A MORTUARY ANALYSIS OF THE DAWENKOU CEMETERY SITE, SHANDONG, CHINA by ANNE UNDERHILL KINGSCOTT B.A., Duke University, 1977 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Department of Anthropology and Sociology) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October 1983 Anne Underhill Kingscott In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements fo r an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y a v a i l a b l e for reference and study. I further agree that permission for extensive copying of t h i s t h e s i s for s c h o l a r l y purposes may be granted by the head of my department or by h i s or her representatives. I t i s understood that copying or p u b l i c a t i o n of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of ^/ftj/W^-L*^ CT/VJL ^rd^rtry^ The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date E-6 (3/81) ABSTRACT This study i s concerned with the development of s o c i a l ranking i n Shandong province, China, and i t s environs, during the late N e o l i t h i c period. The Dawenkou cemetery has been considered by Chinese and western archaeologists as represent-ing one of the e a r l i e s t ranked societ i e s i n t h i s region. There i s some disagreement regarding the degree of status d i f f e r e n t i a -t i o n r e f l e c t e d by the mortuary remains at Dawenkou. Opinions vary as to whether the s i t e shows i n c i p i e n t ranking, or a more f u l l y developed system, or whether there was a hereditary r u l i n g class. The primary goal of t h i s study i s to provide a greater understanding of the nature and degree of status d i f f e r e n t i a -t i o n represented at Dawenkou, by means of an in-depth mortuary analysis u t i l i z i n g current archaeological methods. The methodology upon which t h i s study i s based i s outlined i n Chapter 2 . Four analyses are included: an evaluation of the three r e l a t i v e chronological periods at the s i t e (Chapter 3 ) , an analysis to estimate the sex of unsexed b u r i a l s on the basis of grave goods (Chapter 4 ) , an exploratory assessment of s o c i a l subgroup a f f i l i a t i o n (Chapter 5 ) , and the analysis of status d i f f e r e n t i a t i o n (Chapter 6 ) . I conclude that an increase through time i n the degree of ranking i s represented at the s i t e . The Early period b u r i a l s r e f l e c t elements of both an achieved and an ascribed (ranked) system of status d i f f e r e n t i a t i o n . The Late period b u r i a l s appear to r e f l e c t a highly developed ranked society. I propose that the Early period b u r i a l s represent part of a s o c i a l system i n which members of a descent group were ranked. Also, the Late period b u r i a l s represent members of a descent group that constitutes one status l e v e l i n a regional status system. Mortuary analyses of other Dawenkou Culture s i t e s roughly contemporaneous to the Early and Late periods at Dawenkou could reveal whether sim i l a r changes through time are apparent. It i s l i k e l y that ranking f i r s t developed i n the eastern seaboard region at an e a r l i e r date than previously considered. The secondary goal of t h i s study i s to make a methodo-l o g i c a l contribution to mortuary analysis. I argue that status may be symbolized by energy expenditure, grave goods, or both. Since the b u r i a l s i n a cemetery r e f l e c t more than one s o c i a l system through time, an analysis of status should emphasize change through time. F i n a l l y , some of the techniques employed in t h i s study should have u t i l i t y for other mortuary analyses: the Simple Inspection Method to estimate sex (Chapter 4 ) , and Ward's Method of c l u s t e r analysis and Torgerson's Metric Multidimensional Scaling for an analysis of status (Chapter 6 ) , or for s e r i a t i o n (Chapter 3 ) . i v TABLE OF CONTENTS Page ABSTRACT . i i TABLE OF CONTENTS i v LIST OF FIGURES v i i LIST OF APPENDICES x ACKNOWLEDGEMENTS x i i CHAPTER 1 Introduction 1 1.1. The Dawenkou Site and the Research Problem 1 1.2. C l a s s i f i c a t i o n and Description of Dawenkou Culture 5 1.3. C u l t u r a l Evolution i n the Late Neolithic Period 14 CHAPTER 2 Methodology 17 2.1. General Approaches i n Mortuary Analysis 17 2.2. The Processual Approach 19 2.3. The Symbolist Approach 36 2.4. The Approach Followed i n t h i s Mortuary Analysis 43 2.5. Method to Determine Status Levels . . 4 8 2.6. Consideration of Natural and Cu l t u r a l Factors that may have Affected the Archaeological Record at Dawenkou . . 54 CHAPTER 3 Chronological Analysis 58 3.1. Introduction . 5 8 3.2. The Methods to Derive the Chronological Periods 61 3.3. Analysis 65 3.3.1. Introduction 65 3.3.2. Method 6 7 3.3.3. Data 74 3.3.4. Results 78 3.3.5. Multidimensional Scaling 79 3.3.6. Cluster Analyses . 84 3.4. Conclusion 89 V Page CHAPTER 4 Analysis to Estimate Sex 95 4.1. The Problem 95 4.2. The Analysis 99 4.2.1. Introduction 99 4.2.2. Method . . . 100 4.2.3. The Simple Inspection Method: Data 104 4.2.4. The Simple Inspection Method: Results . . . ' 107 4.2.5. Discriminant Analysis: Data . . . I l l 4.2.6. Discriminant Analysis: Results . 113 4.3. Conclusion 119 CHAPTER 5 The Nature of Social Subgroup A f f i l i a t i o n at Dawenkou 126 5.1. Method 126 5.2. Spatial Location of Graves 130 5.3. Correlation of Grave Orientation, Grave Form, and Body Disposition with Grave Location 131 5.4. Correlation of Ceramic Style with Spatial Location 14 3 5.5. Interpretation of the Potential Social Subgroup A f f i l i a t i o n -Related Variables 145 5.5.1. Introduction 145 5.5.2. Orientation 145 5.5.3. Grave Form 149 5.5.4. Body Disposition . . . 151 5.6. Concluding Propositions . 153 5.6.1. Argument for a Descent Group at Dawenkou 15 3 5.6.2. Kinship or Residential .Groups at Dawenkou 156 5.6.3. The Mortuary Population at Dawenkou .157 CHAPTER 6 Analysis of Status D i f f e r e n t i a t i o n . . 160 6.1. Introduction 160 6.1.1. Procedure 160 6.1.2. Test Implications 161 6.2. Analysis of Status, Early period . . 163 6.2.1. Data 163 6.2.1.1. The Multivariate Analyses . . . 163 6.2.1.2. High Status Unique Items . . . 169 6.2.1.3. Energy Expenditure 172 6.2.2. Results: The Multidimensional Scaling 173 v i Page CHAPTER 6 (cont'd) 6.2.3. The Three Types of Cluster Analysis 176 6.2.4. Interpretation of Status D i s t i n c -tions i n the Early Period . . . . . 181 6.2.4.1. D i s t r i b u t i o n of the Unique Items 181 6.2.4.2. Argument for Four Status Levels 182 6.2.4.3. D i f f e r e n t i a t i o n by Energy Expenditure . 183 6.2.4.4. Body Disposition 188 6.2.4.5. Grave Location 188 6.2.4.6. Test Implications for Achieved versus Ascribed Status 191 6.2.4.7. Multiple Burials 193 6.2.4.8. Conclusions . 195 6.3. Analysis of Status, Late Period . . . 198 6.3.1. Data 198 6.3.1.1. The Multivariate Analyses . . . 198 6.3.1.2. Energy Expenditure 201 6.3.1.3. High Status Unique Items . . . . 202 6.3.2. Results: The Multidimensional Scaling 203 6.3.3. The Three Types of Cluster Analysis 205 6.3.4. Interpretation of Status D i s t i n c -tions i n the Late Period and Assessment of Change Through Time . 209 6.3.4.1. D i s t r i b u t i o n of the Unique Items 209 6.3.4.2. Argument for Three or Two Status Levels 210 6.3.4.3. D i f f e r e n t i a t i o n by Energy Expenditure . . . 216 6.3.4.4. Body Disposition and Grave Location . 217 6.3.4.5. Test Implications 219 6.3.4.6. Conclusions 222 6.4. Implications 227 CHAPTER 7 Conclusions and Suggestions for Future Research . . . 229 7.1. Conclusions Regarding Dawenkou . . . 229 7.2. Future Research 233 7.3. Methodological Conclusions 238 BIBLIOGRAPHY . . . . . 241 APPENDICES 258 GLOSSARY 318 v i i LIST OF FIGURES Figure Page 1-1 The location of Dawenkou and other Ne o l i t h i c s i t e s 7 1-2 Rough chronological relationship of the Dawenkou Culture with other N e o l i t h i c cultures of China 8 1- 3 Rough chronological order of N e o l i t h i c s i t e s i n the eastern seaboard region of China 9 2- 1 Number of dimensional d i s t i n c t i o n s of the s o c i a l persona according to form of subsistence 24 2-2 Average number of dimensional d i s t i n c -tions according to subsistence cate-gory 2 4 2- 3 Aspects of the s o c i a l persona symbolized i n mortuary treatment 26 3- 1 The 115 b u r i a l s i n the chronological analysis from Torgerson's Metric M u l t i -dimensional Scaling, with clusters from Ward's Method 81 3-2 Dendrogram from Ward's Method of c l u s t e r analysis 86 3-3 L i s t of b u r i a l s i n the newly assigned Early and Late periods derived from the chronological analysis, as well as undatable b u r i a l s • . . 90 3- 4 Location of b u r i a l s within the cemetery from each period as defined by the chronological analysis 91 4- 1 D i s t r i b u t i o n of sex-linked a r t i f a c t types 102 4-2 Distance values for the unsexed b u r i a l s from the Simple Inspection Method and assignment of sex, and the distance values for known sexed b u r i a l s . . . . 109 v i i i Figure Page 4-3 Standardized Discriminant Function Co e f f i c i e n t s for the 2 7 variables that d i s t i n g u i s h between known male and female classes of bu r i a l s . . . . 116 4-4 Discriminant scores for the 82 unsexed, single adult b u r i a l s i n the Discrimin-ant Analysis 117 4-5 Known sexed and estimated sexed b u r i a l s from the Simple Inspection Method and Discriminant Analysis . . . 121 4- 6 Spatial location of the known sexed and estimated sexed b u r i a l s i n the cemetery 122 5- 1 Spatial groups of b u r i a l s derived by v i s u a l inspection, Early period . . . 132 5-2 Spatial groups of b u r i a l s derived by vi s u a l inspection, Late period . . . . 133 5-3 The range of grave orientations for Early, Late and undatable b u r i a l s . . 135 5-4 Correlation of grave form, grave orientation, body d i s p o s i t i o n , sex and age with s p a t i a l location of grave, Early period . . . . . . 136 5-5 Correlation of grave form, grave orie n t a t i o n , body d i s p o s i t i o n , sex and age with s p a t i a l location of grave, Late period 141 5- 6 Correlation of grave form, grave orientation, body d i s p o s i t i o n , sex and age with s p a t i a l location of grave, undatable b u r i a l s 142 6- 1 Multidimensional scaling p l o t of Early period b u r i a l s i n the analysis of status, with Ward's Method clusters . . 175 6-2 Dendrogram of Early period b u r i a l s from Ward's Method . . . . . 178 6-3 Distinguishing c h a r a c t e r i s t i c s of the four status l e v e l s , Early period . . . 184 i x Figure Page 6-4 Age and sex composition of the status lev e l s i n the Early period 189 6-5 Location of the four status groups i n the Early period 190 6-6 Multidimensional scaling plot of Late period b u r i a l s i n the analysis of status, with Ward's Method clusters . . . 204 6-7 Dendrogram of Late period b u r i a l s from Ward's Method 207 6-8 Distinguishing c h a r a c t e r i s t i c s of the three status l e v e l s , Late period . . . . 213 6-9 Age and sex composition of the three status le v e l s i n the Late period, and data regarding the one multiple b u r i a l i n the Late period 218 6-10 Location of the three status groups i n the cemetery, Late period 220 X LIST OF APPENDICES A3-1 The functional types, subtypes, and styles of pottery at the Dawenkou s i t e 258 A3-2 The 15 pairs of intrusive b u r i a l s 261 A3-3 Burials included i n the chronological analysis 262 A3-4 The 83 ceramic forms i n the chronological analysis . 263 A3-5 D i s t r i b u t i o n of ceramic categories among the b u r i a l sample 2 65 A3-6 Frequency of occurrence of ceramic categories af t e r lumping 266 A3-7 D i s t r i b u t i o n of ceramic categories a f t e r lumping 267 A3-8 Composition of clust e r s from Ward's dendro-gram i n terms of ceramic styles 268 A4-1 L i s t of sexed and unsexed b u r i a l s : Early, Late, undatable 2 71 A4-2 The 129 a r t i f a c t types i n the Discriminant Analysis 272 A5-1 Orientation of graves, Early period . . . 274 A5-2 Orientation of graves, Late period . . . 275 A5-3 D i s t r i b u t i o n of ceramic s t y l e s , Early period 276 A5-4 D i s t r i b u t i o n of ceramic s t y l e s , Late period 281 A6-1 The 79 b u r i a l s i n the Early period analysis of status, with data on age and sex, grave form, grave size 285 A6-2 The 2 3 variables and t h e i r attributes i n the Early period analysis of status . . . . . 287 A6-3 D i s t r i b u t i o n of serving stands, Early period 289 A6-4 D i s t r i b u t i o n of pottery vessels, Early period 290 A6-5 D i s t r i b u t i o n of stone tools, Early period . . 291 A6-6 D i s t r i b u t i o n of bone tools, Early period . . 292 A6-7 Di s t r i b u t i o n of pig s k u l l s , Early period . . 293 x i F i g u r e Page A 6 - 8 D i s t r i b u t i o n o f d e e r t e e t h , E a r l y p e r i o d . . 294 A 6 - 9 D i s t r i b u t i o n o f raw m a t e r i a l p i e c e s , E a r l y p e r i o d 295 A 6 - 1 0 G r a v e a r e a o f E a r l y p e r i o d b u r i a l s . . . . 296 A6-11 The m o r t u a r y a t t r i b u t e s o f t h e s e v e n c l u s t e r s f r o m Ward's M e t h o d , E a r l y p e r i o d b u r i a l s . . 2 9 7 A 6 - 1 2 M u l t i p l e b u r i a l s f r o m t h e E a r l y p e r i o d : a g e , s e x , g r a v e g o o d s , e n e r g y e x p e n d i t u r e . 302 A6-13 The 32 b u r i a l s i n t h e L a t e p e r i o d a n a l y s i s o f s t a t u s , w i t h d a t a on age and s e x , g r a v e f o r m , g r a v e s i z e 303 A6-14 The 2 8 v a r i a b l e s a nd t h e i r a t t r i b u t e s i n t h e L a t e p e r i o d a n a l y s i s o f s t a t u s 304 A6-15 D i s t r i b u t i o n o f s e r v i n g s t a n d s , L a t e p e r i o d 306 A6-16 D i s t r i b u t i o n o f t a l l stemmed c u p , L a t e ' p e r i o d 307 A6-17 D i s t r i b u t i o n o f p o t t e r y v e s s e l s , L a t e p e r i o d 308 A6-18 D i s t r i b u t i o n o f s t o n e t o o l s , L a t e p e r i o d . 309 A6-19 D i s t r i b u t i o n o f bone t o o l s , L a t e p e r i o d . . 310 A6-20 D i s t r i b u t i o n o f d e e r t e e t h , L a t e p e r i o d . . 311 A6-21 D i s t r i b u t i o n o f raw m a t e r i a l p i e c e s , L a t e p e r i o d 312 A6-22 G r a v e a r e a o f L a t e p e r i o d b u r i a l s . . . . 313 A6-2 3 The m o r t u a r y a t t r i b u t e s o f t h e e i g h t c l u s t e r s f r o m Ward's M e t h o d , L a t e p e r i o d b u r i a l s 314 A6-24 D a t a r e g a r d i n g t h e u n d a t a b l e b u r i a l s : a g e , s e x , g r a v e g o o d s , e n e r g y e x p e n d i t u r e . . 317 x i i ACKNOWLEDGEMENTS I am extremely grat e f u l to several individuals for t h e i r valuable comments, encouragement and support throughout the duration of t h i s study. I f e e l very fortunate to have had Dr. Richard Pearson, Dr. R.G. Matson, and Dr. David Pokotylo on my M.A. Committee. My advisor, Dr. Pearson, gave me much help and encouragement throughout the three years of my M.A. program. He generously shared his data, published and unpub-lished papers, and ideas about the Dawenkou s i t e and the Chinese N e o l i t h i c period. His knowledge and enthusiasm for East Asian archaeology have been an i n s p i r a t i o n to me. Dr. Matson was very generous with his time regarding computer analyses and the interpretation of r e s u l t s . Dr. Pokotylo gave me detailed comments for every section of the study. I also thank the Department of Anthropology and Sociology for computer funds. Michael Blake, v i s i t i n g professor at the Department of Anthropology and Sociology during 1983, provided valuable i n -sight into the methodology of mortuary analysis. Dr. N e i l Guppy calmed my panic at the computer terminal on a few occa-sions. Kian Kwok generously helped with t r a n s l a t i o n problems. I am g r a t e f u l to Zou Heng of Beida University, B e i j i n g , and Gao Guangren of the Institute of Archaeology, Chinese Academy of S o c i a l Sciences, for t h e i r information. Clarence Shangraw of the Asian Art Museum, San Francisco, was extremely generous with his time and help. During his v i s i t to t h i s Department x i i i i n 1983, Dr. Lewis Binford gave me hel p f u l comments on method-ology. I could not have produced t h i s thesis without the help of Moira Irvine, Pat Berringer, J u l i e Mandziuk and Rick Clements. I greatly appreciate the time Moira spent drawing several figures. J u l i e and Rick typed four chapters aft e r working a l l day. Pat Berringer typed the remainder of the thesis and gave me valuable advice, and provided much support and encouragement. She i s one of the most generous individuals I have ever met. Dana Lepofsky and Deanna Ludowicz were also very supportive. I e s p e c i a l l y thank Michael for his humor and loving support. My parents, as always, gave me much encouragement. F i n a l l y , I thank Dr. Michael Hammond for his support during my years at Duke University, and for stimulating my i n t e r e s t i n the development of prehistoric, complex s o c i e t i e s . - 1 -CHAPTER 1 INTRODUCTION 1.1. The Dawenkou Site and the Research Problem The Dawenkou s i t e i s a late N e o l i t h i c period s i t e located i n south-central Shandong province. The s i t e was discovered i n 1959 during the construction of a railway track near the town of Dawen on the Dawen River (Gao 1978:31). The f i r s t excava-t i o n of the s i t e took place i n 1959 and the second i n 1974. The 133 b u r i a l s discovered i n the f i r s t excavation are the subject of t h i s mortuary analysis. A detailed description of these b u r i a l s i s provided i n a s i t e report published i n 1974 by the Shandong Pr o v i n c i a l C u ltural Properties Commission and the Jinan Cit y Museum (Dawenkou: Report of the Excavation of a N e o l i t h i c Cemetery. Peking: Wen Wu Press). The b u r i a l s from the 1959 excavation are regarded as c r i t i c a l for understanding the development of s o c i a l ranking during the Ne o l i t h i c period i n the eastern seaboard region. Both western and Chinese archaeologists maintain that the Dawenkou cemetery represents a ranked society, but they disagree over the degree and nature of status d i f f e r e n t i a t i o n represented at the s i t e . Some researchers maintain that the s i t e r e f l e c t s i n c i p i e n t ranking and others, a highly developed system of ranking. The primary goal of t h i s study i s to provide a greater understanding of the s i t e by means of an in-depth mortuary analysis u t i l i z i n g current archaeological methods. My method-- 2 -ology and the outline of t h i s study are presented i n Chapter 2. The secondary goal of t h i s study i s to make a methodological contribution to mortuary analysis, i n terms of o v e r a l l approach and the s p e c i f i c techniques that have been employed to i d e n t i f y status d i s t i n c t i o n s i n a cemetery. The authors of the Dawenkou s i t e report c l a s s i f y the 13 3 b u r i a l s into three periods: Early (75 b u r i a l s ) , Middle (19 b u r i a l s ) , and Late (25 b u r i a l s ) . Fifteen b u r i a l s are regarded as undatable. Samples for radiocarbon dates were not taken during the 1959 excavation. A recent a r t i c l e presents a revis i o n of t h i s r e l a t i v e chronology, stating there are two periods represented at the s i t e : Early and Late. The Early period i s said to represent approximately 600 years (from 3400 - 2800 B.C.) and the Late period, 400 years (2800 - 2400 B.C.) (Wu Ruzuo^ 1982: 268"). The method by which these dates were derived i s not clear. The b u r i a l s from the 1974 excavation represent an e a r l i e r time period than those excavated i n 1959. Apparently, the report for the second excavation i n 19 74 has not been completed and the exact number of b u r i a l s found i s not known (Gao, personal communciation, 1983). Two radiocarbon dates from material found during the 1974 excavations yielded the following recalibrated dates: 1) ZK 469(T10,4; lower H24) : 6155 + 140 (4205 B.C.) and 2) ZK 468(T12,2B; lower H3) : 6210 + 135 (4260 B.C.) (Wu Ren 1982:55). The Dawenkou s i t e also has some Longshan period and Han Dynasty remains (Shandong P r o v i n c i a l C u l t u r a l Properties Commission and the Jinan City - 3 -Museum 1974: plate I I I ) . In The Archaeology of Ancient China, K. C. Chang (1979:160) states that Dawenkou and other Dawenkou Culture s i t e s represent an i n c i p i e n t system of ranking. More recently, Chang (19 80:361, 1983a:513-514) states i t i s not clear i f the Dawenkou Culture represents a l e v e l of c u l t u r a l development which he c a l l s " i n t e r v i l l a g e aggregates" (characterized by evidence for r i c h and poor groups, violence, c r a f t s s p e c i a l i z a t i o n , r e l i g i o u s and v i l l a g e leaders such as chi e f s , and i n t e r v i l l a g e leagues). He believes that Longshan s i t e s of Shaanxi, Henan, and Shandong had reached t h i s l e v e l of c u l t u r a l complexity. Chinese archaeologists view s o c i o - c u l t u r a l evolution i n terms of Marxist h i s t o r i c a l philosophy and disagree whether Dawenkou and other Dawenkou Culture s i t e s r e f l e c t the begin-ning of slave society (Chang 1983b:575). However, a r t i c l e s by the Wen Wu Correspondent (19 78), the Shandong P r o v i n c i a l Museum (1978), the Kao Gu E d i t o r i a l Staff (1979), Zhang (1979), Gao (1978), Wei (1976), Luo and Zhang (1979), Wu (1973), the Nanjing Museum (1978) imply that ranking i s f u l l y developed at Dawenkou, e s p e c i a l l y by the Late period. An increase i n the d i f f e r e n t i a t i o n of b u r i a l s i n terms of grave goods from the Early period to the Late i s noted by the Shandong P r o v i n c i a l Museum (1978) and implied by the Kao Gu E d i t o r i a l Staff (1979). Luo and Zhang (19 79) note a concentration of high status b u r i a l s from both periods i n the northern sector of the s i t e . Pearson's (1981) position on the degree of ranking at - 4 -Dawenkou i s i n t e r m e d i a t e t o t h a t o f Chang (19 79) and the Chinese a r c h a e o l o g i c a l l i t e r a t u r e c i t e d above. Pearson (1981:1086) proposes t h a t the presence of stone beads and stone and jade ornaments i n both male and female graves i n d i c a t e s a system of ra n k i n g . A l s o , the wealth e v i d e n t from some graves suggests the presence of a c h i e f l y o f f i c e . The l a c k o f a complete s e g r e g a t i o n o f wealthy graves may i n d i c a t e the absence of a h e r e d i t a r y r u l i n g c l a s s . Pearson's (19 81) comparative study of s t a t u s d i f f e r e n t i a -t i o n among s e v e r a l s i t e s i n the e a s t e r n seaboard r e g i o n i s the only study u t i l i z i n g c u r r e n t a r c h a e o l o g i c a l methods of mortuary a n a l y s i s t o understand the nature and degree of s t a t u s d i f f e r -e n t i a t i o n at Dawenkou C u l t u r e s i t e s . My study i s an outgrowth of Pearson's (1981) study. I hope the a p p l i c a t i o n of c u r r e n t methods of mortuary a n a l y s i s on the Dawenkou data can pr o v i d e a b e t t e r understanding of the nature and degree of s t a t u s d i f f e r -e n t i a t i o n a t the s i t e , as w e l l as t o r e s o l v e the c o n t r o v e r s y i n the degree of ranki n g noted above. T h i s study i n c l u d e s an e v a l u a t i o n o f s e v e r a l trends through time i n s t a t u s d i f f e r e n t i a t i o n t h a t are concluded by Pearson (19 81). Pearson notes an i n c r e a s e i n the mean number of ceramics i n b u r i a l s (1981:1081, Table 2, and 1082) as w e l l as an i n c r e a s e i n the percentage of b u r i a l s w i t h ceramics (1981: 10 85). A l s o , an i n c r e a s i n g v a r i a t i o n i n the number of ceramics found i n each b u r i a l through time i s suggested (19 81:10 82,1083, Table 4). For the Dawenkou s i t e t here i s a l s o i n c r e a s i n g - 5 -va r i a t i o n i n quantities of tools and ornaments within b u r i a l s (19 81:10 85). An increase i n the quantities of ceramics i n male bu r i a l s over female b u r i a l s through time i s noted for some of the late Dawenkou Culture s i t e s , but t h i s increase i s not apparent i n the Dawenkou s i t e b u r i a l s due to the small number of sexed b u r i a l s (1981:1083). A trend towards s p a t i a l segrega-tion of high status b u r i a l s by the Late period at Dawenkou i s proposed (1981:1084) but i s considered tentative u n t i l more assessment has been made (1981:1085). Pearson (1981:1086) concludes that by the Late Dawenkou Culture period at approx-imately 2 000 B.C. (which includes the Late period at Dawenkou s i t e ) , evidence of pronounced ranking i s not present, nor i s occupational s p e c i a l i z a t i o n . He also proposes that the status of men i n terms of power and wealth was increasing by the Late Dawenkou Cu l t u r a l period and c r a f t s p e c i a l i z a t i o n was develop-ing ( i b i d ) . 1.2. C l a s s i f i c a t i o n and Description of Dawenkou Culture Dawenkou i s regarded i n the Chinese archaeological l i t e r a -ture as part of the Dawenkou Culture, which spans a period of approximately 2000 years (An 1979-80:38, 1981:258, 1982:58). Dawenkou Culture s i t e s , the great majority of which are cemeteries, are located i n the lower Yellow River v a l l e y , primarily i n Shandong and Jiangsu (An 1979-80:38). Other Dawenkou Culture s i t e s have been found i n northern Anhui, eastern Henan and the Liaodong Peninsula ( i b i d ) . The location of Dawenkou and other N e o l i t h i c s i t e s i n the eastern seaboard - 6 -region i s indicated i n F i g . 1-1 (after Pearson 1981:1079). Only the s i t e s i n the northern areas of the map (northern Jiangsu and Shandong) are considered as part of the Dawenkou Culture. I am not aware of a published source that l i s t s the t o t a l number of Dawenkou Culture s i t e s that have been discovered, as well as the number of b u r i a l s from each s i t e and the geographic location of each s i t e . Over 6,000 Ne o l i t h i c s i t e s have been discovered i n China during the past t h i r t y years (An 1979-80:35). The Dawenkou Culture i s considered i n the Chinese archaeo-l o g i c a l l i t e r a t u r e to e x i s t from approximately 4000 B.C. to 2000B.C. F i g 1-2 (from An 1982:58) depicts the rough chrono-l o g i c a l r e l a t i o n s h i p of Dawenkou Culture with other N e o l i t h i c cultures of China. Radiocarbon dates from i n d i v i d u a l Dawenkou Culture s i t e s are discussed by Xia (1979), Wen Wu Correspondent (1978), Wu Ren (1982), Gao (1980), and Pearson (in press:7-8). The dating of the culture i n general i s discussed by An (1979-80). He explains (1979-80:38) that twelve radiocarbon dates have been taken from Dawenkou Culture s i t e s . Two, ZK 90 and ZK 479, show a range of 4494 - 2690 B.C. (calibrated). Four dates ( Z K 317, 319-0, 321, and 361-0) give a range of 2350 - 1905 B.C. (calibrated). An (1979-80:38) believes these late dates may be explained by d i f f i c u l t i e s i n distinguishing Dawenkou Culture s i t e s from Longshan Culture s i t e s . The rough chronological order of s i t e s i n the eastern seaboard region (from Pearson 1981:1081) i s indicated i n F i g . 