o n , either by s p l i t t i n g into more groups or lumping with other groups. Another necessary step before testing the model i s discussing the types of information on pottery presented in reports as well as information on ceramic function. These topics are addressed i n a separate chapter, since they have not been treated extensively in the western l i t e r a t u r e on Chinese archaeology. The chapter consists of four sections. The f i r s t section i s a description of t r a d i t i o n a l terms used to designate shape classes of vessels i n Chinese archaeological reports. The second section describes kinds of ceramic data included i n Chinese Neolithic s i t e reports. The t h i r d section consists of the evaluation of shape classes defined in archaeological reports for Hougang, Baiying, Meishan, and Lujiakou. In the fourth section, hypotheses about vessel function are discussed. 87 TRADITIONAL TERMS FOR DESIGNATING SHAPE CLASSES A r c h a e o l o g i s t s i n C h i n a use a t r a d i t i o n a l s e t of terms t o d e s i g n a t e shape c l a s s e s of c o n t a i n e r s . These terms may r e f e r t o p o t t e r y , b r o n z e , p o r c e l a i n , wooden, o r l a c q u e r v e s s e l s . They o r i g i n a t e from two s o u r c e s : 1) a n c i e n t t e x t s and 2) modern usage (An Z h i m i n 1953:73). A c c o r d i n g t o Chang (1981:158), many a n c i e n t terms d a t e t o t h e Song Dynasty (A.D. 960-1279). Song s c h o l a r s adopted terms from o l d e r , c l a s s i c t e x t s t h a t denoted d i f f e r e n t shapes of b r o n z e v e s s e l s . One such term i s d i n g c a u l d r o n , mentioned i n C h a p t e r 3. Modern terms i n c l u d e b e i cup and wan b o w l . T a b l e 11 l i s t s p u b l i s h e d d e f i n i t i o n s f o r t r a d i t i o n a l terms d e s i g n a t i n g shape c l a s s e s of v e s s e l s . A l l of t h e forms l i s t e d a r e r e p r e s e n t e d i n Longshan s i t e s . As t h e t a b l e i n d i c a t e s , some terms such as hu pot o r i g i n a t e from a n c i e n t t e x t s and a r e s t i l l used t o d a y . G i v e n s u b s t a n t i a l c o n t i n u i t y i n v e s s e l shape o v e r t i m e i n C h i n a , a r c h a e o l o - g i s t s have seen advantages i n u s i n g t h e t r a d i t i o n a l terms t o d e s c r i b e N e o l i t h i c as w e l l as v e s s e l s from h i s t o r i c p e r i o d s i t e s . However, problems have been e n c o u n t e r e d i n u s i n g t h e s e t e r m s . As i n many a r e a s , r e s e a r c h e r s c l a s s i f y v e s s e l s by s u b j e c t i v e assessment of s i m i l a r i t y . A r c h a e o l o g i c a l r e p o r t s do not g i v e an i n d i c a t i o n t h a t q u a n t i t a t i v e methods a r e used i n c l a s s i f i c a t i o n . As a r e s u l t t h e r e i s some c o n f u s i o n r e g a r d i n g d e f i n i t i o n of terms and assignment of v e s s e l s t o c l a s s e s (see Chang 1981:161). A u t h o r s of d i f f e r e n t r e p o r t s use d i f f e r e n t d e f i n i t i o n s 88 Table 11. Published D e f i n i t i o n s f or T r a d i t i o n a l Terms Describing Shapes of Vessels Found in Longshan Si t e s . ancient vessel shapes, terms no longer used: dou Chinese-English Dictionary (1979): ancient stemmed cup or bowl Zhang (1983): hemispherical bowl with high stem and spreading foot, a container to serve food L i Xueqin (1980:11): meat container, may have a deep b e l l y and l i d , or shallow bowl Chang (1978:128): during the Shang Dynasty, vessel of pottery, only, to serve meat dishes ding Chinese-English Dictionary (1979): ancient cooking vessel with two loop handles and three or four legs Zhang (1983): three or four legged cauldron f o r cooking meats and cereals L i Xueqin (1980:8): vessel to cook meat with a deep b e l l y and three or four legs Chang (1981:160): a s p e c i f i c term that refers to a vessel with three s o l i d legs gu Chinese-English Dictionary (1979): ancient wine vessel, beaker, goblet Zhang (1983): goblet with broad l i p , long narrow stem, and f l a r e d base, f o r wine Chang (1981:162): a generic term for a vessel to warm wine gui Chinese-English Dictionary (1979): ancient pitcher with three legs (pottery only) j i a Chinese-English Dictionary (1979): ancient t r i p o d wine vessel with a round mouth Zhang (1983): round t r i p o d vessel for wine with handle and capped columns L i Xueqin (1980:14): wine goblet, moderately large b e l l y , sometimes with l i d Chang (1981:162): a generic term for a vessel to serve wine 89 I i Chinese-English Dictionary (1979): ancient tripod cooking vessel with hollow legs Zhang (1983): cauldron for cooking meats and cereals L i Xueqin (1980:10): a cooking vessel with pouch-like hollow legs Chang (1981:162): a generic term for a food vessel l e i L i Xueqin (1980:16): an urn-like vessel that i s f a i r l y t a l l and narrow xian/yan Zhang (1983): steamer for vegetables and cereals L i Xueqin (1980:10): a tripod steamer with pouch-like legs, upper part i s l i k e a ding and has a rack at the base Chang (1981:161): a s p e c i f i c term for a tripod food vessel zeng Chinese-English Dictionary (1979): an ancient earthen utensil for steaming r i c e zun Chinese-English Dictionary (1979): a kind of ancient wine vessel Zhang (1983): cup for drinking or warming wine L i Xueqin (1980:15): a wine container ( i l l u s t r a t i o n shows a vessel with wide o r i f i c e , carinated body, splayed foot) Chang (1981:161): a generic term for wine vessels zuo Guo (1981:136): a stand for other vessels known from the l a t e r Zhou Dynasty terms denoting ancient and modern vessel shapes: bei Chinese-English Dictionary (1979): cup Zhang (1983): wine vessel, cup bi Chinese-English Dictionary (1979): grate or grid for steaming 90 bo Chinese-English Dictionary (1979): earthen bowl bu Chinese-English Dictionary (1979): vase L i Xueqin (1980:15-6): short and squat urn-like vessel for holding wine ( i l l u s t r a t i o n page 17 shows a large ring foot) gang Chinese-English Dictionary (1979): vat, ja r , crock, for holding water, pickled vegetables, etc. guan Chinese-English Dictionary (1979): j a r , pot, for jam, tea, water, etc. Hansford (1954): a substantial vessel moderately contracted towards the neck, a jar Chang (1978:127): a storage vessel made of pottery during the Shang and Zhou Dynasties gai Chinese-English Dictionary (1979): l i d , cover Hansford (1954): cover he Chinese-English Dictionary (1979): box, case Hansford (1954): a small box with f i t t e d cover, often c i r c u l a r i n shape hu Chinese-English Dictionary (1979): kettle for water, pot (for tea, etc.) Zhang (1983): jar for wine of various shapes - round, rectangular, compressed L i Xueqin (1980:16): a wine container of various shapes - round, rectangular, compressed Chang (1981:161): a sp e c i f i c term for a wine vessel Hansford (1954): describes several d i f f e r - ent shapes and functions such as warming wine, tea lei-bo term from An Jiayuan (1986), called deng l u by most writers to mean strainer, f i l t e r ; from the common meaning of the two words, for example in The Chinese-English Dictionary (1979) 9 1 pan Chinese-English Dictionary (1979): plate, dish, tray Zhang (1983): wide shallow bowl for holding water to use i n washing or ceremonial ablutions, usually with high r i n g foot and handles L i Xueqin (1980:19): a water vessel f o r ceremonial ablutions, used with he p i t c h e r before the middle Zhou period Chang (1981:162): a generic term for a water container pen Chinese-English Dictionary (1979): basin, tub, pot for water, flowers, etc. Hansford (1954): pot, tub Chang (1981:162): a generic term for a water container ping Chinese-English Dictionary (1979): b o t t l e , vase, jar Hansford (1954): vase or b o t t l e ; vessels with narrow necks and swelling bodies wan Chinese-English Dictionary (1979): bowl Hansford (1954): small bowl used for eating or drinking tea weng Chinese-English Dictionary (1979): earthen ja r for water, p i c k l e d vegetables, etc. Hansford (1954): l i k e guan, a j a r moderately contracted towards the neck yu Chinese-English Dictionary (1979): a broad- mouthed vessel f o r holding l i q u i d such as a spittoon 92 for terms. This problem i s compounded by the fact that there i s often a great variety of vessel shapes in ceramic assemblages, as Medley (1976:24) points out. A class of vessels described by one term may include more than one ceramic form. Table 11 shows that one reason for t h i s confusion i s a lack of s p e c i f i c i t y i n the definitions for terms. Chang (1981) points out that only some of the ancient terms were meant to refer to s p e c i f i c forms of bronze vessels; most terms are generic. The other terms should be considered generic, too. The problem of nonspecific terms for describing vessel shape i s not limited to Chinese archaeology. As Fournier (1981) shows, several terms used i n modern English such as "cup" and " j a r " are not precise. Problems of t h i s kind i n c l a s s i f i c a t i o n are inevitable when researchers attempt to apply formal types on a widespread basis. The t r a d i t i o n a l terms should be considered as starting points for establishment of classes that suit p a r t i c u l a r research problems, rather than formal types with standardized d e f i n i t i o n s to f i t a l l cases. Table 11 i l l u s t r a t e s another problem, that the t r a d i t i o n a l terms refer to vessel function as well as morphology. Terms with functional implications are not useful on a widespread basis, because vessel function can change over time and space. I t should not be assumed, for example, that vessels of a given form were used for the same purpose during the Neolithic and h i s t o r i c periods, especially when different raw materials ( i . e . , pottery and bronze) are involved. Functional aspects 93 of the t r a d i t i o n a l terms should be considered as hypotheses on the basis of h i s t o r i c a l analogy, not fact. As the la s t section of t h i s chapter shows, in some cases archaeological data do not provide support for the tr a d i t i o n a l d e f i n i t i o n s of function. Scholars i n the West should r e a l i z e that Chinese researchers have employed a variety of methods for c l a s s i f y i n g vessels using the tr a d i t i o n a l terms; i . e . , subdividing the basic categories of shape. C l a s s i f i c a t i o n procedures used by L i Chi (1956) for pottery at the Shang si t e of Xiaotun provided a standard for l a t e r archaeological reports, as Li Chi himself points out (1977:138). For the Xiaotun pottery, L i Chi developed a paradigmatic c l a s s i f i c a t i o n procedure, dividing the assemblage into successively f i n e r classes with lower body shape as the f i r s t d i v i s i o n , and o r i f i c e size as the second. For the late Neolithic s i t e of Chengziyai, L i Chi et a l . (1956) describe a similar c l a s s i - f i c a t i o n system that i s based on differences i n vessel openings and legs. In 1953, An Zhimin stated that each worker devises his own method of using the t r a d i t i o n a l terms (1953:73). This statement i s s t i l l applicable to recent archaeological reports. Authors of Neolithic s i t e reports often use three terms to describe vessel shape. Xingzhuang are broad categories of shape such as ding and bei. Xing are subtypes of xingzhuang, usually denoted by capital l e t t e r s such as "A" and "B". Shi are subdivisions of xing and referred to as "styles". They are always designated by Roman numerals 94 such as " I " and " I I " . Styles are usually defined by attributes of shape; occasionally attributes of decoration are used as well (Yan 1985:34). C l a s s i f i c a t i o n i s undertaken primarily for one purpose: to arrange vessels i n a detailed chronological series, using the shi styles. I t i s a tool used primarily for one type of research problem: description of culture history. Because t r a d i t i o n a l terms are used on a widespread basis for t h i s single purpose, archaeologists i n China actually work with typologies, or leixingxue, rather than establishing classes to use for s p e c i f i c analyses, or l e i b i e (Yan, personal communication, 1987b). My evaluation of c l a s s i f i c a t i o n procedures concentrates on de f i n i t i o n s of xingzhuang and xing given i n reports. The shi s t y l i s t i c groups, usually defined on the basis of r e l a t i v e l y minor shape attributes, tend to be inappropriate for the research problem at hand. KINDS OF CERAMIC DATA IN SITE REPORTS There are three factors that must be considered when interpreting ceramic data i n Neolithic s i t e reports. F i r s t , formats of describing vessels can vary substantially. Table 12 l i s t s formats of describing vessels i n the Hougang, Baiying, Meishan, and Lujiakou s i t e reports. The authors of the Hougang and Lujiakou reports use similar formats. For example, they group vessels according to xingzhuang, xing, and shi 95 Table 12. Formats of Describing Vessels i n Site Reports. s i t e report method Hougang (Anyang Archaeo l o g i c a l Team, IA, CASS, 1985) pots from a l l phases grouped together into major shape classes (xingzhuang), subtypes (xing), and styles (shi). t o t a l s for each major shape class not stated for individual phases, but to t a l s given for s i t e as a whole Baiying (CPAM of Anyang D i s t r i c t , Henan Province 1983) pots f i r s t separated by phase, then by type of paste and surface color, then by form according to major shape class (xingzhuang) and style (shi ), only. styles can represent major or fine dis t i n c t i o n s i n form, vessels of one shape class may be described in separate sections, t o t a l s of each major shape class given for each phase Meishan (Second Henan Archaeological Team, IA, CASS 1982 ) pots described for each phase according to major shape class (xingzhuang) and style (shi ), only, t o t a l s for quantities of vessels per phase or i n the si t e as a whole not stated 96 Lujiakou (Shandong Archaeo l o g i c a l Team, IA, CASS and the Art Museum of Weifang County, Shandong Province 1985) pots from a l l phases grouped together by major form (xing- zhuang ), subtype (xing), and style ( s h i ) , but major forms are sometimes grouped by type of paste and surface color, and styles may represent major differences i n shape, t o t a l s provided for s i t e as a whole but not for each phase 97 s t y l e s . The authors of the Baiying and Meishan reports do not define xing subgroups of shape. Reports tend to lack information that i s relevant to research goals other than establishment or refinement of culture h i s t o r y . This s i t u a t i o n creates problems f o r the present study. For instance, reports may state the t o t a l number of vessels for a given major form (xingzhuang) i d e n t i f i e d at s i t e s , but not the number of vessels from each phase. This i s the case for Hougang and Lujiakou. As discussed in Chapter 5, i t i s possible to estimate these f i g u r e s . Another important f a c t o r i n i n t e r p r e t i n g descriptions of vessels i n Chinese s i t e reports i s that only whole vessels or sherds with a recognizable form ( i . e . , xingzhuang) are described i n any d e t a i l . While some of these vessels were o r i g i n a l l y recovered i n one piece, most were broken; technicians s k i l l f u l l y reconstructed them with p l a s t e r . Sometimes reports summarize d i f f e r e n t types of paste and decoration found on sherds that cannot be used i n reconstruction. However, they do not provide separate summaries for d i f f e r e n t phases. The most serious problem with respect to t h i s study i s that reports only provide descriptions of vessels that are considered representative of a given s t y l e ( s h i ) , or major form (xingzhuang) i n cases when no styles are defined. For example, a report may give a t o t a l of ten ding cauldrons and describe only three, one vessel from each s t y l e . 98 Thus, samples of vessels described i n Neolithic s i t e reports represent an even greater theoretical distance from an o r i g i n a l systemic context (as defined by Schiffer 1976) than samples described i n most western reports. This situation can be described by four le v e l s , beginning with the systemic context (Table 13). The l a s t two levels pertain to Neolithic s i t e reports i n p a r t i c u l a r . In recent years western archaeologists have made efforts to describe how vessels excavated from sites represent a distorted picture of the o r i g i n a l systemic context (e.g. Deal 1983, 1985; Hayden and Cannon 1983). They have pointed out how discard behavior affects the kinds and quantities of vessels recovered i n archaeological contexts. This topic i s discussed further i n Chapter 6. Vessels described in Neolithic s i t e reports represent an a r t i f i c i a l sample of the archaeological context. Because t h i s study r e l i e s on ceramic data from s i t e reports, conclusions about changes in production and inferred use of labor-intensive vessels must be regarded as preliminary. I t i s not e n t i r e l y clear how well samples of vessels described i n the Hougang, Baiying, Meishan, and Lujiakou s i t e reports represent the population of vessels from the archaeological context. An additional concern i s that most shape classes are represented by very small samples of vessels. This problem i s a l l e v i a t e d to some extent by the fact that I was able to examine several vessels i n addition to those that are l i s t e d i n the reports as representative of styles. In the analysis of shape classes, I describe these vessels as 99 Table 13. Schematic Representation of Levels in Interpreting Samples of Pottery Described i n Chinese Neolithic Site Reports level 1 vessels in use by households: curated, recycled, discarded. systemic several discard areas; context provisional or temporary, and f i n a l (Deal 1983, 1985) le v e l 2 abandonment of households, eventually the whole community; archaeological type of abandonment affects kinds context and quantities of vessels found in excavation (Deal 1983). recovery of whole and p a r t i a l l y broken vessels, and sherds le v e l 3 only vessels that are whole or reconstructed, or large sherds Neolithic with recognizable form s i t e reports (xingzhuang) are counted i n totals level 4 Neolithic s i t e reports only vessels considered as representative of major forms (xingzhuang), subtypes (xing), or styles (shi) are described 100 "unknowns", since they are not mentioned in reports. I incorporate these vessels in my analyses i f the major shape class (xingzhuang) such as wan bowl or bei cup i s obvious. Also, I ensure that the quantity of vessels i n my samples for major shape classes (with the "unknowns" added) matches the figures given in reports. ANALYSIS OF SHAPE CLASSES IN SITE REPORTS Some reports make direct statements about c r i t e r i a archaeologists used to define individual classes of vessels. These statements are important because they refer to the t o t a l range of excavated vessels. They are expressed i n quali t a t i v e terms; refe r r i n g , for example, to a large b e l l y and wide mouth. When appropriate I evaluate these statements by quantitative methods, referring to sp e c i f i c dimensions of vessels such as maximum diameter and rim diameter. In some cases examination of drawings and photographs i n reports i s s u f f i c i e n t to determine whether classes are d i s t i n c t . When there i s no explanation of c r i t e r i a used i n defining shape classes, I am limited to making evaluations on the basis of v a r i a b i l i t y that i s apparent i n the samples of vessels described i n reports. In many cases the classes defined i n reports are d i s t i n c t . The only change necessary, i f any at a l l , i s re-assignment of a few vessels to more appropriate classes. However, i n some cases I conclude there i s no appreciable difference between established classes. Thus i t i s 101 necessary to lump classes together. At times i t i s necessary to re- c l a s s i f y vessels, defining three forms, for example, when there were two o r i g i n a l l y . Methods of analysis There are a variety of approaches to c l a s s i f i c a t i o n of vessels by means of morphological attributes in the western archaeological l i t e r a t u r e . This study i s concerned with establishing e t i c rather than emic classes, or devised (versus folk) classes i n the terminology of Rice (1987). Descriptions of emic c l a s s i f i c a t i o n systems include Kempton (1981) and Kaplan and Levine (1981). Also, this discussion concerns c l a s s i f i c a t i o n of whole vessels rather than sherds. Other archaeological studies with t h i s goal that are p a r t i c u l a r l y relevant here are Sinopoli (1988), Barnes (1986), Froese (1985), Hardy- Smith (1974), and Whallon (1982). The l i t e r a t u r e indicates that there are two c r i t i c a l steps in c l a s s i f i c a t i o n : 1) selection of variables to define classes and 2) selection of s t a t i s t i c a l techniques for establishing classes. Several studies have demonstrated the effectiveness of using ratios describing major vessel proportions i n c l a s s i f i c a t i o n (Barnes 1986, Hally 1986, Sinopoli 1986, Froese 1985, Whallon 1982, and Hardy-Smith 1974). Ratios are more useful i n identifying major differences i n shape rather than single dimensions, p a r t i c u l a r l y when 102 whole vessels are analyzed (Barnes 1986:474; Whallon 1982:151-4, Hardy- Smith 1974:5). Also, i t i s important to id e n t i f y different size classes for each morphological class of vessels using dimensions such as height (Barnes 1986:474, Hardy-Smith 1974:5). This c l a s s i f i c a t i o n method i s adopted here, for reasons of p r a c t i c a l i t y with respect to small samples as well as effectiveness. Some studies employ large numbers of ratios for individual vessels (e.g., Froese 1985, Shennan and Wilcock 1975). This analysis uses measurements taken i n the f i e l d and measurements calculated from scale drawings of vessels i n reports. Time constraints and d i f f i c u l t conditions i n the f i e l d made i t impractical to take large numbers of measurements on each vessel. Likewise, i t i s not feasible to accurately calculate large numbers of dimensions from scale drawings i n reports. Reliable results may be obtained only for major dimensions. Restriction to major dimensions should not undermine the an a l y t i c a l r e s u l t s , however. At least one study concludes that only a few key ratios are needed to distinguish between vessel shapes (Shennan and Wilcox 1975:27). Another complex method of c l a s s i f i c a t i o n considered infeasible for t h i s study involves principles of s o l i d geometry (e.g., Ericson and Stic k e l 1973; Shepard 1976; see Rice 1987:219-20). Ratios and individual dimensions with formal-functional significance are used here to evaluate shape classes and define new ones when necessary (Table 14). There i s a need for more ethnoarchaeological research examining the relationship between form and function. The few 103 Table 14. Ratios and Individual Measurements with Formal - Functional Significance Used i n Analysis of Shape Classes. r a t i o or measurement significance height and maximum diameter (HT, MXD) maximum diameter/ height (MXD/HT) or maximum diameter/ height of o r i f i c e diameter, or body height (MXD/ODHT) size; ethnographic data indicate that many shape classes of vessels have more than one size class (David and Hennig 1972, DeBoer and Lathrap 1979:105, Pastron 1974:103, Longacre 1981:53) tr a n s p o r t a b i l i t y (Rice 1987:226) overall shape and s t a b i l i t y (Hally 1986:278); ethnographic data indicate that t a l l and thin vessels ( i . e . those with smaller values of MXD/HT) tend to have a primary function of long term storage for dry goods or long term storage for l i q u i d staples (Henrickson and McDonald 1983:632,633); r e l a t i v e l y short and squat vessels (with larger values of MXD/HT) tend to be used for temporary storage of dry goods or for cooking (Henrickson and McDonald 1983:632, 631) o r i f i c e diameter/ base diameter (OD/BD) or rim diameter/ base diameter (RD/BD) maximum diameter/ base diameter (MXD/BD) s t a b i l i t y ; Ericson et a l . (1972:89, 90,94) propose that vessels with r e l a t i v e l y large values are more stable s t a b i l i t y ; proposed by Hally (1986:278) 104 o r i f i c e diameter/ height (OD/HT), or rim diameter/height (RD/HT), or o r i f i c e diameter/height of o r i f i c e (OD/ODHT) a c c e s s i b i l i t y ; proposed by Hally (1986:280), ethnographic data indicate functional significance of o r i f i c e size and vessel volume (Smith 1988:914), also that vessels for dry storage tend to have a small o r i f i c e and large volume, and that eating dishes tend to have a large o r i f i c e and small volume (Smith 1985:300,301) o r i f i c e diameter/ maximum diameter (OD/MXD), or rim diameter/maximum diameter (RD/MXD), or neck diameter/maximum diameter (ND/MXD) a c c e s s i b i l i t y ; ethnographic data indicate that unrestricted vessels have larger values and r e s t r i c t e d vessels have smaller values (Rice 1987:212), vessels for eating tend to be unrestricted and have larger values (Rice 1987:236) 105 studies that have been published include Henrickson and McDonald (1983) and Smith (1988, 1985). The functional significance of many ratios and single dimensions i s not clear. Form and function are not d i r e c t l y related; a given vessel form may be used for more than one purpose (Rice 1987:224). I t i s more profitable to define functional types on the basis of attributes other than morphology such as type of paste and presence of residues from use, as discussed i n the f i n a l section of th i s chapter. A number of s t a t i s t i c a l techniques, of varying complexity, have been used i n c l a s s i f i c a t i o n s of vessel shape. Relatively complex techniques such as p r i n c i p a l components analysis and cluster analysis are feasible when large data sets are available, in terms of quantity of cases and variables (see Sinopoli 1986, Froese 1985, Whallon 1982, Shennan and Wilcox 1975, Rice and Saffer 1982). I t i s not p r a c t i c a l to use complex s t a t i s t i c a l techniques i n this analysis, since only small samples of vessels and measurements are available. Furthermore, the purpose of analysis i s evaluation of previously established shape classes on the basis of key c r i t e r i a either mentioned in reports or apparent from observation of vessels. It i s not possible to t o t a l l y r e - c l a s s i f y a s i g n i f i c a n t quantity of vessels that are cl e a r l y representative of the population of vessels recovered during excavation, i . e . , to define a l l major shape classes or xingzhuang such as ding and L i . I t i s more feasible to evaluate existing classes using r e l a t i v e l y simple s t a t i s t i c a l techniques. Also, multivariate techniques 106 such as cluster analysis would not be able to indicate c l e a r l y whether previously established classes should be combined into one class. For example, a cluster analysis of two sub-classes ( i . e . , xing such as A, B) of guan jars would result i n a number of options for grouping vessels. When sample sizes are adequate, I use techniques of exploratory data analysis, or EDA (Hartwig and Dearing 1979, Shennan 1988), primarily scatterplots as well as stem and leaf displays indicating ranges of measurements and median values. Traditional simple s t a t i s t i c a l techniques such as histograms do not describe distributions of measurements as completely (Hartwig and Dearing 1979). More than one study has successfully i d e n t i f i e d shape classes by means of scatterplots (see Barnes 1986, Hardy-Smith 1974). I try a number of combinations of variables and select the plots that most cl e a r l y depict d i s t i n c t groups, ensuring that the resultant classes are lo g i c a l given variation among vessels observed i n the f i e l d and apparent from information i n reports (descriptions, drawings, photographs). Analyses were conducted with the computer programs SYSTAT and SYGRAPH (Wilkinson 1988a, 1988b). When sample sizes of vessels permit, I employ nonparametric significance tests to evaluate existing classes and to define new ones when necessary. Nonparametric significance tests are appropriate when normal di s t r i b u t i o n s cannot be assumed and when sample size i s small (Siegal and Castellan 1988). I adopt a rejection level of 0.01 for the nonparametric tests. I prefer to use a r e l a t i v e l y conservative 107 c r i t e r i o n for rejecting the n u l l hypothesis that there i s no si g n i f i c a n t difference between shape classes, for concluding that different shape classes are present. S t r i c t rules for defining groups are necessary, given that small samples are used in the analyses. Otherwise, perceived differences between vessels could merely represent the range of variation for one shape class. Measurements on individual vessels, whether taken d i r e c t l y or calculated from scale drawings in reports, are rounded to one decimal place ( i . e . , 5.6 cm, 5.7 cm, etc). If vessel shape was uneven, I took the maximum measurement for a given dimension. Some measurements refer to the i n t e r i o r portion of vessels: o r i f i c e diameter, neck diameter, and maximum diameter. When calculating dimensions from scale drawings of vessels i n reports, i t was not necessary to determine height and rim diameter. This information i s usually given i n reports, designated as gao (height) and koujing (rim diameter). These data made i t possible to check the accuracy of calculated measurements from different scales. I compared my calculated measurements with the known ones, a procedure which detected errors i n report measurements i n a few cases. Occasionally I did random checks to compare my calculated measurements with those I had taken in the f i e l d . The majority of measurements used i n the following analyses, however, were taken d i r e c t l y from vessels. 108 Results This section b r i e f l y describes results from the analyses of shape classes defined i n the Hougang, Baiying, Meishan, and Lujiakou s i t e reports. For details describing o r i g i n a l shape classes accepted and new classes established, see Appendix A (Tables 36-37, 39-44; Table 38 gives results of nonparametric significance tests for vessels from Hougang). This discussion focuses on comparing the shape classes defined i n reports to the t r a d i t i o n a l terms as described i n Table 11. The evaluation of shape classes was more successful for Hougang than the other s i t e s . There are larger samples of vessels, including vessels I examined d i r e c t l y and those i l l u s t r a t e d i n the report. Also, the report provides more statements about c r i t e r i a for establishing classes. It was often d i f f i c u l t to evaluate the shape classes at Baiying and Lujiakou, because the primary method of c l a s s i f i c a t i o n i s color and r e l a t i v e texture of paste rather than shape. Vessels i n a given shape class are often described in different sections of these reports. A problem with the analysis of pots from Meishan i s that only a small number of vessels were seen. A shape class (xingzhuang) such as guan jar may refer to one form, or a number of forms (Table 15). These forms often d i f f e r from s i t e to s i t e . I t i s clear that the t r a d i t i o n a l terms should be regarded as generic. My proposed de f i n i t i o n s for major shape classes of vessels in 109 Table 15. Summary of Results from Analysis of Shape Classes at Hougang, Baiying, Meishan, and Lujiakou. ("old" = o r i g i n a l class(es) accepted, "new" = new classes established; "(x)" refers to the number of shape and size classes indicated by each term; * denotes size classes, only) xingzhuang (shape class) bei bi bo bu dou g_ai gang 2H guan gui he hu j i a l e i lei-bo l i Hougang new (3) new (2) new (1) old (1) old (1) old (2) old (5) new (2) new (3*) old (1) old (1) old (2) Baiying new (5) old (1) old (1) new (2) old (1) new (11) new (10) old (1) old (1) new (3) old (1) old (1) Meishan new (6) old (1) old (1) old (1) old (2) old (1) old (1) old (1) old (1) old (1) new (2) Lujiakou new (4) old (1) old (1) new (7) old (1) old (4) new (9) old (1) old (1) old (1) 110 pan panxingqi pen Ping pingdipen quanzupan sanzupan shenfupen shuangfupen sizumin old (1) old (1) wan weng xian/yan y_u zeng zhe (gu) fupen zun zuo other jars, no necks other necked jars new (2) old (1) new (2*) new (2) old (1) old (1) new (5) old (1) new (1) old (1) old (1) old (1) new (3) old (1) old (1) new (4) old (1) old (1) old (1) new (3) old (1) old (1) old (2) new (6) old (1) old (1) old (1) new (1) old (3) old (3) new (10) old (4) new (2) old (1) new (2) other pitchers new (4) 111 Longshan sites are given i n Table 16. The table describes common var i e t i e s or subclasses of xingzhuang ( i l l u s t r a t e d i n Figure 4). Some terms usually refer to a wider range of forms than others, such as bei cup, gai l i d , guan j a r , pingdipen basin, pen container, and wan bowl. In other words, these terms often refer to a variety of proportions, requiring subdivision into different shape classes. One factor contributing to th i s conclusion i s that r e l a t i v e l y large samples of vessels were available to evaluate these classes. If other classes were represented by larger samples, a s i m i l a r l y wide range of variation might be apparent. Cultural factors are probably important as well. These forms are more common on sites than other forms, r e f l e c t i n g r e l a t i v e l y high rates of production, use, and discard. The morpho- l o g i c a l v a riation may also represent a r e l a t i v e l y large number of separate producing units. This p o s s i b i l i t y i s discussed i n Chapter 6. Many terms incorporate more than one size of vessel, too. Different size classes of vessels are common in t r a d i t i o n a l societies and often indicate different functions. This pattern of subsuming more than one shape under a given t r a d i t i o n a l term characterizes reports of other late prehistoric s i t e s . L i Chi et a l . (1956) state that the Chengziya s i t e i n Shandong contains several v a r i e t i e s of dou stemmed dish, ding tripod, bei cup, guan j a r , gai l i d , pen container, and wan bowl. Keightley (1985a) describes how the terms dou stemmed dish, xian tripod, and ding tripod refer to more than one shape of vessel from the e a r l i e r phases at the Dahe s i t e . 112 Table 16. Proposed D e f i n i t i o n s f o r of Vessels (Xingzhuang) in Longshan Major Shape Classes S i t e s . bei a class of vessels with great v a r i a t i o n in shape; forms are s i m i l a r to modern cups, most form are tube-shaped, with rim diameter/base diameter r a t i o close to 1.0; handles optional b i a class of vessels with straight walls, r e l a t i v e l y wide rim diameter and base diameter, and small c i r c u l a r holes pierced through the base; again there are variants i n terms of vessel s i z e , and size and shape of holes. Two other forms are i n s i t e s : 1) a vessel with straight walls, r e l a t i v e l y wide rim diameter and base diameter, and saw- tooth shaped ridges completely around the base, and 2) a small bowl with holes pierced through the lower body and base bo a small bowl with a bulge (usually a point of maximum body diameter) at the middle or lower portion of the body that may be rounded or sharp i n p r o f i l e (a c a r i n a t i o n p o i n t ) , the rim tends to be wide i n comparison to the base bu a short, r e l a t i v e l y wide vessel i n terms of maximum diameter, bulbous in shape, with a f l a r i n g rim ding a globular vessel with a moderately wide o r i f i c e , f l a r i n g rim, round base, and 3 s o l i d feet of various shapes and s i z e s ; there i s v a r i a t i o n i n degree of g l o b u l a r i t y (maximum diameter/height) dou a stemmed dish; there i s v a r i a t i o n i n depth and size of dish, and in height and width of stem 113 gai a l i d , a v a r i e t y of shapes and sizes; including f l a t l i d s , shallow-bodied l i d s , and l i d s with deep bodies that are bowl- l i k e i n shape; a va r i e t y of shapes of handles (one or two per vessel) gang a large j a r , o v e r a l l the major class of vessels with the largest vessels i n terms of height and maximum diameter, a r e l a t i v e small base and wide o r i f i c e , handles optional; two v a r i e t i e s : 1) short, wide neck and 2) no neck gu a vessel shaped l i k e a modern beaker with a t a l l body, splayed foot and d i s t i n c t l y wide, f l a r i n g rim guan a jar with a f l a t base, i l l - d e f i n e d shoulder, f l a r i n g rim, p a r t i a l l y c o n s t r i c t e d o r i f i c e , the point of maximum diameter i s at the mid-body, no handles; more than one size c l a s s ; reports apply t h i s term to other shapes of jars as well, some with necks gui a r e l a t i v e l y large legged vessel with a large, wide spout, a large handle (more than one v a r i e t y ) ; three legs that are e i t h e r r e l a t i v e l y t h i n and s o l i d , or wide and hollow (mammiform) he a short, wide vessel with a r e l a t i v e l y small base and wide o r i f i c e , incurving rim, point of maximum diameter i s at the shoulder hu a jar with a long, wide neck and small f l a r i n g rim, the o r i f i c e diameter i s s l i g h t l y smaller than the neck diameter, the point of maximum body diameter i s approximately mid-height and the change in p r o f i l e i s abrupt, handles optional 114 j i a a t r i p o d with r e l a t i v e l y wide hollow legs, vessel body i s r e l a t i v e l y globular, wide o r i f i c e and f l a r i n g rim; there are variat i o n s i n height of rim, shape of body (in terms of maximum diameter/ height), and shape of base ( r e l a t i v e l y f l a t or round) l e i a jar with a r e l a t i v e l y t a l l , narrow neck; r e l a t i v e l y small p e d e s t a l - l i k e base, the point of maximum diameter i s at the shoulder and i s carinated i n p r o f i l e ; four wide, f l a t handles l e i - b o a bowl with small base, wide o r i f i c e and incurving rim, the point of maximum diameter i s at the o r i f i c e , there are large, wide engraved l i n e s that radiate out from the i n t e r i o r base up the sides, covering the en t i r e inside portion of the vessel; another, less common va r i e t y i s a t a l l and narrow vessel, cup-shaped l i a large, globular t r i p o d vessel with short neck and wide f l a r i n g rim, the hollow legs are wide and t a l l , bag-shaped or mammiform, and the legs are t a l l i n r e l a t i o n to the body pen a jar that i s r e l a t i v e l y short and squat i n shape, with a wide o r i f i c e , wide f l a r i n g rim, and small base; there are two v a r i e t i e s , those with a d i s t i n c t shoulder, and those without l i k e shenfupen ping a small j a r with a short, wide neck, globular body and r e l a t i v e l y wide base, small rim pingdipen l i k e other pen, t h i s vessel has a r e l a t i v e l y wide rim diameter and o r i f i c e diameter; however there i s a wide base and the vessel i s r e l a t i v e l y shallow (low i n height); there i s more than one var i e t y i n terms of o r i f i c e diameter, base diameter, and height 115 quanzupan a large, r e l a t i v e l y shallow dish on a wide stem or ri n g foot; shape of dish varies -- some are very shallow, or p l a t e - l i k e sanzupan a r e l a t i v e l y wide and shallow vessel with o u t f l a r i n g walls on three short legs, legs may be hoop-shaped or s o l i d shenfupen s i z u m m a r e l a t i v e l y t a l l j a r with a small base, large o r i f i c e and wide f l a r i n g rim; no d i s t i n c t shoulder, the point of maximum diameter i s the rim rather than an area of the body, handles optional the body of t h i s vessel i s e s s e n t i a l l y a basin or pingdipen, there are 4 short and wide feet wan a vessel with short, o u t f l a r i n g walls and a r e l a t i v e l y small base, the rim i s the point of maximum diameter; there i s more than one vari e t y i n terms of rim diameter, base diameter, and height; and more than one size c l a s s weng a large j ar with a r e l a t i v e l y narrow and d i s t i n c t neck, d i s t i n c t shoulder that i s the point of maximum diameter, small base, small f l a r i n g rim, more than one size c l a s s ; handles optional xian/yan) yu a t a l l t r i p o d vessel with hollow or bag-shaped legs; two d i s t i n c t portions: bottom portion or the legs, and the upper portion that i s shaped l i k e a guan j a r ; these vessel parts are connected; the space between these two parts may have had a small b i grate a small vessel the size of a cup that i s r e l a t i v e l y short and squat, there i s a bulge at the lower portion of the body approximately equal i n siz e to the rim diameter, r e l a t i v e l y wide base and small f l a r i n g rim 116 zeng a jar with a wide o r i f i c e and wide, f l a r i n g rim; i l l - d e f i n e d shoulder, o v e r a l l shape i s r e l a t i v e l y short and squat; there are several holes pierced through the base; some vessels have holes pierced through the lower body as well zhefupen l i k e other pen, t h i s vessel has a r e l a t i v e l y wide rim and o r i f i c e diameter, and small base diameter; there i s a sharp c a r i n a t i o n point at the lower portion of the vessel, o v e r a l l shape i s short and wide zun a small vessel the size of a cup with a d i s t i n c t l y wide, f l a r i n g rim, wide o r i f i c e , and bulge at mid-height, and a wide base zuo a large vessel with d i s t i n c t l y wide bottom and f l a r i n g walls, there i s va r i a t i o n i n o v e r a l l shape (maximum diameter/height) -- some vessels are short and squat, others are more t a l l and narrow; no base 117 Figure 4 . Major forms of vessels in Longshan s i tes , (not to scale) 118 Figure 4 continued. tripod jar tri pod 119 u r e 4 c o n t i n u e d . shenfupen sizumin • sanzupan deep belly jar four-footed vessel three-footed dish wan bowl weng necked jar xlan (yan) tripod composite steamer 120 Figure 4 continued. zeng steamer zun zuo container stand 121 In terms of methodology, the analyses indicate the u t i l i t y of comparing v i s u a l observations with r e s u l t s from s t a t i s t i c a l analyses. S t a t i s t i c a l techniques are useful for supporting o v e r a l l patterns observed by eye. Analyses may indicate that more than one grouping i s possible, and r e s u l t s should match v i s u a l observations. On the other hand, s t a t i s t i c a l analyses can demonstrate that what appear to be separate groups instead represent the range of v a r i a t i o n f o r one group. For sake of convenience, I describe s p e c i f i c r e s u l t s f or s i t e s i n the following order: j a r s , open forms, legged vessels, cups, and l i d s . At every s i t e there i s a wide v a r i e t y of j a r forms without necks. At Hougang, there are three clear size classes of guan j a r s : small (xiao guan, N=5), medium (guan, N=18), and large (shenfu guan, N=15). These classes, defined in the report, are i d e n t i f i a b l e by a s c a t t e r p l o t of height with maximum diameter (Appendix A, Figure 5). My r e s u l t s agree with the o r i g i n a l c l a s s i f i c a t i o n with the exception of one pot. A Kruskal-Wallis test indicates that the three size classes are s i g n i f i c a n t l y d i f f e r e n t i n terms of height (p=0.000). Sample size i s inadequate to i d e n t i f y size classes of guan or other jars from the other s i t e s . The Hougang report i d e n t i f i e s subtypes of medium and large guan j a r s . I use seven r a t i o s that q u a n t i t a t i v e l y express the c r i t e r i a i t c i t e s as important i n forming groups. Nonparametric te s t s indicate that none of the subtypes are s i g n i f i c a n t l y d i f f e r e n t from one another. 122 The Baiying, Meishan, and Lujiakou reports use the term "guan" to refer to a variety of shapes of j a r s , with or without necks. The Baiying and Meishan reports do not use several terms for jars that are in the Hougang report such as pen, ping, or hu. Sample size i s s u f f i c i e n t to examine variation i n size of guan jars, class eight (N=ll), at Baiying by a scatterplot (Appendix A, Figure 6). One pot i s d i s t i n c t l y larger than the others, but other size classes are not clear. In the Lujiakou report, the term "pen" i s used to refer to a wide range of shapes, including open forms and jars . Sample size i s s u f f i c i e n t to examine three types of open forms at s i t e s : pingdipen basin, wan bowl, and quanzupan pedestalled dish. Four ratios are thought to express the c r i t e r i a used in the Hougang report to i d e n t i f y two subtypes of pingdipen basin (N=24). A Mann-Whitney test using "known" vessels indicates no s i g n i f i c a n t difference between the two groups. Several "unknown" basins may be added to the sample i n order to make a scatterplot with two r a t i o s , o r i f i c e diameter/height and o r i f i c e diameter/base diameter (Appendix A, Figure 7). For c l a s s i f i c a t i o n on the basis of vessel proportions for t h i s type of shape, ratios with o r i f i c e diameter are preferable to those with rim diameter, since rim form tends to have s t y l i s t i c rather than formal- functional significance (Henrickson and McDonald 1983:635). The scatterplot indicates no d i s t i n c t clustering of vessels to j u s t i f y d i v i s i o n into subclasses. Another scatterplot, of rim diameter with 123 height, shows that there are c l e a r l y two sizes of basins at Hougang: one pot i s d i s t i n c t l y larger than the rest (Appendix A, Figure 8). A similar scatterplot for the pingdipen basins at Baiying (N=7) seems to id e n t i f y two clusters of vessels (Appendix A, Figure 9). However, a Mann-Whitney test indicates no sig n i f i c a n t difference between groups. I t also shows that one vessel i s d i s t i n c t l y d ifferent i n terms of o r i f i c e diamter/height. This i s a function of size of o r i f i c e , as a separate scatterplot for differences among vessels i n terms of size indicates (Appendix A, Figure 10). It i s also possible to examine morphological variation i n wan bowls from Hougang, Meishan, and Lujiakou. Two ratios are thought to express c r i t e r i a for identifying subtypes of bowls (A, B, C) at Hougang: rim diameter/base diameter and rim diameter/height. A Kruskal-Wallis test indicates that the three subtypes for 23 "known" pots are s i g n i f i c a n t l y different from one another i n terms of rim diameter/base diameter (p=0.003). When 21 "unknown" bowls are added to the sample (giving a t o t a l of N=44), and another scatterplot generated, i t can be seen that variation i s primarily in terms of th i s same r a t i o . However, the scatterplot shows no d i s t i n c t subgroups of bowls (Appendix A, Figure 11). Also, I found that s i g n i f i c a n t results with nonparametric tests could be achieved by assigning pots to more than one set of shape classes from the scatterplot. Therefore I see no j u s t i f i c a t i o n for dividing the bowls into subclasses. A separate scatterplot of rim 124 diameter with height shows that there i s a wide range of sizes of bowls at Hougang (Appendix A, Figure 12). Bowls at Meishan and at Lujiakou may be evaluated by the same ra t i o s . The scatterplot for Meishan (N=ll) shows no clear clustering of vessels (Appendix A, Figure 13). Again, there i s a wide range of variation i n terms of size, but no d i s t i n c t breaks i n the plot (Appendix A, Figure 14). A scatterplot combining fiv e "known" and 13 "unknown" bowls from Lujiakou shows one pot as c l e a r l y separated from the others. It i s assigned to a different class, since i t i s q u a l i t a t i v e l y d i f f e r e n t . There i s a wide range of sizes of bowls at thi s s i t e as well (Appendix A, Figures 15, 16). Two subtypes of quanzupan pedestalled dish (N=7) defined i n the Hougang report may be evaluated by examining variation i n one r a t i o , rim diameter/height of dish (RD/BHT). A Mann-Whitney test indicates no sig n i f i c a n t difference i n the two subtypes. However, my observations of these vessels and i l l u s t r a t i o n s i n the report indicate that two different subclasses should be defined i n terms of this r a t i o . There i s a clear break i n the range of values for the r a t i o to j u s t i f y establishment of two classes: 1) shallow or p l a t e - l i k e dish (N=2), and 2) r e l a t i v e l y deep dish (N=5). The Lujiakou report i d e n t i f i e s three subtypes (A, B, C) of ding tripods. These are evaluated using two ratios thought to express the c r i t e r i a stated i n the report: o r i f i c e diameter/maximum diameter and maximum diameter/height of o r i f i c e diameter. The scatterplot (N=7) 125 shows that the groups i d e n t i f i e d in the report are not d i s t i n c t l y d i fferent from one another (Appendix A, Figure 17). There i s no j u s t i f i c a t i o n for establishing other subgroups. A scatterplot of six bei cups at Baiying with rim diameter/height and rim diameter/base diameter shows that one pot i s separated from the others (Appendix A, Figure 18). Again, i t i s not ideal to define a separate shape class on the basis of only one vessel. However, i l l u s t r a t i o n s i n the report indicate that the cup i s q u a l i t a t i v e l y different from the others in terms of shape. F i n a l l y , a scatterplot (Appendix A, Figure 19) shows that one gai l i d at Baiying i s c l e a r l y larger i n size than the others (N=16, two with inadequate data for the p l o t ) . HYPOTHESES ABOUT VESSEL FUNCTION This section describes archaeological evidence for vessel function at Hougang, Baiying, Meishan, and Lujiakou, and at other Longshan Period s i t e s . There i s limited information on type of paste, residues from use, context of deposition, ethnographic analogy, and shape from Longshan sites -- c r i t e r i a that are commonly used in the West to id e n t i f y vessel function (Rice 1987). I t appears that f i v e general functional types of vessels are present in s i t e s : 1) cooking, 2) serving and eating food, 3) preparing, holding, and drinking alchoholic beverages, 4) storage, and 5) r i t u a l . The conclusions reached i n th i s 126 discussion are hypotheses that require t e s t i n g with adequate samples of vessels from i n d i v i d u a l assemblages. On the basis of h i s t o r i c a l analogy, the following major shape classes of vessels should have cooking as a primary function (see Table 11): ding t r i p o d , L i t r i p o d , xian t r i p o d , zeng perforated j a r , and b i grate. On the basis of archaeological data, three other forms were probably used f o r cooking as well during the Longshan Period: guan j a r , gui t r i p o d , and j i a t r i p o d . Therefore the t r a d i t i o n a l functional i n t e r p r e t a t i o n s f or these three forms may not be t o t a l l y accurate: guan as a storage j a r , gui as a serving pitcher, and j i a as a vessel for holding wine. Type of paste i s regarded by archaeologists i n China as the most important c r i t e r i o n f o r i d e n t i f y i n g cooking vessels (Yan, personal communication, 1987b). Vessels used f o r cooking should have a r e l a t i v e l y coarse paste ( j i a sha), rather than a f i n e paste (ni z h i , meaning no or small inclusions v i s i b l e ) . Presence of soot i s not always a r e l i a b l e i n d i c a t o r , since soot may not survive postdepositional processes ( i b i d ) . Archaeologists have associated cooking vessels with coarse paste f o r over 30 years; An Zhimin (1953:69) remarks that coarse texture of paste helps prevent cooking vessels from cracking when heated ( i . e . , i t helps prevent thermal shock). Zhou et a l . (1982:269) mention that laboratory analysis (apparently from t h i n sections) of sherds in N e o l i t h i c s i t e s indicates that cooking vessels have paste with g r i t i n c l u s i o n s . Few other classes of vessels have t h i s type of paste. 127 Ethnographic data support t h i s interpretation; among the Wa minority- group, coarser paste i s used for cooking vessels (Li Yangsong 1959:250). In the Hougang, Baiying, Meishan, and Lujiakou s i t e reports, the shape classes l i s t e d above tend to have coarse paste. In the guan jar and ding tripod classes, vessels may have coarse or fine paste. Small jars tend to have fine paste. Pots i n s i x classes of guan jars at Lujiakou have fine paste. Pots i n one class of ding at 3aiying and six classes at Lujiakou have fine paste. I t i s l i k e l y that cooking was not the primary function of these vessels. Ding tripods of very fine paste have been found at the Liangchengzhen s i t e as well (Wu 1938:68). My brief observations of vessels suggest that the terms " j i a sha" and "n_i z h i " incorporate a great deal of v a r i a b i l i t y . Examination of existing fractures on vessels from Hougang with a hand lens (16x) revealed a range of types of materials, grain sizes, and density of grains. S i m i l a r l y , Wu (1938:63-4) reports that there i s a wide range of paste texture at Chengziyai, from very g r i t t y to extremely f i n e . A systematic examination of paste composition for different shape classes could provide valuable information on ceramic function as well as di v e r s i t y of raw materials exploited. Some microscopic analysis of paste texture for Neolithic ceramics has been reported. Zhou et a l . (1982:269) f i n d a range of materials represented, including feldspar and quartz. I saw two guan jars from Hougang with large mica fragments. There i s variation for wan bowls at Hougang as well. I saw one bowl 128 with large inclusions, and another with large voids on the surface thay may represent organic temper. Soot has been found on the legs of gui tripods and on guan jars from Longshan sites (Yan, personal communication, 1987b). In par t i c u l a r , there i s soot on large guan jars at Hougang and on medium- sized guan jars at the Kexingzhuang s i t e i n Shaanxi (Zhang and Zhang 1986:47,49). I noted soot on 4 medium-sized guan jars and on 1 large jar at Hougang, as well as one medium-sized guan jar at Baiying. A layer of red burnt earth has been found on some large guan jars from sites belonging to the Wangyoufang cu l t u r a l branch i n eastern Henan (Zhang and Zhang 1986:47). This type of residue seems to be another indication that guan jars were used for cooking during the Longshan Period. At the Kexingzhuang s i t e in Shaanxi, 3 guan jars and one l i _ t ripod were found inside a house near a hearth. A guan jar at the Donghaiyu si t e i n Shandong was found i n a similar location (Institute of Archaeology, CASS 1984:83,103). I t i s not known whether the vessels were deposited through the process of discard after use or at the time of s i t e abandonment. The location of discard i s often a poor indicator of vessel function (Rice 1987:232-3). In either case, there are additional grounds for i n f e r r i n g function of guan ja r s , i . e . , coarse paste and presence of soot. There i s some evidence that guan jars had other functions during the Longshan Period as well. One i s transporting and/or holding water. 129 At the Jiangou s i t e i n Hebei, several guan jars were found at the bottom of a well (Institute of Archaeology, CASS 1984:84). They may have been dropped accidentally during use. Ethnographic analogy provides support for t h i s hypothesis: the Dai people of southern China use guan jars for water (Zhang J i 1959:490). Guan jars and other probable cooking vessels also served as b u r i a l urns for children. These vessels are common in Neolithic sites from the middle reaches of the Huanghe or Yellow River valley (Xu 1989). Ding tripods and xian tripods were used i n addition to guan during the Longshan Period. These vessels were used for domestic purposes before serving as b u r i a l urns; some vessels have soot on them (Xu 1989:334). My observations on b u r i a l urns from Hougang agree with these conclusions. There i s direct evidence for cooking with a vessel at one Longshan s i t e . A guan jar at Baiying contains animal bones (Sui 1988:51). Another type of residue that may be indicative of cooking i s water deposits inside vessels. One gui pitcher at Baiying has t h i s type of deposit (CPAM of Anyang D i s t r i c t , Henan Province 1983:17) and more than one at Lujiakou (Shandong Archaeological Team, IA, CASS, and the Art Museum of Weifang County, Shandong Province 1985). The gui at Lujiakou are solid-legged and r e l a t i v e l y t a l l . Other sites i n Shandong contain a second shape of gui, r e l a t i v e l y wide and with mammiform hollow legs (see Medley 1976:27). Pots of t h i s shape may not have been used for cooking. 130 Several forms of cooking vessels f o r preparing d i f f e r e n t types of foods should be expected (Rice 1987:237). For example, see ethnographic data described by Smith (1985:304) and M i l l e r (1985:58). Therefore several forms from Longshan s i t e s were probably used for cooking a v a r i e t y of foods. Unfortunately there i s a lack of data on subsistence from the Longshan Period, as discussed i n Chapter 2. Cooking jars l i k e guan with a f l a r e d rim are common in other areas of the world. The f l a r e d rims were probably designed with a functional purpose i n mind: for l i f t i n g pots from the f i r e (Woods 1985:168). It i s possible that some b i grates did not function as steaming racks, as commonly assumed. More than one shape and size has been found i n Longshan s i t e s . Both forms of M at Hougang would not have been very suitable f o r placement inside another vessel such as a t r i p o d steamer. For example, class two i s large with saw-tooth applique around the edge. In contrast, the bi_ grate from Baiying i s a small bowl-shaped pot with a perforated bottom. The l e i - b o grooved container may have been used for a s p e c i f i c type of food preparation. According to An Jinyuan (1986:345), these vessels were probably used f o r grinding tuberous plants for food or medicinal purposes. Also, people i n Hunan province continue to use these vessels. There i s more than one form of l e i - b o during the Longshan Period; some are s i m i l a r to pen, bo, and guan (An Jinyuan 1986, Ye 1989). 131 According to the t r a d i t i o n a l d e f i n i t i o n , dou stemmed dishes were used for serving food. This hypothesis i s plausible, as open vessel forms tend to be used for serving and eating food, since vessel contents would be e a s i l y accessible (Rice 1987:240; Henrickson and McDonald 1983:632; Howard 1981:9). Si m i l a r l y , i t i s commonly assumed that wan bowls were used for eating and drinking at meals. This interpretation i s feasible given the open form. The previous analyses show that more than one size of bowl i s present in Longshan s i t e s , probably indicative of more than one function. Different size classes of serving and eating vessels may also relate to size of the consuming group (Rice 1987:240). Small jars could have been used for drinking, too. For example, the Dai of southern China drink water from small hu jars (Zhang J i 1959:490). Other forms of vessels at Longshan sites could have been used for serving food. Food could have been served d i r e c t l y from cooking vessels, especially the tripods. These legged vessels could have been placed e a s i l y on most surfaces. Other r e l a t i v e l y open forms that are suitable for serving include bo bowl, bu container, quanzupan pedestalled dish, and pingdipen f l a t bottomed basin. Of course, these forms and others probably had more than one function. They could have held a variety of substances, including water. I suggest that several forms of gai l i d s at sites were probably used for serving food. Many of these vessels are too large (in terms of rim diameter and height) and too heavy to cover the openings of other vessels such as cooking pots (see Appendix A, Tables 45-48). Only two 132 forms would have been suitable for covering other pots: 1) small, f l a t l i d and 2) small, wan bowl-shaped l i d . Also, some forms are small enough to feasibly cover the o r i f i c e of other vessels, but are elaborate in shape, indicating that they could have been used for serving as well. A number of forms could have been used for holding and/or transporting water, as i n r u r a l India (see M i l l e r 1985:61) or Guatemala (Lischka 1978). For example, necked jars such as ping would have been suitable for l i q u i d s in general. Three forms from late Neolithic sites are commonly assumed to represent vessels for holding or serving alchohol: j i a tripod (previously discussed), zun container, and hu ja r . There has been l i t t l e e f f o r t among western archaeologists to ide n t i f y vessels used for preparing or serving alchohol, or other archaeological evidence for these a c t i v i t i e s . A recent exception i s Moore (1989). This topic has received attention in the Chinese archaeological and ethnographic l i t e r a t u r e . Two writers debate the origins of making alchohol in the Neolithic period. Fang (1964) argues that several forms recorded as used for alchohol during the h i s t o r i c period f i r s t appear in Longshan s i t e s , including zun, l e i , and t a l l stemmed cups, and therefore, represent the onset of alchohol consumption. L i Yangsong (1962) points out that alchohol production probably began at an e a r l i e r date during the Neolithic Period. People could have made alchoholic drinks from more than one grain. 133 More than one form of vessel may have been used for preparing, holding, and drinking alchohol. Any r e l a t i v e l y large cooking vessel could have been used for brewing and fermenting. In the New World, beer i s brewed i n cooking jars in t r a d i t i o n a l societies (Bankes 1985:272; DeBoer 1974:336). Storage jars can be used for brewing (Arnold 1985:150). However, vessels for storing alchohol (on a long or short term basis) tend to have narrow openings (Arnold 1985:150, Chavez 1985:163). Small o r i f i c e diameters or neck diameters should be expected, to prevent the fermented liquids from going f l a t . Therefore the necked jars from Longshan sites would have been suitable for storing alchohol: weng, ping, hu, and l e i . Weng jars have the narrowest necks. Straining the cooked grain before fermentation could have been accomplished by means of a cloth (Chavez 1985:165) or a strainer made of wood or pottery with small holes. I t seems that the holes i n M grates are too large for t h i s purpose. Strainers for making alchohol may yet be found i n Longshan s i t e s . According to Wang (1986:268-9), a huge zun container from the e a r l i e r Neolithic (Dawenkou Culture) s i t e of Lingyanghe in Shandong has a picture of a large strainer for making alchohol. Wang expects that weng jars were used for holding alchohol during the Dawenkou cu l t u r a l period. Alchohol may be drunk from vessels used for d a i l y meals or vessels used solely for alchohol. For example, among the Shipibo-Conibo, beer mugs of a d i s t i n c t shape are used (DeBoer 1974:336). In the Philippines, both types of drinking vessels are used (Solheim 1965:258). 134 Therefore i t i s not possible to conclude with certainty whether a given form of cup was used for drinking alchohol. One indication may be unusually small capacity, rendering i t more feasible to drink alchohol than other l i q u i d s . L i Jianmin (1984:66) proposes that cups that could only hold a small amount of l i q u i d were used for drinking alchohol during the Dawenkou Period i n Shandong. Any cup (or bowl) from Longshan sites could have been used for drinking alchohol: bei or gu, as well as small vessels of other forms such as jars. Vessels used for r i t u a l purposes are perhaps the most d i f f i c u l t to ide n t i f y , especially when primarily morphological data are available. Simple forms used on a daily basis such as a plate, bowl, or water container can be used i n a r i t u a l context as well, as i n present day Kathmandu and Mexico (Birmingham 1975:372; Lischka 1978:230). However, an unusual form may signify a r i t u a l function. David and Hennig (1972:8) describe a bowl with a unique shape used by the Fulani for ceremonial ablutions. Arnold (1985:159) describes more than one unusual form used i n r i t u a l s . Two unusual forms of vessels i n Longshan s i t e s , quanzupan pedestalled dish and zuo stand, may have been used for r i t u a l purposes. CONCLUSIONS The analyses described i n t h i s chapter suggest that the t r a d i t i o n a l terms used for designating shape classes should be regarded 135 as generic and as s t a r t i n g pointing points f o r morphological c l a s s i f i c a t i o n . Some authors use a term to describe one s p e c i f i c form. Others use subclasses such as "A" and "B" to designate s p e c i f i c forms. In reports, there are a v a r i e t y of formats f o r describing ceramic data, too. Most reports describe data that are p a r t i c u l a r l y important f o r establishment of culture h i s t o r y . Archaeological data on function support some assumptions from the t r a d i t i o n a l terms and do not support others. A p r i o r i t y i n future research should be to systematically investigate ceramic function at i n d i v i d u a l s i t e s , u t i l i z i n g more than one l i n e of archaeological evidence. 136 CHAPTER 5. TEST OF THE MODEL INTRODUCTION This chapter describes the tes t of the model of change in production of pottery i n r e l a t i o n to increasing c u l t u r a l complexity, based on the model by Rice (1981). In the ceramic analyses I use shape classes from Hougang, Baiying, Meishan, and Lujiakou that were defined i n Chapter 4. The f i r s t part of t h i s chapter i s a des c r i p t i o n of the test of change i n strategy of producers, i . e . , s i m p l i f i c a t i o n , d i v e r s i f i c a t i o n , or conservatism. There are three analyses: 1) diver- s i t y of shape classes by phase, 2) dimensional standardization, and 3) within-class standardization i n terms of secondary shape features and techniques of decoration. In addition, proposed evidence f o r r e l a t i v e lack of labor input i n production (shaping and decorating) i s discussed. The second part describes the tes t of change in labor-intensive prestige vessels that could have been used i n displays of status at i n t e r - household s o c i a l events. The model proposes that changes i n consumer demand f o r these vessels has an impact on production. Vessels are i d e n t i f i e d that exhibit a r e l a t i v e l y high degree of labor input: having large s i z e , elaborate form, th i n walls and/or p o l i s h i n g . In b r i e f , the analyses suggest that the patterns of change i n ceramic production at Baiying and Lujiakou d i f f e r from the patterns f o r 137 Hougang and Meishan. There i s more evidence for d i v e r s i f i c a t i o n in production at the former two s i t e s . However, there are some s i m i l a r i t i e s between Hougang and Baiying with respect to d i v e r s i f i c a t i o n of labor-intensive wares. In contrast to the other three s i t e s , a pattern of conservatism resulted from most analyses f o r the westernmost s i t e of Meishan. For each s i t e , more than one strategy of production i s evident. The model of change i n production i n r e l a t i o n to increasing c u l t u r a l complexity i s p a r t i a l l y supported. Sample sizes for each analysis are small, and r e s u l t s should be regarded as hypotheses that require more adequate t e s t i n g i n the future. There i s some evidence for d i v e r s i f i c a t i o n i n production over time at three of the four s i t e s , with respect to v a r i e t y of shape classes produced (two s i t e s ) and labor- intensive prestige wares. Although information on labor-intensive wares i s e s p e c i a l l y l i m i t e d , the a v a i l a b l e data (for Hougang, Baiying, and Lujiakou) appear to support the hypothesis by Rice (1981) that there should be increasing v a r i e t i e s of prestige wares. However, data on dimensional standardization and within-class standardization do not support her hypothesis that there should be increasing standardization of wares as c u l t u r a l complexity "increases. 138 ANALYSIS OF CHANGE IN PRODUCTION: STRATEGIES OF PRODUCERS D i v e r s i t y of Shape Classes An increase i n d i v e r s i t y of shape classes over time at s i t e s represents the process of d i v e r s i f i c a t i o n , whereby potters respond to consumer demand for an increasing v a r i e t y of wares. A decrease i n d i v e r s i t y of shape classes over time represents s i m p l i f i c a t i o n . Reducing the v a r i e t y of wares i s one method to increase e f f i c i e n c y i n production. A sharp decline i n v a r i e t y of shape classes made over time could s i g n i f y a fa c t o r causing stress on a c u l t u r a l system such as increased warfare, famine, or environmental degradation. Lack of change i n quantity of shape classes produced over time indicates conservatism, or resistance to change. The comparison of change i n quantity of shape classes i n each period r e s u l t s i n a clear pattern of d i v e r s i f i c a t i o n over time f o r Lujiakou. The pattern f or Meishan i s conservatism. For Hougang, there i s evidence f o r s i m p l i f i c a t i o n from one phase to another. For Baiying, there i s d i v e r s i f i c a t i o n from the Early to Middle Period and conservatism from the Middle to Late Period. For a complete l i s t of forms per phase at each s i t e see Appendix B (Tables 49-52). For each s i t e , the t o t a l number of shape classes per phase i s divided by the number of excavated houses per phase (Table 17). This i s a rough method 139 Table 17. Comparison of Total Number of Shape Classes Made During Each Phase at Hougang, Baiying, Meishan, and Lujiakou. Hougang: Baiying Early 31/2 houses =15.50 forms per house Early 15/9 houses = 1.67 forms per house Middle 32/14 houses = 2.29 forms per house Middle 22/8 houses = 2.75 forms per house Late 30/23 houses = 1.30 forms per house Late 40/46 houses = 0.87 forms per house Meishan Lujiakou Early 27/17 houses = 1.59 forms per house Early 20/6 houses = 3.33 forms per house Late 26/16 houses = 1.63 forms per house Late 47/5 houses = 9.40 forms per house note: dividing the t o t a l number of shape classes by number of excavated houses provides a rough method of standardizing values according to the t o t a l area excavated per period 140 of standardizing values for quantity of shape classes by volume of earth excavated for each phase, since the relevant information i s not provided i n the s i t e reports. From the data in Table 17, i t appears that there i s marked s i m p l i f i c a t i o n from the Early ( t r a n s i t i o n a l to the Longshan Period) to the Middle Period at Hougang. However, a s c a t t e r p l o t depicting quantity of houses per phase by quantity of forms per phase would indicate a modest degree of s i m p l i f i c a t i o n over time (R.G. Matson, personal communication, 1990). This method i l l u s t r a t e s how sample size a f f e c t s d i v e r s i t y ( i b i d ) . There i s further support f o r the pattern of modest s i m p l i f i c a t i o n from the Early to Middle Period at Hougang i n that the analyses of dimensional standardization, within-class standardization, and labor input do not r e s u l t in a pattern of marked s i m p l i f i c a t i o n . The pattern of modest s i m p l i f i c a t i o n f or the Middle to Late Period t r a n s i t i o n at Hougang indicated i n Table 17 would be evident on a s c a t t e r p l o t . For Baiying, both Table 17 and a s c a t t e r p l o t would indicate a pattern of modest d i v e r s i f i c a t i o n from the Early to Middle Period. Although Table 17 suggests a pattern of s i m p l i f i c a t i o n from the Middle to Late Period at Baiying, a scatterplot would indicate a pattern of conservatism ( i b i d ) . Comparing quantity of forms at Baiying i s d i f f i c u l t given the great d i f f e r e n c e i n sample size from the Middle to Late Period. Scatterplots would indicate the patterns evident i n Table 17 for Lujiakou and Meishan. 