1-3. There i s an ongoing controversy regarding the c u l t u r a l - 7 -FIGURE 1-1. The location of Dawenkou and other N e o l i t h i c s i t e s i n the eastern seaboard region of China. (from PearBon 1981:1079) - 8 -FIGURE 1-2. Rough chronological r e l a t i o n s h i p of the Dawenkou Culture with other N e o l i t h i c cultures of China. Chronology of the Ne o l i t h i c Period i n the Huanghe (Yellow) and Changjiang (Yangtze) River Valleys upper reaches of the Huanghe middle reaches of the Huanghe lower reaches of the Huanghe lower reaches of the Changjiang middle reaches of the Changjiang 1000 B.C. B A r S 8 Siba Culture Q i j i a Culture Shang Shang Shang Shang Longshan Culture Longshan Culture Majiayao Culture Liangzhu Culture Dawenkou Culture Majiabang Culture Longshan Culture ' 7$uj f a t i n g . —Culture | Daxi Culture Yangshao Culture lYangshao Culture 2000 B.C. 3000 B.C. I Dadiwan Culture Peiligang Culture I I Qlngliangang Culture Cishan Culture I U000 B.C. 5000 B.C. 6000B.C. (from An Zhimin 1982:58) - 9 -FIGURE 1-3. Rough chronological order of Ne o l i t h i c s i t e s i n the eastern seaboard region (adapted from Pearson 1981: 1081). Dafanzhuang Xixiahou Dawenkou Dawenkou Dawenkou Dadunzi Songze L i u l i n L i u l i n L i u l i n Beiyinyangying Maj iabin Yuduncun Yuduncun Yuduncun Late Middle Early Huating layer L i u l i n layer Dawenkou Culture s i t e Dawenkou Culture s i t e Dawenkou Culture s i t e Dawenkou Culture s i t e excavation 1 Dawenkou excavation 2, upper layer Culture excavation 2, lower layer s i t e Layer IB Layer 2 Layer 3 I - 10 -c l a s s i f i c a t i o n of the Neo l i t h i c s i t e s from the eastern seaboard region. K.C. Chang (1979:136) maintains that the culture p r i o r to Dawenkou, the Qingliangang (see F i g . 1-2), should be divided into a type including s i t e s north of the Yangtze River and a type including s i t e s south of the Yangtze. Chang (1979:138) also considers s i t e s designated as Early Dawenkou Culture by Chinese archaeologists as part of the Qingliangang period. Chang (1979:144, 154-55) c l a s s i f i e s Dawenkou Culture s i t e s after approximately 3200 B.C. as part of the Longshanoid c u l t u r a l complex, which includes N e o l i t h i c s i t e s from the North Central Plain area. The Dawenkou s i t e i s considered as part of the Huating Culture, along with the L i u l i n and Huating s i t e s (Chang 1979:160). A multivariate s t a t i s t i c a l test of the c l a s s i f i c a t i o n of Longshanoid cultures including Dawenkou by Lo (1977) resulted i n a r e p l i c a t i o n of the culture c l a s s i f i c a t i o n scheme i n most cases. Huber's (1981:118, 1983:202) c l a s s i f i c a t i o n scheme i s based p a r t i a l l y upon ceramic development. She concludes that s i t e s i n northern Jiangsu and Shandong are part of the same c u l t u r a l complex, but gives a d i f f e r e n t chronological sequence of s i t e s from these provinces. The scheme u t i l i z e d by Pearson (1981:1078, 1080) and Shangraw (1978:12) consists of the Qingliangang Culture (approximately 5000 - 3000 B.C.), includ-ing s i t e s i n northern Zhejiang, Jiangsu, and southern Shandong such as the L i u l i n and Huating s i t e s ; and the Dawenkou Culture (approximately 3000 - 2000 B.C.). As Shangraw (1978:34) points out, terminological confusion - 11 -over the Dawenkou Culture s i t e s has resulted i n d i f f e r e n t terms being used by Chinese archaeologists from d i f f e r e n t provinces. Shandong archaeologists have la b e l l e d the si t e s i n Shandong as Dawenkou Culture s i t e s while those i n Jinagsu have used the terms Qingliangang, L i u l i n and Huating periods. A symposium i n 19 77 described by the Wen Wu Correspondent (1978) attempted to resolve these terminological problems. Archaeologists at t h i s symposium had three d i f f e r e n t views regarding c l a s s i f i c a t i o n of sit e s north of the Yangtze. As explained by the Wen Wu Corres-pondent (19 78), the f i r s t view, by the Nanjing Museum, advocated six periods consisting of certain s t r a t a from various s i t e s . The Shandong Museum advocated eleven periods and Be i j i n g Univer-s i t y held the t h i r d view that there are eight periods. The participants of the symposium also grouped these periods to-gether into larger stages. These stages appear to be derived on the basis of radiocarbon dates, stratigraphy, technological and s t y l i s t i c c h a r a c t e r i s t i c s of ceramics, stone tools and other aspects of material culture; and an assessment of general c u l t u r a l complexity. The Shandong Museum advocated three stages while the proponents of the two stage view disagreed on the d i v i s i o n s of the f i r s t and second stages. Most Chinese investigators appear to advocate the three stage d i v i s i o n of the Dawenkou Culture. According to Gao (personal communication 1983; 1980:49), the 1974 excavation at the Dawenkou s i t e belongs to the f i r s t stage, the Early and Middle periods from 1959 belong to the second, and the Late - 12 -p e r i o d b u r i a l s belong t o the Late stage. The Shandong Museum (19 78) d e s c r i b e s three stages f o r the Dawenkou C u l t u r e : the E a r l y stage c o n s i s t i n g o f b u r i a l s from the 19 74 e x c a v a t i o n a t Dawenkou, as w e l l as one l a y e r of the Dadunzi s i t e , the Middle stage as c o n s i s t i n g o f some graves from the 19 74 e x c a v a t i o n , some graves from Yedian and Dadunzi and the E a r l y and Middle p e r i o d s from the Dawenkou s i t e (1959 e x c a v a t i o n ) , and the Late p e r i o d b u r i a l s from the 1959 e x c a v a t i o n comprising the Late stage, along w i t h b u r i a l s from X i x i a h o u and Yedian. There was a l s o disagreement a t the 19 77 symposium r e g a r d i n g the r e l a t i o n s h i p o f c u l t u r e s n o r t h and south of the Yangtze. Some p a r t i c i p a n t s regarded the no r t h e r n s i t e s as r e l a t e d but c u l t u r a l l y d i s t i n c t from the southern s i t e s , w h i le o t h e r s regarded them as fundamentally the same. Because the term "Dawenkou C u l t u r e " has been used f a i r l y c o n s i s t e n t l y i n the Chinese a r c h a e o l o g i c a l l i t e r a t u r e s i n c e t h i s symposium, i t appears t h a t the m a j o r i t y o f Chinese a r c h a e o l o g i s t s now h o l d the view t h a t the c u l t u r e s n o r t h and south of the Yangtze developed from the e a r l i e r Qingliangang c u l t u r e which was d i s t r i b u t e d i n areas both n o r t h and south of the Yangtze. The major c h a r a c t e r i s t i c s o f s i t e s i n the e a s t e r n seaboard r e g i o n i n c l u d e probable c u l t i v a t i o n of r i c e and ot h e r p l a n t s s u i t a b l e t o the low l y i n g , marshy environment of the r e g i o n , as suggested by the d i s c o v e r y of the Hemudu s i t e i n Zhejiang (Pearson 19 83:124) — although no remains of r i c e have been found i n s i t e s n o r t h of the Yangtze (Chang 1979:136). There i s - 13 -great consistency i n b u r i a l practices, including: s i m i l a r grave orientation and body d i s p o s i t i o n ; s i m i l a r grave good inclusions such as stone t o o l and ceramic forms, ornaments, pig s k u l l s , deer teeth (Chang 19 79:160-162, Pearson, i n press:17-24), animal shaped vessels (Pearson 1983:140); s i m i l a r practices such as tooth extraction and s k u l l deformation primarily for females (Pearson, i n press:26), and keeping a clay b a l l i n the mouth during one 1s l i f e t i m e (Han and Pan 1980). Incised symbols on ceramic sherds at some s i t e s may indicate a form of proto-writing (Chang 1979:161). A description of b u r i a l practices at s p e c i f i c Dawenkou Culture s i t e s i s included i n di f f e r e n t chapters of t h i s study: Chapter 4 ( a r t i f a c t forms associated with the sexes), Chapter 5 (body d i s p o s i t i o n , grave form, or i e n t a t i o n ) , and Chapters 6 and 7 (status-related grave goods). The environment of the region was probably warmer and moister than at present (Pearson, i n press:25, 1983:134), and most s i t e s are located on small r i v e r s or lakes (Pearson, i n press:2). Some r i v e r s i n Shandong may have been suitable for small to medium-sized i r r i g a t i o n projects (Pearson, i n press:24). The only habitation remains of which I am aware at Dawenkou Culture s i t e s are the remains of house foundations from the 1974 excavation at the Dawenkou s i t e (Wen Wu Correspondent 1978) and a reference to a s i t e i n Shandong by Be i j i n g University (19 83: 195). The Dawenkou s i t e , as represented by the 1959 excavation, appears to be the only cemetery s i t e containing a k i l n . The - 14 -k i l n may be associated with the Late period (Gao, personal communication, 19 83) because some Late period sherds were found i n i t . However, the fact that sherds from the Longshan and Shang periods were also found with the k i l n (Shandong Pr o v i n c i a l Cultural Properties Commission and the Jinan City Museum 19 74:114) may indicate the k i l n i s not dated to the Late period. Subsistence and settlement pattern data for the Dawenkou Culture are lacking (Pearson, i n press:30). However, animal bones from Dawenkou Culture s i t e s suggest a moist, forested environment: r i v e r deer, wild pig, wild bovid, species of t u r t l e (Pearson, i n press:25). Animal species mentioned i n the Dawenkou s i t e report include a l l i g a t o r , crane, d i f f e r e n t types of f i s h and birds, oysters, and domesticated pig and chicken (Shandong P r o v i n c i a l C u l t u r a l Properties Commission and the Jinan City Museum 1974:156-158). Some types of stone tools described i n the s i t e report r e f l e c t a g r i c u l t u r a l a c t i v i t i e s such as the s i c k l e , stone spade and adze. Fishing i s suggested by f i s h hooks, weaving and sewing by spindle whorls and bone needles, and hunting by bone arrowheads and knives. 1.3. C u l t u r a l Evolution i n the Late Ne o l i t h i c Period As F i g . 1-2 i l l u s t r a t e s , there are several c u l t u r a l regions during the late N e o l i t h i c period. Chang (1979:155) maintains that by the l a t t e r part of the N e o l i t h i c period (approximately 3200 B.C.) i n several of these regions, cultures became increas-- 15 -ingly s i m i l a r to one another. These cultures comprise the Longshanoid c u l t u r a l complex. The Longshanoid cultures evolved into the Longshan culture i n several regions. Chang (1979:144) characterizes the Longshanoid phase as t r a n s i t i o n a l but basic-a l l y e g a l i t a r i a n and the Longshan phase as a "war-like and ranked society preparatory for the formation of c i v i l i z a t i o n and the state". An (1979-80:45) disagrees with Chang's (1979) c l a s s i f i c a -t i o n of s i t e s into the Longshanoid c u l t u r a l complex on the grounds that each culture developed i n a d i f f e r e n t manner. However, he states that the late N e o l i t h i c cultures from various regions are characterized by much in t e r a c t i o n i n terms of "exchange and influence on the one hand, and integration and u n i f i c a t i o n on the other" ( i b i d ) . Chang (1981b:155) also em- J. phasizes the extent of regional i n t e r a c t i o n during the Long-shanoid phase, saying that c u l t u r a l systems i n various regions evolved side by side with those i n other regions. It i s implied that ranking developed i n more than one region during the Longshanoid phase. Investigators have noted a number of c u l t u r a l c o n t i n u i t i e s i n b u r i a l practices from the Dawenkou Culture to the Shang period and have debated whether the Dawenkou Culture had an influence i n the development of the Shang culture (Thorp 1980: 51, Chang 1980:345-46, 354, Chang 19 83:509-10). These c u l t u r a l t r a i t s include the log tomb, second l e v e l platform, several forms of a r t i f a c t s , and some types of raw material such as jade. The log tomb i s also present i n some Longshan s i t e s (Cheng - 16 -1982:21). H i s t o r i c a l texts also suggest that c u l t u r a l evolution i n the Dawenkou or Qingliangang culture area was related to the development of the state i n China (Fried 1983: 488). However, Zou (personal communication 1983) maintains that Chinese archaeologists now believe the Dawenkou Culture did not d i r e c t l y influence the development of the Shang state. A new culture subsequent to the Longshan i n Shandong has recently been discovered, to which the Dawenkou Culture i s d i r e c t l y related ( i b i d ) . I t i s l i k e l y that new discoveries i n Shandong and elsewhere w i l l continue to modify our understand-ing of s o c i o - c u l t u r a l evolution i n the eastern seaboard region of China. - 17 -CHAPTER 2 METHODOLOGY 2.1. General Approaches i n Mortuary Analysis During the past twenty years, mortuary s i t e s have been analyzed primarily i n order to provide an understanding of the s o c i a l systems represented at these s i t e s . The two major concerns of mortuary analyses have been: 1) status d i f f e r -e n t i a t i o n among bu r i a l s i n a cemetery or i n other types of disposal areas and 2) s o c i a l relationships among bu r i a l s i n terms of kinship groups or r e s i d e n t i a l groups. Some recent mortuary analyses have been concerned with understand-ing the i d e o l o g i c a l component of culture represented by b u r i a l remains. V a r i a b i l i t y i n grave good inclusions, grave form, human s k e l e t a l remains, and i n s p a t i a l patterning of grave goods, grave form, or human remains are the major aspects of mortuary treatment that have been assessed i n studies concerned with either s o c i a l organization or ideology. Human sk e l e t a l remains analyzed by physical anthropologists have enhanced studies of status or s o c i a l subgroups. Results of these studies have indicated a relat i o n s h i p between stature and s o c i a l status (e.g., Haviland 1967, Buikstra 1976), degenerative j o i n t disease and status (Tainter 19 80) and enamel defects and status (Cook 1981). Human s k e l e t a l remains have also provided information on b i o l o g i c a l distance - 18 -(e.g., Buikstra 1976) and paleodemography (Chapman and Randsborg 1981:19-20). The other aspects of material c u l -ture noted above are focused upon i n t h i s study. The primary concern of most mortuary studies has been status d i f f e r e n t i a t i o n . An investigation of s o c i a l subgroups i s included i n only some of these studies. Since the nature of status d i f f e r e n t i a t i o n represented at Dawenkou i s the research problem addressed i n t h i s study, general approaches to analyses of status d i f f e r e n t i a t i o n are emphasized here. However, t h i s chapter includes some discussion of general approaches to analyses of s o c i a l subgroups because, as argued below, an understanding of the nature of status d i f f e r e n t i a t i o n represented at a s i t e i s not complete without consideration of the nature of s o c i a l subgroups represented. There are currently two general approaches to the study of status d i f f e r e n t i a t i o n from mortuary remains i n the archaeological l i t e r a t u r e : 1) the processual approach, which has the goal of reconstructing the s o c i a l organization of the community represented by the b u r i a l s i n a mortuary s i t e and 2) the symbolist approach, which has the goal of understand-ing status d i s t i n c t i o n s i n a society within the broader context of the symbolism expressed by material culture and the ideology of that society. Many of the a r t i c l e s following the symbolist approach have appeared i n the. l i t e r a t u r e since the comprehen-sive review of: archaeological mortuary analysis by Chapman and Randsborg (19 81). - 19 -The m a j o r i t y o f mortuary s t u d i e s i n the a r c h a e o l o g i c a l l i t e r a t u r e f o l l o w the f i r s t approach (e.g. T a i n t e r 1978, 1977a, 1977b, 1975a, 1976, 1973; T a i n t e r and Cordy 1977, Shennan 1975, Wright 1978, Decker 1969, Braun 1977 and 1979, Van de Velde 1979, R o t h s c h i l d 1979, King 1969, G o l d s t e i n 1981 and 1980, Hatch 1976, Peebles 1971, 1972, 1974; Peebles and Kus 1977, and M a i n f o r t 19 77). The proponents of the second approach i n c l u d e Leach (1977), Hodder (1982a and b, 1980), Pader (1982, 1980), Shennan (1982), Pearson (1982), Shanks and T i l l e y (1982), and Blackmore e t a l . (1979). These authors m a i n t a i n t h a t the p r o c e s s u a l approach i s t o t a l l y inadequate f o r understanding s t a t u s d i f f e r e n t i a t i o n (Hodder 1980:161). I m a i n t a i n t h a t elements of both approaches are u s e f u l and t h a t elements of both are p r o b l e m a t i c . The approach of the mortuary a n a l y s i s presented here i s based upon elements from each. 2.2. The P r o c e s s u a l Approach The p r o c e s s u a l approach i s based mainly upon the c r o s s -c u l t u r a l ethnographic t e s t s of Saxe (1970) and B i n f o r d (1971), and to a l e s s e r e x t e n t , from T a i n t e r (1975b, 1973) and G o l d s t e i n (1976). These t e s t s , d e s c r i b e d below, are important because they i n d i c a t e t h a t d i f f e r e n t s o c i a l s t a t u s e s of i n d i v i d u a l s and the g e n e r a l s o c i a l o r g a n i z a t i o n i n the s o c i e t y of the deceased are r e f l e c t e d by mortuary remains ( G o l d s t e i n 1981:54). For the context of t h i s mortuary a n a l y s i s , " s t a t u s " i s d e f i n e d as " r e l a t i v e rank i n a h i e r a r c h y of p r e s t i g e " - 20 -(Webster's Seventh New C o l l e g i a t e D i c t i o n a r y 1969, page 856). One can assume t h a t d i f f e r e n t i a l treatment of an i n d i v i d u a l i n death r e f l e c t s h i s or her d i f f e r e n t i a l treatment i n l i f e ( i b i d ) . Although c o n c l u s i o n s from these t e s t s need to be confirmed from f u r t h e r t e s t i n g with ethnographic and archaeo-l o g i c a l data (Chapman and Randsborg 1981:23; B a r t e l 1982:52), they are s t i l l u s e f u l f o r c u r r e n t mortuary a n a l y s e s . More r e c e n t ethnographic s t u d i e s have i n d i c a t e d t h a t aspects of o r g a n i z a t i o n may not be d i r e c t l y or t o t a l l y expressed by b u r i a l remains. However, the c r o s s - c u l t u r a l r e g u l a r i t i e s d e r i v e d from the t e s t s of Saxe (1970), B i n f o r d (1971), T a i n t e r (19 73, 1975b) and G o l d s t e i n (19 76) p r o v i d e a means by which to assess v a r i a b i l i t y i n mortuary remains. No c r o s s - c u l t u r a l ethnographic t e s t s r e g a r d i n g mortuary r i t u a l have been pub-l i s h e d by r e s e a r c h e r s who advocate the s y m b o l i s t approach. The c o n c l u s i o n s from Saxe (19 70), B i n f o r d (19 71), and T a i n t e r (1973, 1975b) have been u t i l i z e d by r e s e a r c h e r s f o l l o w i n g the processual.. approach to formulate t e s t i m p l i c a t i o n s f o r achieved s o c i a l s t a t u s versus a s c r i b e d s o c i a l s t a t u s , as w e l l as t o i d e n t i f y v a r i o u s s t a t u s l e v e l s i n the s o c i e t y i n which the deceased had l i v e d . S axe 1s (19 70) t e s t based upon three modern s o c i e t i e s of v a r y i n g s o c i a l complexity (the Kapauku Papuans of New Guinea, the A s h a n t i of West A f r i c a and the Bontoc I g o r o t of Luzon, P h i l i p p i n e s ) and B i n f o r d ' s (19 71) t e s t based upon 4 0 non-state o r g a n i z e d s o c i e t i e s from the Human R e l a t i o n s Area F i l e s both i n d i c a t e t h a t mortuary remains r e f l e c t the - 21 -" s o c i a l persona" of the deceased, or the "composite of the s o c i a l i d e n t i t i e s maintained i n l i f e and recognized as appropriate for consideration after death" (Binford 1971:17). During l i f e , the various s o c i a l i d e n t i t i e s of an i n d i v i d u a l are expressed i n d i f f e r e n t s i t u a t i o n s . At the death of an i n d i v i d u a l , the l i v i n g decide which i d e n t i t i e s of the deceased are the most important and should be expressed i n mortuary r i t u a l . According to Saxe (19 70:4-9), the determining factors in t h i s decisions are the rig h t s and duties of the l i v i n g to the deceased. The aspects of the s o c i a l persona that are symbolized as well as the form of the symbols themselves vary from culture to culture (Binford 1971:16-17). Saxe's (19 70) study involved the te s t i n g of eight hypotheses, most of which received p a r t i a l support. The two hypotheses firmly supported are 1) The Components of a Given Disposal Domain Cooperate i n a P a r t i t i o n i n g of the Universe, the Resultant Combinations Representing D i f f e r -ent Social Personae (Saxe 1970:65). 2) In a Given Domain, the P r i n c i p l e s Organizing the Set of Social Personae (Produced by the Cooperative P a r t i t i o n i n g of the Universe of Disposal Components) are Congruent with Those Organizing Social Relations in the Society at Large (Saxe 1970:66). Saxe's (19 70) eighth hypothesis refers to s o c i a l subgroup a f f i l i a t i o n and i s discussed i n Chapter 5. Goldstein's (19 76) ethnographic test, a re-examination of Saxe's eighth hypo-- 22 -thesis, i s also discussed i n Chapter 5. Binford (.1971) tested and received support for three hypotheses. Binford*s (19 71) study i s the most useful ethnographic study for mortuary analyses to date because i t considers a number of aspects of the s o c i a l persona and of mortuary treatment with material correlates. Binford con-cludes from the testing of these hypotheses that ...the form and structure which characterize the mortuary practices of any society are conditioned by the form and complexity of the organizational c h a r a c t e r i s t i c s of the society i t s e l f (Binford 1971:23). The f i r s t hypothesis tested by Binford (1971) i s ...there should be a high degree of isomorphism between a) the complexity of the status struc-ture i n a s o c i o - c u l t u r a l system and b) the complexity of mortuary ceremonialism as regards d i f f e r e n t i a l treatment of persons occupying the d i f f e r e n t status positions (Binford 1971:18). The complexity of the mortuary r i t u a l i s measured by the number of dimensions of the s o c i a l persona i n a society's mortuary practices. These various dimensions are age, sex, s o c i a l p osition or status, sub-group a f f i l i a t i o n , cause of death, and location of death. Due to l i m i t a t i o n s of the data set, complexity of s o c i o - c u l t u r a l system i s measured by form of subsistence (whether hunter-gatherers, s h i f t i n g a g r i c u l t u r a l i s t s , s e t t l e d a g r i c u l t u r a l i s t s , and p a s t o r a l i s t s ) . - 23 -A greater number of d i s t i n c t i o n s of the s o c i a l persona i s found to be symbolized i n the mortuary r i t u a l of s e t t l e d a g r i c u l t u r a l i s t s , generally accepted as having the most complex form of s o c i o - c u l t u r a l system than the other forms of subsistence (Binford 1971:18). Binford's (1971:20) resu l t s are shown i n Figures 2-1 and 2-2. Hodder (19 80:168) c r i t i c i z e s the testing of t h i s proposition on the basis of Binford's measure of s o c i e t a l complexity - which Binford (1971:18) admits i s s i m p l i s t i c . Binford's (1971) second hypothesis i s : We would predict that age and sex should serve more commonly as bases for mortuary d i s t i n c -tion among hunter and gatherers; while among a g r i c u l t u r a l i s t s , s o c i a l p o s i t i o n , as varying independently of age and sex as well as sub-group a f f i l i a t i o n , should more commonly serve as the basis for d i f f e r e n t i a l mortuary t r e a t -ment (Binford 1971:20). Binford (1971:20) concludes from his tabulations (shown i n Figure 2-1) that there i s a marked difference i n the number of cases from the t o t a l in which age and sex are symbolized i n the mortuary r i t u a l of a g r i c u l t u r a l i s t s versus hunters and gatherers, and i n which s o c i a l position i s symbolized. This hypothesis i s the basis of the t e s t implications employed i n processual mortuary studies for achieved versus ascribed s o c i a l p o s i tion. Binford's (19 71) t h i r d hypothesis i s employed i n processual mortuary studies to indicate status d i s t i n c t i o n s - 24 -FIGURE 2-1. Number of dimensional d i s t i n c t i o n s of the s o c i a l persona symbolized i n mortuary practices according to form of subsistence. dimensional hunters & s h i f t i n g s e t t l e d p a s t o r a l i s t s d i s t i n c t i o n s gatherers a g r i c u l t u r a l i s t s a g r i c u l t u r a l i s t s conditions of 1 0 i death X o j_ l o c a t i o n of 1 i 0 0 death X age 2 1 7 1 sex 12 4 10 3 s o c i a l p o s i t i o n 6 5 11 0 s o c i a l a f f i l i a t i o n 4 3 10 1 t o t a l cases 15 8 14 3 (from Binford 1971:20, Table 2) FIGURE 2-2. Average number of dimensional d i s t i n c t i o n s according to subsistence category. subsistence category average number of dimensional d i s t i n c t i o n s per category (1) hunters & gatherers 1.73 (2) s h i f t i n g a g r i c u l t u r a l i s t s 1.75 (3) s e t t l e d a g r i c u l t u r a l i s t s 3.14 (4) p a s t o r a l i s t s 1.66 (from Binford 1971:20, Table 3) - 25 -or levels i n the society of the deceased: ...the locus of mortuary r i t u a l and the degree that the actual performance of the r i t u a l w i l l i n t e r f e r e with the normal a c t i v i t i e s of the community should vary d i r e c t l y with the number of duty status relationships obtaining between the deceased and other members of the community (Binford 1971:21). This hypothesis i s supported and other correlations are found between the various c h a r a c t e r i s t i c s of the s o c i a l persona and d i f f e r e n t types of mortuary treatment. Figure 2-3 indicates the manner i n which age and other aspects of the s o c i a l persona may be distinguished i n mortuary r i t u a l . Social position i s expressed by the greatest variety of aspects of mortuary treatment, and many d i f f e r e n t aspects may symbolize s o c i a l p osition i n one society (Binford 1971:22). Social position tends to be distinguished by the form and quantity of grave goods, p a r t i c u l a r l y " s t a t u s - s p e c i f i c " 'badges' of o f f i c e and quantities of grave goods, and the location of interment (Binford 1971:23). Tainter (1978:121) disagrees with Binford (1971:23) that status i s often r e f l e c t e d by grave goods. Tainter's ethnographic survey of mortuary practices revealed that grave goods r e f l e c t status i n less than f i v e percent of the cases (Tainter 1974:125). Tainter's (1973, 1975b) ethnographic te s t expands upon Binford's (1971) t h i r d hypothesis. Tainter (1973:6) proposes that the amount of community involvement and the degree of - 26 -FIGURE 2-3. Aspects of the s o c i a l persona symbolized i n mortuary treatment. condition l o c a t i o n age sex s o c i a l s o c i a l of death of death posit i o n a f f i l i a t i o n (1) preparation * >^ . . 