141 Comparison of d i v e r s i t y i n number of forms per functional class over time i s not as straightforward (Table 18; for details on sp e c i f i c shape classes at each s i t e see Appendix B, Tables 53-56). For example, i t i s not clear whether there i s a decline over time i n quantity of shape classes made in every hypothesized functional category at Hougang, or whether only a few functional categories are involved. S i m i l a r l y , i t i s not known whether there i s indeed an increase in forms of cups produced over time at Hougang and at Baiying, or open forms at Baiying. On the basis of data regarding quantity of houses excavated from each phase, r e l a t i v e l y small areas were dug for the early periods at Hougang and Baiying. Reports do not specify the volume of earth excavated per period. Dividing the figures i n Table 18 by quantity of houses per phase as an estimate for volume of excavated earth results i n figures with values less than 1.0. Although t h i s method of comparison i s certainly not ideal, the figures suggest that for every inferred functional category at Hougang and Baiying, quantity of shape classes does not decline over time. From ca. 2500-2200 B.C. (Middle to Late Period at Hougang, Early to Middle at Baiying), there may be an increase rather than decrease i n d i v e r s i t y of a few forms. For example, there may be an increase i n types of bowls at Hougang and bowls, cups, l i d s for serving food, and small pitchers at Baiying. Since the quantity of houses excavated at Meishan and at Lujiakou per phase i s approximately the same, the figures in Table 18 constitute an accurate basis for comparison. One change at Meishan i s r e l a t i v e l y 142 Table 18. Change Over Time i n Quantity of Shape Classes By Hypothesized Functional Category at Hougang, Baiying, Meishan, and Lujiakou. note: values i n parentheses are derived from d i v i s i o n by number of houses per phase, as a rough method of standard- i z i n g by amount of earth excavated per phase. An approx- imately equal number of houses per phase was excavated at Meishan and Lujiakou. "E"= Early, "M"= Middle, "L"= Late hypothesized Hougang Baiying Meishan Lujiakou functional category cooking tripods E 6 E 4 E 3 E 2 (3.0) (0.44) M 5 M 3 (0.36) (0.38) L 6 L 6 L 3 L 3 (0.26) (0.13) other cooking E 4 pots (2.0) M 4 (0.29) L 2 (0.09) E 0 E 3 E l (0) M 1 (0.13) L 3 L 3 L 2 (0.07) other jars with E 4 E 0 wide o r i f i c e for (2.0) (0) storage, long or short term M 6 M 4 (0.43) (0.50) (0.17) (0.09) L 2 143 j a r s , narrow E 2 E l o r i f i c e , f o r (1.0) (0.11) l i q u i d s M 3 M l (0.21) (0.13) L 3 L 3 L 3 L 6 (0.13) (0.07) bowls, for E 4 E l eating, drinking, (2.00) (0.11) serving M l M 2 (0.07) (0.25) L 5 L 2 (0.22) (0.04) other open forms, no pedestal or stem; for preparing, serving food L 2 L 6 L I L 1 2 (0.09) (0.13) E 2 E 3 E 4 E 4 (1.0) (0.33) M 3 M 3 (0.21) (0.38) open forms with E 5 E 0 pedestal or stem (2.50) (0) fo r serving, eating food M 5 M 0 (0.36) (0) L 4 L 2 L 2 L l l (0.17) (0.04) 144 cups f o r E O E l E 4 E l d r i n k i n g (0) (0.11) M l M 2 (0.07) (0.25) L 2 L 2 L 5 L 3 (0.09) (0.11) l i d s f o r c o v e r i n g E O E l E 0 E l o t h e r v e s s e l s (0) (0.11) M l M l (0.07) (0.13) L 0 L I L 0 L 2 (0) (0.02) l i d s f o r s e r v i n g E 3 E 2 E 2 E 0 f o o d (1.50) (0.22) M 3 M 3 (0.21) (0.38) L 2 L 6 L 2 L 2 (0.09) (0.13) s m a l l p i t c h e r s E 0 E l E 0 E 0 (0) (0.11) M 0 M 2 (0) (0.25) L 0 L I L 0 L 0 (0) (0.02) 145 d i s t i n c t , a decline in quantity of open forms with no pedestal or stem. At Lujiakou, there i s a marked increase i n quantity of open forms produced over time, as well as an increase in nearly every i n f e r r e d functional category. Comparisons of d i v e r s i t y of forms produced over time at s i t e s must take three factors into account: percentage of s i t e area excavated, ceramic u s e - l i f e , and depositional behavior. In addition, formats for describing ceramic data i n reports must be considered. For example, i t i s important to discuss whether some functional categories of vessels could be under-represented at s i t e s . One reason for the r e l a t i v e l y large number of shape classes recovered at Baiying may be the r e l a t i v e l y high percentage of s i t e area excavated (Table 19, see also Table 4, Chapter 1). I suggest l a t e r that a c u l t u r a l f a c t o r , i . e . , s o c i a l demand, may be responsible for the pattern as well. As discussed i n Chapter 1, the low quantity of vessels reported at Baiying i n comparison to Hougang i s probably due to the fact that fewer vessels were reconstructed. There i s too l i t t l e information about Meishan to determine whether the r e l a t i v e l y low number of shape classes i s a r e s u l t of small excavated area. The small quantity of vessels recovered from Lujiakou may be explainable by the low percentage of s i t e area excavated. One reason f o r the r e l a t i v e l y small percentage of non-tripod cooking pots at Baiying may be that "nicer looking" vessels were a p r i o r i t y i n reconstruction. The low number of bowls reported at Baiying 146 Table 19. Quantities of Forms by Hypothesized Functional Category at Hougang, Baiying, and Lujiakou. (note: there i s i n s u f f i c i e n t information f or Meishan) functional category Hougang Baiying Lujiakou cooking tripods 59 pots (13.0%) 38 pots (20.3%) 6 pots (5.0%) other cooking pots 115 pots (25.3%) 18 pots (9.6%) 11 pots (9.2%) other jars with wide o r i f i c e , f o r storage 45 pots (9.9%) 16 pots (8.6%) 5 pots (4.2%) jars with narrow o r i f i c e , f o r l i q u i d storage 35 pots (7.7%) 10 pots (5.4%) 11 pots (9.2%) bowls, f o r eating, drinking, serving 84 pots (18.5%) 8 pots (4.3%) 28 pots (23.3%) other open forms, no pedestal or stem, f o r preparing, serving food 34 pots (7.5%) 34 pots (18.2%) 37 pots (30.8%) open forms with stem or pedestal, f or eating, serving food 38 pots (8.4%) 9 pots (4.8%) 11 pots (9.2%) cups f o r drinking 8 pots (1.2%) 14 pots (7.5%) 6 pots (5.0%) 147 small p i t c h e r s f o r serving l i q u i d s 0 5 pots (2.7%) 0 l i d s f o r covering other v e s s e l s 1 pots 17 pots 4 pots (0.2%) (9.1%) (3.3%) l i d s f o r serving food 35 pots 17 pots 1 pot (7.7%) (9.1%) (0.8%) other forms (zuo stand) 1 pot 1 pot (0.2%) (0.5%) TOTAL NUMBER OF FORMS TOTAL NUMBER OF POTS percentage of s i t e area excavated 44 454 1.81% 61 188 5.45^ 49 134 0. 91% 148 i s misleading, because one form of gai l i d (for covering other vessels) i s equivalent to wan bowls at Hougang. The r e l a t i v e l y low number of cooking vessels at Lujiakou i s explainable by the small area excavated. Quantities of cooking pots at Hougang are probably a more accurate representation of the systematic context than quantities at the other s i t e s . Ethnoarchaeological studies indicate that t r a d i t i o n a l societies tend to produce a r e l a t i v e l y large number of cooking vessel forms, and in large quantities (Rice 1987:295). These vessels often break during use (Rice 1987:297-9), and they are replaced r e l a t i v e l y quickly (Rice 1987:303). Necked vessels that may have been used for storing and serving alchohol such as hu, weng, and l e i are not common at s i t e s . The low number of forms and vessels may not represent the frequency with which drinking events took place. In the Philippines, for example, pots for alchohol tend to have a long life-span and are replaced less often. Consumers desire old vessels that enhance the flavor of the alchohol (Solheim 1965:256). According to one member of the Kalinga, wine jars " w i l l l a s t forever i f you are careful" (Longacre 1981:63). The kinds and quantities of cups and pitchers produced at sites may have been higher than Table 19 indicates. These forms tend to have r e l a t i v e l y t h i n walls and break ea s i l y . Also, thin-walled pots tend to break into many pieces (Rice 1987:291). These factors, combined with r e l a t i v e l y small size, may have inhibited discovery at sites and reconstruction of certain types of vessels by archaeologists. 149 Prestige vessels used for social display must be under-represented at s i t e s . Hougang, Baiying, Meishan, and Lujiakou do not represent "Pompei" type archaeological s i t e s . Instead, they appear to represent gradual abandonment with no intention of returning, a process described by Deal (1983:210) in his ethnoarchaeological study of discard behavior. Few vessels were found in houses at these s i t e s , as mentioned i n Chapter 1 (see Table 5). The majority were found in test or open excavation areas with no c u l t u r a l features. These vessels may represent dispersed or broadcast dumping, and the vessels i n p i t s , dumping in discrete areas (see Deal 1983:198). Vessels with r e l a t i v e l y high value would have been taken by the occupants at the time of abandonment (Deal 1983:210). Some patterns evident i n Tables 18 and 19 may be explained by cu l t u r a l factors. At Baiying and Lujiakou, the two easternmost s i t e s , there are several forms of vessels that could have been used i n displays of status at inter-household events. At Baiying and at Lujiakou, there i s a higher percentage of cups i n r e l a t i o n to other forms than at Hougang (Table 19). The possible increase in d i v e r s i t y of shape classes at Baiying from the Early to Middle Period (Table 18) involves forms that could have been used in social displays such as cups, l i d s for serving food, and small pitchers. S i m i l a r l y , the marked increase i n quantity of open forms over time at Lujiakou may have been caused by increased soc i a l demand for display vessels. Consumers attempting to enhance t h e i r positions of status may demand greater v a r i e t i e s of 150 vessels that can be used for the same function, a process that M i l l e r (1982) observes in India. There may be inaccuracies due to my methods of interpreting the formats of describing vessels in reports. In the Hougang report, the number of vessels excavated per phase i s sometimes not clear. On the basis of patterning i n style numbers for individual xingzhuang or major shape classes, at times I had no choice but to estimate the number of vessels from each shape class per phase. In the report, low style numbers tend to be associated with the Early Period, and high style numbers, with the Late. The Lujiakou report i s especially unclear i n this regard. However, my interpretation of an increase i n variety of shape classes produced over time agrees with the conclusion reached by the Shandong Archaeological Team, IA, CASS and the Art Museum of Weifang County, Shandong Province (1985:348). Dimensional Standardization Sample size i s s u f f i c i e n t to assess change i n dimensional standardization with s t a t i s t i c a l techniques "for a few classes at Hougang, Meishan, and Lujiakou (Table 20). At Hougang, there are three or more vessels per phase i n four shape classes: large guan ja r s , medium-sized guan jars, medium-sized pingdipen basins, and wan bowls (class one). At Meishan, there i s an adequate sample of guan jars and 151 Table 20. Analysis of Dimensional Standardization. site Hougang form large guan jar sample Middle, N=8 Late, N=5 dimension OD BD MXD ODHT medium size guan jar pingdipen basin, medium-sized Early, N=3 Middle, N=7 Late, N=8 Early, N=5 Middle, N=5 Late, N=13 OD BD MXD ODHT OD BD HT wan bowl, class one medium-sized Early, N=8 Middle, N=10 Late, N=12 RD BD HT Meishan Lujiakou guan j a r ding t r i p o d ding t r i p o d class seven Early, N=3 Late, N=5 Early, N=5 Late, N=6 Early, N=3 Late, N=4 OD OD MXD OD MXD BHT 152 ding tripods. There are enough ding tripods (class seven) for analysis at Lujiakou. For Hougang and Lujiakou, I use primarily measurements of vessels taken i n the f i e l d . For Meishan, a l l measurements are from scale drawings in the report. Since these data sets consist mainly of whole (reconstructed) vessels, i t i s possible to assess more than one dimension per vessel. For Meishan, however, most of the jars and tripods i l l u s t r a t e d i n the report are fragmentary. Therefore fewer dimensions could be analyzed. The small number of vessels available for analysis precludes s t a t i s t i c a l techniques used i n other studies of ceramic production such as c o e f f i c i e n t of variation (Longacre et a l . 1988, T o l l 1981) and histograms with parametric significance tests (Davis and Lewis 1985, Hagstrum 1986). I use a method more suitable for small samples: inspection of boxplots depicting the range of variation (interquartile range) for individual measurements over time. The i n t e r q u a r t i l e range i s more appropriate than the t o t a l range for comparative purposes since i t i s more resistant to sample size. The boxplots were constructed by SYSTAT and SYGRAPH (Wilkinson 1988a, 1988b). A s i g n i f i c a n t reduction over time i n the range of values for a given measurement would indicate an increase i n dimensional standard- i z a t i o n . This pattern would represent the process of s i m p l i f i c a t i o n i n production, or efforts to increase ef f i c i e n c y . Conversely, a s i g n i f i c a n t increase i n range of values would indicate a decrease i n 153 dimensional standardization over time, or an increase i n morphological d i v e r s i t y for a shape class. The analysis of each shape class assesses change i n major dimensions. For jars, basins, and tripods, I use o r i f i c e diameter rather than rim diameter. On the basis of ethnoarchaeological data, size of rim can vary substantially on vessels, even those made by spe c i a l i s t s (Rice 1988:6). Analyses of dimensional standardization should not be conducted unless variation i n size can be assessed. If different size classes are not recognized, a n a l y t i c a l results may be inaccurate (Rice 1988:3). A recent ethnoarchaeological study indicates that i n this type of situat i o n , variation i n dimensions i s actually less than an analysis would show (Longacre et a l . 1988:108). For t h i s reason, i t i s not appropriate to conduct analysis of dimensional standardization on wan bowls (class one, N=12) at Meishan or those at Lujiakou (class one, N=17). Scatterplots of height by rim diameter (Appendix A, Figures 14, 16) show that there i s considerable variation in size of bowls at each s i t e . Sample size i s too small i n each case to i d e n t i f y d i s t i n c t size classes with confidence. The scatterplots show that there are probably three to four size classes of bowls at each s i t e . Creation of size classes from these samples would make analysis of dimensional standardization impossible, since sample size would be too small. 154 However, i t i s possible to subdivide the sample of bowls from Hougang (N=44) and result i n an adequate sample for analysis. Although the scatterplot i n Appendix A (Figure 12) for these bowls does not indicate clear breaks in the d i s t r i b u t i o n , the wide range of values for rim diameter indicates that size classes should be defined. On the basis of the scatterplot, a stem and leaf display of values for rim diameter, and my observations i n the f i e l d , I define three size classes of bowls at Hougang: small (N=ll, rim diameter less than or equal to 9.0 cm), large (N=3, rim diameter more than or equal to 18.9 cm), and medium (N=30, with intermediary values for rim diameter). Sample size i s s u f f i c i e n t to use the medium-sized vessels i n analysis of dimensional standardization. For each shape class tested, the boxplots do not indicate a clear pattern of increasing standardization or reduction i n i n t e r q u a r t i l e range, for any dimension (Appendix B, Figures 20-26). Indeed, some boxplots suggest the converse, or increasing d i v e r s i t y i n range of values. Larger samples of vessels are necessary to determine whether there i s support for a pattern of d i v e r s i f i c a t i o n . A change i n degree of standardization should be evident from boxplots of more than one dimension for a given shape class. The small sample of vessels analyzed here suggests that potters adopted a strategy of conservatism during the Longshan Period, since there i s no evidence for a clear change i n standardization over time. 155 A pattern of conservatism i s explainable in l i g h t of the inferred functional categories of vessels examined. Sample size was s u f f i c i e n t to examine cooking vessels (large and medium-sized guan ja r s , ding tripods), basins (pingdipen) that could have been used for a wide variety of purposes such as serving or preparing food or l i q u i d s , and wan bowls for eating and drinking. On the whole, these are common shape classes at s i t e s . They would have been among the set of vessels used most often by people. The results here support the prediction made by Rice (1984:245-6) that there w i l l be lack of change in production of wares used for basic needs. Consumers do not demand changes i n vessel forms that are important in d a i l y a c t i v i t i e s and that already function adequately. Also, there may have been less consumer demand for changes in production of vessels used i n private (household) contexts. Thus, there i s no indication that competition between potters, i n response to increasing population size and density, i n t e n s i f i e d over time, r e s u l t i n g i n efforts to produce wares more e f f i c i e n t l y . Increasing dimensional standardization i s one indicator of change in mode of production, as discussed i n Chapter 3. Other categories of data described i n Chapter 6 do not indicate change i n mode of production over time, either. However, below I suggest that there was concern with e f f i c i e n c y i n production of these shape classes. Some vessels appear to have evidence for r e l a t i v e lack of labor input i n shaping and decorating. Thus, 156 vessels i n these shape classes, were, on the whole, non-prestige wares (as defined i n t h i s study). Relative Lack of Labor Input in Production At Hougang and Lujiakou, there are three types of evidence that may indicate r e l a t i v e lack of labor input i n production: shaping, decorating, and f i r i n g (Tables 21-22). Some wan bowls from each period at Hougang have evidence f o r lack of labor input i n shaping. There i s a lump i n the center of the i n t e r i o r base; potters did not take time to smooth i t over. Since reconstruction of vessels with p l a s t e r probably often covers the evidence, i t i s not possible to i d e n t i f y whether there i s an increase or decrease i n the number of bowls of t h i s type over time. Some ding tripods at Lujiakou have incomplete i n c i s e d l i n e s that should have continued around the circumference of vessels (in comparison to i n c i s e d l i n e s on other vessels), and an uneven surface color i n d i c a t i v e of r e l a t i v e lack of e f f o r t i n f i r i n g . There i s addit i o n a l evidence for r e l a t i v e lack of labor input i n production at Hougang f o r large guan jars and medium-sized pingdipen basins. I saw one large jar from the Late Period and two basins (one from the Middle Period, one from the Late) with incompletely i n c i s e d l i n e s . In the Hougang and Meishan reports, there i s information on change over time i n q u a l i t y of decorative techniques f o r a l l vessels with 157 Table 21. Suggested Evidence f o r Relative Lack of Labor Input i n Vessel Forming f o r Wan Bowls, Class One, Hougang. lump of clay i n center of i n t e r i o r base Early Middle Late present absent 6 13 t o t a l number of vessels 10 15 19 Table 22. Suggested Evidence f o r Relative Lack of Labor Input i n Decoration and F i r i n g f o r Ding Tripods, Class Seven, Lujiakou. type of evidence Early Late uneven f i r i n g present absent uncompleted i n c i s e d l i n e s present absent t o t a l number of vessels 1 5 8 impressed decoration. Impressed decoration i s placed with more care i n the Early Period at Hougang, and i t i s neatly applied and r e l a t i v e l y deep. Pots from the Middle and Late Periods have r e l a t i v e l y crude, more shallow impressed decoration (Anyang Archaeological Team, IA, CASS 1985:79, 81). Thus i t appears that by the Middle Period, potters decided to spend less time in decorating guan jars presumably f o r reasons of e f f i c i e n c y , even though they d i d not form them more e f f i c i e n t l y . A s i m i l a r trend in q u a l i t y of impressed decoration over time i s present among the Meishan vessels. In the Early Period, impressed decoration i s r e l a t i v e l y neat and deep. By the Late Period, t h i s type of decoration i s not as c a r e f u l l y applied (Second Henan Archaeological Team, IA, CASS 1982:472-3). In addition, vessel shapes are said to have become "les s orderly" over time at Meishan, perhaps i n d i c a t i n g that they were produced more quickly than previously. There are unevenly and low f i r e d vessels i n the Late Period as well (Second Henan Archaeological Team, IA, CASS 1982:447, 473). The change i n r e l a t i v e care i n execution of impressed decoration from the Early to Middle Period at Hougang correlates with the marked decline in d i v e r s i t y of shape classes. There i s no evidence f o r extreme stress on the c u l t u r a l system. A more p l a u s i b l e explanation i s that vessels with impressed decoration, and perhaps other classes as well, were valued l e s s by consumers over time. Some vessels may have been used f o r d i s p l a y during the Middle and Late periods at Hougang, as 159 discussed l a t e r i n this chapter. These vessels do not have impressed decoration. The decline i n variety of shape classes from the Early to Middle Period at Hougang may represent the opposite process that M i l l e r (1982) describes. A decrease i n variety of shape classes over time may signal a decline i n demand for a variety of vessels for use i n social displays. The more widespread decline i n care i n production by the Late Period at Meishan may indicate that pottery vessels, as a class of materials, were considered less valuable than previously, p a r t i c u l a r l y for social display at inter-household events. Within-class Standardization This section describes variation i n secondary shape features and techniques of decoration within shape classes. An increase i n var i e t i e s of decorative techniques over time, for example, indicates d i v e r s i f i - cation. Conversely, a decrease i n var i e t i e s of decorative techniques s i g n i f i e s s i m p l i f i c a t i o n . Although there i s r e l a t i v e l y l i t t l e relevant information for each s i t e , the patterns that emerge tend to support the conclusions reached i n the analysis of change i n d i v e r s i t y of shape classes. Details on t h i s analysis are provided i n Appendix B, Tables 57-73. 160 For Hougang, there i s adequate information to assess d i v e r s i t y i n shape of leg for xian tripods and ding tripods, as well as handles for xian tripods. Again, i t i s necessary to standardize the figures by t o t a l number of houses excavated per phase, as a rough measure of volume of excavated earth per phase. Unfortunately, sample size i s so small (even when categories are lumped) that d i v i s i o n by quantity of houses results i n very small figures that are d i f f i c u l t to compare. Some patterns can be seen from the raw data for Hougang i n the tables (Appendix B, Tables 57-59). In more than one case, increasing d i v e r s i t y i n secondary shape features i s d i r e c t l y related to sample size (handles for xian tripods, for example). However, for shape of leg, xian tripods, there seems to be s i m p l i f i c a t i o n over time. Even though the number of excavated houses increases markedly from the Early to Middle and Late Periods, the t o t a l number of vari e t i e s i n leg shape does not. The same pattern may apply to shape of leg for ding tripods. Four comparisons of d i v e r s i t y i n decorative techniques over time at Hougang are possible: large guan j a r s , medium-sized guan jars, xian tripods, and wan bowls, class one (Appendix B, Tables 60-63). There appears to be a pattern of s i m p l i f i c a t i o n from the Middle to Late Period: large jars, xian tripods, and bowls. S i m i l a r l y , s i m p l i f i c a t i o n may characterize medium-sized jars from the Early to Middle Period. Two comparisons for Baiying suggest a pattern of s i m p l i f i c a t i o n from the Middle to Late Periods: handles and decorative techniques for ding tripods (Appendix B, Tables 64, 65). Although the number of 161 excavated houses increases markedly from the Middle to the Late phase (a jump from 8 to 46), the v a r i e t i e s of handles and decorative combinations do not. The analysis of d i v e r s i t y of shape classes indicated a pattern of s i m p l i f i c a t i o n from the Middle to Late Period at Baiying as well. Comparison between phases at Meishan and at Lujiakou i s more straightforward, since an approximately equal number of houses was excavated from each phase. For Meishan, each of four comparisons suggests the same pattern, conservatism over time: decorative techniques on guan jars, dou stemmed dishes, and ding tripods; and shape of leg f o r ding tripods (Appendix B, Tables 66-69). The same pattern was obtained, then, for the analyses of d i v e r s i t y i n shape classes and within-class standardization. For Lujiakou, three of four comparisons suggest a pattern of d i v e r s i f i c a t i o n , again an agreement of the f i r s t analysis conducted: 1) decorative techniques f o r ding tripods, 2) decorative techniques f o r wan bowls, and 3) handles for ding tripods (Appendix B, Tables 70-73). Cl e a r l y , l a rger sample sizes should be used to support or re j e c t these conclusions. It would be he l p f u l to cal c u l a t e d i v e r s i t y formally by a technique such as the Shannon-Weaver D i v e r s i t y Index used i n other studies of ceramic production (Rice 1981, T o l l 1981). Also, i t i s necessary to evaluate observed differences i n d i v e r s i t y by means of sig n i f i c a n c e t e s t s (Rice 1989:112). It would be worthwhile to assess d i v e r s i t y of other types of secondary shape features and decorative techniques on vessels from 162 Longshan s i t e s . When examining vessels i n China, I noted several other types of variation that could not be investigated further due to inadequate sample size. For example, guan jars (medium and large) at Hougang vary i n placement of impressed decoration. Impressions may cover the complete lower part of the body, or stop before the base. There i s also variation in quantity of incised lines that go around the circumference of vessels. Some jars from Hougang have what appears to be a l i d shelf. The report states that the inner edge of the rim on these vessels becomes a ridge (Anyang Archaeological Team, IA, CASS 1985:68). Lid shelves are known for large jars from the Wangyoufang region in Henan as well (Zhang and Zhang 1986:46-7). F i n a l l y , some variation i n type of paste may stem from social causes and be relevant to the model of ceramic change. White wares are rare and appear at approximately the same time period i n northern Henan, ca. 2300-2100 B.C.: during the Late Period at Hougang (Anyang Archaeological Team, IA, CASS 1985:56, 81) and the Middle Period at Baiying (CPAM of Anyang D i s t r i c t , Henan Province 1983:27). There i s very l i t t l e information on white wares i n the reports. One shape class made of t h i s ware (a k a o l i n i c clay) i s a small gui pitcher. Vessels made from t h i s q u a l i t a t i v e l y different and r e l a t i v e l y r e s t r i c t e d clay source may have been prestige wares. They are rare within the Hougang and Baiying assemblages. White wares had been used in Shandong since the Dawenkou Period (Underhill 1983). During the Longshan Period, people in the Hougang II region may have sought to 163 enhance t h e i r positions of status by introducing a new ware used in another region. The Late Formative Period i n the Valley of Guatemala may be an appropriate analogue. Rice (1977) shows that e l i t e s probably c o n t r o l l e d production and exchange of rare, f i n e l y made white wares. In the case of Hougang and Baiying, however, there i s no published information on context of deposition or exchange. Summary of Changes i n Strategies of Production Similar r e s u l t s were obtained i n three separate analyses: d i v e r s i t y of shape classes, dimensional standardization, and within- class standardization (Table 23). The westernmost s i t e , Meishan, i s characterized by a pattern of conservatism over time, or resistance to change, on the basis of a l l three analyses. The pattern f o r Hougang i s s i m p l i f i c a t i o n i n d i v e r s i t y of shape classes and within-class standardization, and conservatism i n dimensional standardization. There i s more than one pattern for Baiying. Unlike Hougang but s i m i l a r to Lujiakou i n Shandong, there i s a pattern of d i v e r s i f i c a t i o n f o r v a r i e t y of shape classes from the Early to the Middle Period. There i s no information on change i n dimensional standardization. There i s a more clear pattern of increase in v a r i e t y of shape classes produced over time 164 Table 23. Summary of Change i n Strategy of Pottery Production at Hougang, Baiying, Meishan, and Lujiakou. (note: "E"= Early, "M"= Middle, "L"= Late Period) s i t e and d i v e r s i t y dimensional within-class region of shape standard- standard- classes i z a t i o n i z a t i o n Hougang II Hougang E - M modest s i m p l i f i - cation conservatism for j a r s , bowls, basins s i m p l i f i c a t i o n ? (1 comparison) M - L Baiying E - M modest s i m p l i f i - cation modest d i v e r s i f i cation same as above no data s i m p l i f i c a t i o n ? (5 comparisons) conservatism? (2 comparisons) M - L conserva • tism no data s i m p l i f i c a t i o n ? (2 comparisons) Wangwan III Meishan E - L conserva- tism conservatism j a r s , tripods conservatism? (4 comparisons) Liangcheng Lujiakou E - L d i v e r s i f i - cation, e s p e c i a l l y open forms conservatism tripods d i v e r s i f i c a t i o n ? (3 of 4 comparisons) 165 at Lujiakou compared to the other three, more western s i t e s . But there i s a pattern of conservatism i n dimensional standardization. There are i n t e r e s t i n g implications of these r e s u l t s . One i s the v a r i a t i o n i n patterns for the d i f f e r e n t s i t e s , underscoring the point that there was regional d i v e r s i t y during the Longshan Period. There i s v a r i a t i o n within one region, Hougang II, as well. Each s i t e exhibits v a r i a t i o n , supporting the hypothesis of Rice (1984) that d i f f e r e n t components of ceramic production systems do not always change i n the same manner. S i m i l a r i t i e s in r e s u l t s for d i f f e r e n t analyses pertaining to one s i t e are encouraging, suggesting that the small samples of vessels examined here provided r e l a t i v e l y accurate patterns. The analysis of within-class standardization gave p a r t i c u l a r l y t entative r e s u l t s , but r e s u l t s which tend to support the analysis of d i v e r s i t y i n shape classes. S i m i l a r l y , there i s r e l a t i v e l y l i t t l e information a v a i l a b l e f o r the Early Period at Hougang, but two analyses gave s i m i l a r r e s u l t s . Meishan belongs to the Wangwan III type in western Henan, the region i n which state formation probably f i r s t took place. It i s commonly assumed that culture becomes inc r e a s i n g l y complex i n a l l subsystems during the process of state formation. However, the analyses conducted here suggest that the system of pottery production may not have changed s i g n i f i c a n t l y over time. There i s no c l e a r evidence f o r i n t e n s i f i c a t i o n or increasing e f f o r t s to produce vessels more e f f i c i e n t l y . The only possible i n d i c a t i o n of increasing e f f i c i e n c y i n 166 production i s the potential evidence for increasing incidence of lack of labor input. Si m i l a r l y , there i s no evidence for d i v e r s i f i c a t i o n , another important aspect of increasing specialization i n ceramic production (Rice 1981:220). A serious problem i s that there i s r e l a t i v e l y l i t t l e information on ceramic vessels at Meishan i n comparison to other s i t e s . It i s necessary to examine more vessels from the Wangwan I I I region to v e r i f y these findings. There i s a more complete picture of change i n ceramic production in the Hougang II region. A pattern of modest s i m p l i f i c a t i o n characterizes the change from the Early ( t r a n s i t i o n a l to the Longshan Period) to the Middle Period at Hougang. There may have been a decline in demand for a variety of vessels by consumers for social display. From the Early to Middle Period as wel l , i t appears that there was a decline i n labor input for applying impressed decoration. ) Analyses either indicate a pattern of si m p l i f i c a t i o n or conservatism for the Middle to Late Period at Hougang. For each s i t e , I suggest that the shape classes with the largest samples of vessels represent functional types that were common and regarded as non-prestige wares: xian and ding tripods, guan ja r s , and wan bowls. Some vessels at Hougang, for example, exhibit evidence of lack of labor input i n production. For each s i t e there were i n s u f f i c i e n t data to assess within-class standardization and dimensional standardization for other wares. The patterns summarized i n Table 23 may not characterize change i n 167 production of prestige wares. The case of Baiying indicates that patterns of change i n prestige wares may d i f f e r from non-prestige wares. Potters may have d i v e r s i f i e d production of vessels that could be used i n s o c i a l displays from the Early to the Middle Period: cups, serving l i d s , and p i t c h e r s . Some aspects of ceramic production i n the Hougang II region may have been caused by c u l t u r a l i n t e r a c t i o n with Shandong, such as d i v e r s i f i c a t i o n of shape classes that could be used for display at Baiying and the introduction of white wares at Hougang and Baiying. These suggested patterns, and those described i n the following section, appear to support the hypothesis proposed by Rice (1981) that d i v e r s i f i c a t i o n of ceramic production i n a context of increasing c u l t u r a l complexity should occur p r i m a r i l y i n the area of status-related vessels. ANALYSIS OF CHANGE IN LABOR-INTENSIVE (PRESTIGE) VESSELS The model suggests that two types of displays with containers were common in p r e h i s t o r i c chiefdoms, largesse and conspicuous consumption. Displays of largesse tend to be undertaken by higher status people who have resources to give food and drink away to others. Generosity i s an important method of maintaining and increasing supporters. Large vessels symbolize the generosity of the donors. Displays of largesse, 168 on a smaller scale, may be undertaken by f a m i l i e s at inter-household s o c i a l events such as marriage ceremonies and funerals. In displays of conspicuous consumption, people display personal consumption of food and drink with vessels that are elaborate i n shape, have t h i n walls, or have a number of decorative techniques. When there i s an increase i n status competition and pottery vessels are prestige objects, there should be an increase i n degree of v a r i a t i o n i n labor input among vessels. People may attempt to surpass the displays of others by acquiring: 1) the same types of vessels, but e x h i b i t i n g a higher degree of labor input, 2) vessels e x h i b i t i n g new types of labor-intensive techniques, and/or 3) acquiring an increasing v a r i e t y of shape classes e x h i b i t i n g labor-intensive techniques. Conversely, when the competitive system declines, the opposite pattern w i l l occur. It i s possible to i d e n t i f y the presence of four types of labor- intensive vessels at Hougang, Baiying, Meishan, and Lujiakou: 1) very large, 2) elaborate i n form, 3) thin-walled, and 4) polished (Table 24). Unfortunately, i t i s not possible to i d e n t i f y the t o t a l number of labor- intensive vessels at these s i t e s on the basis of information presented in s i t e reports. However, as reports tend to describe at l e a s t one vessel representing each ceramic category (Chapter 4), the v a r i e t i e s of labor-intensive vessels that are present can be i n f e r r e d . In my estimation, Table 24 l i s t s the t o t a l quantity of very large vessels that were excavated at s i t e s . It should l i s t most of the thin-walled and 169 Table 24. Labor-intensive Vessels That May Have Been Used For Display at Hougang, Baiying, Meishan, and Lujiakou. (note: "E"= Early, "M"= Middle, "L"= Late Period; *= several decorative techniques present on one vessel) s i t e and region very large in size t h i n - walled elaborate form polished surface Hougang II Hougang E: 2 zuo stands * l i d s f o r serving: 3 forms, 4 pots pots i n several shape classes M: 1 basin polished pingdipen 1 j a r guan (medium- sized) l i d s f o r serving: 3 forms, 5 pots pots i n several shape classes Baiying E: 1 zuo stand *, polished; 1 l i d , 1 basin polished 1 j ar guan (medium- sized) 1 cup bei l i d s f o r serving: 2 forms, 6 pots l i d s f o r serving: 2 forms, 2 pots pots i n several shape classes pots i n several shape classes M: 2 pitchers 1 cup bei l i d s f o r serving: 3 forms, 4 pots pots i n several shape classes 170 l i d s f o r serving: 6 forms, 7 pots W a n g w a n III M e i s h a n L i angcheng L u j i a k o u E: 1 cup bei 1 t r i p o d d ing E: h u jars (quantity not clear) L; l i d s f o r serving: 2 forms, 3 pots l i d s f o r serving: 2 forms, 3 pots gui pitchers with long spouts: 1 f orm, 3 pots gui pitchers with long spouts: 1 f orm, 1 pot; l i d s f o r serving: 1 form 1 pot ; 1 lei j ar same as above, and 1 * quanzu- pan with painted human figures pots i n several shape classes pots i n several shape classes pots i n several shape classes pots i n several shape classes 171 elaborately shaped vessels. There are probably several polished vessels in several shape classes at each s i t e , most of which are not described in reports. Two s i t e s , Hougang and Baiying, have r e l a t i v e l y large vessels. At both of these site s there are large zuo stands and one extremely large pingdipen basin. The basin at Baiying i s p a r t i c u l a r l y large, with a rim diameter of 77.4 cm. The basin at Hougang i s 60.7 cm i n rim diameter. At Baiying there i s also an extremely large l i d shaped l i k e a sombrero at 56.0 cm i n rim diameter. The scatterplots for size of basins at Hougang and Baiying (Appendix A, Figures 8, 10) i l l u s t r a t e the extent to which these vessels d i f f e r from others. The zuo stand at Baiying, completely reconstructed, i s 45.5 cm i n rim diameter and 29.4 cm in height. Many of these vessels also exhibit labor-intensive decorative techniques. The basins are highly polished, and the zuo stands have several techniques: painting, carved out holes, polishing, and i n c i s i n g . A l l of these vessels belong to the e a r l i e r Longshan Period: the Early and Middle Period at Hougang, and the Early Period at Baiying. These large vessels may represent display of largesse i n giving food or drink to other people. Large ceramic basins used for t h i s purpose are known in Peru (Tschopik 1950:215-6). I suggest that largesse was a social t a c t i c used at more than one community i n the Hougang II region during the e a r l i e r Longshan Period. There i s also one d i s t i n c t l y large guan jar from the Late Period at Baiying, with a height of 55.8 cm (Appendix A, Figure 6). It i s 172 possible that this vessel was also used in displays. However, jars of t h i s size are not an unusual occurrence i n northern Henan. At Hougang, gang jars are present in every period and are approximately the same height. They may have been used for storage i n private (household) rather than public contexts. As mentioned in the f i r s t section of t h i s chapter (Table 18), I suggest that there are several forms of l i d s too elaborate i n form and/or too large and heavy to function as covers of other vessels. These are present i n each period at Hougang, Baiying, Meishan, and Lujiakou. From the Late Period at Baiying, there are also two large l i d handles of elaborate shapes, one of which i s at least 22.6 cm t a l l . From the Late Period at Meishan, there are two elaborate l i d handles, one of which i s shaped l i k e the head of a wolf. There are other forms of elaborately shaped vessels at Lujiakou. The l e i necked jar i s only present i n the Late Period, but gui pitchers with exaggerated (long) spouts are i n each period. Vessels with thin walls are present i n more than one period and region, but to a greater extent at Baiying and Lujiakou. The small pitchers from Baiying are on display at the Puyang work station. They are extremely f i n e l y made and very thin-walled, at ca. 1.0 mm. The Early Period sherd from a cup of uncertain shape i s said to be "eggshell" t h i n (CPAM of Anyang D i s t r i c t , Henan Province 1983). At Lujiakou, some hu necked jars are thin-walled. 173 There are vessels i n a number of shape classes that are polished. Some classes exhibit variation i n degree of luster on vessels as well. A few bei cups, zun cups, and dou dishes from the Middle and Late periods at Baiying have a high degree of luster (CPAM of Anyang D i s t r i c t , Henan Province 1983:17, 19, 28). Unfortunately, i t i s not possible to describe variation i n polishing among vessels i n any d e t a i l . When vessels were examined i n the f i e l d , extensive smearing of plaster from the reconstruction process prevented accurate comparison of vessels. I suggest that thin-walled, polished, and elaborately shaped vessels were used i n displays of conspicuous consumption throughout the Longshan Period. Individual-sized vessels such as cups could have been used, as well as serving vessels such as necked jars that could have contained alchohol. The large and elaborate l i d s could have been used in either displays of largesse or conspicuous consumption. The same hypothesis can be offered for polished open forms. The unusual painted quanzupan dish from the Late Period at Baiying may represent display of r i t u a l status instead. In most cases, i t i s not clear whether displays of largesse or conspicuous consumption became more i n t e n s i f i e d over time as a result of increasing status competition. However, there i s some indication that displays of conspicuous consumption may have i n t e n s i f i e d over time i n the Hougang II region. Vessels became increasingly thin-walled from the Early to the Middle Period at Baiying (CPAM of Anyang D i s t r i c t , Henan 174 Province 1983:37), and from the Middle to Late Period at Hougang (Anyang Archaeological Team, IA, CASS 1985:81). As previously discussed, these phases are roughly contemporaneous. Also, white wares were introduced at t h i s time. I t appears that there i s a change i n character of display vessels in the Hougang II region over time. Large vessels seem to drop out by the l a t e r Longshan Period. I suggest that people i n status competition decided to adopt a different t a c t i c , displaying a variety of shape classes exhibiting labor-intensive production as well as personal consumption of food and drink. Again there i s evidence for a decline in pottery production over time at Meishan. In the Late phase, extremely thin-walled vessels are not present. Also, the number of vessels with polishing declines over time. Vessels from only the Early Period have a high degree of polishing (Second Henan Archaeological Team, IA, CASS 1982:273-4). In contrast, there i s an increase i n polishing for vessels from Lujiakou (Shandong Archaeological Team, IA, CASS and the Art Museum of Weifang County, Shandong Province 1985:348). One sherd of eggshell thickness (ca. 1 mm) was found, of unspecified date. Displays of status with pottery may have continued with greater intensity i n Shandong and northern Henan than i n western Henan during the late Longshan Period. The patterns describe here appear to be present at other, less well documented sites from the Longshan Period. Very large containers occur at three sites dating to the e a r l i e r Longshan Period (Table 25). 175 Table 25. Very Large Vessels at Other Sites from the Longshan Period. cultural region / site Taosi characteristics of vessels Taosi (Xiangfen County, Shanxi) Early Period dates ca. 2500-2300 B.C. (Zhang and Zhang 1986:51, Gao et a l . 1984:28) Wangwan III Wangwan (near Luoyang c i t y , Henan) layer I I I , one C14 date f o r s i t e , ZK 126, 2390+/145 B.C. (Zhang and Zhang 1986:53) Hougang II Xiapanwang (Cixian County, Hebei) (one C14 date f o r s i t e , ZK 200-1, 2515+/145 B.C. applies to F l (Zhang and (Zhang 1986:52) 1 wooden, red painted quanzupan stemmed dish i n Early Period grave M3015, a large and the r i c h e s t grave at the s i t e (178 grave goods), rim diameter of dish= 63.6 cm, height= 22.0 cm (Shanxi Archaeological Team, IA, CASS and the Linfen D i s t r i c t C u l t u r a l Bureau 1983:31,35,39) 1 pedestalled pen dish, polished, on display at the Archaeological Work Station, Luoyang, Henan, rim diameter= ca. 54.7 cm, height= ca. 30.6 cm, and 1 guan j a r , rim diameter= ca. 41.5 cm, height ca. 65.0 cm; 1 large zuo stand (polished, i n c i s e d , with carved out holes) on display at the National History Museum, B e i j i n g , rim diameter= ca. 45.0 cm, height= ca. 28.0 cm 1 pingdipen basin of same shape class as the large basin at Hougang) from house F l , rim diameter= 61.2 cm, height= 14.0 cm (Department of Cu l t u r a l R e l i c s , Hebei Province 1975:92) 176 S i g n i f i c a n t l y , the grave at Taosi shows an association between large vessels (wooden, i n th i s case) and high status users or owners (see Shanxi Archaeological Team, IA, CASS and the Linfen D i s t r i c t Cultural Bureau 1983:31, 35, 39). Thus, displays of largesse may have occurred in more than one area, not just northern Henan. A large pingdipen basin has been reported for a t h i r d s i t e i n the Hougang II region, Xiapanwang (Department of Cultural R e l i c s , Hebei Province 1975:92). Other types of labor-intensive vessels have been found that must have been used i n displays of status. Thin-walled, polished, and t a l l (exaggerated height) vessels are reported at other sites from the Wangwan I I I , Hougang I I , and Liangcheng regions (Table 26). Information on polishing i s limit e d , but i t appears that most of the vessels i n question were polished. Sites from Henan have yielded thin-walled vessels of more than one shape class, p a r t i c u l a r l y cups. In the Liangcheng region of Shandong, the only form of thin-walled vessel i s the well-known "eggshell" (dan ke tao) t a l l stemmed cup (gao bing b e i ) . These vessels, of remarkable thinness (as low as 0.3 mm) are associated primarily with mortuary s i t e s . They are rare i n habitation contexts; the s i t e of Yaoguanzhuang i s one example (see Institute of Archaeology, Shandong Province e t N a l . 1981:22). "Eggshell" t a l l stemmed cups represent the climax of ceramic technology during the Neolithic period (Yan 1986, Du 1982). Thin-walled vessels were probably made f i r s t i n Shandong, ca. 2400 B.C., and next i n the adjacent Hougang II region. They are 177 Table 26. Labor-intensive Vessels that May Have Been Used fo r Display of Conspicuous Consumption at Other Sites from the Longshan Period, by Cul t u r a l Region. type of vessel Wangwan III Hougang II Liangcheng thin-walled cups - v a r i e t i e s of bei short 1, Wang- chenggang, Period IV, polished, ca. 1900 B.C. 1, Dahancun no date given less than 2.0mm thick t a l l stem- gao bing bei 1, Wadian, Period I I , C14 date for s i t e WBS2-24 2080+/135 B.C., (associated period not stated), reported as ca. 0.1mm thick, also polished, i n c i s e d , with small carved out holes several i n b u r i a l s at si t e s ca. 2400- 1800 B.C. most pots 0.3-0.5mm thick, polished, some with small carved out holes; at Chengzi, Sanlihe thin-walled beakers - v a r i e t i e s of gu 1 polished bei shaped l i k e a gu, Niuzhai, no clear date 1 7 8 thin-walled l i d s - v a r i e t i e s of Jgai 1, Wadian, period not stated 1, trumpet - shaped, ca. 24.5 cm t a l l , polished, at Yanzhuang, ca 1.0 mm thick, ca. 2175 B.C. 1, same shape as above, ca. 28.0 cm t a l l , b r i g h t l y polished, 1.2- 1.8 mm thick, at Gelawang, lat e Longshan Period t a l l beakers- v a r i e t i e s of gu 2, Wadian, Period I I , polished, broken heights 21.5 cm, 14.5 cm; 2 from Period I I I , b r i g h t l y polished, elaborate in shape, 25.2 cm, 29.0 cm in height 179 polished, 1, Wangchenggang Periods I I , IV, both ca. 20 cm t a l l , (ca. 2455, 1900 B.C.) tal l pitchers- v a r i e t i e s of gui, exaggerated necks and/or spouts several i n b u r i a l and other s i t e s ca. 2400- 1800 B.C., vessels up to 42.0 cm t a l l (Yao- guanzhuang, middle-late Longshan Period References: 1) Wadian (Yuxian County, Henan), some vessels on display at the Dengfeng (Gaocheng) Archaeological Research Station, and described by the Henan Province C u l t u r a l Research I n s t i t u t e and the Department of History at Zhengzhou University, Archaeology Specialty 1983:41,43; for dating see Zhang and Zhang 1986:53 2) Wangchenggang (Dengfeng County, Henan), several vessels on display at the Dengfeng (Gaocheng) Archaeological Research Station, and described by the Henan Province C u l t u r a l Research I n s t i t u t e and the Archaeological Section of the Museum of Chinese History 1983:13,16; for dating see Zhang and Zhang 1986:53 3) Niuzhai (near Zhengzhou c i t y , Henan), Cu l t u r a l Relics Work Team, Bureau of Culture, Henan Province 1958:22-3,26 180 4) Yanzhuang (near Zhengzhou c i t y , Henan), Museum of Zhengzhou City 1983:5), for dating see Zhang and Zhang 1986:53 5) Gelawang (near Zhengzhou c i t y , Henan), F i r s t Cultural Relics Work Team, Bureau of Culture, Henan Province 1958:46,47 6) Dahancun (near Anyang c i t y , Henan), Anyang Archaeological Team, IA, CASS 1990:55,56 7) Chengzi (Zhucheng County, Shandong), Archaeological Group of the Changwei Area and the Museum of Zhucheng County 1980 8) Sanlihe (Jiaoxian County, Shandong), Institute of Archaeology, Shandong Province 1988 9) Yaoguanzhuang (near Weifang c i t y , Shandong), Institute of Archaeology, Shandong Province et a l . 1981:14-7 181 present i n the Early Period at Baiying but appear to be more common by the Middle Period. From there they spread further west. I t i s possible that consumers attempted to emulate high status people i n adjacent regions. Using copies of display vessels made i n other areas i s a common strategy i n social competition ( M i l l e r 1982:193). Again, similar forms of vessels that may have been used for social displays are found i n more than one si t e within c u l t u r a l regions. For example, i n the Wangwan I I I region, a trumpet-shaped, polished, t h i n - walled l i d i s present at Yanzhuang (Museum of Zhengzhou City 1983:5) and at Gelawang ( F i r s t Cultural Relics Work Team, Bureau of Culture, Henan Province 1958:46, 47). "Eggshell" t a l l stemmed cups are found in several sites in Shandong. Conclus ions on Changes i n L a b o r - i n t e n s i v e ( P r e s t i g e ) V e s s e l s Large, thin-walled, polished, and elaborate vessels are found i n a number of sites i n more than one phase during the Longshan Period. S i g n i f i c a n t l y , similar patterns are found i n more than one si t e within the same cu l t u r a l subregion. I suggest that persons i n status competition emulated the displays of others -- people i n the same settlement system as well as i n other communities or even other regions. People in northern Henan, p a r t i c u l a r l y at the s i t e of Baiying, may have attempted to emulate the display t a c t i c s of people i n Shandong. This 182 process may have i n t e n s i f i e d from ca. 2300-2200 B.C. (by the Middle Period at Baiying and the Late at Hougang). It appears that display vessels continued to be used at Meishan, even though there i s potential evidence for decline in production of prestige wares by the Late Period. Other sites from the Wangwan I I I region have yielded labor-intensive vessels that may have been used for display as well. An exquisite, very t a l l , well-polished g_u beaker was found from Wadian. Thus i t appears that prestige wares continued to be important during the la t e r Longshan Period i n this region. One reason for differences i n patterns between sites i n the Wangwan I I I region may be that metal vessels were beginning to replace ceramic vessels as prestige goods. In the future a larger sample of vessels from Longshan sites should be examined for variation i n labor input, enabling a more precise comparison of vessels by the production step index (Feinman et a l . 1981). It i s necessary to document change i n production of (labor- intensive) prestige wares more c l e a r l y . This analysis suggests that a l l three types of change driven by increasing status competition cited previously may have taken place: increasing degree of labor input for the same type of shape classes (Baiying), using vessels with new types of labor-intensive techniques (Hougang, Baiying), and acquiring a greater variety of vessels with labor-intensive techniques (Baiying, Lujiakou). These changes represent an increase i n var i e t i e s of ceramic 183 categories over time as predicted by my revised version of the model Rice (1981). 184 CHAPTER 6. CHANGE IN MODE OF PRODUCTION AND ACCESS TO GOODS INTRODUCTION This chapter addresses two topics. The f i r s t i s archaeological evidence f o r mode of pottery production. As discussed i n Chapter 3, the following modes may be represented i n chiefdoms: household production, simple household industry, complex household industry, and i n d i v i d u a l workshop industry. The chapter addresses whether there i s evidence to support the hypothesis r a i s e d i n the Chinese and western l i t e r a t u r e that workshop production was present during the l a t e N e o l i t h i c period of northern China. Also, i t attempts to determine whether there i s evidence f o r a change i n mode over time at s i t e s , as predicted by the model. As discussed i n Chapter 3, patterns of standardization and d i v e r s i t y of ceramic vessels can be useful i n recognizing change i n mode of production. Two other categories of archaeological data that can be informative as well are d i r e c t evidence f o r pottery production ( k i l n s , t o o l s , etc.) and techniques used i n production (for shaping, f i r i n g ) . The l i m i t e d data on ceramic vessels, d i r e c t evidence f o r production, and s p e c i f i c techniques suggest that there i s no evidence f o r change i n mode of production at any s i t e . Also, the mode that i s most l i k e l y i s 185 complex household industry. Other Longshan s i t e s have s i m i l a r evidence for pottery production. The second topic addressed here i s patterns of access to c r a f t goods at Hougang, Baiying, Meishan, and Lujiakou. One important issue i s whether there i s evidence f o r d i f f e r e n t i a l access to goods for large s i t e s with walls versus smaller s i t e s without walls. In chiefdoms, r e l a t i v e l y large settlements with substantial a r c h i t e c t u r a l features are usually centers i n which residents have r e s t r i c t e d access to a range of c r a f t goods. As discussed i n Chapter 2, Hougang i s reported as much larger i n size than the other three s i t e s and has remains of a surrounding w a l l . I compare the types and quantities of ceramic and nonceramic goods found at the four s i t e s -- p o t e n t i a l prestige items and u t i l i t a r i a n items. Another important issue i s whether there i s evidence f or change i n patterns of access to goods at an i n t r a - s i t e l e v e l . The analyses i n Chapter 5 suggest that display a c t i v i t y with pottery vessels i s present throughout the Longshan Period i n more than one region. Also, there may have been an increase i n i n t e n s i t y of display a c t i v i t y over time i n the Hougang II and Liangcheng regions. For example, i t appears that vessels become in c r e a s i n g l y thin-walled over time i n the Hougang II region. In chiefdoms, status competition often involves more than one category of material goods. It i s important to investigate whether other goods indicate status competition as well, and whether there i s evidence f o r change i n degree of competition over time. 186 F i r s t , I describe v a r i a t i o n i n d i v e r s i t y and quantity of a r t i f a c t s at each s i t e , p o t e n t i a l prestige items and u t i l i t a r i a n objects. Then I discuss v a r i a t i o n i n size and construction material of houses at the four s i t e s . There i s much more information for Hougang and Baiying than Meishan and Lujiakou. On the whole, there i s l i t t l e i n d i c a t i o n that the occupants of Hougang had access to a greater v a r i e t y and quantity of goods than the other s i t e s . However, d i f f e r e n t i a t i o n i n housing at Hougang and Baiying becomes more marked over time. Unfortunately, since few ceramic and nonceramic goods were recovered from houses, there i s no independent data on status of consumers of hypothesized prestige items. Conclusions about possible changes i n access to goods at s i t e s i s discussed rather than d e t a i l s about consumption of goods. EVIDENCE FOR MODE OF PRODUCTION Since Chapter 3 describes how ceramic a t t r i b u t e s r e f l e c t i n g degree of standardization and d i v e r s i t y can indicate mode of production, t h i s section outlines proposed expectations concerning the other two relevant categories of data, d i r e c t evidence for ceramic production and ceramic technology. Then i t describes the ava i l a b l e data f o r s i t e s . Table 27 summarizes the expected archaeological evidence f o r change i n mode of production i n c h i e f l y s o c i e t i e s . Change i n productive mode i s more r e a d i l y i d e n t i f i a b l e by change i n d i r e c t evidence f o r production. If only ceramic vessels and techniques used i n production 187 Table 27. Hypothesized Archaeological Change i n Mode of Production. Indicators f o r type of evidence household to simple household industry simple household industry to complex household industry complex household industry to in d i v i d u a l workshop industry change i n di r e c t evidence for pottery production: t o o l s , k i l n s , pots i n houses change from evidence f o r prod- uction i n most houses or several areas to a few no change; evidence f o r prod- uction i n only a few houses or areas change from production in few houses to no houses; remains of workshops change i n technology change i n same techniques as above to increase e f f i c i e n c y i n shaping, f i r i n g change i n characteri s t i e s of ceramic products increases i n degree of standardi- zation and/or d i v e r s i t y in terms of size, shape, decoration same as above same ; r e s u l t i s high degree of standard- i z a t i o n and d i v e r s i t y 188 are examined, i t may not be possible to ascertain whether there i s a change i n mode. Instead, the data may indicate evidence for increasing e f f i c i e n c y i n production with no change i n mode. Ideally, a l l three categories of data should be examined i n assessing mode of production. Unfortunately, i t i s often d i f f i c u l t to identify direct evidence for ceramic production from archaeological remains. Researchers need to be aware of po t e n t i a l l y useful data during excavation (Stark 1985, Deal 1988). Analysts can predict expected categories of data on the basis of production techniques used by modern t r a d i t i o n a l societies in general (van der Leeuw 1984) and i n sp e c i f i c areas. For example, researchers have used ethnoarchaeological data from the American southwest (Sullivan 1988) and Mesoamerica (Stark 1985, Deal 1983, 1988) to predict useful categories of data. I review evidence for pottery production i n Chinese Neolithic and early h i s t o r i c sites i n order to ide n t i f y relevant classes of data that may be present at Hougang, Baiying, Meishan, and Lujiakou. It i s often profitable to examine ceramic vessels for s p e c i f i c techniques used i n production (Santley et a l . 1989:111). Here I discuss general information from the archaeological l i t e r a t u r e on techniques of production. Ideally, household production should be recognizable by material evidence for production i n or near each house i n a settlement. Craft spe c i a l i z a t i o n within a settlement should be recognizable by remains of production i n a limited number of areas, for both simple and complex 189 household industry (see Tosi 1984:23-4). A change to workshop production should be recognizable by the appearance of workshops. There should be a concentration of permanent f a c i l i t i e s for different steps i n production such as preparing paste, shaping, decorating, and f i r i n g (Santley et a l . 1989:109). A number of factors may complicate archaeological recognition of pottery production, especially when c l e a r l y i d e n t i f i a b l e workshops are absent. The presence of portable items such as tools in an area does not necessarily indicate that production took place i n that area (Stark 1985:167). Immovable features such as k i l n s provide more r e l i a b l e evidence for production. Some steps in production involving clay soaking p i t s and wheels, etc. may take place i n courtyards adjacent to houses, as i n Indian v i l l a g e s with complex household industry ( M i l l e r 1985:209). Conclusions about location of production are more r e l i a b l e when a range of archaeological data i s examined (Stark 1985:177). In most studies, production areas are located by ide n t i f y i n g tools, k i l n s , and wasters. In recent years other c r i t e r i a have been used such as r e l a t i v e concentration of residues from production over a s i t e (Tosi 1984:23, Santley et a l . 1989), presence of raw material storage (Deal 1988, 1983), and i d e n t i f i c a t i o n of tools from use-wear analysis (Deal 1988). However, the present study i s limited to identifying tools and k i l n s as described i n archaeological reports. This study also considers another potential indicator of ceramic production proposed by Deal (1983:111): presence of a great d i v e r s i t y 190 and quantity of vessels i n houses where potters reside. This c r i t e r i o n should apply to the household and household industry modes. An obvious problem i n using the c r i t e r i o n i s the d i f f i c u l t y i n determining whether a great v a r i e t y and quantity of vessels i n a house indicates a high status family rather than the household of a potter. A high status family should symbolize i t s s o c i a l p o s i t i o n by more than one category of material remains. A house with a great v a r i e t y and quantity of vessels should be r e l a t i v e l y large i n size, b u i l t with construction materials of r e l a t i v e l y good q u a l i t y , and/or contain other prestige goods. I suggest that a change i n mode of production may also be indicated by technological changes to increase e f f i c i e n c y , i . e . , to produce a greater number of vessels i n a shorter time period. Greater speed can be achieved by increased use of moulds as well as wheels (Arnold 1985:209). For example, using moulds to make a greater number of shape classes can increase e f f i c i e n c y i n production. Another solution i s to make changes i n k i l n design. For example, the size of a f i r i n g chamber l i m i t s the number of vessels that can be f i r e d at one time (Kramer 1985:81). A cl e a r increase i n size of f i r i n g chamber could indicate e f f o r t s at greater e f f i c i e n c y , i . e . , by f i r i n g a greater quantity of vessels at one time. However, size of baking chamber also r e l a t e s to size of the vessels being f i r e d . 191 Evidence f o r Pottery Production i n Chinese N e o l i t h i c Sites There are several types of evidence for pottery production from Neolithic and early h i s t o r i c period s i t e s . Large features include p i t (updraft) k i l n s at several sites (Feng et a l . 1982, Zhou et a l . 1982), a stone turntable from the Yangshao s i t e Banpo (Zhao 1989), and p i t s for preparing paste from a Shang s i t e (described by Shangraw 1978:139). Wheels or p i t s have not been reported from Longshan s i t e s . In addition, pottery, stone, and bone tools for shaping (trimming, smoothing, adjusting shape) and decorating (stamping, incising) vessels have been i d e n t i f i e d (Zeng 1985, An Zhimin 1982). Unfortunately most of the tools pictured by Zeng (1985) and An Zhimin (1982) are rather i n d i s t i n c t i n shape. Some of the tool forms are i l l u s t r a t e d in Longshan s i t e reports. However, i t i s not possible to ascertain whether most of these tools were used for pottery production or exclusively for other tasks. Therefore I only record the presence of d i s t i n c t i v e tool forms from Hougang, Baiying, Meishan, and Lujiakou. The tool form that i s most c l e a r l y associated with pottery production i s the p a i z i or "beater". This tool i s shaped l i k e a mushroom, i . e . , wide and f l a t at one end, and i s small, often about 192 6 cm t a l l . P a i z i made from both pottery and stone are present i n Longshan s i t e s . Zeng (1985:72) i l l u s t r a t e s two types of pottery "beaters", one that has a plai n end and one that i s covered with decorative l i n e s . Scholars have argued that the l a t t e r type was used to make impressed decoration on pots, as an alternative to carved wooden paddles (Feng et a l . 1982:39-40). Most p a i z i from Longshan Period sites are pla i n at the end. Baiying contains two covered with decorative lines that are s l i g h t l y different i n shape than other p a i z i . A possible alternative technique for making impressed decoration on Longshan vessels i s discussed below. Like Cheng (1959:93), I think that the majority of p a i z i were used as anvils i n shaping. Some scholars i n China, however, use the term dian to denote anvils (Feng et. a l 1982:40). Small, mushroom-shaped tools of pottery and stone have been i d e n t i f i e d as anvils i n several areas of the world. Rye and Evans (1976) report that s p e c i a l i s t potters i n Pakistan use more than one size class of these tools for a l t e r i n g the shape of wheel-thrown vessels. Twenty years e a r l i e r , Foster (1956) observed s p e c i a l i s t Pakistani potters using anvils of t h i s kind for thinning the walls and enlarging the size of wheel thrown Vessels. M i l l e r (1985:207-9, 222) also notes that Indian potters use mushroom-shaped anvils with wooden paddles for t h i s purpose. Facets on the i n t e r i o r walls of vessels are evidence for use of anvils (Rice 1987:136). In my estimation, wheel-thrown bowls and basins 193 at Hougang have facets of t h i s sort, r e f l e c t i n g r a pid a l t e r a t i o n of shape by beating a f t e r throwing. Table 28 describes the evidence for pottery production at Hougang, Baiying, and Meishan. The lack of evidence at Lujiakou i s probably r e l a t e d to the small percentage of the s i t e area excavated. P a i z i are the most common tools at s i t e s but other types of tools have been found as w e l l : pottery (zao) c h i s e l and po l i s h i n g sherds. It i s l i k e l y that tools made of pottery described as knives or c h i s e l s , etc. were used for trimming vessels during the shaping process. The edges would not have been sharp enough for other tasks. Ye and Yu (1984:81), however, argue that pottery tools such as knives were used f o r a g r i c u l t u r a l tasks during the Longshan Period. The authors of the Hougang report maintain that 21 sherds show evidence of being used as p o l i s h i n g tools (Anyang Archaeological Team, IA, CASS 1985:71), but only two are described. When I examined vessels f i r s t h a n d from Hougang and Baiying, I d i d not see any of the c h a r a c t e r i s t i c s t r i a t i o n s from burnishing on vessels (see Rice 1987:138), even with a hand lens (16x). Because lustrous black and grey vessels have an extremely smooth appearance, i t i s l i k e l y that p o l i s h i n g was usually achieved by other means. At the I n s t i t u t e of Archaeology i n Jinan, Shandong, I was fortunate to meet Zhong Huanan, an experienced potter who has been r e p l i c a t i n g N e o l i t h i c vessels from several periods. He believes that Longshan potters polished vessels before and a f t e r f i r i n g with a soft material such as cl o t h or fur (Zhong Huanan, personal communication, July 1987). Leather could have 194 Table 28. Direct Evidence for Pottery Production at Hougang, Baiying, and Meishan by Excavation (T) Area. s i t e pottery c h i s e l p a i z i t o o l f o r p o l i s h i n g ceramic mould k i l n Hougang Early pottery 1 in T2 1 i n p i t T1-T2 Middle 2 sherds i n p i t s , T12, T12A part of k i l n wall T16 Late stone, 1 i n T16 Baiying Early 1, T32 Middle Late 1 i n 14 pottery, 3: T15, T18, and unclear Meishan Early 5 i n cor- ner, T2 Late stone, 1 i n T l 195 been used as well (personal communication, Rice 1990). These materials, of course, would not normally be preserved i n s i t e s . Another type of tool that has been discovered, at Hougang, i s a pottery mould for forming the legs of l_i tripods. The mould was f i r s t reported i n 1961 (Anyang Archaeological Team, IA, Academica Sinica 1961:64-5). In a later report, the Anyang Archaeological Team, IA, CASS (1982:578) states that the mould i s 11 cm t a l l and was found i n p i t H12. According to t h e i r f i n a l report for Hougang published i n 1935, t h i s p i t belongs to the Early Period. The mould i s on display at the Xiaotun Research Station. Moulds were commonly used for shaping vessel legs during the Neolithic period (Feng et a l . 1982:38), but i t i s not clear i f moulds were used for making the bodies of vessels. I suggest that the bodies of jars with impressed decoration such as guan might have been made by using moulds. The impressed decoration such as cordmarks are deep, v e r t i c a l , and extremely regular for the entire length of the vessels. Also, the marks do not overlap with one another. This type of surface could not have been achieved by beating with cord-wrapped paddles or engraved p a i z i . A ceramic or wooden mould lined with cord, basketry, or fabric could have been used. This technique would also f a c i l i t a t e removal of leatherhard vessels from moulds without sticking (Rice, personal communication 1990). Some scholars suggest that moulds may have been used to make vessels with impressed decoration (Zhou et a l . 1982:272, Wu 1938). Wu 196 (1938:133-4) maintains that a "belt-mould" rather than a true mould was used to create the deep, regular impressions on jars. He concludes after experimentation that a curved s t r i p of wood, bamboo, or pottery about 2 inches wide and 10 inches long was used when forming large jars by c o i l i n g on a turntable (Wu 1938:31-2). Further experimentation and detailed examination of vessels w i l l help determine techniques used for making various shape classes. The Anyang Archaeological Team, IA, CASS (1985:56) states that the bodies of large vessels such as guan, pen and xian steamers were made by moulds (mu z h i ) . The rims of these jars were wheel turned dun x i u ) , and the bases, wheel thrown. The largest vessels such as large guan and gang were c o i l made (ni tiao pan zhu), with wheel-turned rims. Most smaller vessels were made by fast wheel dun z h i ) . Sim i l a r l y , the CPAM of Anyang D i s t r i c t , Henan Province (1983:27) states that large forms at Baiying such as xian and j i a tripods were made by moulds and the wheel. In both reports the implication i s that pottery making techniques are consistent over time for a l l classes of vessels. Also, there i s agreement among scholars that potters made several shape classes by assembling separately fabricated parts throughout the Longshan Period, whether the parts were made by a mould, the wheel, or the c o i l method. My br i e f observations on shaping methods at sites support these conclusions. Kilns were found at Baiying and Meishan (Table 28), but unfortunately, none of the k i l n s from Baiying are described i n the 197 report. Only one k i l n from the Early Period i s mentioned, Y3. Given the designation, probably two other k i l n s were found at the s i t e as well. A l l three extremely large vessels from the s i t e described in Chapter 5 (the basin, zuo stand, and l i d ) were found in the k i l n , i n addition to a small pitcher. The report does not give the l o c a t i o n of the k i l n i n r e l a t i o n to the Early Period houses. The Second Henan Archaeological Team, IA, CASS (1982:431) describes four of f i v e k i l n s from Period I at Meishan. A l l of the k i l n s were found i n the northwest corner of excavation area (grid) T2. No information on the location of t h i s area within the s i t e or i n r e l a t i o n to other features such as houses i s given. The chronological order i n which the k i l n s were used i s Y3, Y2, and YI with Y4. Two of the k i l n s , Y3 and Y4, are incomplete. There i s no relevant information f o r Y3. The i l l u s t r a t i o n s and descriptions provided indicate no s t r u c t u r a l change in YI, Y2, and Y4 over time. The size of the f i r i n g chamber i s nearly the same f o r Y2 and Y4, with a maximum diameter of 1 m. Y2 i s s l i g h t l y bigger, at 1.4 m. The Anyang Archaeological Team, IA, CASS (1985:71) mentions that a small piece of a k i l n wall was found at Hougang from a Middle Period stratum in a general excavation area. It i s a burnt clump of grass mixed with mud 15 cm long that originated i n the f u e l chamber. 