5 "g (2) treatment (3) d i s p o s i t i o n 2 1 - - 2 2 2 1 3 - 2 1 1 - 1 - 3 1 3 - 9 3 - 7 - 8 15 16 5 - - 9 (4) form J5 > (5) o r i e n t a t i o n oo (6) l o c a t i o n JJ (7) form only 3 5 (8) quantity only ** (9) form and quantity (from Binford 1971:22. Table 4) - 27 -a c t i v i t y d i s r u p t i o n i n mortuary r i t u a l corresponds t o the amount of energy expended i n the mortuary r i t u a l . The h i g h e r the s o c i a l rank of an i n d i v i d u a l , the g r e a t e r p o r t i o n of the community and degree of a c t i v i t y d i s r u p t i o n and the g r e a t e r amount of energy expenditure ( i b i d ) . T h i s h y p o t h e s i s i s i n i t i a l l y t e s t e d from the a r c h a e o l o g i c a l and e t h n o h i s t o r i c data from the Kaloko cemetery, Hawaii ( T a i n t e r 1973) and i n more d e t a i l from a sample of 103 ethnographic s o c i e t i e s ( T a i n t e r 1975b). Energy expenditure i s r e f l e c t e d by s i z e and e l a b o r a t i o n of grave, method of h a n d l i n g and d i s p o s a l of the corpse, and the nature of grave goods ( T a i n t e r 1976:95). I t i s claimed t h a t the hypothesis i s c l e a r l y supported and t h a t one may c o n s i d e r t h a t d i f f e r e n t amounts of energy expenditure i n mortuary r i t u a l correspond t o d i f f e r e n t l e v e l s o r grades of r a n k i n g ( i b i d ) . The c o n c l u s i o n s from T a i n t e r ' s (1973, 1975b) ethnographic t e s t s have been c h a l l e n g e d by Braun (1981) and K i r c h (1980). Braun (19 81:411) p o i n t s out t h a t T a i n t e r d i d not t e s t whether d i f f e r e n c e s i n energy expenditure can occur among i n d i v i d u a l s who do not d i f f e r i n s o c i a l s t a t u s (although T a i n t e r 19 81:419 i n s i s t s t h a t he d i d ) . K i r c h (1980) t e s t s T a i n t e r ' s h y p o t h e s i s l i n k i n g energy expenditure and s o c i a l rank w i t h e t h n o h i s t o r i c and a r c h e o l o g i c a l data from a Tongan s o c i e t y mortuary s i t e . K i r c h (19 80:306) concludes t h a t i n Tongan s o c i e t y , energy expenditure r e f l e c t s r e l a t i v e s o c i o p o l i t i c a l s t a t u s more than s o c i a l rank. K i r c h (1980:304-5) f i n d s t h a t a hi g h rank-i n g c h i e f has a s m a l l e r grave monument than a c h i e f of lower rank. H i s t o r i c a l data i n d i c a t e t h a t a d i r e c t c o r r e l a t i o n - 28 -between s o c i e t a l rank and s o c i o - p o l i t i c a l status was not present i n Tongan society (Kirch 1980:305). Some in d i v i d u a l s of high s o c i e t a l rank did not have much p o l i t i c a l power, while others with a great deal of p o l i t i c a l power were of lower s o c i e t a l rank ( i b i d ) . Recent ethnographic studies have also indicated that status d i f f e r e n t i a t i o n can be inf e r r e d from mortuary remains. O'Shea (1981:49) concludes from his ethnohistorical and archaeological study of three Plains Indian mortuary prac-t i c e s that aspects of s o c i a l ranking known from ethnohistoric data are c l e a r l y observable from the archaeological remains of Plains Indian mortuary practices. In the Plains cemeteries, ranking i s symbolized i n an obvious manner, by various symbols of wealth and increased energy expenditure (O'Shea 19 81:49-50). However, the t o t a l range of mortuary treatment i s not observ-able from the archaeological data. Contrary to Binford (19 71:22), O'Shea (19 81:49) concludes that d i s t i n c t i o n s of s o c i a l subgroup a f f i l i a t i o n are not c l e a r l y observable from archaeological remains of mortuary practices. Chapman and Randsborg (1981:8) mention a recent study which indicates that s o c i a l status i n r u r a l Hungarian cemeteries has been symbolized by form and color of grave markers. Some authors (Ucko 1969, Leach 1977, Orme 1981) deny the potential of archaeological mortuary data for i n d i c a t i n g information about status d i f f e r e n t i a t i o n or any other aspect of s o c i a l organization. Ucko (1969:265-266) states that grave goods may be offered but they may not be placed i n - 29 -graves with the deceased. Also, the quantity and qual i t y of grave goods may not r e f l e c t r e l a t i v e status of indivi d u a l s (Ucko 1969:266-267), nor grave size (Ucko 1969:296-299). However, as Chapman and Randsborg (19 81:8-9) point out, Ucko (1969:270) actually confirms the conclusions of Binford's (1971) and Saxe's (1970) studies by saying: . . . i n the vast majority of cases known ethno-graphically, a culture or society i s not charac-te r i z e d by one type of b u r i a l only, but that, on the contrary, one society w i l l undertake several d i f f e r e n t forms of b u r i a l , and that these forms w i l l often be correlated with the status of the deceased (Ucko 1969:270). Ucko (1969:270) adds that location of graves p a r t i c u l a r l y indicates status of i n d i v i d u a l s . Orme (1981:235) c i t e s a few examples i n which the mortuary r i t u a l i s more a r e f l e c t i o n of s o c i a l status of l i v i n g r e l a t i v e s rather than that of the deceased. As Bartel (1982:47) points out, both Orme (1981) and Ucko (1969) sele c t c e r t a i n ethnographic cases which show that mortuary r i t u a l does not r e f l e c t s o c i a l status. Their studies would be more useful i f a c r o s s - c u l t u r a l sample of mortuary practices had been included ( i b i d ) . The same c r i t i c i s m can be applied to Leach (1977:162). It i s .clear from the ethnographic studies just described that s o c i a l status and other aspects of the s o c i a l persona are not expressed by the same aspect(s) of mortuary treatment i n a l l s o c i a l groups. For example, status may be expressed - 30 -by wealth i n terms of grave goods, energy expenditure, or both. Also, d i f f e r e n t aspects of the s o c i a l persona are considered important i n d i f f e r e n t cultures. One study of modern mortuary practices indicates that cause of death i s expressed more i n mortuary r i t u a l than s o c i a l status, while another study concludes that s o c i a l and r e l i g i o u s a f f i l i a t i o n tends to be expressed i n times of p o l i t i c a l or economic stress (Chapman and Randsborg 19 81:8). Grave goods do not seem to express status i n Danish Viking graves (Randsborg 19 81:112) and i n a large group of b u r i a l s from the Moche Valley representing 3500 years (Donnan et a l . 1978:312). Huntington and Metcalf (1980:1) point out that there i s a tremendous amount of v a r i a t i o n i n mortuary r i t u a l throughout the world. However, the mortuary programs of in d i v i d u a l s o c i e t i e s display a large amount of redundancy. This conclusion has been made by Bartel (19 82:55) on the basis of his 19 73 study of 2 7 ethnographic s o c i e t i e s from the Human Relations Area F i l e s and Tainter (1978:114) on the basis of Saxe's (1970) formal analysis of mortuary treatment of the three s o c i e t i e s mentioned previously. Thus, d i f f e r e n t s o c i a l statuses, for example, symbolized i n mortuary r i t u a l should be i d e n t i f i a b l e by archaeological data due to re p e t i t i o n of certain mortuary treatment(s). For example, high status may be consistently symbolized i n a cemetery by large quantities of a r t i f a c t s and a p a r t i c u l a r form of a r t i f a c t . - 31 -Some processual mortuary studies investigate status d i s t i n c t i o n s on the basis of only wealth i n terms of grave goods (such as Rothschild 1979, Peebles 1972, 1974; Shennan 19 75) or only energy expenditure (e.g. Tainter 19 77a and b, 1975a, Tainter and Cordy 1977, Tainter 1973, Tainter 1976). These studies assume that status i s expressed i n the same manner in a l l s o c i e i t e s . Wealth and energy expenditure as well as other variables Binford i d e n t i f i e s as r e f l e c t i n g status (body treatment, body preparation, body d i s p o s i t i o n , grave location) (1971:22) should be regarded as only p o t e n t i a l status indicators at a mortuary s i t e . This conclusion i s also reached by Whittlesey (19 78:106). The problem with mortuary studies that follow the processual approach i s that the derived status d i s t i n c t i o n s are interpreted as representing the major status le v e l s that existed i n one community at one point i n time (M. Blake, personal communication, 1983). This manner of interpreta-tion ignores the f a c t that most cemeteries contain i n d i v i d u a l s who were interred continuously throughout a c e r t a i n length of time. The i n d i v i d u a l s from each chronological period within a cemetery do not represent one s o c i a l system but any number of s o c i a l systems. This point i s recognized i n the reviews by Chapman and Randsborg (19 81:15) and Braun (1981:409), the ethnohistorical and archaeological study by O'Shea (1981:40), and i n the processual mortuary studies by Doran (1973:150-151), Jones - 32 -(1980:193), Chapman (1977:30), MacDonald (19 80:38-39), Shennan (1975:280), Van de Velde (1979:46), Gruber (1971:64-65), Hodson (1977:403, 1979:25), Greber (1979b:36), and Goldstein (1981:56-57). Goldstein (1981:56-57) points out that i f the time factor i s not considered, d i f f e r e n t perceived rank l e v e l s may be due to changes through time i n mortuary treatment. O'Shea (19 81:45) notes a change through time i n Pawnee mortuary treatment i n the means by which a certa i n s o c i a l d i s t i n c t i o n was symbolized and i n the s o c i a l d i s t i n c -tion i t s e l f . There was a great change i n a l l aspects of Arikara mortuary treatment, e s p e c i a l l y i n the types of grave goods (O'Shea, 1981:48). O'Shea (1981:51) concludes that s i m i l a r changes i n mortuary treatment through time should be expected at any mortuary s i t e . O'Shea (19 81:52) states that his study supports Binford's (1971:23) conclusion that mortuary treatment r e f l e c t s the organization of the society in which the deceased had l i v e d but that the p o s s i b i l i t y that the mortuary treatment changed through time must also be considered. The state of rapid change c h a r a c t e r i s t i c of the Plains s o c i e t i e s i s probably not t y p i c a l of p r e h i s t o r i c s o c i e t i e s , but some change i n s o c i a l organization should be expected ( i b i d ) . Braun (19 81:409) points out that the longer a cemetery i s i n use, the more l i k e l y there w i l l be changes in s o c i a l organization and i n mortuary r i t u a l . Although the mortuary studies c i t e d above acknowledge the time factor, i t i s ignored by many researchers when inter p r e t i n g the status l e v e l s represented at mortuary s i t e s - 33 -w i t h i n each c h r o n o l o g i c a l p e r i o d (such as Shennan 1975, T a i n t e r and Cordy 1977, T a i n t e r 1975a, 1977a and b, 1973; Greber 1979a, Wright 1978, Peebles and Kus 1977, R o t h s c h i l d 1979, Spencer 1982, B i n f o r d 1972, Peebles 1972, 1974, B u i k s t r a 19 76). Some r e s e a r c h e r s t h a t address the problem of the time f a c t o r make unwarranted assumptions about the p o p u l a t i o n u s i n g the cemetery under study. Peebles and Kus (19 77:431) assume t h a t the composition o f a p o p u l a t i o n u s i n g a cemetery ( i n terms of s i z e , age and sex) does not change s i g n i f i c a n t l y through time, so t h a t the age and sex popula-t i o n of the s i t e r e f l e c t s t h a t o f the p o p u l a t i o n through time. Shennan (1975:283) es t i m a t e s the s i z e o f the l i v i n g community from Branc and i t s age and sex d i s t r i b u t i o n a t an average p o i n t i n time, assuming t h a t the p o p u l a t i o n s i z e and age and sex s t r u c t u r e d i d not change. Greber (1979a:43) assumes the i n d i v i d u a l s w i t h i n the Seip mound are contemporaneous. I argue t h a t such assumptions should not be made u n t i l an e f f o r t has been made to determine the c h r o n o l o g i c a l r e l a t i o n s h i p s of b u r i a l s w i t h i n the same broad c h r o n o l o g i c a l p e r i o d . Only a few a r c h a e o l o g i c a l mortuary s t u d i e s have made t h i s attempt. These s t u d i e s have attempted to determine i f d i f f e r e n t s p a t i a l areas o f a cemetery r e p r e s e n t d i f f e r e n t p e r i o d s of time w i t h i n one c h r o n o l o g i c a l p e r i o d ( G o l d s t e i n 1981:66, 1980:122-123; T a i n t e r 1976:102, MacDonald 1980:38-39, Shennan 1975:280, Doran 19 73:150-151, Jones 1980: 193, and Hodson 1979:25). A few mortuary s t u d i e s do c o n s i d e r - 34 -change through time i n the system of status d i f f e r e n t i a t i o n i n a cemetery with more than one chronological period (Braun 1977, Wright 1978, Mainfort 1977, Hatch 1977, Tainter 1977a and b). Pearson (19 81) discusses change through time i n status d i f f e r e n t i a t i o n within the Dawenkou s i t e and within the Dawenkou Culture region as a whole. Even fewer mortuary studies have attempted to explain the c u l t u r a l processes responsible for change within a s i t e or c u l t u r a l region (e.g., Wright 1978, Pearson 1981). The order i n which the b u r i a l s within a cemetery were interred i s one aspect of the processes that affected the archaeological record at a cemetery. Consideration of these processes has been lacking i n processual mortuary studies with the goal of understanding s o c i a l ranking (Chapman and Randsborg 1981:11 and Tainter 1978:109). I maintain that these processes include natural and c u l t u r a l factors that may have created s i t e disturbance and three factors c h a r a c t e r i s t i c of the c u l t u r a l group(s) using the cemetery: 1) the order of interment of b u r i a l s as discussed above, 2) the nature of the s o c i a l subgroup(s) using the cemetery, and 3) the nature of the mortuary r i t u a l i t s e l f - whether the mater-i a l forms chosen to symbolize p a r t i c u l a r aspects of the s o c i a l persona are preserved i n the archaeological record and whether they are c l e a r l y recognizable. The mortuary studies c i t e d previously that follow the processual approach attempt to reconstruct the status d i s t i n c -tions i n one p a r t i c u l a r society. I t i s i m p l i c i t l y assumed - 35 -that the people buried i n the cemetery are from one community. It i s known from ethnographical and archaeological data that there i s great v a r i a b i l i t y i n the association between s e t t l e -ments and cemeteries (Chapman and Randsborg 1981:15). A cemetery may represent one whole community, subsection(s) of one community, subsection(s) of more than one community, or more than one whole community. As O'Shea (1981:40) points out, consideration of the t h i r d c u l t u r e - s p e c i f i c type of process mentioned e a r l i e r has e s p e c i a l l y been lacking i n the l i t e r a t u r e . His study i s a valuable contribution to the understanding of aspects of mortuary r i t u a l that are observable i n the archaeological record. O'Shea (1981:49-50) concludes that although s o c i a l status tends to be c l e a r l y r e f l e c t e d i n the mortuary remains from mortuary r i t u a l , s o c i a l subgroup a f f i l i a t i o n tends to be symbolized by materials which do not preserve well or i n a subtle manner. O'Shea (19 81:52) maintains there are r e g u l a r i t i e s i n the formation processes of mortuary s i t e s , and although there tends to be a d i s t o r t i o n between s o c i a l organization as interpreted from archaeological mortuary data (especially s o c i a l subgroup a f f i l i a t i o n ) and the actual s o c i a l organization of a society, t h i s d i s t o r t i o n i s predictable. Levine's (19 77) study of New Guinean mortuary r i t u a l s also indicates that aspects of the s o c i a l persona are not d i r e c t l y expressed i n mortuary r i t u a l (Chapman and Randsborg 19 81:14). - 36 -2.3. The Symbolist Approach The mortuary studies following the symbolist approach also maintain that the goal of s o c i a l reconstruction i s problematic, as well as the lack of consideration of processes that affected the archaeological record of a cemetery. The symbolist approach i s c l e a r l y expressed i n Hodder (1982a, 1980). Hodder (1980) c a l l s for a t o t a l l y new approach in mortuary analysis than that employed i n processual mortuary analyses. Hodder's (19 82a, 19 80) ethnoarchaeological work among the Nubia i n Sudan demonstrates (l i k e the work of O'Shea 19 81) that s o c i a l organization may not be d i r e c t l y expressed by archaeological mortuary data. There tends to be "a d i s t o r t i o n and structured disjunction" between mortuary patterning and patterning c h a r a c t e r i s t i c of the l i v i n g society ( i b i d ) . The material culture of a society may eithe r repre-sent or misrepresent the s o c i a l organization of the society, depending upon the ideology and symbolic codes of the society (Hodder 1982:210). Thus, the degree to which the material remains from mortuary r i t u a l r e f l e c t the s o c i a l organization of a society depends upon that society's ideology. The mortuary studies following the symbolist approach have made the informative conclusion that depending on the p a r t i c u l a r c u l t u r a l attitudes towards death, status d i s t i n c -tions or other aspects of the s o c i a l persona may or may not be symbolized in the material remains from mortuary r i t u a l . In some Nubian societies, many personal items of the deceased - 37 -are broken and placed on top of the b u r i a l mound instead of inside i t (Hodder 1982a:163). The Lozi i n Zambia do not symbolize high status by material goods (1982a:120). The fear of impurity created by death causes B r i t i s h gypsies to break objects and not place them i n graves (Hodder 1980:167). Thus, the lack of d i f f e r e n t i a t i o n of status or of any other aspect of the s o c i a l persona by mortuary treatment does not indicate that t h i s d i f f e r e n t i a t i o n did not occur i n l i f e (Hodder 1980:166). Unlike O'Shea (1981:52), Hodder (1982a: 207-208) maintains that the nature i n which the archaeological record r e f l e c t s a transformation of s o c i a l organization cannot be predicted, because d i f f e r e n t s o c i e t i e s have d i f f e r e n t ideologies and symbolic codes which continually change with time. However, Hodder concedes that archaeological mortuary data do r e f l e c t some aspects of s o c i a l organization: While b u r i a l behavior may d i s t o r t and invert, i t does not t o t a l l y hide. There w i l l always be some aspect of the s o c i e t a l organization which can be picked out i n the gross cross-c u l t u r a l reviews as being r e f l e c t e d i n b u r i a l (Hodder 1980:168). Hodder's (19 80:168) solution to the problem of deter-mining the degree to which the mortuary data from a s i t e r e f l e c t s the s o c i a l organization of the s o c i a l group(s) using the cemetery i s to i d e n t i f y the ideology of the society with regard to death: - 38 -A new approach to b u r i a l must not expect simple correlations between s o c i a l organization and b u r i a l . Rather, i t must i d e n t i f y the way i n which p r e v a i l i n g attitudes to death can be derived from d i f f e r e n t conceptions of the l i v i n g p r a c t i c a l world ( i b i d ) . Hodder (1982:208-9) points out that status d i f f e r e n t i a t i o n i n any society i s more complex than has been r e a l i z e d i n most mortuary studies. An understanding of s o c i a l ranking requires consideration of attitudes towards domination, power and authority i n a society ( i b i d ) . Hodder (1980:168) c a l l s for new ethnoarchaeological studies which explain the r e l a t i o n -ship between the ideology of s o c i e t i e s and t h e i r mortuary r i t u a l s . A diachronic ethnoarchaeological study of the type advocated by Hodder (19 80:168) has been reported by Pearson (1982). Pearson's (1982) conclusions regarding the mortuary practices of Vi c t o r i a n and modern England are p o t e n t i a l l y applicable to p r e h i s t o r i c mortuary s i t e s . Mortuary r i t u a l may r e f l e c t i d e a l , not actual, r e l a t i o n s of power within a society, and l i v i n g s o c i a l groups may manipulate the statuses of deceased indiv i d u a l s i n order to elevate t h e i r own statuses or for other purposes (Pearson 1982:112). I maintain that u n t i l more i n d i v i d u a l or c r o s s - c u l t u r a l ethnoarchaeological studies of the kind advocated by Hodder (19 80:168) have been made, the only resort for the researcher i s to make the most of the ethnographic and archaeological - 39 -data a v a i l a b l e . A l s o , although s t a t u s r e l a t i o n s h i p s are most l i k e l y to be more complex i n l i f e than t h a t apparent from mortuary treatment, i t has been e s t a b l i s h e d t h a t mortu-ary remains do tend t o r e f l e c t s t a t u s r e l a t i o n s h i p s a t l e a s t i n p a r t . The k i n d of ethnographic study advocated by Hodder may be as c o n t r o v e r s i a l as those p r e s e n t l y employed i n mortuary analyses because informants' accounts of mortuary p r a c t i c e s tend to be u n r e p r e s e n t a t i v e of the t o t a l range of v a r i a b i l i t y i n the mortuary p r a c t i c e s of a s o c i e t y (O'Shea 1981:43). Hodder (1982a, 1980) does not p r o v i d e a method which can i d e n t i f y c u l t u r a l a t t i t u d e s towards death on the b a s i s of a r c h a e o l o g i c a l data. Leach (1977:169) a l s o maintains t h a t c u l t u r a l i d e o l o g y r e g a r d i n g death should be assessed and does not p r o v i d e a method. Other mortuary s t u d i e s of the s y m b o l i s t approach p r o v i d e methods w i t h which one can attempt to determine symbolic codes and c u l t u r a l i d e o l o g y r e g a r d i n g death with a r c h a e o l o g i c a l data. However, these methods are not based upon c r o s s - c u l t u r a l r e g u l a r -i t i e s i n the i d e o l o g y of death. A l s o , some of these s t u d i e s do not r e c o g n i z e t h a t mortuary s i t e s r e f l e c t a number of changing s o c i a l systems through time. Pader (19 82, 19 80) maintains t h a t the s p a t i a l p a t t e r n i n g of a r t i f a c t s w i t h i n graves, p a r t i c u l a r l y items of dress or b o d i l y adornment, symbolizes s o c i a l r e l a t i o n s w i t h i n a s o c i e t y . A summary of items of dress or b o d i l y adornment as symbols i n the mortuary r i t u a l of ethnographic s o c i e t i e s i s i n c l u d e d i n Pader (1982). Pader (1982,1980) s t u d i e s the - 40 -v a r i a b i l i t y i n the above type of patterning with regard to age, sex, s k e l e t a l p o s i t i o n , and s p a t i a l location of graves within two Anglo Saxon cemeteries. Her study i s synchronic. Pader maintains that a study of s o c i a l ranking based upon the r e l a t i v e richness of b u r i a l s alone i s limited (Pader 1982:54, 1980:143). A study of ranking which ignores symbolism r e f l e c t e d by the s p a t i a l component (either within graves or between graves) ignores the c u l t u r a l context of any derived status d i s t i n c t i o n s (Pader 1980:170). The c u l t u r a l context i s also indicated by v a r i a b i l i t y i n the other aspects of mortuary treatment at a s i t e ( i b i d ) . Pader (1980:156) concludes that the male and female sexes within the two cemeteries are symbolized i n d i f f e r e n t manners, so that a comparison of the quantity and qu a l i t y of a r t i f a c t types between the two sexes would be misleading. Like Pader (1982, 1980), Shanks and T i l l e y (1982:152) maintain that neglect of the symbolism r e f l e c t e d by mortuary treatment neglects the s o c i a l context of derived status d i s t i n c t i o n s . They assess the treatment and arrangements of bones from skeletons i n European N e o l i t h i c barrows. Com-parison of the archaeological patterning with ethnographic data suggests that d i f f e r e n t i a l access to power by various s o c i a l subgroups i s being symbolized (Shanks and T i l l e y 1982:151). An aspect of mortuary treatment regarded i n some mortuary analyses following either the symbolist or processual - 41 -approach as e s p e c i a l l y important for understanding s o c i a l context i s s p a t i a l l o cation of grave. The s p a t i a l component i s considered important by Pader (1982, 1980 - within graves as well as between graves) and p a r t i c u l a r l y i n the processual mortuary study described i n Goldstein (1980, 1981 - between graves only). Chapman and Randsborg (1981:14) state that s p a t i a l patterning i s another neglected topic i n mortuary analysis. Goldstein (1980:9-10) maintains that s p a t i a l location of graves within a cemetery provides information on the nature of derived status d i s t i n c t i o n s and on how these d i s t i n c t i o n s operated within the society i n which the deceased had l i v e d . Like Pader (1982:170), Goldstein (1981: 56) asserts that a study of s o c i a l ranking based exclusively upon a r t i f a c t s i s incomplete and disregards the s o c i a l context in which the derived status d i s t i n c t i o n s should be considered: What does each group or status type mean? How do the groups re l a t e to each other? What are the functions of each group, and what are the functional relationships between groups? While many of these questions may not be e a s i l y or r e l i a b l y answered, current mortuary analysis does not even approach or attempt to ask these questions. Can we r e a l l y say... that a culture in which we have determined seven s o c i a l group-ings i s more complex than one i n which we fi n d s i x groupings? Some mortuary analyses following either the symbolist or processual approach examine mortuary s i t e s i n terms of - 42 -the regional s o c i a l context as well as the l o c a l s o c i a l context. These symbolic studies include Shennan (19 82), Hodder (1982b) and Blackmore et .al. (1979). Processual mortuary studies with th i s approach are Goldstein (1980, 1981), Peebles (1974), and Seeman (1979). The comparative study by Pearson (19 81) has been mentioned. The regional s o c i a l context i s yet another neglected aspect of mortuary analysis (Chapman and Randsborg 1981:23). Goldstein (1980, 1981) and Peebles (19 74) apply the same techniques that c i d e n t i f y d i f f e r e n t status l e v e l s to a number of s i t e s i n the s p e c i f i c c u l t u r a l region under study. Seeman (19 79) provides an illuminating interpretation of the r i t u a l and economic function of mortuary s i t e s within the Hopewell culture region. Shennan (1982:160) suggests that d i f f e r e n t ideologies in d i f f e r e n t regions of the