198 Evaluation of Change i n Mode of Production The t o o l s , k i l n s , and vessels from the four Longshan Period s i t e s examined i n t h i s study show no evidence of change i n mode of production over time. Most l i k e l y , the household industry mode characterizes the en t i r e period. However, there may be differences between d i f f e r e n t types of s i t e s . The i n d i v i d u a l workshop industry mode could be represented at centers of settlement. In his c r o s s - c u l t u r a l study of c r a f t production, Clark (1986:31) found that there i s a highly s i g n i f i c a n t c o r r e l a t i o n between community size or s o c i a l density and d i v e r s i t y of modes of production. Larger samples of data on d i r e c t evidence f o r production, technology, and c h a r a c t e r i s t i c s of ceramic products are necessary in order to substantiate these conclusions. Every s i t e except Lujiakou contains d i r e c t evidence for pottery production; however, only a small portion of Lujiakou was excavated (see Table 4, Chapter 1). Further excavation should uncover relevant data. At the other three s i t e s , there i s evidence suggesting that production took place i n a l i m i t e d number of areas at each s i t e i n each period. However, no t o o l or k i l n was found c l e a r l y associated with an i n d i v i d u a l house. Table 28 l i s t s the excavation (T) areas containing evidence f o r production. As discussed, the presence of portable tools i n an area 199 does not indicate that production took place i n that area. A stronger case can be made when there i s an association of more than one category of evidence, such as the T1-T2 area from the Early Period at Hougang, as well as the T16-T12 area from the Middle Period. The data at hand do not indicate a s i g n i f i c a n t change over time i n number of areas i n general with evidence for production at any s i t e . Very few pots were found i n houses at Hougang, Baiying, Meishan, and Lujiakou. The exception to t h i s pattern i s one house from the Middle Period at Baiying, F55. Given the r e l a t i v e l y large quantity and d i v e r s i t y of pots found i n i t , F55 may have been occupied by a potter, according to the c r i t e r i o n proposed by Deal (1983). This house i s located in T4, the area where the pottery c h i s e l was found. Although F55 i s the largest house at the s i t e , i t i s made of common construction material and does not contain any non-ceramic prestige goods. Therefore, the house may represent the residence of a potter rather than an e l i t e family. Another p o s s i b i l i t y i s that house F55 represents an abandoned structure used by households i n the community for dumping vessels a f t e r use. This pattern of deposition i s seen at Jomon s i t e s i n Japan (Kobayashi 1974:166). I examined other archaeological reports for evidence of ceramic production at s i t e s . Table 29 l i s t s evidence for production such as tools and k i l n s at s i t e s by c u l t u r a l region. These data suggest that production f o r at least some wares took place in most communities. There i s r e l a t i v e l y l i t t l e information for Shandong. 200 Table 29. Evidence for Ceramic Production at Other Sites from the Longshan Period. cultural region / site Kexingzhuang II Kexingzhuang (Changan County, Shaanxi), period(s) (Xu et a l not stated 1982:21-2) T. kilns 3, one in a house anvils other aosi Taosi (Xiangfen County, Shanxi) period(s) not stated (Xu et a l . 1982:21-2) Wangwan III Dahecun; Period V (near Zhengzhou c i t y , Henan), early Longshan Period, (Museum of Zhengzhou City (1979:362) 1, stone, pl a i n end Niuzhai (near Zhengzhou c i t y , Henan), no date available, e a r l i e r than Gelawang s i t e (Cultural Relics Work Team, Bureau of Culture, Henan Province 1958:23,26) 1 rect- angular pottery paizi "beater" ca. 11.0 cm long, function not clear 201 Gelawang (nea r Zhengzhou c i t y , Henan), l a t e Longshan P e r i o d , ( F i r s t C u l t u r a l R e l i c s Work Team, Bureau of C u l t u r e , Henan P r o v i n c e 1958:50-1) 1 r e c t - a n g u l a r c l a y muzhi "mould" same as paizi above Yanzhuang (nea r Zhengzhou c i t y , Henan), l a t e Longshan P e r i o d , (Museum of Zhengzhou C i t y 1983: 4,5,7), one C14 d a t e , WB79-46, 2175+/-105 B.C., (Zhang and Zhang 1986:53) 1 r e c t - a n g u l a r p o t t e r y paizi c a . 13.2 cm l o n g , l i k e t o o l s above H a o j i a t a i (Yancheng County, Henan), number and p e r i o d s n ot r e p o r t e d (Renmin R i b a o 1986) XX Hougang II T a i k o u c u n (Yongnian County, H e b e i ) , no d a t e a v a i l a b l e , ( C u l t u r a l R e l i c s Work Team, Bureau of C u l t u r e , Hebei P r o v i n c e 1962: 638,640) 1, p o t t e r y , p l a i n end Xiapanwang ( C i x i a n County, Hebei"), (Department of C u l t u r a l R e l i c s , H e b ei P r o v i n c e 1975:94), one C14 d a t e , ZK 200-1, 2515+/-145 B.C. (Zhang and Zhang 1986:52) 1, p o t t e r y , e ngraved l i n e s on end 202 Jiangou 7 (Handan County, Hebei), period(s) not stated, Xu et a l . 1982: 21-2) Balizhuang 1 (near Anyang c i t y , Henan), period not clear, (Xu et a l 1982: 21-2), C14 date for early period ZK 756, 2585+/-145 B.C. (Zhang and Zhang 1986:52, Anyang Archaeological Team 1985:82) Wangyoufang Pingliangtai (Huaiyang County, Henan), period(s) not stated, Henan Province Cultural Research Institute and the Cultural Objects Division of the Zhoukou D i s t r i c t 1983:31) Chengziya Shangzhuang (Chiping County, Shandong), late Longshan Period (Institute of Archaeology, Shandong Province 1985:484,503) 3, i n two separate areas, one i n a house 2, pottery, pl a i n end 203 Liangcheng Yaoguanzhuang -- 1, (near Weifang c i t y , pottery, Shandong), middle-late p l a i n Longshan Period end (Institute of Arch- aeology, Shandong Province et a l . 1981: 11-2,41) 204 S i g n i f i c a n t l y , at two s i t e s , evidence for production i s associated with a house. A k i l n was found attached to a house at Kexingzhuang i n Shaanxi, and at the walled s i t e of Pingliangtai i n eastern Henan (see Xu et a l . 1982, Henan Province Cultural Research Institute and the Cultural Objects Division of the Zhoukou D i s t r i c t 1983:31). This pattern indicates more c l e a r l y that a form of household production may characterize most communities during the Longshan Period. Feng et a l . (1982:13) also hypothesize that the remains at Kexingzhuang indicate a type of specialized household production. Scholars have remarked that k i l n design improves over time from the pre-Longshan period to the Longshan Period (Song et a l . 1983:269-70, Medley 1976:28, Feng et. a l 1982:40-2). There i s no evidence for technological changes indicative of efforts toward greater eff i c i e n c y i n production at any of the four sites during the Longshan Period. My observation that the k i l n s at Meishan do not show structural change agrees with a statement made by L i J i a z h i (1984:145) that k i l n s do not vary i n structure during the Longshan Period. Kilns have been found i n other sites from the Longshan Period (Xu et a l . 1982), as indicated i n Table 30. These k i l n s , l i k e those at Meishan, do not exhibit s i g n i f i c a n t structural change over time. There i s no evidence that potters t r i e d to f i r e vessels more e f f i c i e n t l y by increasing the size of the k i l n room (yao wu), or by increasing the quantity or size of individual f i r e flues (huo dao). Differences i n k i l n s among sites appear to be minor. 205 Table 30. Characteristics of Kilns from Other Longshan Period Sites. (derived from Table 1, Xu et a l . 1982:21-2). site diameter of k i l n room width of main flues quantity of main flues branches of flues Kexingzhuang (N=3) 26m, 96- 33, X , X , X 2, x, x 4,x,x Taosi (N=3) .02- 42, x 0.11- 0.16m, x,x 3,x,x 2,x,x Gelawang (N=l) Jiangou (N=7) N=7:x N=7:x 4 k i l n s have 2, N=3:x 2 k i l n s have 3, 1 has 4, N=4 :x Balizhuang (N=l) 0.06- 0.1 Sanliqiao (N=l) 1.3 0.11 Miaodigou (N=l) 0.78- 0.93 0.07 note: 1) "x" denotes no information for a k i l n . 2) Sanliqiao and Miaodigou are located i n far western Henan, Shaanxian County. There i s no date for the Longshan Period remains from Sanliqiao, and the associated c u l t u r a l type i s debated (see Zhang and Zhang 1986 ) . 3) Miaodigou II remains are t r a n s i t i o n a l to the Longshan Period (Yan, personal communication 1987b), ZK 111 2780+/145 B.C. (Institute of Archaeology, CASS 1983:72). 206 F i n a l l y , data on ceramic v a r i a b i l i t y from the four sites as described i n Chapter 5 do not indicate s i g n i f i c a n t change i n degree of standardization over time for any of the shape classes examined. There i s no evidence for increasing dimensional standardization. The pattern for within-class standardization i n terms of secondary shape features or decorative techniques i s not clear; i t i s possible that there i s increasing standardization for a few shape classes at Hougang and Baiying. Another possible indicator of change i n mode i s increase i n d i v e r s i t y i n ceramic categories over time. As Chapter 5 suggests, there i s a notable increase over time i n number of shape classes at Lujiakou. There i s a pattern of d i v e r s i f i c a t i o n for the Early to Middle Period at Baiying with regard to variety of shape classes produced. However, there are no other possible indicators for change i n mode of production at either s i t e . Complex rather than simple household industry i s the mode that most l i k e l y characterizes the majority of settlements during the Longshan Period. There i s substantial ceramic v a r i a b i l i t y at sites with respect to shape, size, and decoration, and some concern with e f f i c i e n c y i s evident, from the probable extensive use of wheels and moulds. Kilns represent a substantial investment i n equipment and f a c i l i t a t e a r e l a t i v e l y large output. Also, a number of vessels exhibit r e l a t i v e l y labor-intensive techniques of production. There i s evidence for technological sophistication i n more than one production step such as levigation to make very fine pastes and 207 complex f i r i n g methods. Black and grey vessels were produced by the imbibing (yin yao) method (Wu 1938:32, Zhou et a l . 1982, Medley 1976:280). Medley explains that t h i s method involves placing damp straw or other similar materials i n the k i l n when a c r i t i c a l f i r i n g temperature i s reached. Then water i s sprinkled on, causing a rapid decline i n temperature and reduction of oxygen inside the k i l n . Some black vessels from Longshan sites are uniform i n color throughout the entire thickness of the walls. According to Zhou et a l . (1982), t h i s effect i s caused by the penetration of carbonaceous p a r t i c l e s into the pores of vessels. Potters during the Longshan Period made several shape classes of vessels by assembling separately made parts. Keightley (1987) argues that t h i s "componential construction" i s technically sophisticated, however, i t does not require a highly organized workshop mode of production. Measurement of different parts can be achieved by simple techniques such as wooden hoops, as Chavez (1985:46) observes i n Peru. In India, potters with a complex household industry mode make cooking vessels from more than one part ( M i l l e r 1985:223-4). The ceramic data do not support the hypothesis of a workshop mode, either. My analyses suggest that there i s a low degree of dimensional and within-class standardization with respect to decorative techniques over time. Song et a l . (1983:262), among others, maintain that potters used a fast wheel operated by two persons during the late Neolithic period. They discuss production methods among t r a d i t i o n a l ethnic groups i n China 208 who use a wheel i n which one person turns, and the other shapes. They imply that only men, working i n workshops, were potters during the Longshan Period. A l l l i n e s of evidence examined i n t h i s study, however, indicate that complex household industry i s more l i k e l y as the predominant mode of production than i n d i v i d u a l workshop industry. PATTERNS OF ACCESS TO GOODS Since the Hougang s i t e i s much larger than the other three s i t e s and has a surrounding wall, i t probably represents a center of a settlement hierarchy (Chapter 2). If Hougang was a center, i t should have had a resident e l i t e stratum who displayed t h e i r status by greater consumption of goods. In chiefdoms, centers should have a greater d i v e r s i t y of ceramic and other goods than subsidiary settlements (Feinman et. a l 1981, T o l l 1981, Costin and Earle 1989). Wares ex h i b i t i n g extensive labor input and other prestige goods should be present. Only a broad comparison between s i t e s can be made here since each s i t e was part of a separate settlement system. Also, only two s i t e s , Hougang and Baiying, belong to the same c u l t u r a l region (Hougang II i n northern Henan and southern Hebei). However, i t does not appear that the occupants of Hougang had greater access to labor-intensive vessels than the other three, non- walled s i t e s (Chapter 5, Table 24). S i m i l a r l y , with the exception of 209 the Early ( t r a n s i t i o n a l ) Period, Hougang does not have a greater d i v e r s i t y of shape classes than the other s i t e s (Chapter 5, Table 17). There are few nonceraraic p o t e n t i a l prestige goods at these s i t e s . L i t t l e i s known at present about status competition and use of prestige goods during the Ne o l i t h i c period. On the basis of h i s t o r i c a l and archaeological data from the early dynastic period, i t appears that some goods with prestige value began to have importance during the lat e N e o l i t h i c period. As discussed i n Chapter 2, these goods include items of jade, bronze, and turquoise. There i s d i f f e r e n t i a t i o n i n size and type of construction material for housing as well. Potential prestige goods found at Hougang, Baiying, Meishan, and Lujiakou are jade items, bronze fragments, and ornaments (stone, s h e l l , bone, horn, ceramic). Hougang does not have a greater quantity or va r i e t y of these items than the other s i t e s (Table 31, and Appendix C, Table 74). Lujiakou has the greatest number of items (standardizing values per house, and lumping items of a l l materials). There i s some information on change over time i n access to po t e n t i a l prestige goods at two s i t e s , Baiying and Meishan. At Baiying, there may be an increase i n access to goods from the Early to Middle Period. This pattern corresponds with the suggested increase i n production of display vessels at the s i t e . However, i t appears there i s a decline from the Middle to Late Period. There i s a much greater number of houses from the Late Period at Baiying, but r e l a t i v e l y small 210 Table 31. Diversity and Quantity of Nonceramic Potential Prestige Goods at Hougang, Baiying, Meishan, and Lujiakou. sit e jade bronze ornaments Hougang 2 forms 7 forms (39 houses) 4 items 56 items t o t a l quantity of pieces, a l l materials: 60/39 houses = 1.54 items per house Baiying Early -- -- 2 forms (9 houses) 14 pieces Middle -- -- 2 forms (8 houses) 30 pieces Late 2 forms -- 8 forms (46 houses) 6 pieces 26 pieces t o t a l quantity of pieces, a l l materials: 76/63 houses = 1.21 items per house Meishan Early -- -- 5 forms (17 houses) 8 pieces Late -- traces on 4 forms (16 houses) 2 crucibles 8 pieces t o t a l quantity of pieces, a l l materials: 18/33 houses = 0.55 items per house Lujiakou -- -- 2 forms (11 houses) 25 pieces t o t a l quantity of pieces, a l l materials: 25/11 houses = 2.27 items per house 211 quantities and kinds of items. There i s no evidence of change i n access to p o t e n t i a l prestige goods at Meishan over time. Since few goods of any kind were found i n houses at these s i t e s , a pattern of gradual s i t e abandonment (Deal 1983, Chapter 4) may explain depositional patterns of nonceramic as well as ceramic items. Nonceramic prestige goods were found in a v a r i e t y of areas (open t e s t areas, storage p i t s ) within each s i t e , suggesting that they were used by a number of households rather than by only a few, e l i t e f a m i l i e s . S i m i l a r l y , there i s no i n d i c a t i o n that only a few houses at any s i t e had access to labor-intensive vessels at any time period. These vessels were found i n a va r i e t y of areas within each s i t e . However, one unusual pot at Baiying (Late Period), the quanzupan pedestalled dish with painted human fig u r e s , was found in a house. The occupants of t h i s house could have had high r i t u a l status. However, the house i s not large and was not b u i l t with c o s t l y materials. A s i m i l a r pattern i s observable with respect to u t i l i t a r i a n a r t i f a c t s at s i t e s (Table 32; and Appendix C, Table 75). There i s no i n d i c a t i o n that the occupants of Hougang had access to a greater d i v e r s i t y and quantity of u t i l i t a r i a n goods than people at the other s i t e s . Again, Lujiakou has higher values (standardized per house) than the other s i t e s . The data also suggest that there was an increase i n quantity and v a r i e t y of u t i l i t a r i a n goods at Baiying from the Early to Middle Periods, but a decrease from the Middle to Late, agreeing with the pattern for prestige goods. At Meishan as well, there appears to be 212 Table 32. Diversity and Quantity of U t i l i t a r i a n Tools at Hougang, Baiying, Meishan, and Lujiakou. Hougang 23 forms/39 houses= 0.59 forms per house 259 items/39 houses= 6.64 items per house Baiying Early Period: 14 forms/9 houses= 1.56 forms per house 49 items/9 houses= 5.44 items per house Middle Period: 21 forms/8 houses= 2.63 forms per house 112 items/8 houses= 14.0 items per house Late Period: 36 forms/46 houses= 0.78 forms per house 235 items/46 houses= 5.11 items per house s i t e as a whole: 71 forms/63 houses= 1.13 forms per house 396 pieces/63 houses= 6.29 items per house Meishan Early Period: 12 forms/17 houses= 0.71 forms per house 29 items/16 houses= 1.71 items per house Late Period: 21 forms/16 houses= 1.31 forms per house 60 items/16 houses= 3.75 items per house s i t e as a whole: 33 forms/33 houses= 1.0 forms per house 89 items/33 houses= 2.70 items per house Lujiakou 24 forms/11 houses= 2.18 forms per house 178 items/11 houses= 16.18 items per house 213 an increase i n general access to u t i l i t a r i a n goods from the Early to Late Period. Housing Published data are s u f f i c i e n t to investigate d i f f e r e n t i a t i o n i n house size and construction material at Hougang and Baiying. At Hougang, there are 39 separate c i r c u l a r houses. One of the houses, F12, i s occupied during two phases, 3 and 4 of the Late Period. At Baiying, there are 62 c i r c u l a r houses, and 1 rectangular house from the Late Period. Details on housing are provided i n Appendix C (Tables 76-77). At Hougang and Baiying, there are three types of wall material that vary i n c o s t l i n e s s of construction: 1) mud or earthen (tu), 2) wattle-and-daub (mu g_u duo ni ), and 3) adobe (tupi ). Adobe i s known as a c o s t l y construction material among the Highland Maya (Blake 1988:51). As mentioned i n Chapter 2, large adobe houses were probably occupied by e l i t e s at the large walled s i t e of P i n g l i a n g t a i i n the Wangyoufang c u l t u r a l region. Poorer households are associated with wattle-and-daub i n Mexico (Blake 1988:51) and other areas. Changes i n housing at each s i t e are evident when f l o o r area i s compared with type of construction material (Table 33). At Hougang, there i s greater v a r i a t i o n i n the Late Period: adobe houses, are, on the average, larger than earthen or wattle-and-daub houses. Also, the range of v a r i a t i o n i n f l o o r area i s greater during t h i s period. Thus, there 214 Table 33. Va r i a t i o n i n Average Size (Floor Area) of Houses by Di f f e r e n t Types of Construction Material at Hougang and Baiying. SITE Hougang Early Middle Late Baiying Early Middle Late EARTHEN WATTLE-AND-DAUB ADOBE 10.17 m2 17.02 15.89 15.70 14.36 16.60 23.07 21.08 11.64 12.12 9.72 12.17 11.25 15.89 215 i s increasing d i f f e r e n t i a t i o n i n housing over time. The same pattern i s evident at Baiying. Adobe houses do not appear u n t i l the Late Period, and they are the largest i n size, on the average. In contrast to the pattern for ceramic and nonceramic items, the adobe houses at Hougang provide some indication of a center of settlement. There i s a greater quantity of these houses than at Baiying, and average f l o o r area i s larger. Thus i t appears that status differences were symbolized more overtly i n housing than i n prestige or u t i l i t a r i a n a r t i f a c t s at these s i t e s . There are other indications for increasing d i f f e r e n t i a t i o n i n housing at Hougang and Baiying over time. At each s i t e , a q u a l i t a t i v e l y different type of house appears i n the Late Period: one house with planked fl o o r i n g at Hougang, and one rectangular house at Baiying. Burial sites from Shandong also provide information on d i f f e r e n t i a l access to goods during the Longshan Period. A brief examination of two s i t e s , Sanlihe and Chengzi, also indicates a pattern of a range of people having access to prestige items rather than just a few e l i t e s (Tables 34, 35). Labor-intensive "eggshell" t a l l stemmed cups are buried with individuals of both sexes and representing a range i n ages. Mortuary treatment varies from s i t e to s i t e as well. At Chengzi, only males have the cups (Archaeological Group of the Changwei Area and the Museum of Zhucheng County 1980). At Sanlihe, both females and males have the cups (Institute of Archaeology, Shandong Province 1988). 216 Table 34. Characteristics of Burials with T a l l Stemmed "Eggshell" Cups or gao bing bei at Chengzi (derived from Table 2, Archaeological Group of the Changwei Area and the Museum of Zhucheng County 1980:381-4). note: () indicates t o t a l number of graves by sex i n sample age group by period 25-30 Early (1 male) Middle (--) Late (--) 30-35 Early (--) Middle (1 male) Late (--) 35-40 number of eggshell cups t o t a l number of other vessel forms wooden c o f f i n and/or second level plat- form in grave platform and c o f f i n platform and c o f f i n Early (1 male) 1 Middle (1 male.) 1 Late (--) platform both absent about 45 Early (--) Middle (--) Late (1 male) 12 platform and c o f f i n t o t a l number of graves with eggshell cups: Early 4/15, Middle 3/10, Late 3/6 (age and sex unknown for 5 graves, 56 graves with no pottery and unknown date, t o t a l number of graves = 87) 217 Table 35. Characteristics of Burials with T a l l Stemmed "Eggshell" Cups or gap bing bei at Sanlihe (derived from Table 2, Institute of Archaeology, Shandong Province 1988:173-84). note: F= female, M= male, () indicates t o t a l number of graves by sex i n sample, x= absent, *= present age group by period 20-30 number of eggshell cups Early (IF) Middle (IM, IF) M:l, F:l Late (IF) 30-35 Early (2M) Middle (2M) Late (IM) 1,1 2,1 2 40-50 Early (IF) Middle (2M, 2F) Late (IM) M:3,l F : l , l t o t a l number of other vessel forms second level platform in grave 13 x M:l, F:2 M:unclear, F:x 1,8 5,0 5 x, x X, X M:10,l F:8,8 M:x,X F : * * 218 55-60 Early (IM) 1 1 x Middle (4F) 1,1/4,1 1,8,16,6 x,*,x,* Late (0) t o t a l number of graves with eggshell cups: Early 6/14, Middle 15/32, Late 5/10 (age and sex unknown for 6 graves, 42 graves with no pottery and unknown date, t o t a l number of graves= 98) 219 As Pearson (1988:13-4) points out, status d i f f e r e n t i a t i o n i s not the only s o c i a l dimension expressed i n mortuary r i t u a l . Only some of the graves with eggshell cups at Chengzi and Sanlihe have other p o t e n t i a l indicators of high status or wealth such as second l e v e l platforms (ercengtai), jade items, or a high d i v e r s i t y of vessel forms. At both s i t e s , however, older i n d i v i d u a l s tend to have the cups. The association of age and rank known from the early dynastic period and symbolized by use of vessels (Cooper 1982) may have begun i n the late N e o l i t h i c period. Individual f a m i l i e s may have decided how much they wanted to invest i n the funerals of deceased r e l a t i v e s , depending upon t h e i r s o c i a l ambition (see Orme 1981:234-5, Trinkhaus 1984). B u r i a l of labor- intensive pots creates a condition of s c a r c i t y and ensures that high value i s maintained (Arnold 1985:162-3). As Keightley (1985b) suggests, these cups and others may have been used by mourners at funerals before b u r i a l . CONCLUSIONS This chapter has examined two separate topics, mode of production and access to goods. The f i r s t section examined three categories of data i n order to i n f e r whether there i s evidence f o r change i n mode of ceramic production during the Longshan Period. It described t e s t implications involving d i r e c t evidence f o r production, technology, and 220 ceramic products. The data suggest that there was no change i n mode over time at any s i t e . This pattern i s incongruent with the p r e d i c t i o n in the model that organization of labor becomes more complex as s o c i o p o l i t i c a l complexity increases. I suggest that the complex household industry mode i s more probable than a workshop mode, contrary to the proposition made by other researchers. The second section of the chapter examined i n t e r - s i t e and i n t r a - s i t e v a r i a t i o n with respect to a r t i f a c t s and housing. The large walled s i t e of Hougang i s not markedly d i f f e r e n t from the other s i t e s i n terms of quantity and d i v e r s i t y of a r t i f a c t s , an unexpected r e s u l t f o r a probable center of settlement. However, a comparison of Hougang with contemporary s i t e s in the same settlement system may indicate that the occupants had access to a greater quantity and v a r i e t y of goods. At a l l four s i t e s , p o t e n t i a l prestige goods were found i n a number of areas, i n each period. It i s l i k e l y that a number of households i n each period had access to these goods. There i s information on change over time i n access to a r t i f a c t s at Baiying and Meishan. There i s some i n d i c a t i o n that there was an increase over time i n access to p o t e n t i a l prestige goods and u t i l i t a r i a n goods from the Early to Middle Period at Baiying. For Meishan, there i s only an increase over time i n quantity and v a r i e t y of u t i l i t a r i a n goods. There i s evidence for increasing d i f f e r e n t i a t i o n over time at Hougang and Baiying i n terms of house f l o o r area and c o s t l i n e s s of construction 221 material. This pattern suggests that there i s increasing s o c i a l d i f f e r e n t i a t i o n as symbolized i n housing at these two s i t e s . F i n a l l y , the analyses on i n t r a - s i t e v a r i a t i o n i l l u s t r a t e the importance of considering s i t e formation processes and excavation strategies when making interpretations about access to c r a f t goods. Researchers should consider that s i t e formation processes may vary for d i f f e r e n t categories of material culture. Smith (1987:313) maintains that serving vessels constitute the most r e l i a b l e archaeological i n d i c a t o r of household wealth without adequately considering how processes of deposition and s i t e abandonment. Few vessels or a r t i f a c t s of any kind were recovered from houses at Hougang, Baiying, Meishan, and Lujiakou. More information on patterns of deposition of goods at N e o l i t h i c s i t e s i s necessary. Some disposal patterns may represent patterns of consumption such as status competition, but others may not such as dumping vessels i n abandoned houses at Jomon s i t e s i n Japan (Kobayashi 1974). A f a c t o r relevant to Longshan s i t e s i s that because sherds are not always used for reconstruction of vessels, often they are not described i n reports. 222 CHAPTER 7. CONCLUSIONS INTRODUCTION The general goal of t h i s study was to address how c r a f t production changes i n r e l a t i o n to increasing c u l t u r a l complexity. The study examines how systems of pottery production may change i n r e l a t i o n to evolving chiefdoms. A revised version of the important model outlined by Rice (1981) i s presented and tested with ceramic data from Longshan s i t e s located i n the Huanghe or Yellow River v a l l e y of northern China. The s p e c i f i c goal of the d i s s e r t a t i o n was to address the question: "How do systems of pottery production change during the Longshan Period?" My model, a f t e r Rice (1981), makes the hypothesis that there should be an increase i n d i v e r s i t y of ceramic categories as s o c i o p o l i t i c a l complexity increases. Also, there should be evidence f o r increasing standardization of shape classes and change i n mode of ceramic production. However, the model focuses on describing changes i n production that may take place i n complex chiefdoms as well as s o c i a l factors that may cause d i f f e r e n t types of ceramic change. On the basis of l a t e r publications by Rice (1984, 1987) and ethnographic data, the model describes three a l t e r n a t i v e strategies of ceramic producers and consumers: d i v e r s i f i c a t i o n , s i m p l i f i c a t i o n , and 223 conservatism. It outlines how production of two categories of vessels, labor-intensive or hypothesized prestige vessels and non-prestige vessels, may change. For example, increasing population s i z e and density may cause potters to adopt a strategy of s i m p l i f i c a t i o n , to produce c e r t a i n classes of non-prestige vessels more e f f i c i e n t l y . In t h i s type of s o c i a l context, potters may also decide to produce a greater d i v e r s i t y of wares i n response to increasingly varied consumer demand (Rice 1984). The model also explains how consumer demand for labor-intensive vessels f o r use in public displays of status can cause ceramic change. This part of the model i s my own compilation and i n t e r p r e t a t i o n of ethnographic sources describing use of containers i n ranked s o c i e t i e s . On the basis of these data, two types of s o c i a l displays with pottery should have been common in p r e h i s t o r i c chiefdoms, largesse and conspicuous consumption (Chapter 3). The rest of the model synthesizes observations and hypotheses made by others about change i n ceramic production. Ceramic data from four s i t e s with r e l a t i v e l y c l e a r dating of phases and complete s i t e reports were used to test the model. These s i t e s are Hougang (northern Henan), Baiying (northern Henan), Meishan (west-central Henan), and Lujiakou (eastern Shandong). Data on ceramic v a r i a b i l i t y from whole and reconstructed vessels were c o l l e c t e d from archaeological work stations and museums i n Henan and Shandong provinces during a period of six months i n 1987. Professor Yan Wenming of B e i j i n g University provided invaluable information and advice. 224 The following analyses were conducted i n order to test the model: di v e r s i t y of shape classes over time, dimensional standardization, within-class standardization, and i d e n t i f i c a t i o n of labor-intensive vessels (Chapter 5). In addition, data on ceramic v a r i a b i l i t y , tools for production, and ceramic technology were assessed for the analysis of change in mode of production (Chapter 6). The interpretations made i n this study should be regarded as hypotheses that can guide future research rather than firm conclusions, since small samples of vessels were used i n the analyses. The assessment of variety of shape classes produced over time provides the most r e l i a b l e r e s u l t s , since most of the excavated vessels are included i n the tabulations. Other analyses must rel y upon a sample of the excavated vessels (Chapter 4). Sample size i s pa r t i c u l a r l y a problem for the analysis of within-class standardization. Two topics are discussed in more d e t a i l below: 1) evaluation of the model i n l i g h t of results obtained from the preliminary test and 2) assessment of cul t u r a l change during the Longshan Period. RESULTS AND EVALUATION OF THE MODEL The model of ceramic change i s p a r t i a l l y supported. There i s a pattern of d i v e r s i f i c a t i o n for some components of the ceramic production systems represented by Hougang, Baiying, Meishan, and Lujiakou. There is evidence for d i v e r s i f i c a t i o n with respect to t o t a l number of shape 225 classes produced from the Early to Middle Period at Baiying and from the Early to Late Period at Lujiakou. At both of these s i t e s , the increase p r i m a r i l y involves open forms. A process observed by M i l l e r ( 1 9 8 2 ) i n India may provide an explanation f o r t h i s increase: consumers desire a var i e t y of vessel shapes f o r displays of status. For Lujiakou, there i s also some i n d i c a t i o n of a decrease over time i n within-class standardization with regard to non-prestige wares. There i s other p o t e n t i a l evidence f o r d i v e r s i f i c a t i o n i n production of non-prestige wares from the Middle to Late Period at Baiying. It appears that the l e i - b o engraved bowl was introduced during the Late Period. This type of bowl may have been used f o r processing tubers, as discussed i n Chapter 4. The introduction of t h i s form la t e i n the occupation of the s i t e may indicate that potters responded to consumer demand f o r a wider va r i e t y of vessels used f o r preparing food. It i s unfortunate with respect to i t s important p o s i t i o n i n the model that information on change i n production of labor-intensive vessels i s quite l i m i t e d . However, reports tend to describe at least one example of each ceramic category found at s i t e s . The data at hand suggest that there i s some evidence f o r d i v e r s i f i c a t i o n i n production of prestige wares. Very large, thin-walled, elaborately shaped, and elaborately decorated vessels have been found. Type of display a c t i v i t y may have changed over time i n the Hougang II and Taosi regions, as indicated by the decline i n very large vessels by the l a t e Longshan Period. Displays of largesse may have been 2 2 6 replaced by displays of conspicuous consumption. Also, there may have been an increase in v a r i e t i e s of labor-intensive vessels from the Middle to Late Period at Hougang and during a roughly contemporary phase at Baiying (from the Early to Middle Period). White wares were introduced at t h i s time, and i t appears that there was an increase i n degree of labor input f o r some vessels as well. For example, there i s evidence for i n c r e a s i n g l y thin-walled vessels over time. These changes provide support f o r the hypothesis presented by Rice (1981) that there should be increasing v a r i e t i e s of status-related wares i n a context of increasing c u l t u r a l complexity. For most s i t e s more than one pattern emerged, supporting the p r e d i c t i o n of Rice (1984) that there should be d i f f e r e n t types of changes f o r d i f f e r e n t components of productive systems. The pattern for d i v e r s i t y of shape classes over time at Hougang was s i m p l i f i c a t i o n , and for Meishan, conservatism. D i v e r s i f i c a t i o n characterized only one phase at Baiying. A pattern of conservatism resulted from the analysis of dimensional standardization for each s i t e examined (Hougang, Meishan, Lujiakou). Results were variable for the analysis of within-class standardization. There i s l i t t l e evidence from the four s i t e s to support the hypothesis derived from Rice (1981) that there should be increasing standardization of vessels, p a r t i c u l a r l y those that served u t i l i t a r i a n purposes. Although there i s a pattern of s i m p l i f i c a t i o n f o r the Middle to Late Period t r a n s i t i o n at Hougang with respect to within-class 227 standardization and variety of ceramic categories produced, there i s a pattern of conservatism f o r dimensional standardization. There i s evidence f o r conservatism i n dimensional standardization f o r non- prestige vessels from Meishan and Lujiakou as well. This pattern may be explained by the fact that potters and consumers often r e s i s t change i n production of wares that are needed on a d a i l y basis (Rice 1984). However, there i s some i n d i c a t i o n that e f f i c i e n c y was a concern i n production of vessels used f o r basic needs. For example, some vessels have decorative l i n e s that were r a p i d l y and incompletely applied. Larger samples of vessels from Longshan s i t e s could y i e l d d i f f e r e n t r e s u l t s f o r analyses of change i n within-class and dimensional standardization. Given the emphasis on reconstruction of vessels from Longshan s i t e s , analysis of within-class standardization i s p o t e n t i a l l y useful f o r future research. With whole vessels, d i f f e r e n t combinations of decorative techniques and areas of placement can be examined. Other analyses should incorporate a greater v a r i e t y of i n f e r r e d functional categories, too. Most classes of vessels with adequate sample size used i n t h i s study probably constitute vessels that served basic needs such as cooking and water storage. It i s necessary to determine, f o r example, whether hypothesized prestige (labor-intensive) vessels become increasingly standardized over time as Rice (1981) p r e d i c t s . Also, the r e s u l t s from examination of standardization and d i v e r s i t y , d i r e c t evidence f o r production (tools, k i l n s ) , and ceramic technology (shaping, f i r i n g ) do not support the pr e d i c t i o n that there 228 should be evidence for change in mode of production over time. Again, larger samples of relevant data from Longshan sites may y i e l d different r e s u l t s . More information on areas of sites sampled for evidence of ceramic production i s necessary. A problem encountered i n any region i s that there are few test implications derived from ethnographic data for iden t i f y i n g change i n mode of production. This study suggests that a "complex household industry" mode (my term) characterizes most communities from the Longshan Period rather than a workshop mode, the assumption of some researchers. In the future when relevant reports are published, i t should be determined whether there are remains of workshops i n centers of settlement. Thus more analysis i s necessary to determine whether mode of ceramic production changes in a context of increasing cul t u r a l complexity. It i s possible that changes i n organization of labor for ceramic production do not occur u n t i l a f ter state formation takes place i n a region. Although the results from this study should be regarded as preliminary due to small sample size and i n s u f f i c i e n t information on the extent to which samples of vessels represent the t o t a l population of excavated vessels, they suggest that the model developed by Rice (1981) and the revised version offered here may not be supported i n a l l geographic areas. One factor that should be considered i s scale of analysis, both temporal and s p a t i a l . This dissertation refers to a large area i n northern China and a r e l a t i v e l y short time span, roughly 500 years (ca. 2500-2000 B.C.). 229 Rice (1981) t e s t s her model on ceramic data from a 1000 year period and a single s i t e i n B e l i z e . She i s concerned with t r a c i n g ceramic change over four developmental stages, culminating with s t r a t i f i e d s o c i e t i e s . If one compares the Longshan Period as a whole with the e a r l i e r N e o l i t h i c period i n northern China, i t could be said that there i s evidence f o r d i v e r s i f i c a t i o n i n ceramic production over time. The analyses on a f i n e r temporal scale described i n t h i s t hesis resulted i n d i f f e r e n t patterns f o r d i f f e r e n t phases. For example, the Early to Middle Period at Baiying i s characterized by d i v e r s i f i c a t i o n i n v a r i e t y of shape classes, and the Middle to Late, conservatism. C u l t u r a l differences between regions, such as northern China versus B e l i z e , may have an impact on r e s u l t s from analysis of ceramic change. This f a c t o r may help explain patterns that r e s u l t e d for the s i t e of Meishan, located i n the region i n which i t state formation probably f i r s t took place, Wangwan III i n west-central Henan. A pattern of conservatism resulted from nearly every analysis f o r Meishan. Also, there i s evidence f o r a decline in the production system with respect to labor-intensive wares. Polished and thin-walled vessels decrease over time. This pattern i s evident at the walled s i t e of Wangchenggang as w e l l , according to L i Xiandeng (1983). My observations of vessels from the l a t e s t phases at the s i t e on display at Gaocheng, Henan agree with t h i s conclusion. The Late Period at Meishan and Period V at Wangchenggang represent a phase that i s t r a n s i t i o n a l to the E r l i t o u 230 Period. Thus they may represent an early phase of state formation (Chapter 2). According to the model, there should be evidence for d i v e r s i f i - cation over time for labor-intensive vessels in the Wangwan III region in conjunction with increasing c u l t u r a l complexity. The model predicts that a decline i n labor-intensive techniques for ceramic vessels i s caused by a decline i n status competition. In this case, however, status competition with containers may not have decreased i n intensity. People may have decided to begin using another type of labor-intensive container during the terminal Longshan Period, bronze vessels. The Wangwan I I I region has the most physical evidence for a developing bronze industry during the Longshan Period (Chapter 2), whether sheet metal vessels were produced (La Plante 1988) i n addition to cast vessels or not. Therefore, i t i s possible that different results from testing the model may be achieved, depending on whether a more prestigious and labor-intensive material for containers was introduced at any point. This was not the case in Belize. A p r i o r i t y for future research should be to test other sites in the Wangwan I I I region. If there was indeed a decline i n use of labor- intensive pottery vessels for display purposes, then i t was probably a gradual one. Labor-intensive vessels are found at other Longshan sites i n the Wangwan I I I region such as Wadian (Chapter 5). However, I suggest that e l i t e s would have competed more intensely for control over production and use of bronze vessels than pottery vessels, once the 231 appropriate technology had been developed. Bronze vessels symbolize more labor input with respect to workers and equipment required for procurement of resources, preparation of materials f o r casting, and a c q u i s i t i o n of a substantial number of s k i l l e d craftsmen i n general. Another p r i o r i t y i n future research should be to obtain more information on v a r i a b i l i t y i n labor-intensive vessels. If larger samples of vessels were a v a i l a b l e , i t would be possible to use the production step index developed by Feinman et a l . (1981). Also, more data on uses of labor-intensive vessels f o r displays of status by e l i t e s and others are necessary. Sites should be examined f o r evidence of a c t i v i t i e s with these vessels such as displays of largesse or conspicuous consumption. The extent to which e l i t e s attempted to control production and/or d i s t r i b u t i o n of labor-intensive vessels by sponsoring s p e c i a l i s t s , etc. should be explored as the relevant data from Longshan s i t e s become a v a i l a b l e . It i s clear that there i s a need f o r more information on function and depositional patterns f o r ceramic vessels from Longshan s i t e s . Functional analysis could help i d e n t i f y categories of vessels that the model predicts should be important i n public displays of status such as vessels f o r cooking, serving, and drinking alchohol. It could also i d e n t i f y classes of non-prestige vessels used on a d a i l y basis by households. Increased recovery of vessels i n houses can provide more information on strategies of consumers. Comparisons should be made of vessels found i n houses with independent information on status of the 232 occupants. For example, Costin and Earle (1989) compare types of ceramics found i n houses and patio areas of e l i t e s with those of commoners ( i d e n t i f i e d from a r c h i t e c t u r a l data) from s i t e s dating to the la t e prehispanic period i n Peru. An important component of the model proposed by Rice (1981) that could not be assessed i n t h i s study i s change i n l o c a t i o n of ceramic production areas and in exchange systems on a regional scale. Change in production and d i s t r i b u t i o n of vessels within and between settlement systems should be a p r i o r i t y i n future research. In Europe, for example, compositional analysis of sherds has i d e n t i f i e d production s i t e s and d i s t r i b u t i o n areas f o r d i f f e r e n t shape classes of vessels (Attas et a l . 1987). Exchange systems may have become more complex over time i n northern China by involving more s i t e s , greater distances, or greater volumes of vessels. Exchange of at l e a s t some classes of vessels must have taken place during the Longshan Period, given that a s p e c i a l i s t mode of ceramic production was probably present. More research needs to be conducted i n order to determine whether e l i t e s attempted to increase control of ceramic production and d i s t r i b u t i o n i n a context of increasing s o c i o p o l i t i c a l complexity. Berman (1986), f o r example, concludes from neutron a c t i v a t i o n analysis of more than one ceramic ware that there i s no evidence f o r development of c e n t r a l i z e d production during the l a t e p r e h i s t o r i c period on the Susiana P l a i n of Iran. 2 3 3 Another topic that should be addressed i n future analyses of vessels from the Longshan Period i s the impact of technological change on ceramic change. This study discusses d i f f e r e n t techniques of production that may have been used such as moulds f or shaping (Chapter 6). Innovations i n shaping, decorating, or f i r i n g techniques as well as discoveries of new raw materials (clays, pastes) may cause potters to change strategies of production. For example, use of c e r t a i n types of clays can f a c i l i t a t e production of larger volumes of vessels at one time. Innovations i n techniques f o r moulds may have the same e f f e c t on production. CULTURAL CHANGE DURING THE LONGSHAN PERIOD This study has emphasized a regional approach i n assessing c u l t u r a l change during the Longshan Period. Although there i s evidence for regional d i v e r s i t y during the period, archaeological research has not focused on the region as a point of comparison (Chapter 2). As previously discussed, the ceramic analyses provided d i f f e r e n t r e s u l t s f o r the three regions examined i n most d e t a i l : Wangwan III (represented by Meishan), Hougang II (Hougang and Baiying), and Liangcheng (Lujiakou). For example, conservatism was more evident at the westernmost s i t e of Meishan, and d i v e r s i f i c a t i o n was more evident at the easternmost s i t e of Lujiakou. 234 Hougang, Baiying, Meishan, and Lujiakou belong to the "eastern" region that Keightley (1987) characterizes as homogeneous i n pottery production. Although I agree that there are s i m i l a r i t i e s among s i t e s , I suggest that the differences between sites are important and should be addressed more thoroughly i n future research. This study notes s i m i l a r i t i e s i n types of labor-intensive hypothesized prestige vessels across a wide region of the Huanghe (Yellow River). One example i s the use of very large vessels, from Taosi i n Shanxi to Hougang and Baiying i n northern Henan. On a smaller scale, there are nearly i d e n t i c a l thin-walled horn-shaped " l i d s " at two s i t e s , Gelawang and Yanzhuang, in the Wangwan I I I region. Another example i s the increase i n thin-walled vessels and the introduction of white wares during roughly the same time period i n the Hougang II region at Hougang (Middle to Late Period) and Baiying (Early to Middle Period). These s i m i l a r i t i e s suggest that there was considerable social interaction between regions during the Longshan Period. As Chang (1986) has pointed out, i t i s necessary to investigate the kinds of soci a l interactions that must have taken place between communities. One type of interaction may have been emulation of "foreign" high status individuals, including their" use of containers for display (Chapter 5). Another could have been exchange of vessels and/or t h e i r contents. This study has also emphasized a processual approach i n examining archaeological data from the Longshan Period. Future research should attempt to obtain more data on processes of change that must have been 235 important such as increasing population size and density, warfare, and increasing s o c i a l d i f f e r e n t i a t i o n (Chapter 2). There i s l i m i t e d information on how status differences were symbolized by material goods at s i t e s such as Hougang, Baiying, Meishan, and Lujiakou (Chapter 6). There are few nonceramic p o t e n t i a l prestige items at these s i t e s such as jade ornaments. One reason may be that mortuary contexts were regarded as more appropriate for displays of status with c r a f t items in general, or with respect to s p e c i f i c kinds of items such as jade (the cemetery at Taosi, for example). However, i t should be p r o f i t a b l e to examine v a r i a t i o n over time i n housing at habitation s i t e s . Analysis of the r e l a t i v e l y abundant data on housing from Hougang and Baiying indicates increasing d i f f e r e n t i a t i o n in terms of f l o o r area and wall construction material. Future research should investigate status d i f f e r e n t i a t i o n on a s i t e and regional basis more thoroughly. Differences i n terms of economic, p o l i t i c a l , and r i t u a l status should be assessed. Also, centers and outlying communities of the same settlement system should be compared with respect to q u a l i t y and quantity of goods i n general. Scholars have suggested, for example, that period II at the walled s i t e of Wangchenggang has yielded ceramic vessels of higher q u a l i t y than other contemporary s i t e s i n the Wangwan III region (Sui 1988:48). This pattern may not characterize a l l areas, since Hougang does not appear to have higher q u a l i t y vessels than other s i t e s in the Hougang II region 236 (Chapter 6). There may be differences over time as individual centers of settlement expand and collapse, too. Temporal scale of analysis may also affect conclusions reached about evidence for increasing c u l t u r a l complexity in a region. If a r e l a t i v e l y fine time scale i s considered such as 100 years, there may be no indication of increasing c u l t u r a l complexity. There may be evidence for some phases but not others. For example, there i s evidence for development of walled settlements during the onset of the Longshan Period i n west-central Henan, ca. 2500 B.C. There i s no indication at present of other changes i n type of settlement during the l a t e r Longshan Period (Chapter 2). More work on recognizing increases i n c u l t u r a l complexity with archaeological data during the Longshan Period i s needed, such as increases i n cen t r a l i z a t i o n and segregation (Flannery 1972). It would be profitable to examine changes in production of other types of crafts during the Longshan Period i n addition to pottery. There may be different patterns of change i n a context of increasing cu l t u r a l complexity for bronze or jade production. Bronze production, for example, may have involved s p e c i a l i s t s from the onset, due to s k i l l s and f a c i l i t i e s needed for extraction and processing of ores. It may have been more d i r e c t l y affected by increases i n c u l t u r a l complexity than pottery production. There may have been intense competition by e l i t e s to control production and d i s t r i b u t i o n of bronze items. 237 F i n a l l y , i t i s hoped that the analysis of shape classes discussed i n Chapter 4 w i l l be informative to others conducting research on Chinese N e o l i t h i c pottery. The terms used to designate shape classes should be regarded as generic and as s t a r t i n g points for a n a l y s i s . Also, there are a v a r i e t y of methods for describing vessels i n reports. This study has attempted to show how methods of ceramic analysis developed i n the West for c l a s s i f i c a t i o n and f o r other purposes such as understanding change in systems of production can be h e l p f u l i n examining Chinese N e o l i t h i c vessels. 238 BIBLIOGRAPHY Appadurai, Arjun 1986 Introduction: Commodities and the P o l i t i c s of Value. IN The Social L i f e of Things, edited by Arjun Appadurai, pp. 3-63. New York: Cambridge University Press. An Jiayuan 1986 Lei Bo Xiao Y i . (Some Notes on the Lei Bo). Kaogu 1986:4:344-347. An Jinhuai 1983a Jinnian l a i Henan Xia Shang Wenhua Kaogu de Xin Shouhuo. (Results of Recent Archaeological Research on Xia and Shang Culture i n Henan). Wenwu 1983:3:1-7. 1983b Shilun Dengfeng Wangchenggang Longshan Wenhua Chengzhi yu Xiadai Yangcheng. (A Preliminary Discussion of the Longshan Culture City at Wangchenggang in Dengfeng and the Xia Period Site at Yangcheng). IN Zhongguo Kaoguxuehui D i s i c i Nianhui Lunwenji. (A Collection of Papers from the Fourth Year of Chinese Archaeological Meetings), pp. 1-6. Beijing: Wenwu Press. 1987 personal communication, Zhengzhou, Henan; June 1987. 1989 Shilun Henan Diqu Longshan Wenhua de Shehui Xingzhi. (A Preliminary Discussion of the Nature of Society i n the Henan Longshan Culture). Zhongyuan Wenwu 1989:1:20-24. An Zhimin 1953 Taoqi. (Pottery). Wenwu Cankao Z i l i a o 1953:1:66-94. 1982 Gudai de Caomian Taoju. (Ancient Pottery Tools). IN Zhongguo Xinshiqi Shidai Lunji. (Essays on the Neolithic of China), edited by the Institute of Archaeology, CASS, pp.195-203. Beij i n g : Wenwu Press. 1988a Archaeological Research on Neolithic China. Current Anthropology 29(5 ):753-759. 1988b Zhongguo de Liqian Nongye. (Prehistoric Agriculture i n China). Kaogu Xuebao 1988:4:369-381. Anyang Archaeological Team, IA, Academica Sinica 1961 1958-1959 Nian Yin Hsu Fajue Jianbao. (Preliminary Report of the Excavation at Yin Hsu, Anyang 1958-1959). Kaogu 1961:2:63-76. 239 Anyang Archaeological Team, IA, CASS 1982 Anyang Hougang Xinshiqi Shidai Y i z h i de Fajue. (Excavation of the Neolithic Site at Hougang, Anyang). Kaogu 1982:6:565-581. 1985 1979 Nian Anyang Hougang Yizhi Fajue Baogao. (Report of the Excavation at Hougang i n Anyang, 1979). Kaogu Xuebao 1985:1:33-88. 1990 Anyang Dahancun Nangang Y i z h i . (Excavation of the South Mound at Dahan V i l l a g e , Anyang). Kaogu Xuebao 1990:1:43-68. Archaeological Group of the Changwei Area and the Museum of Zhucheng County 1980 Shandong Zhucheng Chengzi Yizhi Fajue Baogao. (Report on the Excavation of the Chengzi Site at Zhucheng, Shandong). Kaogu Xuebao 1980:3:329-385. Arnold, Dean 1985 Ceramic Theory and Cultural Process. New York: Cambridge University Press. Attas, Michael, John Fossey, and Leo Yaffe 1987 An Archaeometric Study of Early Bronze Age Pottery Production and Exchange i n Argolis and Korinthia (Corinthia), Greece. Journal of F i e l d Archaeology 14:77-90. Balfet, Helene 1965 Ethnographical Observations i n North Africa and Archaeological Interpretation: The Pottery of the Maghreb. IN Ceramics and Man, edited by Frederick Matson, pp. 161-177. Chicago: Aldine. Bankes, George 1985 The manufacture and Circulation of Paddle and Anvil Pottery on the North Coast of Peru. World Archaeology 17(2):269- 277. Barnes, Gina 1986 The Structure of Yayoi and Haji Ceramic Typologies. IN Windows on the Japanese Past: Studies i n Archaeology and Prehistory, edited by Richard Pearson, G. Barnes and K. Hutterer, co-editors, pp. 449-476. Ann Arbor: Center for Japanese Studies, University of Michigan. 240 Beaudry, Marilyn 1984 Ceramic Production and D i s t r i b u t i o n i n the Southeastern Maya Periphery. Late C l a s s i c Painted Serving Vessels. Oxford: BAR International Series 203. Benco, Nancy 1987 The Early Medieval Pottery Industry at al-Basra, Morocco. Oxford: BAR International Series 341. 1988 Morphological Standardization: An Approach to the Study of Craft S p e c i a l i z a t i o n . IN A Pot for a l l Reasons. Ceramic Ecology Revisited, edited by Charles Kolb and Louana Lackey, pp. 57-72. Philadelphia: Laboratory of Anthropology, Temple University. Berman, Judith 1986 Ceramic Production and the Development of Complex P o l i t i e s i n Late P r e h i s t o r i c Southwestern Iran. Unpublished PhD D i s s e r t a t i o n , Department of Anthropology, C i t y University of New York. Birmingham, Judy 1975 T r a d i t i o n a l Potters of the Kathmandu Va l l e y : An Ethnoarchaeological Study. Man 10(3):370-336. Blake, Michael 1988 Household Features and Social Process i n a Modern Maya Community. IN Ethnoarchaeology Among the Highland Maya of Chiapas, Mexico, edited by Thomas A. Lee, J r . and Brian Hayden, pp. 45-60. Papers of the New World Archaeological Foundation No. 56, Provo, Utah. Brumfiel, Elizabeth and Timothy Earle 1987 S p e c i a l i z a t i o n , Exchange, and Complex S o c i e t i e s : an Introduction. IN S p e c i a l i z a t i o n , Exchange, and Complex Societies, edited by Elizabeth Brumfiel and Timothy Earle, pp. 1-9. New York: Cambridge University Press. Cai Feng 1987 personal communication, Jinan, Shandong; July, 1987. Carneiro, Robert 1979 The Chiefdom: Precursor of the State. IN T r a n s i t i o n to Statehood i n the New World, edited by Grant Jones and Peter Kautz, pp. 37-79. New York: Cambridge University Press. 241 Chang, K.C. 1978 Early Chinese C i v i l i z a t i o n . Anthropological Perspectives. Cambridge: Harvard University Press. 1980 Shang C i v i l i z a t i o n . New Haven: Yale University Press. 1981 Archaeology and Chinese Historiography. World Archaeology 13(2):156-169. 1983a Sandai Archaeology and the Formation of States in Ancient China: Processual Aspects of the Origins of Chinese C i v i l i z a t i o n . IN Origins of Chinese C i v i l i z a t i o n , edited by David Keightley, pp. 495-521. Berkeley: University of C a l i f o r n i a Press. 1983b Art, Myth, and R i t u a l . The Path to P o l i t i c a l Authority in Ancient China. Cambridge: Harvard University Press. 1986 The Archaeology of Ancient China. Fourth E d i t i o n . Cambridge: Harvard University Press. Chang, K.C., C.C. Lin, M. Stuiver, H.Y. Tu, M. Tsukada, R.J. Pearson, T.M. Hsu. 196 9 Fengpitou, Tapenkeng, and the Prehistory of Taiwan. New Haven: Yale University Publications i n Anthropology, No. 73. Chavez, Karen 1985 T r a d i t i o n a l Pottery of Raqch'i, Cuzco, Peru: A Preliminary Study of i t s Production, D i s t r i b u t i o n , and Consumption. Nawpa Pacha 22/23:161-210. Cheng Te-k'un 1959 Archaeology i n China. Volume One. P r e h i s t o r i c China. Cambridge, England: W. Heffer and Sons Ltd. The Chinese-English Dictionary 1979 The Commercial Press, Ltd. Hong Kong. Clark, John 1986 Craft S p e c i a l i z a t i o n and Cu l t u r a l Complexity. Manuscript i n the Department of Anthropology, University of Michigan, Ann Arbor. 242 Cooper, Eugene 1982 The Potlatch i n Ancient China: P a r a l l e l s i n the So c i o p o l i t i c a l Structure of the Ancient Chinese and the American Indians of the Northwest Coast. History of Religions 22(2) :103- 128. Costin, Cathy 1986 From Chiefdom to Empire State: Ceramic Economy among the Prehispanic Wanka of Highland Peru. Unpublished PhD Dissertation, Department of Anthropology, University of C a l i f o r n i a , Los Angeles. Costin, Cathy Lynne and Timothy Earle 1989 Status D i s t i n c t i o n and Legitimization of Power as Reflected i n Changing Patterns of Consumption in Late Prehispanic Peru. American Antiquity 54(4 ):691-714. CPAM of Anyang D i s t r i c t , Henan Province 1983 Tangyin Baiying Henan Longshan Wenhua Cunluo Yi z h i Fajue Baogao. (Excavation Report of the Longshan Culture Community Site Baiying, at Tangyin, Henan). Kaoguxue Jikan 1983:3:1-50. Cultural Relics Work Team, Bureau of Culture, Hebei Province 1962 Hebei Yongnian Xian Taikoucun Y i z h i Fajue Jianbao (Preliminary Report on the Excavation of the Site at Taikou V i l l a g e , Yongnian County, Hebei). Kaogu 1962:12:635-640,645. Cultural Relics Work Team, Bureau of Culture, Henan Province 1958 Zhengzhou Niuzhai Longshan Wenhua Yizhi Fajue Baogao. (Report on the Excavation of the Longshan Culture Site Niuzhai i n Zhengzhou). Kaogu Xuebao 1958:4:19-26. Dai Yingxin 1988 Shenmu Shimao Longshan Wenhua Yuqi. (Jades of the Longshan Culture Site Shimao i n Shenmu County, Shaanxi). Kaogu Yu Wenwu 1988:5/6:239-250. Davenport, William 1986 Two kinds of Value i n the Eastern Solomon Islands. IN The Social L i f e of Things, edited by Arjun Appadurai, pp. 3-63, New York: Cambridge University Press. David, Nicholas and Hilke Hennig 1972 The Ethnography of Pottery: A Fulani Case Seen i n Archaeological Perspective. Reading, Massachusetts: Addison - Wesley Modular Publications 21. 243 Davis, Jack and Harriet Lewis 1985 Mechanization of Pottery Production: A Case Study from the Cycladic Islands. IN Prehistoric Production and Exchange. The Aegean and Eastern Mediterranean, edited by A. Bernard Knapp and Tamara Stech, pp. 79-92. Los Angeles: In s t i t u t e of Archaeology, University of C a l i f o r n i a Monograph XXV. DeBoer, Warren 1974 Ceramic Longevity and Archaeological Interpretation: An Example from the Upper Ucayali, Peru. American Antiquity 39(2): 335-343. 1984 The Last Pottery Show: System and Sense i n Ceramic Studies. IN The Many Dimensions of Pottery. Ceramics i n Archaeology and Anthropology, edited by S.E. van der Leeuw and A.C. Pritchard, pp. 527-571. CINGVLA VII. Amsterdam: Institute for Pre- and Proto-history, University of Amsterdam. DeBoer, Warren and Donald Lathrap 1979 The Making and Breaking of Shipibo-Conibo Ceramics. IN Ethnoarchaeology, edited by Carol Kramer, pp. 102-138. New York: Columbia University Press. Deal, Michael 1983 Pottery Ethnoarchaeology Among the T z e l t a l Maya. Unpublished PhD Dissertation, Department of Archaeology, Simon Fraser University, Burnaby, B r i t i s h Columbia, Canada. 1985 Household Pottery Disposal in the Maya Highlands: An Ethnoarchaeological Interpretation. Journal of Anthropological Archaeology 4:243-291. 1988 An Ethnoarchaeological Approach to the I d e n t i f i c a t i o n of Maya Domestic Pottery Production. IN Ceramic Ecology Revisited, 1987: The Technology and Socioeconomics of Pottery, edited by Charles C. Kolb, Part I, pp. 111-142. Oxford: BAR International Series 436. Department of Cultural Relics, Hebei Province 1975 Cixian Xiapanwang Yizhi Fajue Jianbao. (Preliminary Report on the Excavation of the Xiapanwang Site, Cixian County). Kaogu Xuebao 1975:1:73-116. Di e t l e r , Michael 1988 Driven by Drink: The Social Uses of Imported Wine and Implications for Economic and P o l i t i c a l Change i n French Iron Age Societies. Paper presented at the 53rd Annual Meeting of the Society for American Archaeology, Phoenix, Arizona, A p r i l 1988. 244 Douglas, Mary and Baron Isherwood 1979 The World of Goods. Towards an Anthropology of Consumption. London: Allen Lane Penguin Books Ltd. Du Zaizhong 1982 Shilun Longshan Wenhua de "Dan Ke Tao". (A Preliminary Discussion on the "Eggshell Pottery" from the Longshan Culture). Kaogu 1982:2:176-181. 1987 personal commmunication, July 1987, Weifang City Museum, Shandong. Earle, Timothy 1981 Comments on Rice. Current Anthropology 22(3 ):230-231. 1987a Chiefdoms i n Archaeological and Ethnohistorical Perspective. Annual Review of Anthropology 16:279-308. 1987b Specialization and the Production of Wealth: Hawaiian Chiefdoms and the Inka Empire. IN Specialization, Exchange, and Complex Soocieties, edited by Elizabeth Brumfiel and Timothy Earle, pp. 64-75, Cambridge University Press. Earle, Timothy and Robert Preucel 1987 Processual Archaeology and the Radical Critique. Current Anthropology 28(4 ):501-538. Ericson, Jonathon, D. Read and C. Burke 1972 Research Design: The Relationships between the Primary Functions and the Physical Properties of Ceramic Vessels and their Implications for Ceramic Distributions on an Archaeological Site. Anthropology UCLA 3(2):84-95. Ericson, Jonathon, and E.G. Stic k e l 1973 A Proposed C l a s s i f i c a t i o n System for Ceramics. World Archaeology 4(3 ):357-367. Fang Yang 1964 Wo Guo Niang J i u Dang Shi Yu Longshan Wenhua. (The Making of Alchohol i n our Country Begins with The Longshan Culture). Kaogu:1964:2:94-97. Fang Yousheng 1986 Shilun Yuxi Longshan Wenhua. (A Preliminary Discussion of the Longshan Culture i n Western Henan). Kaogu Yu Wenwu 1986:1:66-69. 245 Feinman, Gary, Stephen Kowalewski, Richard Blanton 1984 Modelling Ceramic Production and Organizational Change in the Pre-hispanic Valley of Oaxaca, Mexico. IN The Many Dimensions of Pottery, edited by S.E. van der Leeuw and A.C. Pritchard, pp. 296-333. CINGVLA VII. Amsterdam: Albert Egges Van Giffen Institute for Pre- and Proto-history. Feinman, Gary and J i l l Neitzel 1984 Too Many Types: An Overview of Sedentary Prestate Societies i n the Americas. IN Advances i n Archaeological Method and Theory. Volume 7, edited by Michael Schiffer, pp. 39-102. New York: Academic Press. Feinman, Gary, Steadman Upham, and Kent Lightfoot 1981 The Production Step Measure: An Ordinal Index of Labor Input i n Ceramic Manufacture. American Antiquity 46(41:871-884. Feng Xianming, An Zhimin, An Jinhuai, Zhu Boqian, Wang Qingzheng, editing group 1982 Zhongguo Taoci Shi. (A History of Chinese Pottery and Porcelain), edited by the Chinese S i l i c a t e Industry I n s t i t u t e . B e i j i n g : Wenwu Publishing House. F i r s t Cultural Relics Work Team, Bureau of Culture, Henan Province 1958 Zhengzhou Gelawangcun Yi z h i Fajue Baogao. (Report on the Excavation of the Site at Gelawang V i l l a g e , Zhengzhou). Kaogu Xuebao 1958:3:41-62. F i r t h , Raymond 1965 Primitive Polynesian Economy. Second'Edition. London: Routledge and Kegan Paul Ltd. Flannery, Kent 1972 The Cultural Evolution of C i v i l i z a t i o n s . Annual Review of Ecology and Systematics 3:399-426. Foster, George 1956 Pottery-Making i n Bengal. Southwestern Journal of Anthropology 12:395-405. 1965 The Sociology of Pottery: Questions and Hypotheses Arising from Contemporary Mexican Work. IN Ceramics and Man, edited by Frederick Matson, pp. 43-61. Chicago: Aldine. Fournier, Robert 1981 I l l u s t r a t e d Dictionary of Pottery Form. New York: Van Nostrand Reinhold Co. 246 Frankenstein, Susan and M.J. Rowlands 1978 The Internal Structure and Regional Context of Early- Iron Age Society i n South-western Germany. B u l l e t i n of the Institute of Archaeology 15:73-112. Franklin, Ursula 1983 The Beginnings of Metallurgy in China: A Comparative Approach. IN The Great Bronze Age of China: A Symposium, edited by George Kuwayama, pp. 94-99. Seattle: University of Washington Press. Friedman, Jonathon and M.J. Rowlands 1977 Notes Towards an Epigenetic Model of the Evolution of ' C i v i l i s a t i o n ' . IN The Evolution of Social Systems, edited by Jonathon Friedman and M.J. Rowlands, pp. 201-276. Pittsburgh: University of Pittsburgh Press. Froese, P a t r i c i a 1985 Pottery C l a s s i f i c a t i o n and Sherd Assignment. IN Decoding Prehistoric Ceramics, edited by Ben A. Nelson, pp. 229-242. Carbondale: Southern I l l i n o i s University Press. Gao T i a n l i n , Zhang Daihai, and Gao Wei 1984 Longshan Wenhua Taosi Leixing de Niandai Yu Fenqi. (Chronology and C l a s s i f i c a t i o n of the Taosi Type of Longshan Culture). Shiqian Yanjiu 1984:3:22-31. • Goldman, Irving 1970 Ancient Polynesian Society. Chicago: University of Chicago Press. Goody, Jack 1982 Cooking, Cuisine and Class. A Study i n Comparative Sociology. Cambridge: Cambridge University Press. Gosden, Chris 1989 Debt, Production, and Prehistory. Journal of Anthropological Archaeology 8:355-387. Gregory, C.A. 1982 G i f t s and Commodities. London: Academic Press. Guo Baojun 1981 Shang Zhou Tongqi Qun Zonghe Yanjiu. (A Comprehensive Study of the Shang and Zhou Bronze Vessel Group). Bei j i n g : Wenwu Press. 247 Haas, Jonathan 1982 The Evolution of the Prehistoric State. New York: Columbia University Press. 1986 China and the New World: Comparisons and Contrasts. Paper prepared for the Conference Ancient China and Social Science Generalizations, A i r l i e House, V i r g i n i a , June 21-26, 1986. Hagstrum, Melissa 1985 Measuring Prehistoric Ceramic Craft Specialization: A Test Case i n the American Southwest. Journal of F i e l d Archaeology 12(1):65-75. 1986 The Technology of Ceramic Production of Wanka and Inka Wares from the Yanamarca Valley, Peru. IN Ceramic Notes No. 3. Papers i n Ceramics Analysis, edited by Prudence Rice, pp. 1-29. Gainesville: Florida State Museum, University of Florida. 1988 Ceramic Production i n the Central Andes, Peru: An Archaeological and Ethnographic Comparison. IN A Pot For A l l Reasons. Ceramic Ecology Revisited, edited by Charles Kolb and Louana Lackey, pp. 127-145. Philadelphia: Laboratory of Anthropology, Temple University. Hally, David 1986 The Id e n t i f i c a t i o n of Vessel Function: A Case Study from Northwest Georgia. American Antiquity 51(2):267-295. Han Rong 1989 Shilun Chengziya Leixing. (A Preliminary Discussion on the Chengziya Cultural Type). Kaogu Xuebao 1989:2:137-159. Hansford, S. Howard 1954 A Glossary of Chinese Art and Archaeology. London: The China Society. Hao Tiechuan 1983 Longshan Wenhua Luan Zang Keng Shigushen Fen Bianxi. ( I d e n t i f i c a t i o n and Analysis of the Status of the Remains from the Mixed Bu r i a l P i t s of the Longshan Culture). Zhongyuan Wenwu 1983:4:39-42. Hardy-Smith, Anne 1974 Post-Medieval Pot Shapes. A Quantitative Analysis. Science and Archaeology 11:4-15. 248 Hartwig, Frederick and Brian Dearing 1979 Exploratory Data Analysis. Sage University Paper Series on Quantitiative Applications i n the Social Sciences, No. 16. Beverly H i l l s , C a l i f o r n i a : Sage Publications. Hayden, Brian and Aubrey Cannon 1983 Where the Garbage Goes: Refuse Disposal i n the Maya Highlands. Journal of Anthropological Archaeology 2:117-163. 