European Bronze Age resulted in the development of d i f f e r e n t types of ranking systems from region to region. Hodder (19 82b:175) maintains that changing design of ceramics and of other types of material culture r e f l e c t changing ideologies regarding regional expression of power and domination. In another study of regional ceramic design, Blackmore et a l . (19 79:108) suggest that d i s t i n c t i v e l y d i f f e r e n t designs from b u r i a l and s e t t l e -ment s i t e s symbolize competing ethnic groups. - 43 -2.4. The Approach Followed i n th i s Mortuary Analysis This analysis of status d i f f e r e n t i a t i o n u t i l i z e s test implications for achieved versus ascribed status derived from the cro s s - c u l t u r a l r e g u l a r i t i e s concluded by Saxe (1970), Binford (1971), and Tainter (1973, 1975b). The p a r t i c u l a r techniques u t i l i z e d to i d e n t i f y the various status le v e l s or d i s t i n c t i o n s represented among the Dawenkou b u r i a l s are discussed i n the following section. V a r i a b i l i t y i n as many aspects of mortuary treatment that have been considered as p o t e n t i a l l y status-related i s assessed. Variables that r e f l e c t wealth (such as qual i t y and quantity of grave goods) and energy expenditure (grave form and size) are included. Other variables included are body di s p o s i t i o n and s p a t i a l location of grave (see Figure 2-3). The Dawenkou s i t e report does not provide a description of body treatment or body preparation. My exploratory study of status d i f f e r e n t i a t i o n at Dawenkou conducted i n 19 81 suggested that both wealth and energy expenditure symbolize status d i s t i n c t i o n s at Dawenkou. There i s considerable v a r i a t i o n in quantity of grave goods (from none to over 100 items), qua l i t y of grave goods (from u t i l i t a r i a n stone tools and ceramic vessel types to jade ornaments and items of carved i v o r y ) , and grave size i n area (from less than 5.0m2 to 13.Om^), and grave form (from simple p i t s to log tombs). The status d i s t i n c t i o n s derived i n the analysis of - 44 -status are interpreted with consideration of the fact that the Dawenkou cemetery represents a number of s o c i a l systems through time. I t i s es p e c i a l l y important to consider the time factor for Dawenkou because as mentioned i n Chapter 1, the cemetery may have been used for 1000 years (Wu 1982). The number of derived status l e v e l s i s not interpreted as being the exact number of status l e v e l s which existed i n the society of the deceased. Instead, emphasis i s placed upon d i s t i n c t changes in the number and nature of status d i s t i n c -tions i n the cemetery through time. The exact number of status l e v e l s at one point i n time i s also not interpreted l i t e r a l l y due to the probable d i s t o r t i o n of status and of other aspects of the s o c i a l persona as re f l e c t e d i n the cemetery. Some researchers have attempted to determine the degree of ranking represented i n a cemetery at one point i n time (Peebles and Kus 1977, Buikstra 1976, Tainter 1977a, 1977b, 1978; Tainter and Cordy 1977, Shephard 1979, Hatch 1976, Brown 1971). Buikstra (1976:35-37) describes the degree of ranking i n Middle Woodland s i t e s i n terms of Fried's (1967) "ranked" and " s t r a t i f i e d " s o c i e t i e s . Peebles and Kus (1977), Hatch (19 76), and Brown (1971:102) attempt to determine whether a chiefdom i n Service's (19 75) sense i s represented at a s i t e . Some researchers have c r i t i c i z e d the character-i z a t i o n of s i t e s i n terms of evolutionary typologies (Hodder 1982a:208, Goldstein 1981:54, and Tainter 1978:115-117). In these studies i t i s argued that s i t e s are pigeon-- 45 -h o l e d i n t o e v o l u t i o n a r y types on the b a s i s o f a few t r a i t s and t h a t such c l a s s i f i c a t i o n i s d e s c r i p t i o n i n s t e a d o f e x p l a n a t i o n . T a i n t e r (1977a, 1977b, 1978), T a i n t e r and Cordy (1977), and Shephard (1979) g i v e a c a l c u l a t i o n of the amount and degree of o r g a n i z a t i o n . These c a l c u l a t i o n s have been c r i t i c i z e d by B i n f o r d (personal communication, 1983), Braus (1981:408-412) and G o l d s t e i n (1981:55). The measurements of the amount and degree of o r g a n i z a t i o n assume t h a t s t a t u s r e p r e s e n t e d i n every cemetery i s r e f l e c t e d by energy expenditure (see T a i n t e r 1978:134). A l s o , by i n c l u d i n g i n the c a l c u l a t i o n s the number of i n d i v i d u a l s i n every rank l e v e l ( i b i d ) , i t i s assumed t h a t the cemetery d i r e c t l y r e f l e c t s one e n t i r e s o c i a l system. I argue i t i s more f e a s i b l e w i t h a r c h a e o l o g i c a l data to i d e n t i f y degree of o r g a n i z a t i o n i n a r e l a t i v e i n s t e a d of an a b s o l u t e manner. T e s t i m p l i c a t i o n s f o r d i f f e r e n t degrees of ran k i n g d e r i v e d from ethnographic data have not been developed. Whether there i s a change through time i n the degree of s t a t u s d i f f e r e n t i a t i o n a t Dawenkou i s assessed i n t h i s study. A q u a l i t a t i v e change i s concluded i f there i s a change from achieved t o a s c r i b e d s t a t u s , and a q u a n t i t a t i v e change i s concluded i f there i s a c l e a r d i f f e r e n c e i n the number of s t a t u s d i s t i n c t i o n s through time. My e x p l o r a t o r y assessment of s t a t u s d i f f e r e n t i a t i o n conducted i n 19 81 suggested t h a t there i s a g r e a t e r number of high s t a t u s b u r i a l s i n the Late p e r i o d (as d e f i n e d by the authors of the s i t e r e p o r t ) - 4 6 -i n terms of grave goods and energy expenditure, located i n more diverse s p a t i a l areas of the cemetery. In the f i n a l chapter of t h i s study, there i s a discussion of the nature of status d i f f e r e n t i a t i o n represented at Dawenkou in comparison to other s i t e s i n the Dawenkou.Culture. Cultural processes such as those suggested by Pearson ( 1 9 8 1 ) that may have been responsible for change i n status d i f f e r e n t i a t i o n through time are included. The analysis of status i s conducted i n Chapter 6. Three preliminary analyses are described i n Chapters 3 , 4 and 5 . The f i r s t , i n Chapter 3 , i s a chronological analysis of the b u r i a l s . As mentioned i n Chapter 1 , there i s a disagreement between the chronological periods given i n the s i t e report (Early, Middle, Late) and reported i n the recent Chinese archaeological l i t e r a t u r e (Early, Late). The chronological relationships among b u r i a l s must be understood before the analysis of status i s undertaken; otherwise, any concluded v a r i a b i l i t y i n status d i s t i n c t i o n s may be a factor of the temporal dimension. The methods that were employed to derive the two chronological schemes are discussed and an e f f o r t i s made to evaluate the two schemes. The second preliminary analysis, i n Chapter 4, i s an attempt to i d e n t i f y the sex of unsexed b u r i a l s on the basis of the types of grave goods i n known sexed b u r i a l s . For an unknown reason, only 3 0 of the 1 2 7 single b u r i a l s i n the cemetery are sexed. Whether status d i f f e r e n t i a t i o n cross-cuts age and sex categories or whether i t i s on the basis - 47 -of age and sex categories i s a key test implication that has been employed i n processual mortuary analyses to indicate ascribed versus achieved status d i s t i n c t i o n s . An e f f o r t must be made to estimate the sex of the remaining b u r i a l s i n order to adequately assess whether status d i s t i n c t i o n s are ascribed (indicative of a ranking system) as Chinese and western archaeologists have proposed for Dawenkou. Another method by which female b u r i a l s could be d i s -tinguished from male b u r i a l s i s i d e n t i f i c a t i o n of the bur i a l s with tooth extraction and s k u l l deformation noted by Pearson (in press:26). Female b u r i a l s could also be i d e n t i f i e d by determination of skeletons that had clay b a l l s i n t h e i r mouths during t h e i r l i f e t i m e s . Pearson (19 81:1084) notes that female skeletons from Dawenkou Culture s i t e s tend to have the clay b a l l more than male skeletons. Unfortunately, information regarding tooth extraction, s k u l l deformation, and the presence of clay b a l l s for each b u r i a l i s not provided i n the s i t e report. The t h i r d analysis, i n Chapter 5, i s the assessment of s o c i a l subgroup a f f i l i a t i o n represented i n the cemetery. An attempt i s made to understand the s o c i a l context i n which status d i s t i n c t i o n s were a part. Consideration of the number of portion(s) of communities represented at the cemetery i s included. Some aspects of mortuary treatment at Dawenkou may r e f l e c t s o c i a l a f f i l i a t i o n as well as s o c i a l p osition (see Figure 2-3): body d i s p o s i t i o n , grave form, s p a t i a l - 48 -location of grave. V a r i a b i l i t y i n these aspects of mortuary treatment i s assessed i n both Chapters 5 and 6. The s p a t i a l location of graves within the cemetery i s regarded as es p e c i a l l y useful for understanding the nature of s o c i a l subgroup a f f i l i a t i o n . The s p a t i a l patterning of a r t i f a c t s and skeletons within graves i s not assessed here due to time constraints. Study of th i s patterning at Dawenkou i s l i k e l y to be informative. There appear to be some regular-i t i e s i n the placement of certain a r t i f a c t types i n r e l a t i o n to various body parts (Pearson 1981:1080). 2.5. Method to Determine Status Levels Despite the large number of mortuary analyses i n the l i t e r a t u r e that follow the processual approach, a l i m i t e d number of techniques have been employed to determine the status d i s t i n c t i o n s or l e v e l s represented i n cemeteries. These methods have included: i n t u i t i v e assessment based upon the quantity and qua l i t y of a r t i f a c t s (Whalen 1983, Buikstra 1976, Larson 1971, Gruber 1971, Alekshin 1983, Jacobsen and Cullen 1981, Binford 1972, Greber 1979b, Milisauskas 1978), i n t u i t i v e assessment and various significance tests (Rathje 1970, King 1969, Peebles 1971, Saxe 1971), the c o e f f i c i e n t of v a r i a t i o n , a measure of r e l a t i v e v a r i a b i l i t y (Pearson 1981), formal analysis (Brown 1971, Decker 1969), ranking of b u r i a l s or the assignment of wealth scores on the basis of quantity and q u a l i t y of grave goods (Shennan 1975, Shephard - 49 -1979, Rathje 1973, Greber 1979a, Winters 1968, Blackmore et a l . (19 79), and multivariate s t a t i s t i c a l techniques (monothetic - d i v i s i v e or pol y t h e t i c , in an R or Q mode). The monothetic - d i v i s i v e techniques have included c l u s t e r analysis with the information s t a t i s t i c or with the sum of chi-square measure of association (Tainter 1975a, Goldstein 1980, 1981; Peebles 1972, 1974; Jones 1980, Mainfort 1977, Hatch 1976). Cluster analysis i s the poly-t h e t i c method which has been employed to the greatest extent. The polythetic techniques have included Average Linkage cl u s t e r analysis (Shennan 1975, Tainter 1975a, Rothschild 19 79), Complete Linkage c l u s t e r analysis (Tainter 19 75a, Mainfort 1977, Hatch 1976), Single Linkage c l u s t e r analysis (Mainfort 1977, Hodson 1977), c l u s t e r analysis by Ward's Error Sum of Squares Method (Peebles 1974, 1972; Hatch 1976), cl u s t e r analysis by Program Mode (Peebles 19 72), factor analysis (Tainter 1975a, Bayard 1983), p r i n c i p a l components analysis (Braun 1977, 1979; Van de Velde 1979, O'Shea 1981), and c a l c u l a t i o n of the amount and degree of organization by the information s t a t i s t i c (Tainter 1977a and b, 1978; Tainter and Cordy 19 77, Shephard 19 79). For a large data set such as Dawenkou, multivariate s t a t i s t i c a l techniques are more appropriate than i n t u i t i v e assessment, formal analysis, or assignment of wealth scores. These techniques can group b u r i a l s that co-vary i n terms of the numerous variables considered as p o t e n t i a l l y status-related. Orton and Hodson (19 81) point out the s t a t i s t i c a l - 50 -d i f f i c u l t i e s i n employing wealth score measures. The tech-niques chosen for the analysis of status i n Chapter 6 are three types of polythetic c l u s t e r analysis techniques (Average Linkage, Complete Linkage and Ward's Method) and multidimensional s c a l i n g . The above techniques are described in d e t a i l i n Chapter 3. These techniques are employed i n a Q mode to group b u r i a l s of similar status. Most mortuary analyses i n the l i t e r a t u r e are conducted i n a Q mode, but some (e.g. Braun 19 79) have found i t useful to employ an R mode analysis to i d e n t i f y co-varying sets of a r t i f a c t types or grave forms. Factor analysis or p r i n c i p a l components analysis i s not suitable for the Dawenkou data set because the p a r t i c u l a r variables regarded as p o t e n t i a l l y status-related had to be coded on a presence/absence basis. The variables do not occur to a great enough extent among the b u r i a l s to allow the use of frequencies. Polythetic cluster analysis techniques are chosen for the analysis of status because they have been considered useful for mortuary s i t e s and for other subjects of archaeo-l o g i c a l research. In some mortuary analyses, i t i s asserted that monothetic - d i v i s i v e techniques are better able to c l a s s i f y b u r i a l s or a r t i f a c t types than polythetic techniques (Tainter 19 75a, Goldstein 19 80, 19 81; Jones 19 80, Shephard 1979). The opinion i n Tainter (1975a) i s based upon a com-parison of monothetic - d i v i s i v e c l u s t e r i n g techniques with - 51 -polythetic techniques on the same data set. Shephard (1979: 62) states that he compared several other types of clustering methods with the monothetic - d i v i s i v e methods, but he does not name these types. Tainter's (1975a) comparison of techniques has been thoroughly c r i t i c i z e d and considered inconclusive by Braun (1981:405). The opinion i n Goldstein (1980:48, 1981:62-63) i s based upon Tainter's (1975a) faulty comparison of techniques and Peebles' (1972, 19 74) use of monothetic - d i v i s i v e methods for the Moundville s i t e . However, Peebles (19 72, 19 74) maintains that both the mono-the t i c - d i v i s i v e and polythetic types of cl u s t e r analysis yielded adequate r e s u l t s . Peebles (1972:10) states that the res u l t s from Ward's Method and Program Mode f u l f i l l e d a p r i o r i expectations, but some of the v a r i a b i l i t y i n the data set was not as apparent as i n the results from the monothetic d i v i s i v e analysis. Peebles (19 74:167-16 8) concludes that "each of the analyses captured s l i g h t l y d i f f e r e n t aspects of the same sets of behavior" and that "each was valuable for the s l i g h t difference i n t h e i r r e s u l t s " . Comparisons of monothetic - d i v i s i v e and po l y t h e t i c c l u s t e r i n g techniques have also been made by Mainfort (1977) and Hatch (1976). Mainfort (1977:78, 97) concludes that a polythetic c l u s t e r i n g method (Single Linkage) yielded superior r e s u l t s , but as i n Peebles (19 74), both types of methods emphasized d i f f e r e n t aspects of the v a r i a -b i l i t y i n the data set. Hatch (19 76) employes a monothetic -- 52 -d i v i s i v e method along with Ward's Method and Complete Linkage but does not comment on the effectiveness of one method over the other. The re s u l t s from the studies above do not indicate that polythetic c l u s t e r i n g techniques have no u t i l i t y for mortuary analysis i n comparison to monothetic - d i v i s i v e techniques. Polythetic methods df c l a s s i f i c a t i o n are regarded by Braun (1981:405) as more appropriate than monothetic - d i v i s i v e techniques for i s o l a t i n g redundantly symbolized status d i s -t i n c t i o n s i n a data set. Also, polythetic methods do not share the l i m i t a t i o n of the monothetic - d i v i s i v e methods, that define a given group on the basis of a unique set of features. Serious errors i n c l a s s i f i c a t i o n r e s u l t when an item that does not possess the one feature used to make a primary d i v i s i o n i s moved to a d i s t a n t l y related grouping, even i f the item i s related to the items of another grouping in every other feature (Sneath and Sokal 19 73:13). Three types of polythetic c l u s t e r i n g techniques (Average Linkage, Complete Linkage and Ward's Method) are employed so that the patterning interpreted r e f l e c t s the v a r i a b i l i t y i n the data set and i s not a r e f l e c t i o n of the mathematical c h a r a c t e r i s t i c s of the p a r t i c u l a r techniques used (Matson and True 1974:72, Braun 1981:406). Patterning which i s i d e n t i f i e d by more than one technique i s regarded as c h a r a c t e r i s t i c of the data set. Interpretations from mortuary studies based upon only one type of multivariate - 53 -c l a s s i f i c a t i o n technique should be regarded as tentative u n t i l they are substantiated by another type of c l a s s i f i c a t i o n technique (e.g. Rothschild 1979, Goldstein 1980, 1981). Peebles and Kus (1977:43 8) and Shennan (19 75:283) employ more than one type of cl u s t e r i n g technique but do not specify the types. To my knowledge, no mortuary analysis i n the l i t e r a t u r e has u t i l i z e d multidimensional scaling as a technique to i d e n t i f y status d i s t i n c t i o n s . Rothschild (1979:6 72) mentions she w i l l use i t i n a future study. Multidimensional scaling has been regarded i n the archaeological l i t e r a t u r e as capable of i n d i c a t i n g important re l a t i o n s between items i n a data set and as a comparable technique with c l u s t e r analysis (Johnson 1972). The res u l t s from the scaling and c l u s t e r i n g techniques i n Chapter 6 are compared in order to inte r p r e t the important status d i s t i n c t i o n s within the Dawenkou cemetery. The use of c l u s t e r i n g and scaling to i d e n t i f y status d i s t i n c -tions has poten t i a l u t i l i t y for other mortuary s i t e s (Binford, personal communication 19 83). In Chapter 6, i t i s demonstrated that the p a r t i c u l a r multivariate s t a t i s t i c a l techniques employed a f f e c t the status d i s t i n c t i o n s among the b u r i a l s which r e s u l t . Also, in t e r p r e t a t i o n of cl u s t e r i n g and scaling r e s u l t s involves a degree of s u b j e c t i v i t y on the part of the researcher. This i s recognized by many researchers and i s discussed i n Chapter 3. Therefore, there i s a methodological reason as - 54 -well as a t h e o r e t i c a l one for not interpreting the resultant status d i s t i n c t i o n s as the exact d i s t i n c t i o n s which existed in the society of the deceased. 2.6. Consideration of Natural and Cultural Factors that may have Affected the Archaeological Record at Dawenkou A natural factor that may have affected the archaeological record at Dawenkou i s changing of the course of the Dawen River immediately adjacent to the cemetery, and subsequent flooding of a portion of the s i t e . The r i v e r i s shown i n photo plate 1 of the s i t e report. I t i s not clear whether any flooding has taken place since the interment of the b u r i a l s during the Early and Late periods. I t i s possible that a portion of the b u r i a l s o r i g i n a l l y l a i d during the Early and Late periods was not discovered by the excavators of the s i t e i n 1959, due to flooding or some other factor. Additional graves may have been discovered since 1959, too. A recent a r t i c l e on the Chinese N e o l i t h i c period by An (19 82) implies that more b u r i a l s have been found at the Dawenkou cemetery. The English t r a n s l a t i o n of An's a r t i c l e by the editors of China Reconstructs states: "So far some 1500 gravesites have been uncovered at Dawenkou" (An 1982: 62). Zou Heng (personal communication, 1983, interprets t h i s unclear t r a n s l a t i o n to mean that 1500 graves have been found at the Dawenkou s i t e . I believe I am aware of the greater portion of the Chinese archaeological l i t e r a t u r e - 55 -available to the West on the Dawenkou cemetery. I have not found any information on additional b u r i a l s discovered i n the cemetery except for the b u r i a l s excavated i n 19 74 that belong to a time period e a r l i e r than those from the 1959 excavation mentioned i n Chapter 1. Thus, I have not been able to determine whether any of the 1500 additional b u r i a l s (provided Zou's inte r p r e t a t i o n i s correct) are contemporary with those from the Early and Late periods of the 1959 excavation. If some or a l l of the 1500 b u r i a l s at Dawenkou are contemporary with the 133 included i n t h i s mortuary analysis, the 13 3 b u r i a l s may not be completely representative of the b u r i a l population as a whole. It i s not clear whether disturbance of the 13 3 b u r i a l s by l a t e r occupations occurred, e i t h e r . I t appears that the Longshan and Shang periods are represented only by a few pottery shards and vessels (shown i n photo plates 110 and 111 of the s i t e report). I t does not seem that the use of the s i t e during these two periods affected the 13 3 b u r i a l s included i n t h i s mortuary analysis. However, intensive land use may have disturbed the archaeological record at Dawenkou. Five b u r i a l s are noted i n the s i t e report as being disturbed. These b u r i a l s are L3, E27, X39, L46, and L77. The time period(s) i n which these b u r i a l s were disturbed as well as whether both natural and c u l t u r a l factors are responsible are not known. Only the extent of the d i s t u r -bance i n L77 i s described as major. I t i s believed that - 56 -th i s b u r i a l o r i g i n a l l y contained a great number'.and'variety of grave goods (Shandong P r o v i n c i a l Cultural Properties Commission 19 74:14 7). The b u r i a l s described as r i c h i n grave goods i n the s i t e report are not said to be disturbed. A few b u r i a l s are described as having incomplete skeletons; lacking a head or some other body part. I t appears that the disturbance of these b u r i a l s was lim i t e d to the skeletons and did not a f f e c t the grave goods i n association. The s p e c i f i c excavation methods employed in 1959 at Dawenkou are not explained i n the s i t e report and cannot be assessed for t h e i r r e l i a b i l i t y . I t i s not known whether any problems i n the r e l i a b i l i t y of reporting e x i s t , e i t h e r . The publication of the Dawenkou s i t e monograph was interrupted during the recent c u l t u r a l revolution i n China (Goodrich 19 83: 16-17). F i n a l l y , the photographs of graves, skeletons and a r t i f a c t s i n the s i t e report indicate that conditions of preservation at Dawenkou are good and that on the whole, the cemetery as excavated i n 1959 i s an accurate r e f l e c t i o n of the material remains r e s u l t i n g from mortuary r i t u a l during the Early and Late periods. Although the t o t a l e f f e c t s of natural and c u l t u r a l factors that may have affected the archaeological record at Dawenkou are not well understood, a mortuary analysis of the 133 b u r i a l s excavated i n 1959 i s considered worthwhile. The 1959 b u r i a l s are better reported than those excavated i n 1974, and they have been considered by western and Chinese - 5 7 -archaeologists as c r i t i c a l for understanding the development of ranking in the eastern seaboard region. A mortuary analysis of the 1 3 3 b u r i a l s u t i l i z i n g current archaeological methods should be able to contribute to a more complete understanding of the nature of status d i f f e r e n t i a t i o n at the s i t e and i n the Dawenkou Culture region as a whole. The complete description of each b u r i a l in the s i t e report allows such an analysis to be made. The 1 3 3 b u r i a l s described i n the 1 9 7 4 s i t e report may be the most complete source of information on the Dawenkou s i t e for many years. The problem of uncertainty over the representativeness of an excavated s i t e i s not new to East or Southeast Asia (e.g. MacDonald 1 9 8 0 : 3 2 on the Bang s i t e i n Thailand and a l l other s i t e s i n Southeast Asia). As MacDonald ( 1 9 8 0 : 3 2 ) points out, as long as the researcher r e a l i z e s the re s u l t s of his or her study are tentative and subject to new interpretation as more data becomes available, study of s i t e s with the above problem i s worthwhile. - 58 -CHAPTER 3 CHRONOLOGICAL ANALYSIS 3.1. Introduction In t h i s chapter, an attempt i s made to evaluate the two c h r o n o l o g i c a l . sequences that have been suggested for the Dawenkou b u r i a l s . Without such an evaluation, i t would be necessary to regard a l l of the bu r i a l s as belonging to the same period. Thus, i t would not be possible to deter-mine whether the temporal dimension affects any of the status-related v a r i a b i l i t y interpreted from the multidimen-sional scaling and cl u s t e r analyses. If the bu r i a l s could be assigned to two or three chonological periods, changes i n status d i f f e r e n t i a t i o n through time could be assessed. Before discussing the methods I believe were employed to derive the two chronological schemes, i t i s necessary to give a b r i e f description of the ceramic c l a s s i f i c a t i o n system upon which they are based. The ceramic vessels i n the 1974 s i t e report are c l a s s i f i e d into 17 functional types, with a varying number of associated subtypes. The subtypes are subdivided into s t y l e s . The h i e r a r c h i c a l r e l a t i o n s h i p of the functional types, sub-types, and styles can be seen i n Figure A3-1. The majority of the functional types, subtypes and styles are described and i l l u s t r a t e d i n Chapter 5 of the s i t e report. Represent-- 59 -ative photographs are provided elsewhere i n the report. The descriptions, drawings and photographs suggest that the func-t i o n a l types are distinguished by morphology and infe r r e d function. Most of these terms are s t i l l i n use. Chinese archaeologists have used the functional types to describe both ceramic and bronze vessel forms from a variety of time periods. The terms for the functional types of pottery vessels used today by Chinese archaeologists were apparently derived from catalogues of ancient bronze vessels compiled during the Song Dynasty (A.D. 960-1279) (Chang 1981a: 158-159), Song antiquarians used the terms for ancient bronze vessels given i n c l a s s i c h i s t o r i c a l texts or from i n s c r i p t i o n s on the bronzes (Chang 1980:23). The use of the ancient terms has created confusion i n the meaning of some vessel types (Chang 1981a:158,161) . The terms: from the c l a s s i c texts are not always c l e a r , and the terms from the i n s c r i p t i o n s are of varying le v e l s of inclusiveness (Chang 1980;23). Chinese archaeologists currently c l a s s i f y bronze vessels into general types according to t h e i r presumed functions and describe them by the t r a d i t i o n a l terms whenever possible (Chang 1980:24). It seems that ceramic vessels are c l a s s i f i e d i n a s i m i l a r manner. The subtypes from Dawenkou appear to be distinguished by morphology. The term for a subtype describes a prominent physical feature of a functional type. The subtypes are not - 60 -always mutually exclusive. For example, Figure A3-1 shows that the c a i ("painted") subtype for many of the functional types i s not distinguished by morphology. The vessels i n t h i s category could have any number of shapes. Another example from Figure A3-1 i s the subtype wu b i ("no lugs") for the hu functional type ("storage vessel"). A vessel of t h i s subtype could have any number of the physical features described for the other subtypes of storage vessel. Another po t e n t i a l problem i s that the q i ta ("other") category contains a great variety of vessel forms. The st y l e numbers indicate f i n e r physical features of the subtypes. However, the difference between some of the styles i s not obvious from the photographs. The functional types, subtypes and styles as described here appear to form a standard c l a s s i f i c a t i o n system used by Chinese archaeologists for other Dawenkou Culture s i t e s , and perhaps for Ne o l i t h i c s i t e s i n general (e.g., The Shandong Archaelogical Team 1964, The Nanjing Museum 1964, and the Shandong P r o v i n c i a l Museum 1972) . Four ceramic ware types are i d e n t i f i e d i n the report, but these types are not included i n the c l a s s i f i c a t i o n system described above. These types are: red ( j i a sha or "sand tempered" and n i or n i z h i , or " f i n e l y levigated clay") (C. Shangraw, personal communication, 1933), grey (sand tempered or of fine c l a y ) , black, and white (sand tempered or of fine c l a y ) . The significance of these ware types i n terms of - 61 -status d i f f e r e n t i a t i o n i s discussed i n Chapter 6. 