1984 The Structure of Material Systems: Ethnoarchaeology in the Maya Highlands. Society for American Archaeology Papers No. 3, Washington, D.C. . Henan Province Cultural Research Institute and the Archaeological Section of the Museum of Chinese History 1983 Dengfeng Wangchenggang Y i z h i de Fajue. (Excavation of the Wangchenggang Site at Dengfeng). Wenwu 1983:3:8-20. Henan Province Cultural Research Institute and the Cultural Objects Division of the Zhoukou D i s t r i c t 1983 Henan Huaiyang Pingliangtai Longshan Wenhua Chengzhi Shijue Jianbao. (Preliminary Report on the Test Excavations at the Longshan Culture City Site of Pingliangtai i n Huaiyang, Henan). Wenwu 1983:3:21-36. Henan Province Cultural Research Institute and the Department of History at Zhengzhou University, Archaeology Specialty 1983 Yuxian Wadian Yiz h i Fajue Jianbao. (Preliminary Report on the Excavation of the Wadian Si t e , Yuxian County). Wenwu 1983:3:37-48. Henrickson, Elizabeth and M. McDonald 1983 Ceramic Form and Function: An Ethnographic Search and An Archaeological Application. American Anthropologist 85(3):630- 643. Howard, Hilary 1981 In the Wake of Dist r i b u t i o n : Towards an Integrated Approach to Ceramic Studies in Prehistoric B r i t a i n . IN Production and Dist r i b u t i o n : A Ceramic Viewpoint, edited by H. Howard and E. Morris, pp. 1-30. BAR International Series 120. Huber, Louisa F. 1983 The Relationship of the Painted Pottery and Lung-shan Cultures. IN The Origins of Chinese C i v i l i z a t i o n , edited by David Keightley, pp. 177-216. Berkeley: University of Ca l i f o r n i a Press. 249 Institute of Archaeology, CASS 1983 Zhongguo Kaoguxue Zhong Tanshisi Niandai Shuji. (Chronological Data on Radiocarbon Dates i n Chinese Archaeology). Be i j i n g : Wenwu Press. 1984 Xin Zhongguo de Kaogu Faxian He Yanjiu. (Archaeological Excavation and Research i n New China). Beijing: Wenwu Press. Institute of Archaeology, Shandong Province 1985 Chiping Shangzhuang Xinshiqi Shidai Y i z h i . (The Neolithic Site of Shangzhuang at Chiping). Kaogu Xuebao 1985:4:465-505. 1988 Jiaoxian Sanlihe. (The Sanlihe Site at Jiaoxian). Be i j i n g : Wenwu Press. Institute of Archaeology, Shandong Province; Shandong Provincial Museum; Shandong Team, Institute of Archaeology, CASS, and Department of Culture, Changwei D i s t r i c t , Shandong Province 1981 Shandong Yaoguanzhuang Yizhi Fajue Baogao. (Report on the Excavation of the Yaoguanzhuang Site, Shandong). Wenwu Z i l i a o Congkan 1981:5:1-83. Johnson, Allen and Timothy Earle 1987 The Evolution of Human Societies. From Foraging Group to Agrarian State. Stanford, C a l i f o r n i a : Stanford University Press. Kaiser, Timothy 1984 Vinca Ceramics: Economic and Technological Aspects of Late Neolithic Pottery Production i n Southeast Europe. Unpublished PhD Dissertation, Department of Anthropology, University of C a l i f o r n i a , Berkeley. Kamp, Kathryn 1987 Affluence and Image: Ethnoarchaeology i n a Syrian V i l l a g e . Journal of F i e l d Archaeology 14(3 ):283-296. Kaplan, Flora and David Levine 1981 Cognitive Mapping of a Folk Taxonomy of Mexican Pottery: A Multivariate Approach. American Anthropologist 83(4) :868-884. Keightley, David 1985a Pot Makers and Users i n the Central Plains: Cultural Interaction in the Chinese Neolithic, paper prepared for the Annual Meeting of the American H i s t o r i c a l Association, New York, New York, December 28, 1985. 250 1985b Dead but not Gone: The Role of Mortuary Practices in the Formation of Neolithic and Early Bronze Age Chinese Culture, ca. 8000 to 1000 B.C. paper prepared for the Conference, Ritual and the Social Significance of Death in Chinese Society, The Sun Space Ranch Conference Center, Oracle, Arizona, January 2-7, 1985 1987 Archaeology and Mentality: The Making of China. Representations 18:91-128. Kempton, W i l l e t t 1981 The Folk C l a s s i f i c a t i o n of Ceramics. A Study of Cognitive Prototypes. New York: Academic Press. Kobayashi, Tatsuo 1974 Behavioral Patterns Reflected i n Pottery Remains - The Jomon Period. A r c t i c Anthropology XI Supplement:163-170. Kramer, Carol 1985 Ceramic Ethnoarchaeology. Annual Review of Anthropology 14:77-102. Kristiansen, K r i s t i a n 1984 Ideology and Material Culture: an Archaeological Perspective. IN Marxist Perspectives i n Archaeology, edited by Matthew Spriggs, pp. 72-100. New York: Cambridge University Press. Krotser, Paula 1974 Country Potters of Veracruz, Mexico. Technological Survivals and Culture Change. IN Ethnoarchaeology, edited by Christopher Donnan and C.W. Clewlow, J r . , pp. 131-146. Institute of Archaeology, University of C a l i f o r n i a , Los Angeles, Archaeological Survey Monograph IV. Lacovara, Peter 1985 The Ethnoarchaeology of Pottery Production in an Upper Egyptian V i l l a g e . IN Ancient Technology to Modern Science. Ceramics and C i v i l i z a t i o n Volume 1, edited by W.D. Kingery, pp. 51-60. Columbus, Ohio: The American Ceramic Society, Inc. La Plante, John 1988 Ancient Chinese Ritual Vessels. Some Observations on Technology and Style. Early China 13:247-273. Leeuw, Sander van der 1977 Towards a Study of the Economics of Pottery Making. Ex Horreo 4:68-76. 251 1984 Dust to Dust: A Transformational View of the Ceramic Cycle. IN The Many Dimensions of Pottery. Ceramics i n Archaeology and Anthropology, edited by S.E. van der Leeuw and A.C. Pritchard, pp. 707-778. CINGVLA VII. Institute for Pre- and Proto-history, University of Amsterdam. L i Chi 1956 Pottery of the Yin and Pre-Yin Period: A C l a s s i f i e d and Descriptive Account with a Corpus of A l l the Main Types. Archaeologica Sinica, no. 2. Hsiao-t' un: The Yin-Shang Site at Anyang, Honan. Vol. 3. A r t i f a c t s . Fasc. 1, pt. 1. Taipei: Academia Sinica, Institute of History and Philology, 1956. 1977 Anyang. Seattle: University of Washington Press. L i Chi, editors i n chief, Liang Ssu-yung, and Tung Tso-pin, editors; Fu Ssu-nien, Wu Chin-ting, Kuo Pao-chun, and Liu Yu-hsia 1956 Ch'eng-tzu-yai: The Black Pottery Culture Site at Lung- shan-chen i n Li-ch'eng-hsien, Shantung Province. New Haven: Yale University Press. translated by Kenneth Starr L i Jiafang and Gao Guangren 1979 Dianxing Longshan Wenhua de Laiyuan, Fazhan j i Shehui Xingzhi Chutan. (Preliminary Research on the Origin, Development and Nature of Society of Longshan Culture). Wenwu 1979:11:56-62. translated Chui-mei Ho, IN Chinese Archaeological Abstracts, 2. Prehistoric to Western Zhou. Monumenta Archaeologica 9, edited by Albert Dien, Jeffrey Riegel, and Nancy Price, pp. 191-197, Los Angeles: The Institute of Archaeology, University of C a l i f o r n i a , Los Angeles. L i Jianmin 1984 Dawenkou Muzang Chutude J i u q i . (Wine Vessels Excavated from the Dawenkou B u r i a l s ) . Kaogu Yu Wenwu 1984:6:64-68. L i J i a z h i 1984 The Evolution of Chinese Pottery and Porcelain Technology. IN Ancient Technology to Modern Science. Ceramics and C i v i l i z a t i o n . Volume 1, edited by W.D. Kingery, pp. 135-162. Columbus, Ohio: American Ceramic Society. L i Kuang-chou 1981 K'en-ting: An Archaeological Natural Laboratory near Southern Tip of Taiwan, unpublished PhD Dissertation, Department of Anthropology, State University of New York, Binghampton. 252 L i Xiandeng 1983 Dengfeng Gaocheng Wangchenggang Y i z h i de Chubu Fenxi. (A Preliminary Analysis of the Wangchenggang S i t e , Dengfeng, Gaocheng). IN Zhongguo Kaoguxuehui D i s i c i Nianhui Lunwenji (A C o l l e c t i o n of Papers from the Fourth Year of Chinese Archaeological Meetings), edited by the Chinese Archaeology E d i t i n g Group, pp. 7-12. B e i j i n g : Wenwu Press. L i Xueqin 1980 The Wonder of Chinese Bronzes. B e i j i n g : Foreign Languages Press. L i Yangsong 1959 Cong Wazu Zhitao Tantao Gudai Taoqi Zhizuoshangde Jige Wenti. (Some Problems of Ancient Pottery Making as Seen from an Inquiry of Pottery Manufacture among the Wa N a t i o n a l i t y ) . Kaogu 1959:5:250-254. 1962 Dui Woguo Niangjiu Qiyuan de Tantao. (An Investigation Regarding the Origins of Wine Making in Our Country). Kaogu 1962:1:41-44. L i Zhiyan and Cheng Wen 1984 Chinese Pottery and Porcelain. B e i j i n g : Foreign Languages Press. Lischka, Joseph 1978 A Functional Analysis of Middle C l a s s i c Ceramics at Kaminaljuyu. IN The Ceramics of Kaminaljuyu, Guatemala, edited by Ronald Wetherington, pp. 223-278. University Park: The Pennsylvania State University Press. Longacre, William 1981 Kalinga Pottery. An Ethnoarchaeological Study. IN Patterns of the Past, edited by Ian Hodder, G. Isaac, N. Hammond, pp. 49-66. Cambridge: Cambridge University Press. 1985 Pottery U s e - l i f e Among the Kalinga, Northern Luzon, the P h i l i p p i n e s . IN Decoding P r e h i s t o r i c Ceramics, edited by Ben Nelson, pp. 334-346. Carbondale: Southern I l l i n o i s University Press. Longacre, William, K. Kvamme, and M. Kobayashi 1988 Southwestern Pottery Standardization: An Ethnoarchaeological View From the P h i l i p p i n e s . The Kiva 53(2) :101-112. 253 Maisels, Charles 1987 ' ' Models of Social Evolution: Trajectories from the Neolithic to the State. Man 22(2):331-359. Malinowski, Bronislaw 1922 Argonauts of the Western P a c i f i c . London: George Routledge and Sons, Ltd. . Matson, R.G. 1990 personal communication, Department of Anthropology and Sociology, University of B r i t i s h Columbia, Vancouver, B r i t i s h Columbia, June 1990. Medley, Margaret 1976 The Chinese Potter. A P r a c t i c a l History of Chinese Ceramics. New York: Charles Scribner's Sons. Milanich, Jerald, Ann Cordell, Vernon Knight, Timothy Kohler, and Brenda Sigler-Lavelle. 1984 McKeithen Weeden Island. The Culture of Northern Fl o r i d a , A.D. 200-900. Orlando: Academic Press. M i l l e r , Daniel 1982 Structures and Strategies: An Aspect of the Relationship Between Social Hierarchy and Cultural Change, IN Symbolic and Structural Archaeology, edited by Ian Hodder, pp. 89- 98. New York: Cambridge University Press. 1985 Artefacts as Categories. A Study of Ceramic V a r i a b i l i t y i n Central India. New York: Cambridge University Press. Moore, Jerry 1989 Pre-Hispanic Beer in Coastal Peru. American Anthropologist 91(3):682-695. Museum of Zhengzhou City 1979 Zhengzhou Dahecun Yiz h i Fajue Baogao (Report on the Excavation of the Site at Dahe V i l l a g e , Zhengzhou). Kaogu Xuebao 1979:3:301-375. 1983 Zhengzhou Yanzhuang Longshan Wenhua Yiz h i Fajue Jianbao. (Preliminary Report on the Excavation of the Longshan Culture Site Yanzhuang, Zhengzhou). Zhongyuan Wenwu 1983:4:1-8. Nelson, Ben 1981 Ethnoarchaeology and Paleodemography: A Test of Turner and Lofgren's Hypothesis. Journal of Anthropological Research 37(2) :107-129. 254 Netting, Robert McC. 1982 Some Home Truths on Household Size and Wealth. American Behavioral S c i e n t i s t 25(6 ): 641-662. N i c k l i n , Keith 1971 S t a b i l i t y and Innovation in Pottery Manufacture. World Archaeology 3(1):13-48,94. Oliver, Douglas 1955 A Solomon Island Society. Kinship and Leadership Among the Siuai of Bougainville. Boston: Beacon Press. Orme, Bryony 1981 Anthropology for Archaeologists: An Introduction. Ithaca, New York: Cornell University Press. Pankenier, David 1985 Mozi and the Dates of Xia, Shang, and Zhou: A Research Note. Early China 9-10:175-183. Pastron, Allen 1974 Preliminary Ethnoarchaeological Investigations Among the Tarahumara. IN Ethnoarchaeology, edited by Christopher Donnan and CW. Clewlow, J r . , pp. 93-114. Institute of Archaeology, University of C a l i f o r n i a , Los Angeles, Archaeological Survey Monograph IV. Peacock, David 1981 Archaeology, Ethnology and Ceramic Production. IN Production and Distri b u t i o n : A Ceramic Viewpoint, edited by Hilary Howard and Elaine Morris, pp. 187-194. Oxford: BAR International Series 120. 1982 Pottery i n the Roman World. London: Longman Press. Pearson, Richard 1981 Social Complexity i n Chinese Coastal Neolithic Sites. Science 213(4 ):1078-1086. 1988 Chinese Neolithic Burial Patterns: Problems of Method and Interpretation. Early China 13:1-45. Pearson, Richard and Anne Underhill 1987 The Chinese Neolithic: Recent Trends i n Research. American Anthropologist 89(4 ):807-822. 255 Peebles, Christopher, and Susan Kus 1977 Some Archaeological Correlates of Ranked Societies. American Antiquity 42(3 ):421-429. Reina, Reuben and Robert H i l l 1978 The Traditional Pottery of Guatemala. Austin: University of Texas Press. Ren Mei'e, Yang Renzhang, and Bao Haosheng 1985 An Outline of China's Physical Geography. Beijing: Foreign Languages Press. Renmin Ribao (People's Daily) 1985 Shandong Faxian Siqian Nianqian de Gu Chengbao Y i z h i . (Discovery of a Four Thousand Year Old Ancient Castle Site i n Shandong). People's Daily January 3, 1985. 1986 Henan Fajue Chu Longshan Wenhua Y i z h i . (Excavation of a Longshan Culture Site i n Henan). People's Daily December 18, 1986. Rice, Prudence 1977 Whiteware Pottery Production i n the Valley of Guatemala: Specialization and Resource U t i l i z a t i o n . Journal of F i e l d Archaeology 4(2 ):221-233. 1981 Evolution of Specialized Pottery Production: A T r i a l Model. Current Anthropology 22(3):219-240. 1984 Change and Conservatism i n Pottery-Producing Systems. IN The Many Dimensions of Pottery. Ceramics i n Archaeology and Anthropology, edited by S.E. van der Leeuw and A.C. Pritchard, pp. 231-293. CINGVLA VII. Institute for Pre- and Proto-history, University of Amsterdam. 1987 Pottery Analysis. A Sourcebook. Chicago: University of Chicago Press. 1988 Specialization, Standardization, and Diversity: A Retrospective. Paper prepared for the conference Ceramic Analysis and Social Inference in American Archaeology: The Ceramic Legacy of Anna 0. Shepard, October 31- November 2, 1988. 1989 Ceramic Diversity, Production, and Use. IN Quantifying Diversity i n Archaeology, edited by Robert Leonard and George Jones, pp. 109-117. New York: Cambridge University Press. 1990 personal communication, April-May 1990. 256 Rice, Prudence, and M.E. Saffer 1982 Cluster Analysis of Mixed-Level Data: Pottery Provenience as an Example. Journal of Archaeological Science 9(4):395-409. Richards, Audrey 1939 Land, Labour and Diet i n Northern Rhodesia. An Economic Study of the Bemba Tribe. London: Oxford University Press. Rye, Owen and C l i f f o r d Evans 1976 Traditional Pottery Techniques of Pakistan. F i e l d and Laboratory Studies. Smithsonian Contributions to Anthropology No. 21. Washington, D.C: Smithsonian I n s t i t u t i o n Press. Santley, Robert, P h i l i p Arnold, and Christopher Pool 1989 The Ceramics Production System at Matacapan, Veracruz, Mexico. Journal of F i e l d Archaeology 16(1):107-132. Schiffer, Michael 1976 Behavioral Archaeology. New York: Academic Press. Second Henan Archaeological Team, IA, CASS 1982 Henan Linru Meishan Yi z h i Fajue Baogao. (Excavation of the Meishan Site in Linru, Henan Province). Kaogu Xuebao 1982:4:427-476. Shandong Archaeological Team, IA, CASS and the Art Museum of Weifang County, Shandong Province. 1985 Weixian Lujiakou Xinshiqishidai Y i z h i . (The Neolithic Lujiakou Site of Wei County). Kaogu Xuebao 1985:3:313-351. Shangraw, Clarence 1978 Origins of Chinese Ceramics. China House Gallery/China Institute i n America, Inc. Shanxi Archaeological Team, IA, CASS and the Linfen D i s t r i c t Cultural Bureau 1983 1978-1980 Nian Shanxi Xiangfen Taosi Mudi Fajue Jianbao. (Preliminary Report of the Excavations of the Cemetery Site of Taosi i n Xiangfen, Shanxi, 1978-1980.) Kaogu 1983:1:30- 42. 1984 Shanxi Xiangfen Taosi Yizhi Shouci Faxian Tongqi. (Discovery of Bronze for the F i r s t Time at the Taosi Site i n Xiangfen, Shanxi). Kaogu 1984:12:1069-1071. 257 Sherman, David 1988 Quantifying Archaeology. Edinburgh University Press. Shennan, S.J. and J.D. Wilcock 1975 Shape and Style Variation in Central German B e l l Beakers: A Computer-Assisted Study. Science and Archaeology 15:17-31. Shepard, Anna 1976 Ceramics for the Archaeologist. Washington, D.C: Carnegie I n s t i t u t i o n . Siegel, Sidney, and N. John Castellan, J r . 1988 Nonparametric S t a t i s t i c s for the Behavioral Sciences. Second Edition. New York: McGraw-Hill Book Company. Sinopoli, Carla 1986 Material Patterning and Social Organization: A Study of Ceramics from Vijayanagara, South India. Unpublished PhD Dissertation, Department of Anthropology, University of Michigan. 1988 The Organization of Craft Production at Vijayanagara, South India. American Anthropologist 90(3 ):580-597. Smith, Marion Floyd, J r . 1985 Toward an Economic Interpretation of Ceramics: Relating Vessel Size and Shape to Use. IN Decoding Prehistoric Ceramics, edited by Ben Nelson, pp. 254-309. Carbondale: Southern I l l i n o i s University Press. 1988 Function from Whole Vessel Shape: A Method and an Application to Anasazi Black Mesa, Arizona. American Anthropologist 90(4 ): 912 -923. Smith, Michael 1987 Household Possessions and Wealth i n Agrarian States: Implications for Archaeology. Journal of Anthropological Archaeology 6(4 ):297-335. Solheim, Wilhelm 1965 The Functions of Pottery i n Southeast Asia: From the Present to the Past. IN Ceramics and Man, edited by Frederick Matson, pp. 254-273. Chicago: Aldine. Song Zhaolin, L i Jiafang, Du Yaoxi 1983 Zhongguo Yuanshi Shihui Shi. (The History of Chinese Primitive Society). Beijing: Wenwu Publishing House. 258 Stark, Barbara 1985 Archaeological I d e n t i f i c a t i o n of Pottery Production Locations: Ethnoarchaeological and Archaeological Data in Mesoamerica. IN Decoding Prehistoric Ceramics, edited by Ben Nelson, pp. 158-194. Carbondale: Southern I l l i n o i s University Press. Steponaitis, Vincas 1981 Settlement Hierarchies and P o l i t i c a l Complexity in Nonmarket Societies: The Formative Period i n the Valley of Mexico. American Anthropologist 83:320-363. Sui Yuren 1988 Huanghe 2hongxiayou Longshan Wenhua Chengbao Chutan. (A Preliminary Exploration of the Longshan Culture "Castles" from the Middle and Lower Reaches of the Yellow River). Zhongyuan Wenwu 1988:4:46-52. Sullivan, Alan 1988 Prehistoric Southwestern Ceramic Manufacture: The Limitations of the Current Evidence. American Antiquity 53(1 ):23- 35. Sun Shuyun and Han Rubin 1985 A Preliminary Study of Early Chinese Copper and Bronze A r t i f a c t s . Early China 9/10:261-289. translated by J u l i a Murray. Tian Changwu 1986 The Legends of Yao and Shun and the Origins of Chinese C i v i l i z a t i o n , paper presented at the Conference on Ancient China and Social Science Generalizations, A i r l i e House, V i r g i n i a , June 21-27, 1986. T o l l , H. Wolcott 1981 Ceramic Comparisons Concerning Redistribution i n Chaco Canyon, New Mexico. IN Production and D i s t r i b u t i o n : A Ceramic Viewpoint, edited by Hilary Howard and Elaine Morris, pp. 83-121. Oxford: BAR International Series 120. Tong Enzheng 1989 Youguan Wenming Qiyuan de Jige Wenti- Yi An Zhimin Xiansheng Shangque. (Several Issues Regarding the Origins of C i v i l i z a t i o n - A Discussion with Mr. An Zhimin). Kaogu 1989:1:51-59,32. 259 Tosi, Maurizio 1984 The Notion of Craft Specialization and i t s Representation i n the Archaeological Record of Early States i n the Turanian Basin. IN Marxist Perspectives i n Archaeology, edited by Matthew Spriggs, pp. 22-52. New York: Cambridge University Press. Tregear, T.R. 1970 Economic Geography of China. New York: American Elsevier Publishing Co., Inc. Trinkhaus, K. Mauer 1984 Mortuary Ritual and Mortuary Remains. Current Anthropology 25(5 ): 674-679. Tschopik, J r . , Harry 1950 An Andean Tradition in H i s t o r i c a l Perspective. American Antiquity 15(3 ):196-218. Underhill, Anne P. 1983 A Mortuary Analysis of the Dawenkou Cemetery Site, Shandong, China. Unpublished M.A. thesis, Department of Anthropology and Sociology, University of B r i t i s h Columbia, Vancouver, B r i t i s h Columbia, Canada. 1989 Warfare During the Chinese Neolithic Period: A Review of the Evidence. IN Cultures in C o n f l i c t . Current Archaeological Perspectives, edited by Diana Tkaczuk and Brian Vivian, pp. 229- 237. Calgary, Alberta: The Archaeological Association, University of Calgary. i n press, a T'ao-ssu, Wang-ch' eng-kang, P'ing-liang-t'ai, and Hou-kang: Communities at the Threshold of the Bronze Age. paper prepared for the Conference Ancient Chinese and Southeast Asian Bronze Age Cultures, Edith and Joy London Foundation Property, Kioloa, NSW, Austr a l i a , February 8-12, 1988. in press, b Cultural Change During the Longshan Period of Northern China, paper prepared for the Circum-Pacific Prehistory Conference, August 2-6, 1989, Seattle, Washington. Vandiver, Pamela 1988 The Implications of Variation in Ceramic Technology: The Forming of Neolithic Storage Vessels i n China and the Near East. Archeomaterials 2(2):139-174. 260 Wang Shuming 1986 Tan Lingyanghe Yu Dazhucun Chutude Taozun "Wenzi". (A Discussion of the Characters on the Pottery Zun Unearthed from Lingyanghe at Dazhu V i l l a g e ) . IN Collected Works Discussing P r e h i s t o r i c Cultures of Shandong (Shandong Shiqian Wenhua Lunwenji), edited by the Shandong Province Archaeological Series E d i t o r i a l Department, pp. 249-308. Jinan: Shandong Province C u l t u r a l Bureau. Washburne, Chandler 1961 Primitive Drinking. A Study of the Uses and Functions of Alchohol in P r e l i t e r a t e S o c i e t i e s . New York: College and Uni v e r s i t y Press Publishers. Welch, Paul 1986 Models of Chiefdom Economy: P r e h i s t o r i c Moundville as a Case Study. unpublished PhD Di s s e r t a t i o n , Department of Anthropology, University of Michigan. Wells, Peter 1985 Material Symbols and the Interpretation of Cu l t u r a l Change. Oxford Journal of Archaeology 4(1):9-17. Whallon, Robert 1982 Variables and Dimensions: The C r i t i c a l Step i n Quantitative Typology. IN Essays i n Archaeological Typology, edited by Robert Whallon and James Brown, pp. 127-161. Evanston, I l l i n o i s : Center for American Archaeology Press. Wilkinson, Leland 1988a SYSTAT: The System f o r S t a t i s t i c s . Evanston, I l l i n o i s : SYSTAT, Inc. 1988b SYGRAPH. Evanston, I l l i n o i s : SYSTAT, Inc. Wobst, Martin 1977 S t y l i s t i c Behavior and Information Exchange. IN For the Director: Research Essays i n Honor of James G r i f f e n , edited by C.E. Cleland, pp. 317-342. Anthropological Papers No. 61. Ann Arbor: Museum of Anthropology, University of Michigan. Woods, Ann 1985 Form, Fabric, and Function: Some Observations on the Cooking Pot in Antiquity.. IN Technology and Style. Ceramics and C i v i l i z a t i o n . Volume 2, edited by W.D. Kingery, pp. 157-172. Columbus, Ohio: The American Ceramic Society, Inc. 261 Wright, Henry 1977 Recent Research on the Origin of the State. Annual Review of Anthropology 6:379-397. 1978 Toward an Explanation of the Origin of the State. IN Origins of the State. The Anthropology of P o l i t i c a l Evolution, edited by Ronald Cohen and Elman Service, pp. 49-68. Philadelphia: Institute for the Study of Human Issues. 1984 Prestate P o l i t i c a l Formations. IN On the Evolution of Complex Societies. Essays i n Honor of Harry Hoijer 1982, edited by Timothy Earle, pp. 41-77. Malibu, C a l i f o r n i a : Undena Publications. Wu Chin-ting 1938 " Prehistoric Pottery i n China. London: Kegan Paul, Trench, Trubner and Co., Ltd. Wu Ruzuo and Du Zaizhong 1984 Liangcheng Leixing Fenqi Wenti Chutan. (A Preliminary Study of the Periodization of the Liangcheng Type of Longshan Culture). Kaogu Xuebao 1984:1:1-22. Xu Hong 1989 Lue Lun Wo Guo Qianshiqi Weng Guan Zang. (A Brief Discussion of the Prehistoric Burial Urn i n Our Country). Kaogu 1989:4:331-339. Xu Yuanbang, Liu Suicheng, and Liang Xingreng 1982 Zhongguo Gudai Taoqi Yao. (Ancient Pottery Kilns from China). Kaogu Yu Wenwu 1982:5:1-30. Yan Wenming 1981 Longshan Wenhua he Longshan Shidai. (The Longshan Culture and the Longshan Period). Wenwu 1981:6:41-48. 1982a Jiangou de Tougai Bei he Botoupi Fengsu. (Skull Cap Cups from Jiangou and the Custom of Scalping). Kaogu Yu Wenwu 1982:2:38-41. 1982b Zhongguo Daozuo Nongye de Qiyuan. (Origins of Rice Agriculture in China). Nongye Kaogu 1982:1:19-31,151. 1984 Lun Zhongguo de Tongshibingyong Shidai. (A Discussion of the Cha l c o l i t h i c Period i n China). Shiqian Yanjiu 1984:1:36- 44,35. 262 1985 Kaogu Z i l i a o Zhengli Zhong de Biaoxingxue Yanjiu. (Research i n the Study of Classifying Archaeological Data). Kaogu Yu Wenwu 1985:4:30-40. 1986 Unity and Diversity i n Chinese Neolithic Culture, paper prepared for the Conference Ancient China and Social Science Generalizations, A i r l i e House, V i r g i n i a , June 21-26, 1986. translated by Terry Kleeman. 1987a Zhongguo Shiqian Wenhua de Tongyixing Yu Duoyangxing. (Unity and Diversity i n Chinese Neolithic Culture). Wenwu 1987:3:38-50. 1987b personal communication, Bei j i n g , March- June, 1987. 198 9 The Origins of Agriculture and Animal Husbandry i n China, paper prepared for the Circum-Pacific Conference, Seattle, Washington, August 2-6, 1989. Ye Wenkuan 1989 Lei Bo Yuanliu Kao. (An Investigation of the Origin and Development of the Lei-Bo). Kaogu 1989:5:456-462. Ye Wansong and Yu Fuwei 1984 Henan Longshan Wenhua de Shehui J i n g j i Jichu. (The Economic and Social Foundation of the Henan Longshan Culture). Zhongyuan Wenwu 1984:3:81-85. Zeng Qi 1985 Wo Guo Xinshiqi Shidai de Shengchan Gongju Zongshu. (A Summary of Production Tools from the Neolithic Period i n Our Country). Kaogu Yu Wenwu 1985:5:61-73. Zhang J i 1959 Xishuangbanna Daizu de Zhitao Jishu. (Pottery Making Techniques among the Dai People of Xishuangbanna). Kaogu 1959:9:488-490. Zhang Xinglian 1983 A Glossary of Chinese Archaeology. Beijing: Foreign Languages Press. 263 Zhang Yanhuang and Zhang Daihai 1986 Zhongyuan Diqu Longshan Wenhua de Leixing he Niandai. (The Types and Ages of Longshan Culture i n the Central Plain D i s t r i c t ) . IN Zhongguo Kaoguxue Yanjiu. Xia Nai Xiansheng Kaogu Wushi Nianji Nian Lunwenji. (Research i n Chinese Archaeology. A Collection of Essays to Commemorate 50 Years of Archaeology with Mr. Xia Nai)/ edited by the E d i t o r i a l Committee of Archaeological Research in China, pp. 46-56. Beijing: Wenwu Press. Zhao Liansheng 1987 personal communication, Puyang, Henan; May 1987. Zhao Zhenxi 198 9 Woguo Zhitao Lunpan de Qiyan Yi Zaoqi de Yingyong. (The Origins and Early Period of Use of Wheel Made Pottery i n Our Country). Kaogu Yu Wenwu 1989:4:80-84,90. Zhong Huanan 1987 personal communication, Jinan, Shandong; July 1987. Zhou Ren, Zhang Fukang, Zheng Shuipu 1982 Wo Guo Huanghe Liuyu Xinshiqishidai he Yinzhou Shidai Zhi Tao Gongye de Kexue Zongjie. ( S c i e n t i f i c Conclusions on Ceramic Manufacturing Technology during the Neolithic Period and the Yin-Zhou Periods of the Yellow River Valley of Our Country). IN Collected Works Discussing Chinese Ancient Pottery and Porcelain (Zhongguo Gu Taoci Lunwenji), edited by the Chinese S i l i c a t e Industry I n s t i t u t e , pp. 263-286. Beijing: Wenwu Publishing House. Zou Heng 1987 personal communication, Bei j i n g , A p r i l 1987. 264 APPENDIX A. ANALYSIS OF SHAPE CLASSES IN ARCHAEOLOGICAL REPORTS. s i t e reports: Hougang (Anyang Archaeological Team, IA, CASS 1985) Baiying (CPAM of Anyang D i s t r i c t , Henan Province 1983) Meishan (Second Henan Archaeological Team, IA, CASS 1982) Lujiakou (Shandong Archaeological Team, IA, CASS and the Art Museum of Weifang County, Shandong Province 1985) 265 Table 36. Changes Made i n Shape Classes of Vessels I d e n t i f i e d in the Hougang Report. change i n c l a s s i f i c a t i o n method of evaluation 1. pot moved from the large size of guan to medium size class range of values f or HT and MXD, scatt e r p l o t of HT, MXD; N=38, to i d e n t i f y 3 size classes; Kruskal-Wallis nonparametric s i g n i f i c a n c e test on the classes 1 pot moved from the small guan to the pen class subtypes (xing) A,B, of large (deep) guan jar lumped, then "unknowns"- pots not i d e n t i f i e d as to subtype- added to sample for l a t e r analyses differences i n appearance and range of values f or MXD/HT and OD/MXD; N=5 f o r the new small guan class and N=3 for pen with shoulder class Mann-Whitney nonparametric t e s t using pots i d e n t i f i e d as A or B; for OD/MXD, OD/HT, OD/ODHT, RD/HT, RD/MXD, MXD/ODHT, MXD/HT for guan A: N=5, guan B: N=6; no s i g n i f i c a n t difference between subtypes A and B subtypes A,B,C f o r medium- sized guan jar lumped; "unknowns" added to sample for l a t e r analyses Kruskal-Wallis nonparametric s i g n i f i c a n c e t e s t using pots i d e n t i f i e d as A, B, or C; f o r OD/MXD, OD/HT, OD/ODHT, RD/HT, RD/MXD, MXD/ODHT, MXD/HT; f o r guan A: N=5, guan B: N=5, guan C: N=3; no s i g n i f i c a n t difference between subtypes subtypes A and B for small guan jar lumped 2 classes formed f or pen container c l a s s , o r i g i n a l l y no subtypes range i n values f or OD/BD, MXD/HT and s i m i l a r i t y of pots i n appearance; f o r guan A: N=3, B: N=2 q u a l i t a t i v e differences i n appearance: one class of pots has a d i s t i n c t shoulder, one does not; these differences supported by values of OD/MXD, 266 o r i g i n a l class N=5, one N=2, two N=3 1 pot from bo_ bowl class moved to pen class subtypes A,B,C f o r gang jars considered homo- geneous i n terms of report c r i t e r i a but two other classes formed enti r e c l a s s , N=2, of small weng jars moved to gang with neck class subtypes A and B of pingdipen basins lumped as no d i f f e r - ence found between vessels; "unknowns" added to sample and no separate classes i d e n t i f i e d 4 o r i g i n a l subtypes of wan bowls i n report: A,B,C,D; only subtype D c l e a r l y d i s t i n c t (N=2); 4 other pots d i s t i n c t l y d i f f e r e n t and put in 3 separate classes (three, four, f i v e ) ; c r i t e r i a f o r subtypes A,B,C eval- uated; "unknowns" added to sample and 2 classes i d e n t i f i e d q u a l i t a t i v e differences i n appearance; supported by values f o r MXD/HT, OD/MXD; pen class two with shoulder N=4 q u a l i t a t i v e features of pots, and range of values for OD/BD; gang A: N=3, B: N=l, C: N=2; class one= with neck; class two= no neck range of ND/MXD values; o r i g i n a l gang with neck class N=4 Mann-Whitney nonparametric test for RD/HT, OD/HT, RD/BD, OD/BD; A: N=4, B: N=6; sca t t e r p l o t with OD/HT and OD/BD, N=24; scatt e r p l o t of RD and HT, large size, N=l, medium-sized, N=23 on the basis of q u a l i t a t i v e differences i n appearance; supported by values for RD/HT; Kruskal-Wallis nonparametric te s t f o r RD/BD, RD/HT; class A: N=ll, B: N=8, C: N=4; no difference f or RD/HT; but s i g n i f i c a n t difference f o r RD/BD; no subclasses i d e n t i f i e d by scatt e r p l o t of RD/BD and RD/HT; range of sizes apparent from s c a t t e r p l o t of HT, RD 267 subtypes A and B of xian (yan) steamer lumped as well as "other" class can only assess report c r i t e r i a by appearance of pots; no apparent difference; A: N=4, B: N=4; "other" N=4 subtypes A and B of quanzupan pedestalled dish rejected and two new classes formed Mann-Whitney nonparametric test f o r RD/HT indicates no s i g n i f i - cant difference between A and B, A: N=4, B: N=3; two new classes formed using range of values f o r RD/HT and appearance of pots; class one N=2, two N=5 o r i g i n a l l y no subtypes for bi_ grate; 2 classes formed q u a l i t a t i v e differences i n shape; class one N=l, two N=l o r i g i n a l l y no subtypes for bei cup c l a s s ; 3 formed q u a l i t a t i v e differences i n shape; class one N=2, two N=l, three N=l 268 Table 37. O r i g i n a l Shape Classes Accepted from the Hougang Report. shenfupen j a r , N=4, no subtypes, on basis of appearance weng necked j a r , N=5, s u f f i c i e n t information only f or subtype A, N=5, on basis of appearance ping small necked j a r , no subtypes, N=3, from appearance hu necked j a r , no subtypes, N=2, from appearance bu_container, N=l, from appearance sizumin 4 footed basin, N=3, from appearance wan bowl subtype D, N=2, from appearance - bulge at lower body l i t r i p o d , no subtypes, N=5, from appearance j i a t r i p o d , two subtypes accepted, 1) globular body and round base, N=5, 2) more shallow body and wider rim with f l a t base, N=3; from appearance ding t r i p o d , N=l, from appearance gui t r i p o d , N=4 large sherds; although none pictured in report and none personally examined, class accepted because d i s t i n c t form in most s i t e s dou stemmed dish, two subtypes A and B accepted on basis of appearance and range of values for RD/DHT (rim diameter/dish height); subtype A N=3, subtype B N=3 zhefupen carinated basin, no subtypes, N=3, from appearance zeng perforated j a r , no subtypes, N=2, from appearance 269 zuo stand, no subtypes, N=l, from appearance gai l i d , subtypes A,B,C,D d i s t i n c t l y d i f f e r e n t appearance; A: N=6, B: N=7, C: N=l, D: N=l 270 T a b l e 38. Summary of R e s u l t s from N o n p a r a m e t r i c S i g n i f i c a n c e T e s t s f o r Hougang V e s s e l s . c l a s s e s v a r i a b l e t e s t p r o b a b i l - i t y v a l u e s m a l l , medium, l a r g e guan j a r s i z e c l a s s e s , s m a l l : N=5, medium: N=18, l a r g e : N=15 HT K r u s k a l - W a l l i s 0.000 l a r g e guan j a r s u b t y p e s A and B; A: N=5, B: N=6 OD/MXD OD/HT OD/ODHT RD/HT RD/MXD MXD/ODHT MXD/HT Mann-Whitney 0.018 0.584 0.715 0.855 0 .018 0.045 0.273 medium-sized guan j a r s u b t y p e s A,B,C; A: N= 5, B: N=6, C: N=3 OD/MXD OD/HT OD/ODHT RD/HT RD/MXD MXD/ODHT MXD/HT K r u s k a l - W a l l i s 0.262 0.186 0.106 .084 .327 .028 .025 p i n g d i p e n b a s i n s u b t y p e s A and B, A: N=4, B: N=6 RD/HT RD/BD OD/HT OD/BD Mann-Whitney 0.033 0.136 0.033 0.201 wan bowl su b t y p e s A,B,C; A: N = l l , B: N=8, C: N=4 RD/BD RD/HT K r u s k a l - W a l l i s 0.003 0.142 quanzupan p e d e s t a l l e d d i s h , A: N=4, B: N=3 RD/HT Mann-Whitney ( d i s h h e i g h t ) 0.077 271 Table 39. Changes Made in Shape Classes of Vessels I d e n t i f i e d i n the Baiying Report. change in c l a s s i f i c a t i o n method of evaluation dai l i u guan, 1 j a r with spout moved to pitche r class 2 guan j a r s , styles I,II combined as guan class two q u a l i t a t i v e features: spout, neck. q u a l i t a t i v e features: o r i f i c e i s square i n shape, bulge at lower body 1 guan jar grouped with other guan separated as guan class three q u a l i t a t i v e features: jar with long neck 2 pots o r i g i n a l l y i n separate classes: " t a l l neck small guan" and "guan" lumped into a single c l a s s , guan four 1 guan j a r grouped with other guan separated as guan class f i v e q u a l i t a t i v e features: necked jar on a pedestal q u a l i t a t i v e features, no neck 2 guan jars from d i f f e r e n t classes and styles lumped as guan class s i x 1 guan j a r grouped with other guan separated as guan class seven q u a l i t a t i v e features, no neck q u a l i t a t i v e features, no neck 10 guan jars from a number of d i f f e r e n t styles and paste classes lumped into guan class eight q u a l i t a t i v e features, guan as at Hougang scat t e r p l o t of HT, MXD indicates 2 size classes; medium (N=9), large (N=l) 1 guan jar grouped with other guan separated as guan class nine q u a l i t a t i v e features, no neck 272 1 pot i d e n t i f i e d as weng necked ja r moved to guan class eight 1 pot i d e n t i f i e d as guan moved to weng necked j a r class 9 pingdipen basins grouped into d i f f e r e n t s t y l e s and paste classes; 2 pots separated into basin classes two and three 4 wan bowls separated into classes one, two, three, four 8 j i a tripods from d i f f e r e n t styles and paste classes regrouped i n t o classes one, N=3, two, N=3, three, N=2 10 ding tripods of d i f f e r e n t styles and paste classes combined into one class 1 pot c a l l e d aizugui, short legged gui, moved to other pitche r class 1 pot c a l l e d qufupen, curved b e l l y pen, moved to zun class 3 pots c a l l e d bei moved to pitcher c l a s s , 6 pots from 4 classes i n report (bei cup, danerbei single handle cup, xiaobei small cup, zhitongbei straight bodied cup) regrouped into two d i f f e r e n t c l a s s e s , one, N=l, and two, N=5; q u a l i t a t i v e features, s i m i l a r i t y i n shape, no neck q u a l i t a t i v e features, s i m i l a r i t y to weng at Hougang, neck present s i m i l a r i t y to basins at Hougang, and scatter- plot with OD/HT, OD/BD, Mann-Whitney nonparametric te s t used to test groups; no subclasses i d e n t i f i e d ; q u a l i t a t i v e features q u a l i t a t i v e features; class one l i k e wan at Hougang q u a l i t a t i v e features q u a l i t a t i v e features, s i m i l a r i t y i n terms of body shape; s t y l e of s o l i d legs varies q u a l i t a t i v e differences q u a l i t a t i v e features q u a l i t a t i v e features, spout present, sc a t t e r p l o t with RD/HT and RD/BD 273 1 pot from danerbei separated to class three, 1 pot from bei separated to class four; the class gaozubei kept d i s t i n c t but changed name to class f i v e q u a l i t a t i v e differences 18 gai l i d s from various styles and paste classes, regrouped into ten classes, one to ten qu a l i t a t i v e differences; scatterplot of RD and HT indicates 2 sizes, large N=l; medium N=16 (1 no data) 274 Table 40. Report. O r i g i n a l Shape Classes Accepted from the Baiying daierguan, guan j a r with one handle, N=l, q u a l i t a t i v e l y d i s t i n c t bo bowl, N=l, from appearance, as a l l other classes below xian (yan) t r i p o d , N=5 l i t r i p o d , N=2 gui pitcher, N=3 zeng perforated j a r , N=2 b i grate, N=l zuo stand, N=l dou stemmed dish, N=l quanzupan pedestalled dish, N=l panxingqi plate-shaped vessel, N=l pan plate, N=l l e i - b o engraved bowl, N=l zun container, N=l shuangfupen double b e l l y pen, N=2; same as zhefupen, carinated basin at Hougang 275 Table 41. Changes Made i n Shape Classes of Vessels I d e n t i f i e d i n the Meishan Report. change i n c l a s s i f i c a t i o n method of evaluation 2 pots c a l l e d xiekouguan or slanted mouth guan changed to guan class appearance 12 pots from d i f f e r e n t classes (xiefuwan, slanted b e l l y wan N=6; bo bowl N=4; pen container N=2) changed to wan class then evaluated f or homogeneity, r e s u l t i s clas s one, N=12 appearance indicates wan cl a s s ; s c a t t e r p l o t f o r 11 bowls RD/BD and RD/HT; 2 groups tested by Mann-Whitney s t a t i s t i c (N=5, N=6), no s i g n i f i c a n t difference i n terms of these r a t i o s two pots from slanted b e l l y wan cla s s changed to wan class appearance two ( c u p - l i k e ) , 2 pots from qufuwan or curved b e l l y wan class changed to wan class three; a t h i r d pot from qufuwan class changed to wan class four (bulge i n p r o f i l e ) ; a fourth pot from qufuwan class changed to wan class f i v e (angular) 5 styles of kecaopen, engraved appearance pen container, changed to two classes, one (bowl-like) N=5, and two (cup-like) N=l; change class name to l e i - b o danerbei cup with handle class appearance s p l i t into two classes, one N=l, and two N=2 5 styles of gai l i d s changed to appearance classes one (curved wall) N=3, and two (square i n p r o f i l e ) N=3 276 Table 42. O r i g i n a l Shape Classes Accepted from the Meishan Report. guan j a r s , N=6, t h i s class and a l l others evaluated on basis of appearance, 2 st y l e s lumped yuan fu guan, round b e l l y guan, N=2, accepted but c a l l e d round necked j a r , 2 styles lumped da kou guan, large mouthed j a r , N=2 xiao guan, small j a r , N=l ( d i f f e r e n t i n shape than other guan) gao j i n g xiao guan, t a l l neck small j a r , N=2 small pen, N=2; here, a wide rimmed bowl, unlike pen at Hougang weng, necked j a r , N=5, 5 styles lumped shen fu weng, deep b e l l y necked j a r , N=l; d i f f e r e n t i n shape than weng class above xian (yan) t r i p o d , N=l j i a t r i p o d , N=l ding tr i p o d , N=12, 4 st y l e s lumped gui pitcher, N=3 he container, N=l zeng perforated j a r , N=3 bi grate, N=2 guan zu pan, pedestalled dish; two d i f f e r e n t styles from one period d i s t i n g u i s h two d i f f e r e n t shapes, these classes c a l l e d one and two, one vessel from another period with no st y l e designation added to class two; clas s one N=2, class two N=3 dou stemmed dish, N=7, 2 styles lumped; a l l have narrow stem 277 qu fu pen, curved b e l l y container, N=l, same as zhe fu pen, carinated basin at Hougang gu beaker, N=4, 4 styles lumped zhe fu b e i , bent b e l l y cup, N=l chu xing b e i , pestle-shaped cup, N=3, 2 styles lumped yuan fu b e i , round b e l l y cup, N=2 zun xing b e i , zun shaped cup, N=l 278 Table 43. Changes Made in Shape Classes of Vessels I d e n t i f i e d i n the Lujiakou Report. change i n c l a s s i f i c a t i o n method of evaluation under one paste type report l i s t s 3 styles of guan j a r s ; 7 classes formed: one (N=3), two (N=l), three (N=2), four (N=l), f i v e (N=2), six (N=l), seven (N=l) appearance report defines subtypes (xing A,B,C) of pen; A i s s i m i l a r to pingdipen basin at Hougang, 3 classes formed one (shallow, N=l), two ( f l a r i n g walls, N=4), three (deep, N=l); from xing B, 2 classes formed, class four (N=l), f i v e (N=3); from xing C, 2 classes formed, six (N=l), seven (N=2) appearance (2 "unknown" pots added to class one, 1 to class three); appearance; (1 "unknown" pot added to class four); appearance report l i s t s 5 wan bowls i n 2 s t y l e s ; 13 "unknowns" added to sample; 18 pots evaluated 2 classes formed; one (N=17), two (N=l) differences among vessels noticable from examination sca t t e r p l o t of RD/BD, RD/HT shows 2 groups, s c a t t e r p l o t of HT,RD shows a wide range of sizes report l i s t s 5 subtypes for appearance ding t r i p o d of one paste type (A,B,C,D,E); pots regrouped into class one (N=3), two (N=l), three (N=l), four (N=l), f i v e (N=l), six (N=l, an "unknown" pot) 279 report l i s t s 3 subtypes f or ding t r i p o d of another paste type (A,B,C; N=3,3,l) yu container 2 s t y l e s , changed to classes one (N=l), two (N=l) bei cup class s p l i t into 4 classes, one (N=2), two (N=2), three (N=l), four (N=l) sample size permits evaluation by a scatterplot of OD/MXD and MXD/ODHT; groups not d i s t i n c t appearance appearance 280 Table 44. O r i g i n a l Shape Classes Accepted from the Lujiakou Report. xing (subtype) A,B,C guan j a r f o r one type of paste accepted but c a l l e d guan jar classes eight (N=2), nine (N=2) , ten (N=2); these classes and a l l others evaluated on the basis of vessel appearance pen container xing (subtype) D, but c a l l e d pen class eight, N=l; same as zhe fu pen, carinated basin at Hougang pen container xing E, but c a l l e d pen class nine, N=l hu necked jar,' N=3 xian (yan) t r i p o d , N=l gui pitcher, N=3; 1 "unknown" pot added to c l a s s , too dai l i u bo, bo bowl with spout, N=2 b i grate, N=l san zu pan, three footed pan dish, 4 subtypes or xing accepted, c a l l e d class one (N=2) , two (N=2) , three (N=2), four (N=l) dou stemmed dish, N=4 l e i pedestalled j a r , N=l gai l i d , 4 s t y l e s accepted as d i s t i n c t classes, re-named as class one (N=l), two (N=2), three (N=l), four (N=l) 281 Table 45. Lids at Hougang f o r Covering Other Vessels or f o r Serving Food. shape class rim diameter height hypothesized function: covering other pots gai four, N=l 12.3 cm no data but f l a t hypothesized function: serving food; too big in rim diameter and height to cover other pots at o r i f i c e gai one, N=5 range= 25.0-37.0 range= 8.9-1 (one case no data) median= 28.8 median= 10.0 gai two, N=7 range= 10.5-31.4 range= 2.4-9 median= 16.0 median= 6.5 gai three, N=l 16.7 broken gai f i v e , N=l broken, g r e a t l y curved i n p r o f i l e hypothesized cooking vessels guan j a r s , large, N=15 range= 18.8-25.3 median= 20.0 guan jars, medium, N=18 range= 10.0-17.6 median= 14.8 no data 13.0, 13.0, 19.0 282 I i t r i p o d , N = l (4 cases no data) 16.0 xian/yan t r i p o d , N=7 (5 cases no data) ding t r i p o d , N=l (7 cases no data) range= 23.6-29.0 median= 25.4 11.5 283 Table 46. Lids at Baiying for Covering Other Vessels or for Serving Food. shape class rim diameter height hypothesized function: covering other pots gai one, N=5 (bowl-like) range= 12.0-15.5 cm median= 12.8 range= 4.0-7.5 median= 5.7 hypothesized function: serving food; too big in rim diameter and height to cover other pots at o r i f i c e gai two, N=3 gai three, N=l gai four, N=l gai f i v e , N=l gai s i x , N=2 gai seven, N=l gai eight, N=l gai nine, N=l gai ten, N=l gai eleven, N=l range= 20.9-31.5 median= 23.0 56.0 10.4 21.4 13.4, 14.4 14.0 29.1 14.0 18.6 9.3 9.0-12.9 median= 9.4 11.5 broken (5.0) broken (8.0) 6.0, 5.8 4.6 broken (11.0) 6.0 7.1 1.4 284 hypothesized cooking vessels guan jars, eight, medium, N=5 xian/yan tripod, N=4 (1 case no data) l i tripod, N=2 j i a tripod, one, N=3 two, N=3 three, N=2 ding tripod, one, N=9 two, N=l gui tripod, N=l (3 cases no data) zeng j a r , N=2 range= 16.1-23.0 median= 17.9 range= 24.5-33.0 median= 26.6 14.7, 23.0 range= 13.7-18.3 median= 16.3 range= 19.5-25.3 median= 20.0 15.5, 15.3 range= 11.8-19.3 median= 13.6 21.0 8.3 20.0, 15.1 285 Table 47. Lids at Meishan f o r Serving Food. shape class rim diameter height hypothesized function: serving food l i d s too big i n rim diameter and height to cover other pots at o r i f i c e gai one, N=3 range= 26.0-32.0 no data median= 27.0 cm gai two, N=3 range= 29.5-37.0 median= 30.0 no data hypothesized cooking vessels guan j a r , N=8 range= 17.4-27.0 median= 22.0 ding t r i p o d , N=ll (1 pot no data) range= 10.10-23.0 median= 18.5 20.0 286 Table 48. Lids at Lujiakou for Covering other Vessels or fo r Serving Food. shape class rim diameter height hypothesized function: l i d s f o r covering other pots gai one, N=l gai two, N=l gai three, N=2 9.8 12.8 8.4, 12.0 3.4 3.0 2.7, 3.5 hypothesized function: l i d s for serving food: too big in rim diameter and height to cover other pots at o r i f i c e gai four, N=l 23.5 12.0 hypothesized cooking vessels xian/yan t r i p o d , N=l 18.5 ding t r i p o d , seven, N=7 range= 13.0-21.5 median= 18.0 guan j a r , seven, N=2 15.7, 14.0 287 Figure 5. Guan Jar Size Classes, Hougang (n=38) 35 Height (cm) •= Large (n=l5) •= Medium (n=18) •= Small (n=5) 288 Figure 6. Size of Guan Jars, Class Eight, Baiying (n=ll) 20 30 40 50 60 Height (cm) 289 Figure 7 Pingdipen Basin Class, Hougang (n=24) 0.751 1.50 2.50 3.50 4.50 Orifice Diameter/Height (cm) 290 Figure 8. Pingdipen Basin Size Classes, Hougang (n=24) 30 40 50 60 Rim Diameter (cm) ®Two Basins with the Same Value 291 Figure 9. Pingdipen Basin Class One, Baiying (n = 7) 1.50 E u * v. OJ Q3 iam  1.25 o a> tn o m to E 1.00 o Q a> o o 0.75 2.00 3.00 4.00 5.00 Orifice Diameter/Height (cm) 6.00 2 9 2 Figure 10. S i z e s of Pingdipen Basins, Glass One, Baiying (n=7) 20 30 40 50 60 70 80 Rim Diameter (cm) 293 Figure II. Wan Bowl C l a s s One, Hougang (n=44) i.ooi i i I 1.00 2 . 0 0 3 0 0 4 . 0 0 Rim Diameter/Base Diameter (cm) 294 Figure 12. S i z e s of Wan Bowls Class One, Hougang (n=44) 8 7 ® — 6|- • • • o £ 51 QJ X 4 | " • • • ( § ) • • 1 0 15 2 0 Rim Diameter (cm) ® Two Bowls with the Same Value 295 Figure 13. Wan Bowl Class One, Meishan (n = ll) _ j i i _ 2 3 4 Rim Diameter/Base Diameter (cm) Figure 14. Sizes of Wan Bowls, Class One, Meishan (n=ll) Rim Diameter (cm) 297 Figure 15. Wan Bowl Classes One and Two, Lujiakou (n=!8) 4.00 0.00 1,00 2.00 3.00 4.00 Rim Diameter/Base Diameter (cm) 298 Figure 16. Sizes of Wan Bowls Class One, Lujiakou (n=!8) Rim Diameter (cm) 299 Figure 17 Ding Tripod Class Seven, Lujiakou (n=7) 1.20 1.10 0.90 c I i - B B A B A - - i A 1 — • 0.70 0.80 0.90 I.C Orfice Diameter/Maximum Diameter (cm) 300 Figure 18. Bei Cups Classes One and Two, Baiying ~ ~ (n=6) 0.50 Rim Diameter/Base Diameter (cm) 301 Figure 19. Sizes of Gai Lids, Classes One to Eleven, Baiying (n=l6) E u a> X 10 20 30 40 50 60 Rim Diameter (cm) 302 APPENDIX B. DETAILS ON ANALYSES FOR TESTING THE MODEL OF CHANGE IN SYSTEMS OF CERAMIC PRODUCTION IN RELATION TO INCREASING CULTURAL COMPLEXITY. s i t e reports: Hougang (Anyang Archaeological Team, IA, CASS 1985) Baiying (CPAM of Anyang D i s t r i c t , Henan Province 1983) Meishan (Second Henan Archaeological Team, IA, CASS 1982) Lujiakou (Shandong Archaeological Team, IA, CASS and the Art Museum of Weifang County, Shandong Province 1985) 303 Table 49. D i s t r i b u t i o n of Vessels i n Shape Classes Per Phase at Hougang. *** note: For each phase, the f i r s t figures r e f e r to quantities of vessels i n each shape class a v a i l a b l e f o r ana l y s i s ; the figures i n parentheses r e f e r to my estimates of the t o t a l number of vessels excavated. The t o t a l number of vessels i n each o r i g i n a l shape class, given i n the l a s t column, i s c l e a r l y stated i n the report. In some cases when new classes have been established, the t o t a l number of vessels excavated per class i s not cl e a r . ("sh"= sherds) shape class Early Middle Late t o t a l guan j a r , large 2 (8) 8 (26) 5 (19) (53) guan j a r , medium 3 (17) 7 (22) 8 (14) (53) guan j a r , small 2 (5) 1 (8) 2 (3) (16) shenfupen jar 0 (1) 1 (1 ) 3 (4) (6) pen j a r , one 2 (2) 0 (1? ) 0 (0) (12 to- two 1 (1) 1 (6) 2 (2) t a l ) gang j a r , one 0 (?) 2 (5? ) 0 (0? ) (11 gang j a r , two 0 (?) 1 (2? ) 3 (4?) to- t a l ) weng necked jar 1 (5) 2 (12) 2 (7) (24) ping necked jar 1 (4) 1 (1) 1 (1) (6) hu necked j a r 0 (0) 1 (1) 1 (4) (5) pingdipen basin, (25 medium-sized 5 (6?) 5 (5? ) 13 (13? ) to- large size 0 (0) 1 (1) 0 (0) t a l ) sizumin basin 1 (2) 2 (2) 0 (1) (5) 304 wan bowl, one 10 (?) wan bowl, two 0 (?) wan bowl, three 1 (?) wan bowl, four 0 (?) wan bowl, f i v e 1 (?) bo bowl 1 (1) xian/yan t r i p o d 4 (5) l i t r i p o d 1 ( l ) j i a t r i p o d , one 1 (1) j i a t r i p o d , two 1 (2) ding t r i p o d 1 (?) gui t r i p o d 1 (?) zeng perforated j a r 1 (?) bi_ grate, one 0 (0) bi_ grate, two 1 ( l ) quanzupan, one 2 (4) quanzupan, two 1 (2) dou dish, one 1(2) dou dish, two 1 (1) zhefupen basin 1 (2) b_u_ container 0 (0) zuo stand 1 (2) bei cup, one 0 (0) bei cup, two 0 (0 ) bei cup, three 0 (0) 15 (?) 19 (?) (82 0 (?) 2 (?) to- 0 (?) 1 (?) t a l ) 0 (?) 1 (?) 0 (?) 0 (?) 0 (0) 1 (1) (2) 5 (11) 3 (5) (21) 2 (4) 2 (3) (8) 2 (5) 2 (3) (9) 1 ( 3 ) 1 ( 1 ) (6) 2 (?) 5 (?) (8? 14 legs) 0 (?) 3 (?) (7 sh) 1 ( ?) 0 (?) (7) 1 ( 1 ) 0 ( 0 ) (2 0 (0) 0 (0) to-) t a l ) 0 (0) 0 (0) (16 2 (4) 2 (6) to- t a l ) 1 (3) 1 (5) (10) 1 (3) 1 (2) (6) 1 (5) 1 (2) (9) 1 (1) 0 (0) (1) 0 (0) 0 (0) (1) 0 (0) 2 (5?) (8 0 (0) 1 (2?) to- 1 (1) 0 (0) t a l ) 305 g a i l i d , one 1 (1) g a i l i d , two 2 (2) g a i l i d , t h r e e 0 (? ) g a i l i d , f o u r 0 (0 ) g a i l i d , f i v e 1 (? ) 3 (7) 2 (3) (11) 1 (2) 4 (14) (18) 1 (?) 0 (?) (2) 1 (1) 0 (0) (1) 0 ( ? ) 0 (?) (4) 306 Table 50. D i s t r i b u t i o n of Vessels i n Shape Classes Per Phase at Baiying. *** note: For each phase, the f i r s t figures r e f e r to quantities of vessels i n each shape class a v a i l a b l e f o r ana l y s i s . The figures i n parentheses are the t o t a l number of vessels excavated as stated i n the report. In some cases f o r new classes defined i n Chapter 4, the t o t a l number of vessels excavated per class i s not c l e a r . shape class Early Middle Late Total guan j a r , one 1 (1) 0 (0) 0 (0) (1 ) guan j a r , two 0 (0 ) 0 (0) 2 (2) (2) guan j a r , three 0 (0) 1 (2) 0 (0) (2) guan j a r , four 0 (0) 0 (0) 2 (3) (3) guan j a r , f i v e 0 (0 ) 0 (0) 1 (1) (1 ) guan j a r , six 0 (0) 1 (2) 1 (1) (3) guan j a r , seven 0 (0) 1 (1) 0 (0) (1) guan j a r , eight, medium 0 (0) 3 (4) 7 (15) (19) guan j a r , eight, large 0 (0) 0 (0) 1 (1) (1) guan j a r , nine 0 (0) 1 (1 ) 0 (0) (1) weng necked jar 0 (0) 0 (0) 2 (2) (2) pingdipen basin, one medium-sized 1 (2) 4 (4) 1 (16) (22) one, large 1 (1) 0 (0) 0 (0). (1) pingdipen basin, two 0 (0) 0 (0) 1 (2) (2) pingdipen basin, three 0 (0) 0 (0) 1 (1) (1 ) wan bowl, one 0 (0) 1 (1) 0 (0) (1) wan bowl, two 0 (0) 1 (1 ) 0 (0) (1) wan bowl, three 0 (0) 0 (0) 1 (2) (2) wan bowl, four 0 (0) 0 (0) 1 (3) (3) bo bowl 1 (1) 0 (0) 0 (0) (1) xian (yan) t r i p o d 0 (0) 0 (0) 5 (5) (5) l i t r i p o d 1 (1) 0 (0) 1 (1) (2) 307 j i a t r i p o d , one 0 (0) 0 (0) 3 (7? ) (7? j i a t r i p o d , two 1 (1) 2 (2) 0 (0) (3) j i a t r i p o d , three 2 (2) 0 (0) 0 (0) (2) ding t r i p o d , one 1 (1) 2 (2) 6 (7) (10 ding t r i p o d , two 0 (0) 0 (0) 1 (1) (1) gui t r i p o d 0 (0) 1 (1) 3 (7) (8) zeng perforated j a r 0 (0) 0 (0) 2 (7) (7) b i grate 0 (0) 0 (0) 1 (1 ) (1) quanzupan dish 0 (0) 0 (0) 1 (6) (6) dou stemmed dish 0 (0) 0 (0) 3 (3) (3) shuangfupen basin 1 (1) 1 (1) 0 (0) (2) qufupen basin, one 1 (1) 0 (0) 0 (0) (1) qufupen basin, two 0 (0) 0 (0) 1 (1) (1) qufupen basin, three 0 (0) 0 (0) 1 (1) (1) panxingqi dish 0 (0) 0 (0) 1 (1) (1) pan plate 0 (0) 1 (1) 0 (0) (1) zuo stand 1 (1) 0 (0) 0 (0) (1) l e i - b o engraved bowl 0 (0) 0 (0) 1 (1) (1) zun container 0 (0) 0 (0) 1 (1) (1) bei cup, one 0 (0) 1 (1) 0 (0) (1) bei cup, two 1 (1) 1 (1) 5 (5) (7) bei cup, three 0 (0) 0 (0) 1 (1 ) (1) bei cup, four 0 (0) 0 (0) 4 (4) (4) bei cup, f i v e 0 (0) 0 (0) 1 (1) (1) pitcher, one 1 (1) 0 (0) 0 (0) (1) pitcher, two 0 (0) 0 (0) 1 (1) (1) pitcher, three 0 (0) 2 (2) 0 (0) (2) pitcher, four 0 (0) 1 (1) 0 (0) (1) 308 gai l i d , one 1 (1) 3 (7? ) 1 (7) (17?) gai l i d , two 1 (1) 2 (2) 0 (0) (3) gai l i d , three 1 (1) 0 (0) 0 (0) (1) gai l i d , four 0 (0) 1 (3?) 0 (0) (4?) gai l i d , f i v e 0 (0) 1 (2) 0 (0) (2) gai l i d , six 0 (0) 0 (0) 2 (2) (2) gai l i d , seven 0 (0) 0 (0) 1 (1 ) (1) gai l i d , eight 0 (0) 0 (0) 1 (1) (1) gai l i d , nine 0 (0) 0 (0) 1 (1) (1) gai l i d , ten 0 (0) 0 (0) 1 (1) (1) gai l i d , eleven 0 (0) 0 (0) 1 (1) (1) 309 Table 51. D i s t r i b u t i o n of Vessels i n Shape Classes Per Phase at Meishan. *** note: Only quantities of vessels a v a i l a b l e f o r analysis are given here; the report does not provide figures for the t o t a l number of excavated vessels from each shape c l a s s . shape class Early Late guan jar 3 5 dakouguan, large- 1 1 mouthed j a r round necked j a r 2 0 xiaoguan, small guan 0 1 necked small j a r 0 2 weng necked j a r 2 3 large necked j a r 0 1 wide rimmed bowl 1 1 wan bowl, one 8 4 wan bowl, two 1 1 wan bowl, three 1 1 wan bowl, four 1 0 wan bowl, f i v e 0 1 l e i - b o engraved bowl, one 3 2 l e i - b o , two 1 0 xian/yan t r i p o d 0 1 j i a t r i p o d 1 0 ding t r i p o d 9 7 310 gui t r i p o d 1 2 zeng perforated j a r 2 1 b i grate 1 ! quanzupan pedestalled dish, one 2 0 quanzupan, two 2 1 dou stemmed dish 4 3 qufupen carinated basin 1 0 he container 1 0 danerbei cup with handle, one 1 0 danerbei, two 0 2 gu beaker 2 2 zhefubei, carinated cup 1 0 chuxingbei, p e s t l e - shaped cup 1 2 yuanfubei, round cup 0 2 zunxingbei, zun-shaped cup 0 1 gai l i d , one 1 2 gai l i d , two 2 1 311 Table 52. D i s t r i b u t i o n of Vessels i n Shape Classes Per Phase at Lujiakou. *** note: The f i r s t figures r e f e r to quantities of vessels i n each shape class a v a i l a b l e f o r analysis. The figures i n parentheses r e f e r to the number of vessels excavated. In some cases, p a r t i c u l a r l y when new shape classes have been established, there i s only information on the t o t a l number of vessels excavated. There i s no information f o r each phase. shape class Early Late Total guan j a r , one 2 (2? ) 1 (1? ) guan j a r , two 1 (1) 0 (0) guan j a r , three 0 (0? ) 2 (2? ) guan j a r , four 0 (0) 1 (1) guan j a r , f i v e 1 (2? ) 0 (0? ) guan j a r , six 0 (0) 1 (1) guan j a r , seven 1 (?) 1 (?) guan j a r , eight 1 (?) 1 (?) guan j a r , nine 0 (0) 2 (0) hu necked ja r 2 (2) 1 (1) l e i necked [ j a r 0 (0) 1 (1) pen basin, one 0 (?) 3 (?) pen basin, two 2 (?) 1 (?) pen basin, three 1 (?) 1 (?) pen basin, four 0 (0) (2) pen basin, f i v e 0 (0) 3 (3) pen basin, six 0 (?) 1 (?) pen basin, seven 0 (?) 2 (?) pen basin, eight 0 (0) 1 (1) pen basin, nine 0 (0) 1 (1) pen basin, ten 0 (0) 1 (1) wan bowl, one 3 (?) 14 (?) wan bowl, two 0 (?) 1 (?) (6 f o r classes one-three) (1) (2) (1) (10) (3) (2) (3) (1) (19 f o r classes one-three) (2) (3) (8 f o r classes six, seven) (1) (1) (1) (26 f o r classes one-two) 312 d a i l i u b o , bo bowl 0 with spout xian/yan t r i p o d 0 gui t r i p o d 3 ding t r i p o d , one 1 ding t r i p o d , two 0 ding t r i p o d , three 0 ding tr i p o d , four 0 ding t r i p o d , f i v e 0 ding t r i p o d , six 0 ding t r i p o d , seven 3 b i grate 0 sanzupan, 3 footed pan, one 1 sanzupan, two 1 sanzupan, three 1 sanzupan, four 0 dou stemmed dish 1 yu container, one 1 yu container, two 1 bei cup, one 0 bei cup, two 0 bei cup, three 0 bei cup, four 1 gai l i d , one 0 gai l i d , two 0 gai l i d , three 1 gai l i d , four 0 (0) 2 (2) (2) (0) 1 (1) (1) (3) 1 (1) (4) (?) 2 (?) (12 f o r (?) 1 (?) classes (?) 1 (?) one-six) (?) 1 (?) (?) 1 (?) (?) 1 (?) (?) 4 (?) (12) (0) 1 (1) ( l ) (1) 1 (1) (2) (1) 1 (1) (2) (1) 1 (1) (2) (0) 1 (1) (1) (1) 3 (3) (4) (0) 0 (0) (1) (0) 0 (0) (1) (0) 2 (2) (2) (0) 2 (2) (2) (0) 1 (1) (1) (1) 0 (0) (1) (0) 1 (1) (1) (0) 1 (1) (1) (1) 1 (1) (2) (0) 1 (1) (1) 313 Table 53. D i v e r s i t y Functional Types Per of Shape Classes and Hypothesized Period at Hougang. functional type Early Middle Late cooking tripods 6 forms (xian/yan, l i , j i a one and two, ding, gui) other cooking pots 4 (medium and large guan j a r , zeng j a r , b i grate one and two) other jars with wide 4 o r i f i c e (small guan, shenfupen, pen one and two, gang one, two) storage function j a r s , narrow o r i f i c e 2 (weng, ping, hu) f o r l i q u i d storage bowls (wan one, two, 4 three, four, f i v e , bo) f o r eating, serving, drinking other open forms, 2 no pedestal or stem (pingdipen basin, medium and large s i z e ; zhefupen basin) preparing, serving food 314 open forms with pedestal or stem (dou dish one, two; quanzupan dish one, two; b_u container sizumin footed pot) serving, eating food) cups (bei one-three) f o r drinking l i d s f o r covering other vessels (gai class four) l i d s f o r serving food (gai one, two, three, f i v e ) other (zuo stand) TOTAL NUMBER FORMS (46 at whole s i t e ) 5 5 4 0 1 2 0 1 0 3 3 . 2 1 0 0 31 32 30 315 Table 54. D i v e r s i t y of Shape Classes and Hypothesized Functional Types Per Period at Baiying. functional type Early Middle Late cooking tripods (xian/yan, l i , j i a one, two, three; ding one, two; gui) other cooking pots (guan j a r eight, medium; zeng j a r , bi_ grate) other jars with wide o r i f i c e (guan j a r f i v e , s i x , seven, eight medium (fine paste), large; nine f o r storage jars with narrow o r i f i c e (guan jars with necks - one, two, three, four; weng j a r ) ; l i q u i d storage bowls (wan one, two, three, four; bo) for eating, drinking, serving other open forms with no pedestal or stem (pingdipen basin one, medium and large s i z e ; two, three; shuangfupen basin, qufupen basin one, two, three; pan plate, panxingqi dish, l e i - b o bowl) f o r preparing, serving food open forms with pedestal or stem (dou dish, quanzupan di s h ) ; serving, eating food 4 f orms 316 pitchers (one, two, three, four) cups (bei one, two, three, four, f i v e ; zun container) f o r drinking l i d s (gai one) f o r covering other vessels l i d s (gai classes two - eleven) for serving food other (zuo stand) TOTAL NUMBER OF FORMS (62 at whole s i t e ) 1 2 1 1 2 5 1 1 1 2 3 6 1 0 0 15 22 40 317 Table 55. D i v e r s i t y of Shape Classes and Hypothesized Functional Types Per Period at Meishan. functional type Early Late cooking tripods 3 forms (xian/yan, ding, j i a , gui) other cooking pots 3 (guan j a r , zeng j a r , b i grate) other jars with wide 2 o r i f i c e , f o r storage (dakouguan, xiaoguan) jars with narrow o r i f i c e , 2 f o r l i q u i d storage (weng, round necked jar, large necked j a r , small ja r with neck) bowls for eating, serving, 5 drinking (wan one, two, three, four, f i v e ; wide- rimmed bowl) other open forms, no 4 pedestal or stem, for preparing or serving food (qufupen, he, l e i - b o one and two) open forms with pedestal 3 or stem for serving, eating food (dou, quanzupan one, two) cups f o r drinking 4 (gu, danerbei one, two; zhefubei, chuxingbei, yuanfubei, zunxingbei) 318 l i d s f o r serving food (gai one, two) TOTAL NUMBER OF FORMS (35 at whole s i t e ) Table 56. D i v e r s i t y of Shape Classes and Hypothesized Functional Types Per Period at Lujiakou. functional type Early cooking tripods 2 forms (xian/yan, gui, ding seven) other cooking pots 1 (bi grate, guan j a r seven) other jars with wide 1 o r i f i c e , f o r storage (guan j a r three, f i v e , six) jars with narrow o r i f i c e , 4 for l i q u i d storage (guan ja r one, two, four, eight, nine; hu, l e i ) bowls for eating, serving, 1 drinking (wan one, two, three; dailiubo) other open forms, no stem 4 or pedestal, f o r preparing or serving food (pen basin one-ten; y_u_ container, one, two) open forms with pedestal, 5 stem, or legs f o r serving, eating food (dou dish, sanzupan dish one-four, ding t r i p o d one-six) cups for drinking 1 (bei one-four) 320 l i d s f o r covering other vessels (gai one - three) l i d s f o r serving food (gai two) TOTAL NUMBER OF FORMS (50 at whole s i t e ) Table 57. Variation in Shape of Leg for Xian/Yan Tripods, Hougang. shape of legs awl shaped pouch shaped no data- legs broken t o t a l number of vessels t o t a l number of shapes t o t a l number of excavated houses Early Middle Late 2 2 1 0 1 2 2 2 0 4 5 3 2 3 2 2 14 23 Table 58. Variation in Handles for Xian/Yan Tripods, Hougang. feature Early Middle Late none 4 3 3 2 small handles with loop 0 1 0 2 s o l i d lugs 0 1 0 t o t a l number of vessels 4 5 3 t o t a l number of handle types 1 3 1 t o t a l number of excavated houses 2 14 23 322 Table 59. V a r i a t i o n i n Shape of Leg for Ding Tripods, Hougang. shape of legs Early Middle Late round p i l l a r 0 0 2 long and t r i a n g u l a r 1 0 0 short and t r i a n g u l a r 0 0 1 sawtooth 0 1 1 g u i l i a n ("grimace", or 0 1 1 "funny face" ) t o t a l number of vessels 1 2 5 t o t a l number of shapes 1 2 4 t o t a l number of excavated houses 2 14 23 323 Table 60. Variation i n Decorative Techniques for Large Guan Jars, Hougang. techniques basket marks (impressions), and a few incised lines cordmarks (impressions) cordmarks, long and thin applique, incised l i n e s , and i n c i s i o n around rim basketmarks pl a i n - no decoration checkmarks (impressions) basketmarks and cordmarks cordmarks and i n c i s i o n around rim cordmarks, c i r c u l a r applique, thi n and long applique, polished rim t o t a l number of vessels t o t a l number of decorative combinations t o t a l number of excavated houses per phase Early Middle Late 0 0 1 1 1 1 O i l 0 1 0 0 1 0 0 1 1 0 1 0 1 2 0 0 0 1 2 8 5 2 7 5 2 14 23 324 Table 61. V a r i a t i o n i n Decorative Techniques f o r Medium- Sized Guan Jars, Hougang. techniques cordmarks cordmarks and i n c i s i o n around rim cordmarks, long and thi n applique, i n c i s e d l i n e s p l a i n - no decoration basketmarks checkmarks checkmarks and cordmarks checkmarks and i n c i s i o n around rim t o t a l number of vessels t o t a l number of decorative combinations t o t a l number of excavated houses Early Middle Late 1 4 3 1 0 0 1 0 0 0 2 1 0 1 0 0 . 0 2 0 0 1 0 0 1 1 7 8 3 3 5 2 14 23 325 Table 62. V a r i a t i o n i n Decorative Techniques f o r Xian/Yan Tripods, Hougang. techniques basketmarks cordmarks cordmarks and i n c i s i o n around rim cordmarks and 3 small knobs cordmarks, 3 small knobs, and i n c i s i o n around rim cordmarks, 2 i n c i s i o n s around rim basketmarks and 3 small knobs t o t a l number of vessels t o t a l number of decorative combinations t o t a l number of excavated houses Early Middle Late 1 0 0 3 2 1 0 1 0 0 0 1 0 1 0 0 1 0 0 0 1 _ _ _ 2 4 3 2 14 23 326 Table 63. Variation i n Type of Rim for Wan Bowls, Class One, Hougang. type of rim plai n one i n c i s i o n around rim two incisions around rim t o t a l number of vessels t o t a l number of rim types t o t a l number of excavated houses Early Middle Late 10 10 13 0 3 - 6 0 2 0 10 15 19 1 3 2 2 14 23 327 Table 64. V a r i a t i o n i n Handles for Ding Tripods, Class One, Baiying. feature Early Middle Late panshou loop handle 0 0 1 none 1 2 5 t o t a l number of vessels 1 2 6 t o t a l number of types 1 1 2 t o t a l number of excavated houses 9 8 46 328 Table 65. Va r i a t i o n i n Decorative Techniques f o r Ding Tripods, Class One, Baiying. techniques cordmarks checkmarks cordmarks and checkmarks t o t a l number of vessels t o t a l number of decorative combinations t o t a l number of excavated houses Early Middle Late 0 1 5 1 1 0 0 0 1 1 2 6 1 2 2 9 8 46 329 Table 66. V a r i a t i o n i n Shape of Leg f o r Ding Tripods, Meishan. shape of leg Early Late round p i l l a r 1 1 tr i a n g u l a r 3 2 nipple 2 1 tria n g u l a r , one hole 1 0 tri a n g u l a r , two holes 1 0 b i r d beak 0 1 no data- broken legs 1 2 t o t a l number of vessels 9 7 t o t a l number of shapes 6 5 t o t a l number of excavated houses 17 16 330' Table 67. V a r i a t i o n i n Decorative Techniques f o r Guan Jars, Meishan. techniques Early Late basketmarks 1 2 checkmarks 1 2 basketmarks and several 1 0 i n c i s e d l i n e s basketmarks, polished 0 1 upper shoulder t o t a l number of vessels 3 5 t o t a l number of decorative combinations 3 3 t o t a l number of excavated houses 17 16 Table 68. V a r i a t i o n i n Decorative Techniques f o r Dou Stemmed Dishes, Meishan. techniques on stem Early Late none 3 2 2 i n c i s e d l i n e s 1 0 one ridge 0 1 t o t a l number of vessels 4 3 t o t a l number of decorative combinations 2 2 t o t a l number of excavated houses 17 16 331 Table 69. V a r i a t i o n i n Ding Tripods, Meishan. Decorative Techniques f o r techniques Early Late checkmarks, and in c i s e d 1 0 li n e s on shoulder checkmarks 2 4 basketmarks 1 2 checkmarks and i n c i s i o n 1 0 around rim cordmarks 0 1 t o t a l number of vessels 5 7 t o t a l number of decorative 4 3 combinations t o t a l number of excavated houses 17 16 332 Table 70. V a r i a t i o n i n Shape of Leg for Ding Tripods, Class Seven, Lujiakou. shape of leg Early Late round p i l l a r 1 2 v e r t i c a l s t r i p s , applique 1 1 c h i s e l 1 0 no data- broken legs 0 1 t o t a l number of vessels 3 4 t o t a l number of shapes 3 3 t o t a l number of excavated houses 6 5 Table 71. V a r i a t i o n i n Handles f o r Ding Tripods, Class Seven, Lujiakou. feature Early Late 4 cockscomb handles 1 1 3 cockscomb handles 0 1 2 cockscomb handles 0 1 2 bi_ lugs 0 1 none 2 0 t o t a l number of vessels 3 4 t o t a l number of handle types 2 4 t o t a l number of excavated houses 6 5 333 Table 72. V a r i a t i o n i n Decorative Techniques f o r Ding Tripods, Class Seven, Lujiakou. techniques Early Late p o l i s h i n g 1 0 p l a i n - no decoration 0 1 in c i s e d l i n e on body and 1 0 i n c i s i o n around rim basketmarks 1 0 i n c i s i o n around rim 0 1 2 i n c i s e d l i n e s on body 0 1 3 i n c i s e d l i n e s on 0 1 shoulder, i n c i s i o n around rim t o t a l number of vessels 3 4 t o t a l number of decorative combinations 3 4 t o t a l number of excavated houses 6 5 Table 73. V a r i a t i o n i n Decorative Techniques f o r Wan Bowls, Class One, Lujiakou. techniques Early Late p l a i n - no decoration 3 4 two in c i s e d l i n e s 0 3 t o t a l number of vessels 3 7 t o t a l number of decorative combinations 1 2 t o t a l number of excavated houses 6 5 334 Figure 20. (Page I) Variation in Dimensions by Period, Large Guan Jars, Hougang PERIOD Middle (n»5) * Late (n=8) \ 1 1 I L_ . J I I I J | 14 15 16 17 18 19 2 0 21 Orifice Diameter (cm) PERIOD Middle (n=5) Late (n=8) 8 9 10 Base Diameter (cm) 335 Figure 20.(Page 2) PERIOD Variation in Dimensions by Period, Large Guan Jars, Hougang Middle (n=5) Late (n=8) 20 22 24 26 28 Maximum Diameter (cm) 30 PERIOD Middle (n=5) Late (n=8) 24 26 28 30 32 Height at Orifice Diameter (cm) 34 336 Figure 2l.(Page I) Variation in Dimensions by Period, Medium Size Guan Jars, Hougang PERIOD Early (n=3) Middle (n=8) Late (n=7) 7 8 Base Diameter (cm) 1 0 PERIOD Early (n=3) Middle (n=8) Late (n=7) 8 1 0 II 1 2 13 Orifice Diameter (cm) 14 15 337 Figure 21. (Page 2) Variation in Dimensions by Period, Medium Size Guan Jars, Hougang PERIOD Early (n-3) Middle (n=8) Late (n=7) 10 12 14 16 18 Maximum Diameter (cm) 20 PERIOD Early (n=3) Middle (n=8) Late (n=7) 10 15 20 Height at Orifice Diameter (cm) 338 25 Figure 22. Variation in Dimension by Period, Medium Size Pingdipen Basins,Hougang PERIOD Early (n=5) Middle (n=5) Late (n=l3)a j 12 8 9 Height (cm) 10 PERIOD Early (n=5) Middle (n=5) Late (n=l3) 12 14 16 18 20 22 24 26 28 Orifice Diameter (cm) 30 32 34 PERIOD Early (n=5) Middle (n=5) # L a t e (n=l3) 12 14 16 18 20 22 Base Diameter (cm) 24 26 28 30 3.39 F i g u r e 2 3 . Variation in Dimension by Period, Medium Size Wan Bowls, Class One, Hougang P E R I O D E a r l y (n=8) M i d d l e (n=IO) L a t e ( n = l 2 ) * " P E R I O D H e i g h t (cm) E a r l y ( n = 8 ) • M i d d l e ^ (n=IO) L a t e ( n = l 2 ) * - 10 II 12 13 14 15 R i m D i a m e t e r (cm) 16 P E R I O D 17 18 E a r l y ( n=8 ) M i d d l e (n=IO) i L a t e (n=l2) 10 B a s e D i a m e t e r ( cm) 340 Figure 24. Variation in Dimensions by Period, Guan Jars, Meishan PERIOD Early Late I 1 | _ 10 15 20 Orifice Diameter (cm) 3 4 1 Figure 25. Variation in Dimensions by Period, Ding Tripods, Meishan PERIOD Early (n=5) Late (n=6) _L 10 15 Orifice Diameter (cm) PERIOD Early (n=5) * Late _^ (n=6) # j 1 P I I : I I I 10 15 20 25 30 Maximum Diameter (cm) 342 Figure 26. Variation in Dimensions by Period, Ding Tripods, Lujiakou PERIOD Early (n-3) ' 1 | | * Late ^ _ 9 (n=4) L__J I i i i i i i 8 10 12 14 16 18 Orifice Diameter (cm) PERIOD Early (n=3) Late (n=4) 10 15 Maximum Diameter (cm) 20 PERIOD Early (n=3) Late (n=4) 10 14 16 Body Height (cm) 18 20 343 APPENDIX C. DATA ON NONCERAMIC POTENTIAL PRESTIGE GOODS, UTILITARIAN ARTIFACTS, AND HOUSING AT HOUGANG AND BAIYING. s i t e reports: Hougang (Anyang Archaeological Team, IA, CASS 1985) Baiying (CPAM of Anyang D i s t r i c t , Henan Province) 344 Table 74. Distr i b u t i o n of Potential High Status A r t i f a c t s Among Different Types of Sites, with Location of Deposition by Excavation (T) Area. (note: for Hougang, E= Early, M= Middle, L= Late; numbers indicate layers) jade bronze oracle ornaments bones walled sites Hougang 1 huan 1 E6, stone: r i n g , E6, T9 2 huan (period not in T2; rings; stated for 3 bi_ bone: 6 most disks, j i hairpins; a r t i f a c t s ) 1 i n M5 3 guan p i t T15-16 tubes; ceramic: 41 huan; 1 small molded human or animal head, E7; s h e l l : 1 piece, M5; 2 pierced pieces smaller sites without walls Baiying Early -- -- 2, areas unclear bone: 1 j i _ pin, 1 piece; ceramic: 12 huan; 345 s h e l l : 1 huan Middle Late 3 strings of beads T25, T16, T49; 3 f l a t pieces T44, T21, T4 Meishan Early Late -- traces on 2 cru c i b l e s areas unclear stone: 1 huan; bone: 5 j i ; ceramic: 24 huan 8, stone: 1 i n T28 3 huan; bone: 1 piece, 4 11/ 1 guan, 2 pierced pieces, 2 horn pieces, 2 tooth items; ceramic: 1 small b i r d , 8 huan; s h e l l : 1 huan, 1 round item bone: 2 JL' ceramic: 2 huan, 1 b a l l , 1 human figu r e , 2 birds bone: 2 j i _ ; ceramic: 1 b a l l , 3 huan, 2 birds 2, areas unclear 346 Lujiakou (period not stated f o r most a r t i f a c t s ) stone: I huan, Late; bone: II i i . ; ceramic: 12 huan, 1 animal 347 Table 75. D i v e r s i t y of U t i l i t a r i a n Tools f o r Different Types of S i t e s . (note: the f i r s t f igure i s the number of forms, the second fig u r e i n parentheses i s the quantity of items) stone bone/ horn s h e l l wood pottery walled s i t e s : Hougang (period not stated f o r most items) s i t e s without walls: Baiying Early Middle Late Meishan Early Late Lujiakou (periods not stated f o r most items) 19 (119) 9 (23) 9 (41) 18 (90) 5 (16) 7 (25) 10 (75) 8 (107) 5 (26) 9 (56) 11 (96) 5 (9) 6 (20) 8 (37) 4 (31) 2 (14) 5 (34) 1 (2) 5 (7) 3 (36) 1 (1) 1 (1) 1 (1) 2 (15) 1 (2) 3 (8) 3 (30) 3 4 8 Table 76. V a r i a t i o n i n Construction Material f o r Houses (Fangzi) at Hougang, Including I d e n t i f i c a t i o n of Houses with Costly Materials. type of wall Early Middle Late Late material layer 6 layer 5 layer 4 layer 3 mud/earthen 2 10 5 6 (tu) wattle and daub 0 1 4 3 (mu gu duo ni) adobe 0 2 1 2 (tupi) F8, F18 F12 F12, F15 unknown 0 1 0 3 t o t a l 2 14 10 14 349 Table 77. V a r i a t i o n i n Construction Material f o r Houses (Fangzi) at Baiying, Including I d e n t i f i c a t i o n of Houses with Costly Materials. type of wall material Early Middle Late earthen (shengtu) mud and straw (cao j i n ni) wattle and daub (mu gu ni) adobe (tupi l e i ) unknown 34 10 1, F65 t o t a l 46 350

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
France 8 3
China 7 2
United States 5 1
Japan 4 0
City Views Downloads
Unknown 7 19
Washington 4 0
Beijing 4 0
Tokyo 4 0
Hangzhou 2 2
Paris 1 0
Shenzhen 1 0
Ashburn 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items