3.2. The Methods to Derive the Chronological Periods The three periods defined by the authors of the Dawenkou s i t e report (Early, Middle, Late) were derived from a judge-mental assessment of changes i n the ceramic vessel styles and subtypes. Chapter 7 of the s i t e report. "The Method and Basis of the Periods", describes how t h i s assessment was made. Ceramic forms are inferred as developing out of other forms. The changes i n vessel forms from 15 pairs of b u r i a l s that cut into one another ( l i s t e d i n Figure A3-2) form the basis of the r e l a t i v e chronology. The remaining b u r i a l s i n the cemetery were f i t t e d into the chronological sequence by comparing the vessel forms i n them to the sequence of forms i n the above 15 pairs of b u r i a l s . Except for the r e l a t i o n -ship of the i n t r u s i v e b u r i a l s , stratigraphy was not employed as a basis for establishing the three periods. The s i t e report does not provide any s t r a t i g r a p h i c information for the 195 9 excavation of the cemetery. My interpretation of the method by which the three periods were derived has been confirmed and amplified by C. Shangraw (personal communication 1983). Gao's (1980) a r t i c l e includes a discussion of the three periods and t h e i r derivation. Chapter 7 of the s i t e report indicates that the method to determine the sequence of developing ceramic forms i s based upon i m p l i c i t , i n t u i t i v e c r i t e r i a . The authors discuss - 62 -the evolution of the most important functional types, sub-types and styles but do not explain how these forms were i d e n t i f i e d . I t seems the most important forms are those with the most obvious physical features. The evolution of the various subtypes i s described i n the most d e t a i l . Certain morphological features are described as getting f l a t t e r , more bulging, etc. through time. When functional types or subtypes appear to die out i s also noted. From the d i s t r i b u t i o n of style numbers per period for the subtypes provided i n Table 2 of the s i t e i r e p o r t (page 130), I think that the perceived morphological changes i n the sub-types through time were marked by a d i f f e r e n t s t y l e number. From Table 2 i n the s i t e report, i t i s apparent that on the whole, the lowest s t y l e numbers for a subtype occur i n the Early Period and progressively higher numbers occur i n the Middle and Late periods. The sty l e numbers also indicate developmental changes i n the funtional types that are not subdivided into subtypes. I i n f e r the authors used the vessels i n the 15 pairs of bur i a l s which cut into each other to e s t a b l i s h the general sequence i n which the functional types and subtypes changed. For example, vessels of one subtype from a pair of in t r u s i v e b u r i a l s were used to define " e a r l i e r " and " l a t e r " character-i s t i c s , for..that subtype. Vessels of that subtype from other bu r i a l s were judged as "early" of " l a t e " , or "intermediate" i n time. The i n t u i t i v e comparison of the vessels from the - 63 -remaining b u r i a l s with those from the set of 15 bu r i a l s resulted i n a complete sequence of development for each functional type and subtype, marked by appropriate style numbers. The authors then examined the entire set of vessels for each b u r i a l , concluding that three evolutionary stages were represented. Chapter 7 of the s i t e report mentions that a clear difference between the sets of vessels from the Early and Late b u r i a l s i s noticable. But vessel forms from the Middle period b u r i a l s are considered t r a n s i t i o n a l . The authors imply they are not ce r t a i n about the chronological position of some Middle period b u r i a l s . Two of the b u r i a l s that cut into each other (#'s 31 and 62) are assigned to the Early period, yet they do not contain any pottery vessels. The method by which these b u r i a l s were assigned to the Early period i s not clear. The other (15) bur i a l s i n the cemetery with no pottery vessels are considered undatable. Chapter 7 of the s i t e report includes a summary of the d i s t r i b u t i o n of ware colors and of some production t o o l types per period. Four ware colors are said to have chronological s i g n i f i c a n c e : red wares are predominant i n the Early period, grey wares increase i n the Middle period, black wares increase i n the Late period, and white ware i s only present i n the Late period. This general trend through time i s substantiated i n many N e o l i t h i c s i t e s from Shandong and northern Jiangsu (Shangraw, personal communication 1983, Shangraw 1978:12 and - 64 '-Gao 1980:61). Many of the stone and bone t o o l types i n the report are given s t y l e numbers as well. I i n f e r that these styles were also defined on the basis of i m p l i c i t , i n t u i t i v e c r i t e r i a . The participants of the f i r s t symposium on the Dawenkou culture i n 1977 also i d e n t i f i e d a d i s t i n c t i o n between the ceramic forms i n the Early, Middle and Late period b u r i a l s at Dawenkou (Shandong P r o v i n c i a l Museum 1978:59). However, the Early and Middle periods defined i n the Dawenkou site':.report are grouped together as part of one main evolutionary stage of the Dawenkou Culture and the Late period b u r i a l s into another main stage (ibid:61). The two-period chronological scheme (Early and Late, with the former Middle period as part of the Early period) advocated i n the recent Chinese archaeological l i t e r a t u r e appears to have been based upon an i n t u i t i v e comparison of the ceramic forms at the Dawenkou s i t e with the forms from other Dawenkou Culture s i t e s that have r e l a t i v e chronological sequences based upon stratigraphy, such as the Xixaihou s i t e (Wu Ruzuo 1982:267-269). An independent check of the changes i n ceramic morphology from the Dawenkou bu r i a l s i s necessary i n order to assess whether the two-period or three-period scheme should be accepted for t h i s mortuary analysis. - 65 -3.3. Analysis 3.3.1. Introduction Two types of multivariate quantitative methods are employed here i n an e f f o r t to t e s t the consistency of the r e l a t i v e dating method just described based upon change i n ceramic morphology. Multidimensional scaling and c l u s t e r analysis i n a Q-mode are used to group those b u r i a l s that co-vary i n terms of the presence of absence of ceramic s t y l e s , subtypes, and functional types they contain. The purpose of the analysis i s to determine whether an ordering of b u r i a l s from Early to Late can be generated. I f the results from the analyses show b u r i a l s of the same chronological period (Early, Middle and Late) grouped together, i t can be concluded that the r e l a t i v e dating method based upon changes i n vessel form used by the authors of the s i t e report i s consistent. The re s u l t s could indicate that the more recent two-period chronological scheme should be supported. In t h i s case, the res u l t s would show Early and Middle b u r i a l s grouped together and Late b u r i a l s as d i s t i n c t . The: results may show two or three periods composed of d i f f e r e n t groups of b u r i a l s than i n the s i t e report, or more than three periods. If the results from the analysis do not show bu r i a l s grouped together by chronological period at a l l , neither of the two chronological schemes nor any scheme can be accepted and the b u r i a l s must be regarded as belonging to the same period. - 66 -Since there are no radiocarbon dates or consistent stratigraphic data for the b u r i a l s excavated i n 1959, I w i l l not be able to demonstrate whether changes i n s t y l e s , sub-types or functional types actually represent changes i n time. I can only propose that two or more chronological periods (provided an ordering of b u r i a l s results) are represented at the s i t e . The placement of the 15 pairs of int r u s i v e b u r i a l s within the ordering of bur i a l s r e s u l t i n g from the multi-dimensional scaling and cl u s t e r analyses can server.as~an independent check of the ordering. If the i n t r u s i v e b u r i a l s are grouped as expected ( i . e . , with the b u r i a l s i n each pair i n d i f f e r e n t groups representing d i f f e r e n t time periods), the chronological periods interpreted from the multivariate analyses can be accepted with greater certainty. The r e s u l t s from the multivariate analyses cannot be used to confirm the duration of the time periods proposed by Wu (1982). If I propose that two chronological periods are represented at Dawenkou, I cannot be certain the e a r l i e r period represents 600 years and the l a t e r , 400 years, as i n Wu (1982:268). The analysis w i l l not indicate the rate of change of the various ceramic forms. Also, the p o s s i b i l i t y of abandonment of the cemetery between or within chrono-l o g i c a l periods cannot be precluded. F i n a l l y , i t i s beyond the scope of t h i s analysis to assess the r e l i a b i l i t y of the terms for the functional types and subtypes i n the s i t e report. Even though there may be some problems with these terms, they - 67 -w i l l be accepted for the purposes of t h i s analyses and for the remainder of t h i s mortuary study. An evaluation of these terms would require an independent c l a s s i f i c a t i o n of the ceramic forms from Dawenkou based upon firsthand observation. 3.3.2. Method The s p e c i f i c techniques of scaling and c l u s t e r i n g employed i n t h i s analysis and i n Chapter 6 are Torgerson's metric multidimensional scaling and three variants of c l u s t e r analysis: Complete-Linkage or Farthest Neighbor, Average Linkage (Unweighted-Pair Group Method) and Ward's Method (Error Sum of Squares). The multidimensional sc a l i n g program used here i s from Matson (1975). This program i s based upon Torgerson's (1958) algorithm, the equivalent of Gower's (1966) P r i n c i p l e Coordinates Analysis. The computer program for the cl u s t e r analyses i s from Wood (1974). The scaling and c l u s t e r i n g analyses are based upon a s i m i l a r i t y matrix which describes the s i m i l a r i t y between a l l pairs of items i n the data set. S i m i l a r i t y (transformed into a pseudo-distance measure) i s calculated by means of Jaccard's C o e f f i c i e n t of Association (described by Sneath and Sokal 1973:131, Sokal and Sneath 196 3:129,133; and Doran and Hodson 1975:141). This c o e f f i c i e n t i s useful for presence/absence data because i t does not group items on the basis of shared absences (True and Matson 1970:1201). In the chronological analysis, the b u r i a l s are grouped together only i f they/share - 68 -certa i n pottery s t y l e s , subtypes, or functional types. The matrix for the present analysis i s based upon the presence or absence of 83 ceramic forms i n 115 b u r i a l s . Binary data i s necessary for t h i s analysis because the ceramic forms do not occur with great frequency among the 115 b u r i a l s . As mentioned i n Chapter 2, more than one multivariate quantitative technique should be used i n conjunction with one another to ensure that the ordering of items obtained i s an accurate r e f l e c t i o n of the rela t i o n s h i p between the items, not an a r t i f a c t of one p a r t i c u l a r method (Matson and True 1974:72). Various techniques of multidemensional scaling and of c l u s t e r analysis have been found compatible for comparative purposes (e.g., Johnson 1972, Matson and True 1974. Matson 1974, True and Matson 1970, Pokotylo 1982, Pokotylo 1981, Peacock 1976). Multidimensional scaling can reveal important re l a t i o n s between clust e r s (Sokal and Sneath 1973: 252), The use of more than one technique of c l u s t e r analysis i s desirable because d i f f e r e n t techniques can produce d i f f -erent r e s u l t s (Matson and True 1974). In the chronological analysis, the orderings of b u r i a l s from the three c l u s t e r i n g techniques are compared with one . another. A good f i t between the groups of b u r i a l s r e s u l t i n g from each technique would allow acceptance of the groups with certainty. A good f i t of the groupings from the c l u s t e r analyses with the pattern produced by the multidimensional scaling would allow even greater confidence i n the r e s u l t s . - 69 -Torgerson's metric multidimensional scaling (Torgerson 1958) has been used successfully by Peacock (1976), Pokotylo (1982), Matson and Lipe (1977), and Matson and True (1974), among others. The following description of the technique i s based upon the explanations: provided i n the four works just c i t e d . Multidimensional scaling depicts the rel a t i o n s h i p of items i n n-dimensional geometric space.. The distance between items i n space r e f l e c t s the degree of s i m i l a r i t y between the items. The s i m i l a r i t y or distance matrix calculated by the c o e f f i c i e n t of association chosen forms the basis of a matrix of the products of the distances from the configuration's o r i g i n or centroid. This matrix i s factored, r e s u l t i n g i n dimensions of decreasing order of importance. The f i r s t dimension accounts for the greatest amount of v a r i a b i l i t y i n the data set, the second dimension accounts for the second greatest amount, and so on. A measure i s given which indicates the amount of v a r i a b i l i t y explained by each dimension. If one's data set does not meet the metric assumption required for Torgerson's method (Matson and True 1974:70) , Kruskal's (1964) non-metric method may be used. Among the advantages of metric mulitdimensional scaling over non-metric scaling are that unique solutions are derived, the solutions are invariant under changes i n dimensionality, and less computer time and money i s required (Matson and Lipe 1977:4) . Cluster analysis i s an agglomerative, h i e r a r c h i c a l - 70 -method which depicts the grouping of items i n the form of a dendrogram. Items are successively grouped according to decreasing degrees of s i m i l a r i t y (or increasing degrees of pseudo-distance) u n t i l a l l of the items form one large group. Matson and True (1974) point out that each of the three techniques has i t s own advantages and disadvantages. Complete Linkage c l u s t e r analysis has more conservative rules for the addition of items to extant clusters than Average Linkage, and i t avoids the disadvantage of chaining common to Average Linkage. Chaining i s an "elongate growth of single linkage c l u s t e r s " (Sneath and Sokal 1973:223). Chaining obscures the rela t i o n s h i p between items intermediate or connecting to the clusters with chaining ( i b i d ) . Average Linkage has been a standard method i n the archaeological l i t e r a t u r e (Matson and True 1974:61), and the Unweighted Pair-Group method has been most commonly used i n general (Sneath and Sokal 1973:230). Complete Linkage i s a "robust" technique because s l i g h t changes i n c o e f f i c i e n t s w i l l not create s i g n i f i c a n t l y d i f f e r e n t r e s u l t s (Matson and True 1974:61). But items may j o i n clusters with d i f f i c u l t y , r e s u l t i n g i n small clusters that leave out many less related items (Sneath and Sokal 1973:222-223, 226). Matson and True (1974:61) c i t e Ward's method as yi e l d i n g r e s u l t s which were closest to t h e i r i n t u i t i v e l y derived expectations, i n comparison to the res u l t s from Complete Linkage and Average Linkage. A detailed description of the three c l u s t e r i n g techniques - 71 -i s provided by Sneath and Sokal (1972:222-241) and Matson and True (1974). The major difference between the three tech-niques l i e s i n the c r i t e r i o n which groups an item or a c l u s t e r with extant clusters (Matson and True 1974:54). In Complete Linkage Cluster Analysis, the c r i t e r i o n i s the distance be-tween an item and the farthest member of an extant cl u s t e r (Matson and True 1974:57). In the Average Linkage technique, the distance between an item and the average distance for the extant c l u s t e r i s the c r i t e r i o n ( i b i d ) . In Ward's Method, the c r i t e r i o n i s not a single distance but a measure of t o t a l interpoint distances within a cl u s t e r . An item i s grouped with a cl u s t e r i f i t optimizes the sum of squared interpoint distances within that c l u s t e r (Matson and True 1974: 58) . A c r i t i c a l point regarding the use of multivariate s t a t i s t i c a l methods i s that the interpretation of the res u l t s must be made by the researcher. An interpretation i s not provided by the a n a l y t i c a l r e s u l t s (Matson and True 1974:72). For multidimensional scaling, the archaelogical meaning of the dimensions which r e s u l t must be interpreted. Also, the researcher must decide which dimensions best describe the v a r i a b i l i t y i n the data set. For any variant of cl u s t e r analysis, one must judge which clus t e r s have archaeological meaning. Intuitively-based knowledge of the data set as well as some independent data not included in the multivariate - 72 -analysis are required so that an objective interpretation can be made of the results (Matson and True 1974:72). At present, the use of multivariate analyses i n archaeology requires some subjective judgement on the part of the researcher. As Matson and True pointed out i n 19 74, There seems to be no way to eliminate these (judgemental) choices at the present time, and the r e s u l t s are always subject to re-interpretation as new data are recovered or improved methodologies are developed (Matson and True 1974:72). Seriation u t i l i z i n g a r t i f a c t forms such as ceramics by means of multidimensional sca l i n g has been attempted by a number of researchers (including many with b u r i a l data) since the late 1960's (Orton 1982:85). Authors who u t i l i z e multi-dimensional scaling for s e r i a t i o n purposes include Kendall (1971), Peacock (1976), Drennan (1976a, 1976b), Spencer (1982), and Matson and Lipe (1977), but Kendall's (1971) i s the only study i n which graves are the items being seriated. Torgerson's metric sc a l i n g i s employed by Peacock (19 76) and by Matson and Lipe (1977). Both studies u t i l i z e c l u s t e r analysis i n conjunction with the sc a l i n g . Researchers have found that when an ordering of items r e s u l t s , i t often takes the form of a horseshoe when the items are plotted i n two dimensions (Orton 19 82:85). However, a perfect horseshoe shape i s ra r e l y achieved i n practice (Orton 1982:86). Horseshoe shapes resulted from Torgerson's metric sc a l i n g i n Peacock (1976) and i n Matson and Lipe (1977). - 73 -Researchers must also be aware that the dimensions responsible for the ordering of items may not represent time and that interpretation of the ordering must be j u s t i f i e d (Orton 1982:88). Unless the ordering i s checked with inde-pendent chronological data such as stratigraphy, the ordering must be considered hypothetical (Marquardt 1978:287). As stated previously, the placement of the 15 pairs of intrusive b u r i a l s within the ordering from the scaling and c l u s t e r i n g w i l l serve as an independent, but not conclusive, check of the ordering. Mortuary studies i n which other techniques of s e r i a t i o n are employed are those of MacDonald (1980) , Doran (1971), and Hodson (1977). Marquardt (1978) provides a detailed descrip-t i o n of various s e r i a t i o n techniques. Seriation studies such as t h i s one are based on the proposition that s t y l i s t i c s i m i l a r i t y of ceramics i s related to s o c i a l i n t e r a c t i o n and that similar forms of ceramics are close i n time (Doran 1971:424). Although ethnoarchaeological studies support the notion that ceramic types found i n the archaeological record r e f l e c t time change, more studies are needed to provide a better understanding of the process responsible for the formation of the archaeological record and the e f f e c t upon the ceramic types represented at a s i t e (Marquardt 19 78:199-2 30). Archaeologists have recognized the advantage of using a r t i f a c t types from "closed finds" such as graves for - 74 -s e r i a t i o n (Kendall 1971:215). 3.3.3. Data A t o t a l of 115 b u r i a l s of the 133 i n the cemetery are included i n the chronological analysis: 71 of the 74 from the Early period, a l l 19 from the Middle and a l l 25 from the Late. These b u r i a l s are l i s t e d i n Figure A3-3. Four of these b u r i a l s are disturbed (E27, M4.6, L3, L77) , but the pottery styles or classes remaining should indicate the r e l a t i v e chronological placement of the b u r i a l s . Another b u r i a l , L10, i s only s l i g h t l y disturbed, i f at a l l . I t i s described as containing a form of vessel (shuang b i hu, s t y l e I, " j a r with two lugs") that may have originated from-.the b u r i a l into which L10 intrudes, E26. Since the o r i g i n of the jar i s not clear, i t i s considered here to be from L10. L10 contains a large number of vessels and the questionable placement of one should not change i t s position i n the chronological ordering or i t s r e l a t i v e status position. Seventeen b u r i a l s with no pottery vessels are excluded. Fi f t e e n of these, are the b u r i a l s considered as undatable by the authors of the s i t e report. The other two are E31 and E62, the i n t r u s i v e b u r i a l s with no pottery vessels assigned to the Early period. B u r i a l E108, assigned to the Early period, must be excluded as well. For reasons discussed below, the one type of vessel found i n t h i s b u r i a l cannot be included i n the analysis. - 75 -A t o t a l of 83 mutually exclusive ceramic categories from the t o t a l of 176 known forms given i n Figure A3-1 (excluding the unique animal shaped vessel and the uncertain forms - dou zuo and dou pan) are included i n the analysis. A l i s t of these 83 forms i s provided i n Figure A3-4. These forms include s t y l e s , subtypes not divided into s t y l e s , and functional types not divided into subtypes or s t y l e s . Unfortunately, i t i s not possible to include a l l 176 ceramic forms i n the analysis. Study of the d i s t r i b u t i o n of the 176 ceramic forms among the 116 bu r i a l s (including E108 here) indicates that a . number of forms occur very infrequently. Figure A3-5 shows that 61 forms (34.7%) are present only once among the bu r i a l s and 32 (18.2%) are present only twice. The forms begin to occur with greater frequency thereafter. From the graph i n Figure A3-5, I judged that a ceramic form should be present at least three times among the b u r i a l sample i n order for the multivariate analyses to be via b l e . The above decision required that the forms occurring only once or twice across the b u r i a l sample be either excluded from the analysis or grouped with other forms. If over f i f t y percent (52.9%) of the forms were excluded, the analysis could not adequately test the consistency of the r e l a t i v e dating method based upon morphological s i m i l a r i t y . Therefore, an e f f o r t was made to group ceramic forms that are morphologically si m i l a r . Due to my uncertainty with the c a i ("painted") subtypes - 76 -of vessels noted e a r l i e r , a l l 16 of these categories shown i n Figure A3-1 are excluded from the lumping procedure and from the analysis i n general. Since the only vessel i n E108 i s painted, t h i s b u r i a l must be excluded from the analysis. Morphological s i m i l a r i t y of the vessel forms was based upon the descriptions and drawings i n Chapter 5 of the s i t e report and the representative photographs. Chapter 5 describes certa i n parts of vessels such as rims, bodies, and bases as gradually changing form. Detailed study of the vessel forms tended to support my proposition that the forms closest i n styl e number are those closest i n st y l e . For example, a vessel with a sty l e of I appears more sim i l a r i n form to a vessel (of the same subtype) with a sty l e of II than a sty l e of V. A decision was made to group a vessel form with a form having the nearest st y l e number. For example, i f a vessel with a s t y l e number of II had to be lumped, i t was lumped with the vessels of st y l e number I or I I I , depending on my judge-ment of morphological s i m i l a r i t y . As many o r i g i n a l groups of vessels as possible were maintained. A subtype not divided into style numbers was not lumped with another subtype. If a vessel form did not i n my judgement appear morphologically sim i l a r to the vessels i n either adjacent style category, i t was excluded. The qf t a or "other" subtypes and gai (pot li d ) functional type have more styles excluded than other categories..: It.was necessary for a few entire categories to - 77 -be l e f t out, such as "pen ("basin") and gui wu zu ("jug with no f e e t " ) . I acknowledge my lim i t e d understanding of the Chinese c l a s s i f i c a t i o n procedure. The s i t e report only describes and pictures a representative group of vessels for each subtype and s t y l e . The range of v a r i a t i o n within the sty l e and sub-type categories i s not documented i n the s i t e report. Also, the written descriptions can be confusing to a western reader. For example, a t r a n s l a t i o n of "bird-shaped beak" refers to a spout on a vessel and a "nose" refers to a lug. Although the incorrect lumping of styles may d i s t o r t the r e s u l t i n g ordering somewhat by making the ordering less clear, i t should, not r e s u l t i n a spurious ordering (R.G. Matson, personal communication 1983). There are some discrepancies for the figure given i n various places i n the s i t e report for the t o t a l number of vessels i n certa i n functional types, subtypes, or st y l e s . The to t a l s for some categories given i n Chapter 5 of the s i t e report and i n Table 12, page 135, d i f f e r with my t o t a l s calculated from the descriptions of the b u r i a l s . In each case, I accepted the t o t a l s from the descriptions of the bu r i a l s After the lumping procedure was completed, another graph was made that depicts the frequency of occurrence of the ceramic categories i n the b u r i a l sample. Figure A3-6 shows a more even d i s t r i b u t i o n of the categories: 18.1% of - 7 8 -the categories occur three times among the 115 b u r i a l s (excluding E108 now), 14.5% occur four times, 10.8% occur f i v e times, and 16.9% occur six:times. Figure A3-7 shows that 13.9% of the b u r i a l s have only one ceramic category present a f t e r lumping, 15.7% have two, 9.6% have three, and so on. A smaller percentage of b u r i a l s having only one or two ceramic categories would have been more desirable. How-ever, i t was thought that the multivariate analyses would be viable. The 83 forms r e s u l t i n g from the lumping consist of 112 of the o r i g i n a l 176 formal categories. Sixty-four forms could not be lumped and are excluded from the analysis. 3.3.4. Results The orderings of b u r i a l s r e s u l t i n g from the multidimen-sional scaling and three variants of c l u s t e r analysis show a clear d i s t i n c t i o n between the Early and Late b u r i a l s , with some Middle b u r i a l s grouped with the Early b u r i a l s and others with the Late. This patterning i s that described i n the s i t e report. Therefore, I conclude the changes i n ceramic morph-ology were assessed i n a consistent manner to indicate time change. However, the orderings from the scaling and c l u s t e r i n g support neither the three-period chronological scheme advocated i n the s i t e report nor the two-period scheme proposed i n the recent Chinese 'archaeological l i t e r a t u r e , The orderings indicate a v a r i a t i o n of the two-period chronological scheme. - 79 -They depict a two-period d i v i s i o n of Early and Late, with some former Middle period b u r i a l s placed i n the Early period and some i n the Late. This scheme d i f f e r s from the two-period scheme i n which the former Middle period i s placed e n t i r e l y within the Early period. The orderings produced from the scaling and c l u s t e r i n g techniques are described below. Each of the three c l u s t e r i n g techniques shows the clear d i v i s i o n , between Early and Late period.burials. However, each shows a d i f f e r e n t ordering of bur i a l s within the Early period and within the Late. Some of the differences seem due to the inherent nature of each cl u s t e r i n g technique.. Others may be due to the nature of the data set used i n the analysis. The ordering of b u r i a l s from Ward's Method i s chosen for the comparison with the multi-dimensional scaling r e s u l t s . The placement of the in t r u s i v e pairs of b u r i a l s within the orderings from the scaling and Ward's Method provides independent support that the orderings represent chronological relationships among the b u r i a l s . 3.3.5 Multidimensional Scaling Five dimensions were generated, accounting for 29.56% of the - v a r i a b i l i t y within the data set. This low percentage may be due to the low number of ceramic categories within many of the b u r i a l s , and thus the low number of ceramic categories shared between many b u r i a l s . Figure A3-7 shows that 29.64% of the b u r i a l s only had one to two ceramic - 80 -categories present. The f i r s t dimension accounts for the greatest amount of v a r i a b i l i t y , 9.175%, with the second and t h i r d dimensions .accounting for only 5.597% and 5.748%, respectively.- In theory, the dimensions from Torgerson's metric multidimensional scaling method should account for decreasing amounts of v a r i a b i l i t y . The eigenvalues for the f i r s t three dimensions are 4.782, 2.917, and 2.996. Sl i g h t deviations often r e s u l t i n association with small eigenvalues (R.G. Matson, personal communication 1983). One inequality out of 246,905 was v i o l a t e d i n the program; a low proportion of 0.0000405, in d i c a t i n g that t h i s data set i s indeed metric. Figure 3-1 shows the ordering of the 115 b u r i a l s along Dimensions 1 and 2. Dimension 1 i s interpreted as representing time. The b u r i a l s are divided into an Early sector on the l e f t side of the axis and a Late sector on the r i g h t . The div i d i n g point i s at approximately 0.140 on the X axis. A horseshoe shape i s not c l e a r l y discernable. The former Early period b u r i a l s and some Middle period b u r i a l s tend to be scattered while the former Late period b u r i a l s are i n a more r e s t r i c t e d area, suggesting one half of a horseshoe. This difference i n s p a t i a l positioning and the lack of a good horseshoe shape may be caused by differences i n the number of ceramic categories per b u r i a l within each of the three periods. The Early period b u r i a l s account for most of the bur i a l s with one to two ceramic categories shown i n Figure A3-7-(12 of the 16 having one category and 16 of the 18 b u r i a l s - 81 -FIGURE 3-1. The 115 b u r i a l s i n the chronological analysis from Torgerson's Metric Multidimensional Scaling, Dimensions 1 and 2, clus t e r s from Ward's Method. E 8 2 A EI32# Ell , E 3 4 E 6 5 « E13I E I 2 9 / / ' El 10 EI3 " 3 A DIM 2 E5I + A E I 0 ? A E 5 6 A . E I 0 6 . M2I £ 2 3 * E I 0 3 „ E I09 EI8 , E 7 E32< E , l l O E 8 0 A" //A E l 19 * EII5 O E 7 3 Qa E 5 9 A , E I I 4 C- E I20 E 9 9 u E58 E 4 3 " E 4 I . E 8 9 E I 2 ° E 8 ° E 8 7 ° E 8 1 0 E66 E 7 6 o E 5 3 0 E 4 8 A A EI02 E 5 4 * E I I 2 1 »E6 E 2 9 ^ E 4 5 * O EBI E l 16 E9Ia E 2 8 ^ E 3 3 * E 9 4 M 4 9 ' E I 3 0 M42 E 8 6 E 8 8 E 9 0 A Ml 18 E 5 5 E 6 l " - £ A M 3 6 A j L 7 7 , E 2 7 L I L 6 0 + + L I 7 , L 2 5 £ 2 6 ^ " „ ° L 7 2 j D • ' L I 0 5 -EI0I M69 +3.0 CL6*~T^ DIM, E 3 0 " LI23*> ° L I " L 2 U • „ D M I 6 * D • . L 4 a a LI26 , 0 M 9 6 b L 4 7 M 9 * -2.0 M 9 3 M 4 6 " L I04 L 3 L I 0 0 / ° U 2 4 * D L I 2 5 • M 7 5 o M22 M35 C l u s t e r s I 2 3 4 5 6 7 8 E E a r l y B u r i o i M Middle B u r i o i L L a t e B u r i a l i n t r u d i n g b u r i a l - 82 -having two). Thus, Early period b u r i a l s are either very s i m i l a r or very d i f f e r e n t , not allowing a continuous horse-shoe shape. The plot of bu r i a l s also shows a few bu r i a l s located at the same point i n geometric space. These b u r i a l s are i d e n t i c a l i n composition of ceramic categories. The placement of the pairs of intruding b u r i a l s supports the interpretation that Dimension 1 represents time. The Late period b u r i a l s that intrude into Early b u r i a l s (see Figure A3-2); L10, L15, and L24, are located some distance along Dimension 1 from the b u r i a l s they cut in t o : E26, E33, and E30, respectively. Four of the intruding b u r i a l s were excluded from the analysis because they do not contain pottery: E31, E62 and the undatable b u r i a l s X70 and X133. E18 i s not marked on the plot because i t s i n t r u s i v e pair (E31) was not i n the analysis. This i s true for E71 as well (having the undatable b u r i a l 70 as .its p a i r ) . There i s less spatial' separation of the intruding pairs of b u r i a l s from the same former time period: E54 and E58, L123 and L124, E32 and E61, and E78 and E129. Three of the four former Middle period b u r i a l s that cut into Early period b u r i a l s are s p a t i a l l y separated from the l a t t e r b u r i a l s : M9 and E23, M16 and E61, and M121 and E132. M44 and E43 are barely separated along Dimension 1, and M4 4 i s closer to the majority of the Early period b u r i a l s than E4 3. The locations of these two b u r i a l s do not provide additional support for the ordering derived from the scaling, but they make sense - 83 -when M44 i s considered an Early period b u r i a l . I t i s located well to the l e f t of the 0.140 di v i d i n g l i n e on the X axis. Because E30 and E61 have only one ceramic category present, t h e i r locations on the scaling p l o t are not as secure as that of the other intruding b u r i a l s . Since a l l the other intruding b u r i a l s contain more than two categories, t h e i r locations can be r e l i e d upon. It i s possible that Dimension 2 represents time as well as Dimension 1. For three of the f i v e pairs of Early period in t r u s i v e b u r i a l s i n Figure 3-1, the older b u r i a l s (E43, E58, E12 9) are located i n the upper l e f t quadrant and the younger bur i a l s (M44, E54, E78) are located i n the lower l e f t quadrant. A l l the Early b u r i a l s i n the lower l e f t quadrant may be younger than the Early b u r i a l s i n the upper l e f t quadrant. However, the placement of the bu r i a l s i n the one Late period in t r u s i v e pair does not support the p o s s i b i l i t y that Dimension 2 represents time, with older b u r i a l s i n the upper section of Figure 3-1 and progressively younger b u r i a l s towards the lower section of Figure 3-1. The older b u r i a l , L124, i s located below the younger b u r i a l , L12 3. Due to the fact that the ordering of bu r i a l s from Early to Late i s also: exhibited i n dendrograms of the clus t e r analyses,- as well as the lower percentage of v a r i a b i l i t y accounted for by Dimensions 3 and 4, Dimensions 3 and 4 are not interpreted here. - 84 -3.3.6. Cluster Analyses The ordering from the Complete Linkage technique compares poorly with those from Average Linkage and Ward's. Small groups of b u r i a l s j o i n together only at high distance figures. This problem i s common with Complete Linkage due to the conservative rules of joining items to extant c l u s t e r s . While Early b u r i a l s are i n d i f f e r e n t clusters than Late b u r i a l s , a l l of the Early clusters and Late clusters do not eventually j o i n into one large c l u s t e r each. A macrocluster of Early b u r i a l s and one of Late b u r i a l s (with Middle b u r i a l s i n t e r -spersed) i s apparent i n the dendrograms from both Average Linkage and Ward's Method. The dendrograms from Average Linkage and Ward's Method agree f a i r l y well. I t i s apparent that t h i s agreement r e f l e c t s r elationships among b u r i a l s . The orderings of b u r i a l s within the Early period from Average Linkage and Ward's Method compare poorly but the orderings of b u r i a l s within the Late period compare quite well. This difference may be due to the greater number of ceramic s t y l e s , subtypes and functional types per b u r i a l within the Late period. The Early period cl u s t e r s may not be as robust as the Late clusters because they contain b u r i a l s that have few ceramic categories i n common. Some chaining i s evident i n the Average Linkage dendrogram, but not an excessive amount. The dendro-grams from both methods place the 19 Middle period b u r i a l s i n the same manner: eight b u r i a l s i n the Early period and 11 - 85 -i n the Late. The ordering from Ward's Method i s chosen to: compare with the multidimensional scaling r e s u l t s . Ward's Method has been found to y i e l d r e s u l t s with the same data set that are more sim i l a r to i n t u i t i v e l y derived expectations than those from Average Linkage and Complete Linkage (Matson and True 1974:61). The r e s u l t s from Ward's Method are also favored over those from Average Linkage and Complete Linkage by Peebles (1974:100). The dendrogram for Ward's Method i s shown i n Figure 3-2. The extremely clear break between the Early and Late b u r i a l s i s apparent. Figure 3-2 indicates that no Early and Late bu r i a l s are i n the same c l u s t e r . Also, a l l of the Early period b u r i a l clusters j o i n together, and a l l of the Late j o i n together. The Early and Late period macroclusters do not j o i n u n t i l the 4.3578 distance l e v e l , i n d i c a t i n g the clea r d i s s i m i l a r i t y of the Early and Late period b u r i a l s i n terms of ceramic categories. While a number of solutions could be compared with the multidimensional sc a l i n g p l o t , the Eight Group Solution at 2.0023 distance i s chosen as a man^ -ageble, interpretable one. The ceramic categories represented i n each c l u s t e r are given i n Figure A3-8. The placement of the intruding pairs of b u r i a l s i n the clusters (each b u r i a l i n a pair of d i f f e r e n t c l u s t e r s , at varying distance levels) i s further support that the ordering from Ward's Method r e f l e c t s chronology. These b u r i a l s are marked i n Figure 3-2. The pairs from the former Early and H G s w fD 3" i O NJ & • O hi o 3* o i — • c rt fD H > H O O O = j 3 3 / \ pottery kiln 21 3 6 8 0 C 3 73 66 17 27 r 35. 6C ' C Z T Q 25 £ J i 4 ^ 2 ^ 5 0 Q 85 ^ « 76 8 7 ^ 8 6 ^ 9 ,29 ^ 8 9 0 10?CD*°B_!09 127" » t = > 7 8 ^ 3 ^ 9 4 ^ ^ B5 S2 a O a r-riii a 7 a 102 r i ? J « E 3 - - '23 , 124 S-JI05 e £ T C » iota f->» io«r-i , 2 2 < J c j 1 2 5 113 a ? < ^ 1 0 0 1 2 1 / ^ 1 3 3 C>I5 ^ , 3 2 JOII4 92 10 m I • • i • J CD E A R L Y P E R I O D B U R I A L HI L A T E P E R I O D B U R I A L U N D A T A B L E B U R I A L (adapted from The Shandong P r o v i n c i a l Cultural Properties Cornnission and The Jinan City Museum 1974:4) - 92 -analysis. These b u r i a l s are accepted as belonging to the Early period for the remainder of the study. I conclude the vessel i n b u r i a l 108 must have been examined by the method of s t y l i s t i c comparison which the multivariate analyses show as consistent. I t i s l i k e l y the vessels i n the b u r i a l s (18 and 33) that cut into b u r i a l s 31 and 62 were thought to resemble Early period s t y l e s . Since b u r i a l s 31 and 62 are older than 18 and 33, b u r i a l s 31 and 62 were thought to be Early. The fact that several b u r i a l s within the accepted Early and Late periods are cutting into each other suggest each period i s quite lengthy. I t i s not l i k e l y that mourners would d e l i b e r a t e l y disturb an ancestor's remains when digging a grave for a deceased person. I t appears that no surface features mark the graves. However, intensive land use or use of the s i t e during l a t e r c u l t u r a l periods may have destroyed surface features. There may have been such a great time gap between the b u r i a l s i n question above that the locations of ancestor's graves were forgotten.. The fact that some people i n the Late period were buried before the use of the pottery k i l n i n the cemetery and others were buried l a t e r (mentioned i n Chapter 1) also suggests a lengthy Late period. The lengths of the periods proposed i n the Chinese archaelogical l i t -erature (600 and 400 years for the Early and Late periods, respectively) do not seem u n l i k e l y . The only known chronological relationships of b u r i a l s within each of the newly derived periods are indicated by the - 93 -pairs of i n t r u s i v e b u r i a l s within each period. Four pairs of i n t r u s i v e b u r i a l s i n the new Early period (E32 and E61, M44 and M43, E54 and E58, E78 and E129) and one i n the new Late period (L123! and L124) were included i n the chronological analysis. These pairs are located i n a variety of s p a t i a l areas within the cemetery, suggesting that a l l s p a t i a l areas were used continuously during each period. I t does not appear that d i f f e r e n t s p a t i a l areas represent d i f f e r e n t periods of time as noted by Shennan (1975:280) for the European Neol i t h i c cemetery of Branc and Chapman and Randsborg (1981:15) for a modern r u r a l Hungarian cemetery. Due to the lack of i n f o r -mation on the chronological relationships of b u r i a l s within each period, emphasis i s placed on change through time from j the Early period to the Late for the remainder of t h i s study. The d i s t r i b u t i o n of ceramic ware colors within the newly derived Early period and Late period i s comparable to the d i s t r i b u t i o n for Dawenkou Culture s i t e s mentioned previously. The d i s t r i b u t i o n i s additional support for accepting the newly assigned periods. My calculations indicate the d i s t -r i b u t i o n as follows: Early period: red ware i s 54% of the t o t a l , grey 31%, black 10%, painted 5%; Late period: red ware i s 12%, grey 41%, black 14%, white 31%, painted 1%. The d i s t r i b u t i o n of ware color w i l l be discussed i n more d e t a i l i n Chapter 6. The chronological analysis also indicates that some ceramic classes and forms are temporally s e n s i t i v e . - 94 -The following forms are present only i n the newly assigned Early period (see Figure A3-1): ding c a i ("painted t r i p o d " ) , dou guan shi pan ("serving stand with a guan-type dish"), dou da lou kong ("serving stang with large cut out holes"), dou c a i ("painted serving stand"), guan c a i ("painted j a r " ) , he c a i ("painted spouted vess e l " ) , kui xing q i ("helmet... shaped vessel"), zuo ("stand"), and bo c a i ("painted bowl"). These forms occur only i n the Late period: bei dan ba ("cup with simple handle"), dou shuang ceng pan ("serving stand with double-layered dish"), hu kuan j i a n ("storage vessel with wide shoulders"), ping ("bottle"), gui kong zu ("hollow footed tripod p i t c h e r " ) , and dou tong xing ("tubular shaped serving stand"). - 95 -CHAPTER 4 , ANALYSIS TO ESTIMATE SEX 4.1. The Problem Chapter 2 Outlines the methodological problem created by the low proportion of sexed to unsexed buri a l s i n the b u r i a l population. A key test implication for achieved versus ascribed status i s whether status d i f f e r e n t i a t i o n i s based upon age and sex or i f i t cross-cuts age and sex categories. With only 19 out of 80 sexed single b u r i a l s i n the Early period, nine of 32 single sexed buri a l s i n the Late Period, and two out of 15 i n the undatable group of b u r i a l s , i t would be d i f f i c u l t to test whether status d i f f e r -e n t iation i s based upon sex. A greater proportion of sexed bu r i a l s would also enhance discussion of s o c i a l subgroup a f f i l i t a t i o n represented i n the cemetery. The bu r i a l s confidently estimated as male or female are added to the sample of known males and females for the analyses i n Chapters 5 and 6. In t h i s analysis, two multivariate c l a s s i f i c a t i o n methods are employed: a simple inspection method and discriminant analysis. These methods are described i n d e t a i l i n the next section. Each method c l a s s i f i e s the unsexed adult b u r i a l s into a male or female class on the basis of the a r t i f a c t types included i n the known male and female - 96 -graves. An e x p l o r a t o r y study o f the d i s t r i b u t i o n o f a r t i f a c t types w i t h i n the known sexed graves suggested t h a t some types of a r t i f a c t s ( e s p e c i a l l y p r o d u c t i o n t o o l s ) occur e x c l u s i v e l y o r n e a r l y so wi t h males ,and ot h e r s w i t h females. T h i s s u g g e s t i o n has been made f o r Dawenkou C u l t u r e s i t e s by Pearson (1981:1084), the Shandong Prov-i n c i a l Museum (1978), Luo and Zhang (1979), the Shandong A r c h a e o l o g i c a l Team (1979) and Zhang (1979). Pearson (1981) i n c l u d e s a comprehensive summary of the d i s t r i b u t i o n of s e x - r e l a t e d a r t i f a c t types a t Dawenkou C u l t u r e s i t e s . B i n f o r d ' s (1971) ethnographic study i n d i c a t e s t h a t i n mortuary r i t u a l , sex i s commonly expressed by a r t i f a c t form (see F i g u r e 2-3). These a r t i f a c t forms may be c l o t h i n g , forms t h a t express the p e r s o n a l i t y o f an i n d i v i d u a l , o r t o o l s t h a t r e f l e c t the male-female d i v i s i o n o f labour ( B i n f o r d 1971:22). Only unsexed a d u l t b u r i a l s are i n c l u d e d i n t h i s a n a l y s i s . Since 29 of the 30 sexed b u r i a l s are a d u l t s , the a r t i f a c t s a s s o c i a t e d w i t h these b u r i a l s may be a r e f l e c t i o n of r e l a t i v e (adult) age as w e l l as sex. M u l t i p l e b u r i a l s are excluded from t h i s a n a l y s i s because i t i s d i f f i c u l t t o i d e n t i f y the a r t i f a c t s a s s o c i a t e d w i t h each i n d i v i d u a l . Sexed and unsexed s i n g l e a d u l t b u r i a l s from the undatable group of b u r i a l s are i n c l u d e d i n the a n a l y s i s . B i n f o r d ' s (1971) study i n d i c a t e s t h a t o r i e n t a t i o n o.f the grave may r e f l e c t the sex of an i n d i v i d u a l as w e l l . The v a r i a b i l i t y i n grave o r i e n t a t i o n a t Dawenkou i s d i s c u s s e d i n Chapter 5. The s m a l l number of aged s k e l e t o n s a t Dawenkou a l s o l i m i t s the effectiveness of the analysis of status d i f f e r -e n t i a t i o n . There are only four known subadults i n the Early period and one i n the Late. Six are i n the undatable group o f . b u r i a l s . The other skeletons are described as adult, with no d i f f e r e n t i a t i o n as to r e l a t i v e age. According to Binford's (1971) ethnographic study, age i n mortuary r i t u a l may be r e f l e c t e d by grave location, grave form and body d i s p o s i t i o n (see Fig.2-3). V a r i a b i l i t y i n these three aspects of mortuary treatment i s also discussed i n Chapter 5. Whether r e l a t i v e age i s also r e f l e c t e d by certain a r t i f a c t forms at Dawenkou cannot be tested here due to the small sample of b u r i a l s of known r e l a t i v e age and the paucity of grave goods i n the majority of the subadult graves. This analysis assumes that the sexing of the Dawenkou skeletons was done accurately by the investigators. It i s not clear why only 30 skeletons from single b u r i a l s and eight skeletons from multiple b u r i a l s were sexed. The investigators may have encountered fragmentation of c r i t i c a l s k e l e t a l parts or poorly developed suxual c h a r a c t e r i s t i c s , problems which have affected other cases of sexing (Bender 1979:186). Since some of the sexed skeletons have only a few common a r t i f a c t s present, i t seems the sexing was done on the basis of s k e l e t a l c h a r a c t e r i s t i c s and not on a r t i f a c t types. It i s not l i k e l y i t h a t i n 1959 investigators u t i l i z e d a multi-variate sexing method, preferred today for i t s accuracy over the method of applying single c r i t e r i a to each skeleton - 98 -(Weiss 1973:58). But for the purposes of t h i s mortuary study, i t i s assumed the method employed i s adequate. That subadults were not sexed i s not unusual - other paleodemographic analyses have not been able to sex sub-adults due to inadequate development of sexual c h a r a c t e r i s t i c s (Bender 1979:186). I t i s not known i f the group of sexed skeletons from Dawenkou has the common problem of a bias towards males (Weiss 1973:58). A bias may not have occurred because approximately equal numbers of males and females were i d e n t i f i e d (14 males and 16 females from the single b u r i a l s and four males and four females from the multiple b u r i a l s ) . Since males comprise approximately f i f t y percent of most ethnographic adult populations, a sexed s k e l e t a l population should have a si m i l a r male-female r a t i o , unless c u l t u r a l factors are responsible ( i b i d ) . C u l t u r a l factors that could upset t h i s r a t i o include cases i n which males died i n other t e r r i t o r i e s (Bender 1979:187). I t i s possible the known male-female r a t i o at Dawenkou may not be representative of the male-female r a t i o i n the population as a whole. It i s also not clear why the r e l a t i v e ages of the adult skeletons were not determined. The reason may be the inherent d i f f i c u l t y and uncertainty i n the aging of skeletons (Weiss 1979:59). For the purposes of t h i s mortuary study, i t i s assumed that the majority of subadults i n the cemetery were r e l i a b l y i d e n t i f i e d . Four b u r i a l s i n the cemetery do not contain a skeleton (one i n the Early period and three i n the Late). The authors of the s i t e report suggest that these b u r i a l s were - 99 -prepared for people who died i n another t e r r i t o r y , a practice common to modern f i s h i n g peoples (Shandong P r o v i n c i a l Culture Properties Commission and the Jinan City Museum 1974:7). Since these b u r i a l s contain grave goods, they w i l l be included i n the analysis to estimate sex. 4.2. The Analysis 4.2.1. Introduction None of the extant mortuary studies i n the l i t e r a t u r e has employed a simple inspection method or discriminant analysis to estimate sex. The two studies i n which an attempt i s made to estimate sex on the basis of a r t i f a c t inclusions (Hodson 1977; Shephard 1979) do not make e x p l i c i t use of the patterning of a r t i f a c t s i n known sexed b u r i a l s . Hodson (1977) u t i l i z e s Single Linkage Cluster Analysis i n an R mode, and Shephard (1979) u t i l i z e d Single Linkage Cluster Analysis and p r i n c i p a l coordin-ates analysis i n an R mode. Both studies interpret the "resultant groups of a r t i f a c t s as either male-related or female-related. The interpretations seem made on the basis of modern western concepts of male and female a r t i f a c t s . It i s not clear whether the patterning of a r t i f a c t s from known male and female b u r i a l s contributes to the interpretations. Doran (1973) mentions that sex and age w i l l be estimated in a future study for a sample of European Iron Age b u r i a l s , but apparently the results have not yet been published. - 100 -Hodson (1979:25) mentions four studies i n German that attempt to estimate sex on the basis of a r t i f a c t inclusions but does not describe the methods employed. Since the exploratory study of the Dawenkou sexed b u r i a l s suggest s p e c i f i c male and female related a r t i f a c t s , a multi-variate technique that can best make use of the known sexed group of b u r i a l s should be useful. The simple inspection method and discriminant analysis are two such techniques. Of course, Single Linkage Cluster Analysis and p r i n c i p a l coordinates analysis may work for other data sets. 4.2.2. Method The estimations of sex are based upon a comparison of r e s u l t s from the simple inspection method and the more rigorous method of discriminant analysis. Discriminant analysis i s t h e o r e t i c a l l y well suited for a c l a s s i f i c a t i o n problem such as the one here. However, the simple inspection method from Sneath and Sokal (1973:406) seems equally suitable. The simple inspection method i s u t i l i z e d to c l a s s i f y the unsexed single adult b u r i a l s on the basis of a few a r t i f a c t types that were noted i n the exploratory study as exclusive to the known males or females. The simple inspection method could not identi fy a l l the male or female-lfnked a r t i f a c t forms. Therefore, discriminant analysis i s u t i l i z e d to c l a s s i f y the unsexed b u r i a l s on the b a s i s of a l l the a r t i f a c t forms present i n the known sexed - 101 -b u r i a l s . D i s c r i m i n a n t a n a l y s i s can search the a r t i f a c t forms known as e x c l u s i v e t o e i t h e r sex, the forms noted as shared between the sexes and the forms too numerous f o r study by v i s u a l i n s p e c t i o n t o determine the a r t i f a c t forms t h a t b e s t d i s c r i m i n a t e between the known sexed b u r i a l s . However, d i s c r i m i n a n t a n a l y s i s i s not always more e f f e c t i v e than simple i n s p e c t i o n methods i n c l a s s i f y i n g items (Sneath and Sokal 1973:406). I t has been argued (Thomas 1980:344) t h a t a simple s t a t i s t i c a l method should be chosen over a more complicated one when i t i s known t h a t both methods are e q u a l l y capable of a c h i e v i n g the d e s i r e d r e s u l t s . In cases such as t h i s one, i t i s not c l e a r t h a t the simp l e r method w i l l be as e f f e c t i v e as the more complex method. I argue t h a t a. comparison of r e s u l t s from both a simple and complex method should i n d i c a t e the r e l a t i v e e f f e c t i v e n e s s o f each method f o r the p a r t i c u l a r data s e t u t i l i z e d . A comparison should a l s o i d e n t i f y the r e l a t i o n s h i p s between items i n the data s e t , not r e l a t i o n -s h i p s which are a f a c t o r of one method. In the simple i n s p e c t i o n method, i d e a l types are d e r i v e d from the v a r i a b l e s t h a t d i s t i n g u i s h the known c l a s s e s ( i n t h i s case, the known male and female b u r i a l s ) . Each i d e a l type c o n s i s t s o f a s e t of s e x - l i n k e d a r t i f a c t types. The s e t of s e x - l i n k e d a r t i f a c t s , on the b a s i s of the a r t i f a c t types p r e s e n t i n the known male and female b u r i a l s (see F i g u r e 4-1), i s used t o c a l c u l a t e the s i m i l a r i t y o f the unsexed b u r i a l s t o - 102 -FIGURE 4-1. D i s t r i b u t i o n o f s e x - l i n k e d a r t i f a c t t y p e s A r t i f a c t w h e t s t o n e l l s h i f i s h hook gdu t u s k k n i f e ya' dao s t o n e k n i f e s h f dao d i s k , p e n d a n t b i , z h i i i s t o n e c h i s e l s h x z a b bone c h i s e l gu zafo a r r o w h e a d z u s p o o n , s p a t u l a b i h e a d o r n a m e n t , n e c k o r n a m e n t t o u s h i , j i n g s t i i Known m a l e b u r i a l s Known f e m a l e b u r i a l s (14 t o t a l ) (16 t o t a l ) s m a l l r o u n d s t o n e x i a o s h i b i n g - 103 -the i d e a l t ypes. T h i s c a l c u l a t i o n i s by means of Jac c a r d ' s C o e f f i c i e n t , d e s c r i b e d i n Chapter 3. The taxonomic d i s t a n c e of each unsexed b u r i a l from each i d e a l type i s based on the s e x - l i n k e d a r t i f a c t types. A b u r i a l i s estimated t o be male i f the value f o r i t i s c l o s e r t o the "male" i d e a l type or female i f i t comes c l o s e r t o the "female" i d e a l type. D i s c r i m i n a n t a n a l y s i s c l a s s i f i e s items on the b a s i s of a s e t of independent v a r i a b l e s i n t o one of two or more mutually e x c l u s i v e c a t e g o r i e s or c l a s s e s (Morrison 1969:442). In t h i s case, the a n a l y s i s c l a s s i f i e s the unsexed b u r i a l s i n t o a male or female c l a s s on the b a s i s o f a d i s c r i m i n a n t f u n c t i o n . The d i s c r i m i n a n t f u n c t i o n i s c a l c u l a t e d from the pooled v a r i a n c e s and c o v a r i a n c e s between the v a r i a b l e s from each known c l a s s (Sneath and Sokal 1973:401). Here,, the d i s c r i m -i n a n t f u n c t i o n i s c a l c u l a t e d from the q u a n t i t i e s o f a l l the a r t i f a c t ; types p r e s e n t i n the known male and female c l a s s e s of b u r i a l s . The d i s c r i m i n a n t f u n t i o n i s a s t a t i s t i c a l statemehtcof :-thervaf tables:.:found ;to:"distingUishr between vthe e s t a b l i s h e d groups i n q u e s t i o n ( H e t t i n g e r 1979:456). The d i s c r i m i n a t i n g v a r i a b l e s are weighted and l i n e a r l y combined so t h a t the c l a s s e s are as s t a t i s t i c a l l y d i s t i n c t ..as p o s s i b l e (Klecka 1975:435). The c e n t r o i d f o r each c l a s s r e p r e s e n t s the average of the d i s c r i m i n a n t scores f o r the v a r i a b l e s , o r the average l o c a t i o n o f the v a r i a b l e s i n each c l a s s i n the d i s c r i m i n a n t f u n c t i o n (Klecka 1975:443). The d i s c r i m i n a n t scores f o r each case (here, the unsexed s i n g l e a d u l t b u r i a l s ) - 104 -r e f l e c t the p r o b a b i l i t y for membership into one of the classes. The Direct Method of Discriminant Analysis from the SPSS Subprogram DISCRIMINANT (Klecka 1975) i s u t i l i z e d here. In the Direct Method, a l l independent variables are entered at once and the discriminant function (one i n the case of two known classes) i s derived from the set of variables. Unlike the Step-wise method, the discriminating power of each inde-pendent variable i s not considered. In t h i s version of discriminant analysis, a "tolerance" test i s mandatory, r e s u l t i n g i n the r e j e c t i o n of p e r f e c t l y correlated variables. This step may be s t a t i s t i c a l l y j u s t i f i e d , but i t seems u n j u s t i f i e d i n the present case. 4.2.3. The Simple Inspection Method: Data Thirty sexed b u r i a l s (19 from the Early period, nine from the Late and two undatable burials) and 32 unsexed adult b u r i a l s are included i n the analysis by the simple inspection method. Figure A4-1 l i s t s the sexed and unsexed b u r i a l s for each period and for the undatable b u r i a l s . The small sample size of known male and female b u r i a l s i n each of the two periods precluded a separate analysis for each period. In the i n t e r p r e t a t i o n section, consideration i s taken of the two temporally sensitive a r t i f a c t types included i n the analysis. Eleven a r t i f a c t types are included: nine that are exclusive to the 14 known males.(but not present i n a l l male burials) and two that are exclusive to the 16 known females (but not present among a l l females). A few types are items of clothing but most are production t o o l s . The d i s t r i b u t i o n - 105 -of these 11 types i s shown i n Figure 4-1. Some of these a r t i f a c t types have style numbers assigned to them. Style numbers are not included i n the simple inspection method because the d i s t r i b u t i o n of the various numbers according to sex i s too d i f f i c u l t to ascertain by eye. Ceramic styles are excluded from t h i s analysis for the same reason. Some a r t i f a c t types have the same form but are made of a d i f f e r e n t material (stone and bone c h i s e l , and tusk and stone k n i f e ) . Since these types are purposely separated by the authors of the s i t e report and they are exclusive to the known males, a l l are u t i l i z e d i n t h i s analysis, Ceramic subtypes and functional types were not found to be exclusive to either sex except for a few forms that are rare i n the b u r i a l population. Only 32 of the 85 unsexed single adult b u r i a l s i n the cemetery have one or more of these a r t i f a c t types present. These bu r i a l s include: one undatable b u r i a l , 19 Early period bu r i a l s (one of which has no skeleton), and 12 Late period bu r i a l s (of which two have no skeleton present), The a r t i f a c t types do not occur with great enough frequency across the b u r i a l sample to allow the use of frequency data. Two of the disturbed b u r i a l s , E27 and L3, are included i n the analysis. A l l of the a r t i f a c t types from the male subadult b u r i a l , L117, are included i n the analysis because they occur with male adult b u r i a l s . Thus, these types are not a function of age. I judged that the a r t i f a c t types should be present at least four times among the b u r i a l population i n order for the - 106 -analysis to be viable. The j i a o zhui and b l are lumped together i n order to meet t h i s requirement, as well as the neck and head ornaments (jing shi and tou s h i ) . A known female has both the head and neck ornament types. One type exclusive to males must be rejected (stone hammer, chui) as well as two types exclusive to females (elephant ivory comb, xiang ya shu and stone pendant, huarig) . At the time of the analysis, on the basis of the photographs and drawings i n the s i t e report I interpreted j i a o zhui to indicate a horn pendant. I have learned since then that the term may refer to a net weight (Zou Heng, personal communication 1983). Therefore, the j i a o zhui probably should not have been grouped with the b i (elephant ivory d i s k ) . However, since there are only two cases i n which b l are present i n the b u r i a l sample (both from the Early period), the results from the analysis should be r e l i a b l e . The uneven d i s t r i b u t i o n of male and female . a r t i f a c t s l i m i t s the a b i l i t y of the analysis to i d e n t i f y unsexed b u r i a l s as female. Another l i m i t a t i o n i s that the absence of the nine "male" a r t i f a c t s does not necessarily indicate a female b u r i a l because s i x of the 14 known male b u r i a l s do not contain these a r t i f a c t types, ei t h e r . However, because the proportion of known females without these types i s much higher that the known males without these a r t i f a c t types (16 female buri a l s out of 22 total with no "male" a r t ^ i f a c t types/ versus six male bu r i a l s out of 22 t o t a l ) , i t i s - 107 -more l i k e l y that a b u r i a l with none of these a r t i f a c t types i s female. Therefore the variable "no male a r t i f a c t types" i s included as a twelfth variable i n the analysis. The l a s t two cases included i n the analysis are "ideal female" and "ideal male". The "i d e a l female" i s coded as having the nine male a r t i f a c t types absent and the three female a r t i f a c t categories present (the two given i n Figure 4-1 and the category "no male a r t i f a c t types"). The "ideal male" i s coded as having the nine male a r t i f a c t types present and the three female catagories absent. Coding of the 32 unsexed bu r i a l s showed a d i s t r i b u t i o n of the nine male a r t i f a c t types and two female ones which supported the proposition that at least some of the 11 a r t i -f act types are sex-linked. The unsexed bu r i a l s tend to have either some of the nine male a r t i f a c t types present and none of the female a r t i f a c t types, or vice versa. 4.2.4. The Simple Inspection Method: Results The s i m i l a r i t y matrix generated from Jaccard's C o e f f i c i e n t i s not d i f f i c u l t to inte r p r e t because the unsexed b u r i a l s are c l e a r l y s i m i l a r to either the "ideal male" type or to the "ideal female" type (with s i m i l a r i t y expressed by a low d i s -tance value and d i s s i m i l a r i t y expressed by a value of 1.000). None of the unsexed bu r i a l s have values close to both the "i d e a l male" and the " i d e a l female". The distance values for the sexed and unsexed buria l s as well as the assignment - 108 -of "male" or "female" for the unsexed bu r i a l s are given i n Figure 4-2. A t o t a l of 26 unsexed bu r i a l s appear to be male and s i x , female. The assignment of the six "female" b u r i a l s i s not as r e l i a b l e as that of the 26 "male" b u r i a l s . Four of the "females" (E49, E58, L5 and L60) only contain one female a r t i f a c t , the small round stone (xiao shi bing). This a r t i f a c t type i s only found i n one known female grave. Burials L3 and L47 are more r e l i a b l y "female" since they contain an a r t i f a c t type that i s found: i n three known female graves, the neck or head ornament (jing shi or tou s h i ) . This a r t i f a c t type i s uncommon i n the b u r i a l population and i s r e s t r i c t e d to the Late period. It may be a high status a r t i f a c t type (this p o s s i b i l i t y i s discussed i n Chapter 6). Thus, there may be no a r t i f a c t type to r e l i a b l y i d e n t i f y low status females i n the Late period or any females i n the Early period. As discussed previously, the lack of the male related a r t i f a c t types does not necessarily d i s t i n g u i s h a female b u r i a l . The a r t i f a c t types present i n one of the unsexed bu r i a l s suggests that the small round stones may not be female related a f t e r a l l . One of the estimated "females" (L5) contains an a r t i f a c t type that may be male related, the stone hammer. The majority of the estimated "males" (16 out of 26 new male b u r i a l s , or 61.5%) only contain one male related a r t i f a c t type ( a l l those b u r i a l s with a value of 0,889 i n Figure 4-2). The types present i n these b u r i a l s are bone FIGURE 4-2. Distance values for the unsexed b u r i a l s from the Simple Inspection Method and assignment of sex, and the distance values for known sexed b u r i a l s . Unsexed burials E l l E12 E19 E26 E27 E38 E49* E53 E54 Ideal male 0.889 0.889 0.778 0.444 0.889 0.889 1.000 0.889 0.778 Ideal female 1.000 1.000 1.000 1.000 1.000 1.000 0.333 1.000 1.000 E58* E61 E63 E66 E87 E103 E106 E110 E118 Ideal male 1.000 0.889 0.889 0.889 0.889 0.333 0.667 0.889 0.889 Ideal female 0.333 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 E119 L3* L4 L5* L17 L22 L24 L25 L47* Ideal male 0.889 1.000 0.333 1.000 0.333 0.889 0.778 0.667 1.000 Ideal female 1.000 0.0 1.000 0.333 1.000 1.000 1.000 1.000 0.333 L60* L75 L98 L126 X40 Ideal male 1.000 0.889 0.889 0.889 0.778 Ideal female 0.333 1.000 1.000 1.000 1.000 Note: assigned "females" marked*; assigned "males" unmarked. Burials with lower distance values are closer to the "ideal male" or "i d e a l female". L3 i s i d e n t i c a l to the "ideal female" because i t has a l l of the "female" a r t i f a c t s present. Cont'd. FIGURE 4-2 continued Known females E7 E28 E30 E55 E 6 7 E82 E102 E115 E130 I d e a l m a l e 1 .000 1 .000 1 .000 1 .000 1 .000 1 . 0 0 0 1 . 0 0 0 1 .000 1 .000 I d e a l f e m a l e 0 . 6 6 7 0 . 6 6 7 0 . 6 6 7 0 . 6 6 7 0 . 3 3 3 0 . 6 6 7 0 . 6 6 7 0 . 6 6 7 0 . 6 6 7 i'i' E131 i L10 L72 L105 L 1 2 1 X57 X85 Ideal male 1.000 1.000 1.000 1.000 1.000 1.000 1.000 Ideal female 0.667 0.667 0.333 0.333 0.667 0.667 0.667 The female burials with the lower distance figures have more of the "female" a r t i f a c t types present and are more similar to the "ideal female". A l l of the known females are the maximum distance from the "ideal male". Known males E9 E34 E59 E73 E91 E99 E107 E109 E112 Ideal male 0.333 0.556 0.667 0.889 1.000 1.000 1.000 0.778 1.000 Ideal female 1.000 1.000 1.000 1.000 0.667 0.667 0.667 1.000 0.667 L15 L117 L122 L123 L125 Ideal male 0.889 0.778 1.000 1.000 0.222 Ideal female 1.000 1.000 0.667 0.667 1.000 The male burials with the lower distance figures have more of the "male" a r t i f a c t types present and are more similar to the "ideal male". Eight of the known males are the maximum distance from the "ideal female". Six of the known males are more s i m i l a r to the "ideal female" because they have none of the "male" a r t i f a c t types. Males with a value of 0.889 only have one of the "male" a r t i f a c t s present. - I l l -arrowhead, bone spoon or spatula, and whetstone. Twelve of these 16 b u r i a l s are from the Early period and four are- from the Late. Only f i v e of the 12 estimated "male" b u r i a l s from the Early period have more than one male related a r t i f a c t type present, while one-half of the estimated "male" b u r i a l s from the Late period have more than one male related a r t i f a c t present. Thus, the greater number of a r t i f a c t types i n the Late period b u r i a l s makes the estimation of sex i n these bu r i a l s more r e l i a b l e . None of the "male" i d e n t i f i c a t i o n s i s based s o l e l y upon the presence of the uncertain netweight and stone disk category. 4.2.5 Discriminant Analysis: Data The c l a s s i f i c a t i o n procedure i n the discriminant analysis program allows the i n c l u s i o n of a large number of variables. In t h i s analysis, 129 a r t i f a c t types coded by frequencies and 111 b u r i a l s (29 known sexed b u r i a l s and 82 unsexed) are included. In order that the r e s u l t s be comparable to those from the simple inspection method, a separate analysis i s not done for each chronological period and the 11 possibly sex-linked variables from the simple inspection analysis are included as they were i n that analysis. The 129 a r t i f a c t types are l i s t e d i n Figure A4-2. Relationships among items i n the data set ( i e . , the sex-linked a r t i f a c t forms) should be i d e n t i f i e d whether .data are coded by presence/absence or by frequencies. - 112 -In addition to the 11 o r i g i n a l a r t i f a c t types, a l l other types of grave goods from the known male and female buria l s are included. The netweight and stone disk categories are separated here because I became uncertain about my i n i t i a l i n t e r p r e t a t i o n at t h i s point. Tool and ornament types that are known to be shared between the known male and female buri a l s are included. They are coded by styl e number whenever possible i n order to determine whether any styles are exclusive to either sex. The shared types that cannot be coded by st y l e are included i n case they are found more l i k e l y to be associated with either sex. The ceramic styles (in o r i g i n a l , unlumped form) from the known male and female b u r i a l s are included as well. The chronological analysis of Chapter 3 indicated that the v a r i a b i l i t y i n style numbers i s at least p a r t l y explained by chronology. Discriminant analysis should be able to determine whether any of the styles are also sex-linked. The types and classes of ceramics or of other a r t i f a c t s which are known to be temporally sensitive are considered i n the interpretation of the res u l t s i f any of these forms are found to be important discriminators of sex. A l l a r t i f a c t types are included which are present i n the known sexed buri a l s and i n at least one unsexed b u r i a l . Some of the a r t i f a c t types shared between the sexes may be surprising to western archaeologists; fanglun (spindle whorl) , zhen (bone needle) , j_I (stone and bone hairpin) , and bihuan (stone and bone bra c e l e t ) . If these were u t i l i z e d i n an R mode analysis s i m i l a r to that by Shephard (1979) and - 113 -Hodson (1977), i n c o r r e c t l y interpreted male and female dimensions could r e s u l t . One of the sexed buri a l s included i n the analysis by simple inspection i s excluded here (E28, female with i n f a n t ) . In case any of the a r t i f a c t s i n the b u r i a l were intended for the infant, E28 w i l l be treated as a multiple b u r i a l for t h i s and a l l subsequent analyses. This problem was not recognized e a r l i e r . 82 of the 85 unsexed single adult b u r i a l s ( l i s t e d i n Figure A4-1) contain at least one of the 12 9 a r t i f a c t types from the known sexed b u r i a l s . The ex-cluded b u r i a l s are E43, E71, and X128. Only those a r t i f a c t types or styles from the subadult male b u r i a l (L117) that are found i n other adult graves of known sex are included. Types or styles found exclusively i n L117 that could r e f l e c t age are excluded. Some discrepancies i n the quantities of t o o l and ornament types were noted from Chapters 4 and 6 of the s i t e report (on tools and ornaments, r e s p e c t i v e l y ) , the summary chart 5 page 131, and from the descriptions of the b u r i a l s . As i n the chronological analysis of Chapter 3, the quantities from the b u r i a l descriptions are employed here. 4.2.6. Discriminant Analysis: Results One discriminant function that separates the known male bur i a l s from the known female b u r i a l s i s derived, accounting for 100% of the variance. I t i s unfortunate that 102 of the - 114 -129 o r i g i n a l variables entered i n the analysis are not a part of the discriminant function. These variables were rejected because they f a i l e d the minimum tolerance t e s t , not reaching the minimum tolerance value i n the program of 0.00100. The tolerance l e v e l of a variable i s "the propor-ti o n of i t s within-groups variance not accounted for by other variables i n the analysis", or 'unique variance 1 (Hull and Nie 1981:293). Discriminant analysis searches for independent variables that discriminate between groups and rejects v a r i -ables that are found to co-vary with others among the known sexed b u r i a l s . Only f i v e of the 12 a r t i f a c t types employed i n the analysis by simple inspection achieved the minimum tolerance l e v e l and were included i n the discriminant function: stone c h i s e l , bone arrowhead, bone spoon or spatula, tusk knife and the horn netweight. Both possible female a r t i f a c t s were rejected (the head and neck ornaments and the small round stone). The difference i n the centroid value for the known male b u r i a l s (3.86 313) from the value for the female group of b u r i a l s (-3.60558) indicates that:thejtwo .groups are d i s t i n c t . The standardized discriminant function.coefficients i d e n t i f y the strongest discriminators as stone c h i s e l , r i n g (II), awl (I), adze (medium-sized I I ) , and tusk knife. F a i r l y strong discriminators are bone arrowhead, bone pointed t o o l ( I I ) , and horn netweight. Four of these variables (the c h i s e l , tusk knife, bone arrowhead and horn netweight) are - 115 -known from the simple inspection analysis to be male discriminators. The standardized discriminant function c o e f f i c i e n t s are l i s t e d i n Figure 4-3. Thirteen (15.9%) of the 82 unsexed buri a l s are c l a s s -i f i e d as male and 69 (84%) as female. That i s , 13 b u r i a l s are more l i k e l y to be male than female on the basis of the male or female discriminating a r t i f a c t s they contain. I t appears that the assignment "female" to the majority of the unsexed b u r i a l s i s by default because these b u r i a l s do not contain any male related a r t i f a c t s . From my exploratory study of the d i s t r i b u t i o n of a r t i f a c t s i n the known sexed graves, I suspect there are no strongly female-related discriminating a r t i f a c t s . The discriminant scores for the 82 unsexed b u r i a l s are shown i n Figure 4-4. None of the disturbed b u r i a l s are i d e n t i f i e d as male (L46, L3, E27, L77). Eight of the 13 males were also i d e n t i f i e d by the simple inspection method. Five of the b u r i a l s (E79, E81, E101, E129, and L124) are i d e n t i f i e d as male on the basis of variables not included i n the analysis by simple inspection. The c l a s s i f i c a t i o n of the unsexed b u r i a l s by the d i s -criminant analysis cannot be considered t o t a l l y r e l i a b l e since i t i s not based on the t o t a l range of a r t i f a c t types i n the known sexed b u r i a l s . Figure 4-4 shows that the discriminant scores for the estimated males and females fluctuate a great deal about the centroids for the known male and female groups of b u r i a l s , i n d i c a t i n g a poor c l a s s i f i c a t i o n - 116 -FIGURE 4-3. Standardized Discriminant Function Coe f f i c i e n t s for the 27 variables that distinguish between the known male and female classes of b u r i a l s . L i s t e d i n order of importance, according to absolute value. 1 adze (medium sized II) -13.61991 2 awl 9.86757 3 stone c h i s e l -8.40071 4 tusk knife 6. 63768 5 ring (II) 6.10280 6 pendant or net weight 4.18537 7 painted t o o l (II) 4.04045 8 arrowhead -3.90091 9 spade (III) 2.42101 10 spade (IV) -2.21648 11 spear (I) -2.13114 12 s i c k l e 2.08805 13 spoon or spatula -1.53598 14 lower jaw bone of pig 1.49133 15 ax (II) 1.43660 16 spade (V) 1.34083 17 bracelet (IV) -0.67213 18 hairpi n (II) -0.49056 19 three legged vessel with bent body(III)-0.48811 20 h a i r t i e -0.40644 21 spindle whorl -0.40418 22 hairpi n (IV) 0.39779 23 pig s k u l l -0.23252 24 hairpi n (I) 0.22943 25 ring (I) -0.19595 26 deer teeth 0.19533 27 bracelet (I) -0.05743 - 117 -FIGURE 4-4. Discriminant Scores for the 82 unsexed, single adult b u r i a l s i n the Discriminant Analysis, Direct Method. ( * = male b u r i a l s i d e n t i f i e d by the Simple Inspection Method also.) Assigned males: * L4 L24 L25 E26 L75 E79 E81 E101 E103 E110 L124 L126 E129 1.3178 4.0476 14.0075 55.3965 6.3347 4.1213 22.4919 0.3587 22.9345 46.0268 3.8262 6.0396 18.1390 * * * * * * Assigned females: L2 L3 L5 E6 E8 E l l E12 E14 E16 E17 E18 E19 E20 E21 L22 E2 3 E27 E29 E32 E33 E38 X40 E41 E42 -4.2893 -3.1088 -7.2403 -4.3630 -3.6990 14.4705 -8.3470 -2.8137 -4.8795 -1.7808 -5.1746 66.4834 -3.1088 -2.8137 -7.9781 -3.6253 -7.8306 -3.7728 -6.3550 -4.9532 16.7576 38.1530 -3.1088 -3.9204 Cont'd... - 118 -FIGURE 4-4 c o n t i n u e d A s s i g n e d f e m a l e s : c o n t ' d . E4 4 -3.8966 E45 -3.4777 L46 -3.1088 L47 -7.5354 E48 -3.1088 E49 -4.6581 E51 -3.1088 E52 -3.9941 E53 -9.7499 E54 -4.8057 E56 -12.6999 E58 -4.3630 L60 -3.1088 E61 -3.1088 E62 -2.9613 E63 -4.8795 L64 -3.4039 E65 -5.6172 E66 -6.9452 E76 -3.6253 L77 -2.9613 E78 -3.6253 E80 -2.8137 X83 -2.6662 E84 -3.4039 E86 -3.7728 E87 -116.2092 E88 -3.6253 E90 -3.7728 L93 -1.5595 L96 -3.8466 E97 -2.9613 L9 8 -1.3382 L100 -2.9613 L104 -10.9292 E106 -2.5186 E108 -3.2564 X113 -2.9613 E116 -4.3630 E118 -8.1994 E119 -4.2155 E120 -3.1088 L127 -11.5932 E132 -1.3382 X133 -3.1088 - 119 -of the unsexed b u r i a l s . The majority of a r t i f a c t types are not a part of the discriminant function because they were found to co-vary with other a r t i f a c t types. The c r i t e r i o n i n the discriminant analysis program that co-varying variables be rejected seems nonsensical i n the present case. I t seems l i k e l y that i n mortuary r i t u a l , mourners would purposely place sets of items r e f l e c t i n g the s o c i a l persona of individuals i n graves. Therefore, there i s probably a tendancy for a r t i f a c t types to co-vary with others for the purpose of symbolic communication, not for a r t i f a c t types to be independent of one another. The p a r t i c u l a r data set used i n t h i s analysis may be of poor qu a l i t y for discriminant analysis as well. Dis-criminant analysis i s best suited for data sets that can meet the assumption of a normal d i s t r i b u t i o n (Sneath and Sokal 1973: 404). The 129 a r t i f a c t types i n the sexed b u r i a l s do not occur with great frequency or with much v a r i a t i o n i n quantity across the b u r i a l sample. 4.3. Conclusion The re s u l t s from the simple inspection method cannot be adequately compared with those from the discriminant analysis due to the r e j e c t i o n of the majority of variables considered p o t e n t i a l l y sex-linked. Only f i v e of the 12 variables from the simple inspection method (with horn netweight separated from the stone disk a r t i f a c t type) are included i n the - 120 -discriminant function that distinguishes the male and female classes. The fact that the majority of male bu r i a l s (eight of 13) i d e n t i f i e d by the discriminant analysis were i d e n t i f i e d by the simple inspection method i s some support for the simpler c l a s s i f i c a t i o n method. I t seems l i k e l y that i f more of the o r i g i n a l (12) a r t i f a c t types had been included i n the d i s c -riminant analysis, the number of male buria l s i d e n t i f i e d by both methods would have increased. For the Dawenkou data set, the simple inspection method has proven more useful than discriminant analysis. But mortuary data sets exhibitingrmoreecohtinuous v a r i a t i o n may fi n d discriminant analysis more suitable and more capable of c l a s s i f y i n g unsexed b u r i a l s . Figure 4-5 l i s t s the bu r i a l s for which sex has been estimated. Their spatial, locations i n the cemetery are depicted i n Figure 4-6. Figure 4-6 shows that males and females were buried i n a l l locations of the cemetery during both time periods. Eighteen estimates considered as " r e l i a b l e " are those with two l i n e s of support: either they were i d e n t i f i e d by the simple inspection method and discriminant analysis, or they were i d e n t i f i e d by the simple inspection method and they are supported by a systematic comparison of sex-linked production tools from Dawenkou Culture s i t e s i n Pearson (1981), The eight males i d e n t i f i e d by both c l a s s i f i c a t i o n methods - 121 -FIGURE 4-5. Known sexed b u r i a l s and estimated sexed b u r i a l s , from the Simple Inspection Method and Discriminant Analysis. Early period (9) known males: 9, 34, 59, 73, 91, 99, 109, 112. (9) known females: 7, 30, 55, 67, 82, 102, 115, 130, 131. (11) r e l i a b l y estimated males: 11, 12, 19, 26, 54, 61, 63, 66, 103, 106, 110. (0) r e l i a b l y estimated females: none. (10) f a i r estimated males: 27, 38, 53, 79, 81, 87, 101, 118, 119, 129. (2) f a i r estimated females: 49, 58. (34) adult b u r i a l s of unknown sex: 6, 8, 14, 18, 20, 21, 23, 29, 32, 33, 41, 43, 44, 45, 48, 51, 52, 56, 62, 65, 71, 76, 78, 80, 84, 86, 88, 90, 97, 108, 116, 120, 132. Late period (5) known males: 15, 122, 123, 125, 117. (4) known females: 10, 72, 105, 121. (6) r e l i a b l y estimated males: 4, 17, 24, 25, 75, 126. (0) r e l i a b l y estimated females: none. (3) f a i r estimated males: 22, 98, 124. (4) f a i r estimated females: 3, 5, 47, 60/ (10) adult b u r i a l s of unknown sex: 2, 16, 46, 64, 77, 93, 96, 100, 104, 127. Undatable b u r i a l s (0) known males: none. (2) known females: 57, 85. (1) r e l i a b l y estimated-males: 40. (0) r e l i a b l y estimated females: none. (0) f a i r estimated males or females: none. (4) adult b u r i a l s of unknown sex: 83, 113, 128, 133. - 122 -FIGURE 4-6. Spatial location of known sexed and estimated sexed b u r i a l s i n the cemetery. I I E A R L Y P E R I O D B U R I A L E 3 L A T E P E R I O D B U R I A L £ 3 U N D A T A B L E B U R I A L M A L E A F E M A L E - 123 -are marked i n Figure 4-5. The following f i v e b u r i a l s are considered " r e l i a b l e " for the l a t t e r reason: E l l , E12, E61, E6 3, and E66. These buria l s were i d e n t i f i e d on the basis of only one "male" a r t i f a c t , and the a r t i f a c t types i n question have been i d e n t i f i e d as male related for other Dawenkou Culture s i t e s . These types are the bone arrowhead and whet-stone. Pearson (1981:1036) concludes that bone arrowheads (as well as bone and stone c h i s e l s , other male-related a r t -i f a c t s i d e n t i f i e d by the simple inspection method) occur exclusively with male b u r i a l s from four Middle Dawenkou period s i t e s ( L i u l i n , Dadunzi, Xixiahou, and Dawenkou). Whetstones occur approximately three times more on the average with males than with females from these s i t e s (Pearson 1981:1085). Five other b u r i a l s are also considered r e l i a b l y "male": E19, E54, E106, L17, and X40. Although i d e n t i f i e d as male by only one c l a s s i f i c a t i o n method (simple inspection), these b u r i a l s contain more than one of the probable male-related a r t i f a c t s (see Figure 4-2). "Fai r " estimated males include the f i v e i d e n t i f i e d only by discriminant analysis (E79, E81, E101, E129, L124) and eight i d e n t i f i e d by the simple inspection method on the basis of one a r t i f a c t type, with no other support from the arch-aeological l i t e r a t u r e on Dawenkou Culture s i t e s . These bu r i a l s are: E27, E38, E53, E118, E119, L22 and L98. Each contains a bone spoon dr..spatula. Since a female b u r i a l at the Xixiahou s i t e (roughly contemporaneous to Dawenkou, Pearson 1981:1081) contains a bone spoon or spatula, t h i s - 124 -a r t i f a c t type may not be male-related (The Shandong Arch-aeological Team 1964:104). Unfortunately, none of the females i d e n t i f i e d by either c l a s s i f i c a t i o n method can be considered " r e l i a b l e " . As explained previously, the majority of females seem to have been i d e n t i f i e d on the basis of the lack of male a r t i f a c t s rather than the presence of c l e a r l y female-related a r t i f a c t s . Six of the known male bu r i a l s also lack the male-related a r t i f a c t s . The six "females" i d e n t i f i e d by both the simple inspection method and the discriminant analysis are considered " f a i r " estimates: L3, L47 (with the ornaments) and E49, E53, L5 and L60 (with the small round stone). Pearson's (1981) comparative study provides additional support for the a r t i f a c t types regarded as male-linked i n the simple inspection and discriminant analyses. Besides the arrowhead, c h i s e l , and whetstone types, the stone and tusk knives are shown to be male-linked, although not exclusively (Pearson 1981:1085). The difference i n the d i s t r i b u t i o n of some a r t i f a c t types between t h i s analysis and that of Pearson (1981) ( i . e . , a r t i f a c t types such as the whetstone that are exclusively male at Dawenkou but shared between the sexes i n other Middle Dawenkou period sites) may be a r e f l e c t i o n of chronological differences among s i t e s . Pearson (1981:1084, from Luo and Zhang 1979) notes a general lack of exclusively female tools at Middle Dawenkou period s i t e s . Females have fewer tools for agricul t u r e , - 125 -hunting and maintenance but greater quantities of needles and spindle whorls ( i b i d ) . One possible reason for t h i s lack of female-related a r t i f a c t s may be that females were lower i n status than males during the Middle Dawenkou Culture period. Pearson (1981:1086) suggests that the status of women may have been declining through time i n the eastern seaboard region as a g r i c u l t u r a l systems became more intensive. This p o s s i b i l i t y i s considered i n Chapter 6. Although the simple inspection method and discriminant analysis c l a s s i f i c a t i o n procedures were somewhat problematic, a greater proportion of "sexed" b u r i a l s i n the b u r i a l populations for the Early, Late and undatable b u r i a l s resulted. If a l l of the estimated buri a l s are accepted, the number of "sexed" adult b u r i a l s i n the Early period i s increased from 18 (known) to 41, with 34 of unknown sex remaining. The increase i n the Late period i s from nine (known) to 22, with 10 bu r i a l s of unknown sex. The undatable sexed adult b u r i a l s increase from two (known) to three, with f i v e unsexed remaining. - 126 -CHAPTER 5 THE NATURE OF SOCIAL SUBGROUP AFFILIATION AT DAWENKOU 5.1. Method This chapter i s an exploratory assessment of s o c i a l sub-group a f f i l i a t i o n represented at the Dawenkou cemetery during the Early and Late periods. Four aspects of mortuary t r e a t -ment noted as pot e n t i a l indicators of s o c i a l subgroup a f f i l i a -t i o n (spatial location of grave, orientation of grave, grave form and body d i s p o s i t i o n , from Binford 1971:22 - see Figure 2-3) are assessed here. Figure 2-3 indicates that grave orientation and grave location are the most l i k e l y to r e f l e c t s o c i a l subgroup a f f i l i a t i o n . Binford regards s o c i a l a f f i l i a -tion as including a variety of membership groups such as clans, kindreds, and lineages (Binford 1971:22). Assessment of v a r i a b i l i t y i n ceramic s t y l e among the b u r i a l s i s also assessed here. The methods that have been used i n the l i t e r a t u r e to assess s o c i a l subgroup a f f i l i a t i o n are li m i t e d . Some researchers have attempted to i d e n t i f y the s p e c i f i c type of s o c i a l subgroup represented at a s i t e : Longacre (1970: re type of descent and residence group), Binford (1972:411-412, type of descent and residence group), Saxe (1971:52, type of residence group), Van de Velde (1979:43, type of lineage), Decker (1969:78-79, type of descent group), and Shennan (1975:286, type of descent - 127 -group). S t u d i e s t h a t attempt to i d e n t i f y s p e c i f i c type of s o c i a l subgroup are c r i t i c i z e d by A l l e n and Richardson (19 71) because the m a t e r i a l c o r r e l a t e s t h a t r e f l e c t d i f f e r e n t types of s o c i a l subgroups are not known. T a i n t e r (1978:131) a l s o notes the l a c k o f c r o s s - c u l t u r a l ethnographic s t u d i e s t h a t i n d i c a t e d i f f e r e n t types o f s o c i a l subgroups by , d i f f e r e n t p a t t e r n s o f m a t e r i a l c u l t u r e . Some i n v e s t i g a t o r s ( G o l d s t e i n 1981, 1980; King 1969, T a i n t e r 1976, Chapman 1981, 1977) propose t h a t c o r p o r a t e descent groups are r e f l e c t e d i n a cemetery on the b a s i s o f the ethnographic t e s t s by Saxe (1970) and G o l d s t e i n (1976). In ot h e r mortuary s t u d i e s , v a r i a b l e s from B i n f o r d ' s (19 71) study t h a t may i n d i c a t e s o c i a l a f f i l i a t i o n are assessed on an i n d i v i d u a l b a s i s . These were w r i t t e n by Peebles (1974, 1971), Gruber (1971), Wright (.1978), King (1969), Whalen (1983), and Brown (1971) . G o l d s t e i n (1981, 1980) u t i l i z e s the s p a t i a l o r g a n i z a t i o n of the s i t e as the o r g a n i z i n g framework f o r an assessment of s o c i a l subgroup a f f i l i a t i o n . T h i s approach i s f o l l o w e d here. The method u t i l i z e d here i s to assess whether there are c o r r e l a t i o n s i n s p a t i a l l o c a t i o n of grave, grave o r i e n t a t i o n , grave form and body d i s p o s i t i o n d u r i n g each p e r i o d a t Dawenkou. F i r s t , s p a t i a l c l u s t e r s of graves w i t h i n the cemetery f o r each p e r i o d are determined. Then, the v a r i a b i l i t y i n grave o r i e n t a -t i o n , grave form and body d i s p o s i t i o n i s assessed f o r each s p a t i a l c l u s t e r o f b u r i a l s . S o c i a l subgroup a f f i l i a t i o n i s expected t o be r e f l e c t e d i f the s p a t i a l c l u s t e r s are d i s t i n -- 128 -guished by p a r t i c u l a r grave o r i e n t a t i o n ( s ) , grave form(s) and body d i s p o s i t i o n ( s ) . Goldstein (1980:3) and Petersen (19 81:185) point out that the d i f f e r e n t i a l location of graves in a cemetery implies d i f f e r e n t i a l treatment of individuals in l i f e . O1Shea's (1981) study suggests that separate b u r i a l areas within a cemetery are suggestive of s o c i a l subgroups but additional l i n e s of evidence are required to make t h i s suggestion more r e l i a b l e (1981:50). I d e n t i f i c a t i o n of contrasting sets of d i s c r e t e l y occurring attributes i s better evidence for the representation of s o c i a l subgroups i n a cemetery, although the c u l t u r a l meaning of these sets may be ambiguous (O'Shea 1981:50-51). Covariation of age and sex with s p a t i a l areas i s also assessed here. However, because many types of s o c i a l subgroups are composed of an age and sex d i s t r i b u t i o n s i m i l a r to that i n the t o t a l mortuary population (O'Shea 1981:50), patterning by age and sex alone i s not taken to r e f l e c t s o c i a l subgroups. An e f f o r t i s also made to assess whether patterning i n the potent i a l s o c i a l subgroup a f f i l i a t i o n - related variables i s a r e f l e c t i o n of time. In the l a s t section of t h i s chapter, propositions regarding the nature of s o c i a l subgroup a f f i l i a t i o n at Dawenkou are offered. Spatial areas of graves within the cemetery are also assessed for composition of ceramic s t y l e . Pearson (1981: 10 86) suggests that correlations of a r t i f a c t styles within the Dawenkou cemetery could represent lineages. I am not - 129 -aware of any mortuary studies since Longacre (19 70) that u t i l i z e ceramic st y l e to assess s o c i a l a f f i l i a t i o n . The assumptions and methods i n Longacre (19 70) are c r i t i c i z e d by A l l e n and Richardson (19 71) . Since the rela t i o n s h i p between ceramic style and s o c i a l a f f i l i a t i o n does not have support from c r o s s - c u l t u r a l ethnographic studies, a lack of co r r e l a t i o n of ceramic style with s p a t i a l location of grave i s not to be taken as a lack of representation of s o c i a l a f f i l i a t i o n i n the cemetery. The d e f i n i t i o n of location of grave used i n t h i s assessment d i f f e r s s l i g h t l y from that i n Binford's (1971) study. Binford's (1971:21) d e f i n i t i o n i s "whether the f a c i l i t y was d i f f e r e n t i a l l y placed i n the l i f e space of the community, or i n s p a t i a l l y d i f f e r e n t i a t e d b u r i a l locations". Orientation of grave i n degrees from north i s assessed as well as four types of grave form: simple p i t , p i t with ercengtai ("second l e v e l platform", or ledge), log tomb, and log tomb with second l e v e l platform. Binford's (1971: 21) d e f i n i t i o n of body d i s p o s i t i o n i s " d i s t i n c t i o n s made by d i f f e r e n t i a l d i s p o s i t i o n - placed i n a grave, on a s c a f f o l d , disposed of i n the r i v e r , etc.". In t h i s assessment, a modified d e f i n i t i o n of body di s p o s i t i o n i s employed because the s i t e report describes v a r i a b i l i t y i n terms of the di s p o s i t i o n of skeletons within graves (whether supine or prone, for example). - 130 -5.2. Spatial Location of Graves The only methods that have been employed to derive s p a t i a l patterning within a cemetery of which I am aware are v i s u a l inspection to i d e n t i f y rows of b u r i a l s and l i n e a r regression to check the rows (Goldstein 1981, 1980), nearest-neighbor analysis (Tainter 19 76, Peebles 1971, Shennan 19 75, King 1969, Whalen 1983) and v i s u a l inspection alone (Cole and Harding 1979, MacDonald 1980). The method that I consider most appropriate for the Dawenkou cemetery i s v i s u a l inspec-t i o n . Linear regression i s not appropriate because the v i s u a l l y d i s t i n c t groups of b u r i a l s i n the cemetery are more concentric i n form than l i n e a r . Nearest-neighbor analysis has been considered problematic for mortuary s i t e s with dense groups of b u r i a l s (Goldstein 1981:58). However, there i s a more fundamental reason for deriving s p a t i a l groups of b u r i a l s by v i s u a l inspection. The s i t e map shows some v i s u a l l y d i s t i n c t c l u s t e r s of b u r i a l s from each period. Although some cl u s t e r s are not t o t a l l y clear and may be subject to d i f f e r e n t i n t e r p r e t a t i o n by d i f f e r e n t researchers, clusters apparent by v i s u a l inspection are those that are most l i k e l y to be c u l t u r a l l y s i g n i f i c a n t . As Goldstein (19 81:5 8) points out: I t i s reasonable to assume that the s p a t i a l p r i n c i p l e s used by a society w i l l be f a i r l y d i s t i n c t and apparent. I t i s probable that rules for placement of individuals w i l l have been followed consciously by the members of a society (especially i n a cemetery, barrow or mound), and i t i s u n l i k e l y that these - 131 -people had computers or random-number tables to a s s i s t them in developing a v i s u a l l y incomprehensible pattern. Clusters of b u r i a l s i d e n t i f i e d by v i s u a l inspection for the 85 Early period graves are shown i n Figure 5-1. There are f i v e roughly concentric groups of b u r i a l s l a b e l l e d numerically, f i v e pairs of b u r i a l s l a b e l l e d by l e t t e r s , and three i s o l a t e d b u r i a l s (E48, E14, and E36). The c l u s t e r s of b u r i a l s i d e n t i f i e d for the 33 Late period graves are shown i n Figure 5-2. Four c l u s t e r s of b u r i a l s are l a b e l l e d numerically. The locations of the 15 undatable b u r i a l s can be seen i n Figure 3-4. Two undatable b u r i a l s , X12 8 and X92, are s p a t i a l l y i s o l a t e d from other b u r i a l s i n the cemetery. 5.3. Correlation of Grave Orientation, Grave Form and Body Disposition with Grave Location The majority of Early and Late period and undatable bu r i a l s are oriented roughly east-west. An attempt i s made to determine whether the minor differences i n orientation among b u r i a l s are c u l t u r a l l y s i g n i f i c a n t as suggested from the map of the cemetery. The two exceptions to the east-west trend are E45 (270°) and X12 8 (oriented north-south). Grave orientation i n the Early period ranges from 43° -2 70° from north. There are no sharp breaks i n the d i s t r i -bution of values for the majority of b u r i a l s . From the map of the cemetery, i t appears there are three types of orienta-t i o n : east-west, northeast, and southeast. - 132 -FIGURE 5-1. Spatial groups of b u r i a l s derived by visua inspection, Early Period. CD E A R L Y P E R I O D B U R I A L EH L A T E P E R I O D B U R I A L £3 U N D A T A B L E B U R I A L - 133 -FIGURE 5-2. Spatial groups of b u r i a l s derived by visua inspection, Late Period. I I E A R L Y P E R I O D B U R I A L EH L A T E P E R I O D B U R I A L 2 3 U N D A T A B L E B U R I A L - 134 -As Figure 5-3 indicates, northeast orientation i s a r b i t r a r i l y defined here as 43° - 80°, east-west as 83° -100°, and southeast as 102° - 122°. The skeletons l i e with t h e i r s k u l l s to the east (as shown i n Figure 5-3) . The only exception to th i s rule i s E45 with the orientation of 270°. The skeleton's head i n E45 i s towards the west. Figure A5-1 shows the d i s t r i b u t i o n of orientation for the Early period graves. Figure A5-2 shows the d i s t r i b u t i o n for the Late period b u r i a l s . The majority of Early period b u r i a l s are oriented east-west (52, or 61.2%), while 11 (12.9%) are oriented northeast and 22 (25.9%) southeast. The orienta-tions of some b u r i a l s look s l i g h t l y d i f f e r e n t on the s i t e map than one would expect from the orientation i n degrees given with the descriptions of the b u r i a l s . I assume that the values i n degrees are the more accurate description of grave orientation. The d i s t r i b u t i o n of the three a r b i t r a r i l y defined types of orientation within the various s p a t i a l groups of Early b u r i a l s i s shown i n Figure 5-4. The three types of orienta-tion do not tend to be d i s t r i b u t e d exclusively in d i f f e r e n t s p a t i a l locations. I acknowledge that d i f f e r e n t d e f i n i t i o n s of an east-west, northeast and southeast orientation could r e s u l t i n a s l i g h t l y better (or worse) co r r e l a t i o n of orientation with s p a t i a l l o c a t i o n . The lack of co r r e l a t i o n noted here between orientation type and s p a t i a l location could indicate the orientation types are c u l t u r a l l y i n s i g n i f i -cant or that orientation i s not a r e f l e c t i o n of s o c i a l - 135 -FIGURE 5-3.- The range of grave orientations for the Early, Late, and Undatable b u r i a l s . Orientation varies from 43°-270°. o° 180° ,1 Three types of o r i e n t a t i o n , a r b i t r a r i l y defined (by eye): NE (43° - 80°) E-W (82° - 100°) SE (102° - 125°) Unique o r i e n t a t i o n s : B u r i a l E45 at 270° B u r i a l X128 at 180° See text f o r explanation. FIGURE 5-4. Correlation of grave form, grave orientation, body d i s p o s i t i o n , age and sex with s p a t i a l location of grave (as i n Fig. 5-1), Early period. (Intrusive pairs of buria l s indicated i n F i g . A3-2). KEY overleaf. location grave form grave orientation body disposition age sex SP ER LT TE NE E-W SE 270° S LS RS P A C U RM M FM FF F MU 14 1 1 1 1 1 36 1 1 1 1 1 48 1 1 1 1 1 Pair A 1 1 1 1 2 2 1 1 Pair B 2 2 2 2 1 1 Pair C 1 1 2 1 1 2 1 1 Pair D 2 1 1 2 2 2 Cluster 1 (8 t o t a l : 1 multiple 1 no skel-eton ) 4 3 1 8 8 8 1 4 2 1 1 1) M,F Cluster 2 (12 total) 11 1 3 8 1 8 1 1 1 12 11 1 Cont'd. FIGURE 5-4 continued location grave form grave orientation body disposition age sex SP ER LT TE NE E-W SE 270° S LS RS P A C U RM M FM FF F MU Cluster 3 (4 total) 3 1 4 3 1 4 2 1 1 Cluster 4 (14 t o t a l : 2 multiple) 12 2 1 6 7 16 15 1 12 1 1 1 1 1) F,C 2) ? ? Cluster 5 (34 t o t a l : 3 multiple) 26 4 4 2 17 15 34 3 1 34 4 17 4 7 5 5 DM, F,C 2) ? ? 3) M,F SP simple p i t ER - second l e v e l platform LT log tomb TE - log tomb and second l e v e l platform NE = northeast orientation E-W = east-west orientation SE = southeast orientation S = supine LS l e f t side RS = right side P prone A = adult > C = c h i l d i SU = subadult U — unsexed RM = r e l i a b l y estimated male M = known male FM — f a i r estimated male FF — f a i r estimated female F = known female MU — multiple b u r i a l - 138 -a f f i l i a t i o n i n the E a r l y p e r i o d . The m a j o r i t y o f the 85 E a r l y p e r i o d b u r i a l s are simple p i t s (.68, or 80%), 10 (11.8%) are p i t s w i t h second l e v e l p l a t f o r m s , f o u r are l o g tombs (4.7%) and three (3.5%) are l o g tombs wit h second l e v e l p l a t f o r m s . E a r l y p e r i o d grave form types are shown i n F i g u r e 5-4. I t i s apparent t h a t there i s no c o r r e l a t i o n between grave form and grave l o c a t i o n : each type of grave form i s found i n a v a r i e t y o f s p a t i a l l o c a t i o n s . Four types of s k e l e t a l l a y o u t or body d i s p o s i t i o n occur among the E a r l y p e r i o d b u r i a l s : supine, on the r i g h t s i d e , on the l e f t s i d e , and prone. The m a j o r i t y of b u r i a l s are supine (74, or 88.1%), f o u r are on the r i g h t s i d e (4.8%), f i v e on the l e f t (6.1%), and one i s prone (1.2%). Grave E54 does not c o n t a i n a s k e l e t o n . There i s some v a r i a b i l i t y i n the placement of the hands and f e e t among a l l b u r i a l s i n the cemetery, but i t seems l a r g e l y due to type of s k e l e t a l l a y o u t . Only the more obvious v a r i a t i o n i n s k e l e t a l l a y o u t i s c o n s i d e r e d here. There i s a l s o v a r i a b i l i t y among a l l b u r i a l s i n the d i r e c t i o n i n which the s k u l l faces (north, south or e a s t , i n one c a s e ) . T h i s v a r i a b i l i t y i s not assessed i n t h i s chapter because the d i r e c t i o n o f s k u l l i s not d e s c r i b e d f o r more than h a l f of the b u r i a l s . A l s o , the v a r i a b i l i t y may be due to s k e l e t a l l a y o u t or p o s t - d e p o s i t i o n a l d i s t u r b a n c e o f s k e l e t o n s . The ten E a r l y graves i n which s k e l e t a l p o s i t i o n i s oth e r than supine are - 139 -shown in Figure 5-4. I t appears- from Figure 5-4 that s k e l e t a l layout i s not correlated with s p a t i a l location of grave. Thus, none of the three potential s o c i a l a f f i l i a t i o n -related variables, orientation, grave form, and body d i s p o s i -tion i s correlated with s p a t i a l location of grave i n the Early period. Figure 5-4 also indicates that orientation, grave form, and body d i s p o s i t i o n are not correlated with each other, either. The only exceptions are E45 and E14. In E45, the unique orientation (2 70°) i s correlated with the unique s k e l e t a l layout (prone). This distinguished mortuary t r e a t -ment could r e f l e c t s o c i a l deviance (O'Shea 19 81:43) or a feared or unusual cause of death (Ucko 1969:271). E14 i s the only b u r i a l with an orientation less than 70°,and i t i s i n an i s o l a t e d s p a t i a l l o c a t i o n . Since v i s u a l inspection (see Figure 5-4) indicates there i s c l e a r l y no covariation between any of the Early period p o t e n t i a l s o c i a l a f f i l i a t i o n - r e l a t e d variables, I maintain that s t a t i s t i c a l tests of association are unnecessary. Chi square has been used (Peebles 19 71, 19 74) to test for an association between pote n t i a l s o c i a l a f f i l i a t i o n - r e l a t e d variables. I t i s also not necessary to show with a Goodman-Kruskal s t a t i s t i c (MacDonald 1980:42-44) or a Kolmogrov-Smirnov t e s t (Peebles 19 74:94) that orientation i s not random. As Goldstein (19 81:58) points out, "One need not prove with s t a t i s t i c s that which i s apparent". - 140 -Figure 5-5 shows a lack of co r r e l a t i o n between orientation, grave form and body d i s p o s i t i o n with the s p a t i a l groups of b u r i a l s for the 33 Late period b u r i a l s as well. Orientation i s more homogeneous i n the Late period, ranging from 80° to 114°. Using the a r b i t r a r i l y defined types of orientation, one b u r i a l (3 .0%) i s oriented northeast, f i v e southeast (15.2%) and 2 7 (81.8%) east-west. The southeastern oriented b u r i a l s are not r e s t r i c t e d to a certain s p a t i a l loca-t i o n . Burials having a simple p i t (25, 75 .8%) , a p i t with a second l e v e l platform (1, 3.0%), log tomb ( 3 , 9.1%), and log tomb with second l e v e l platform (4, 12.1%) are not r e s t r i c t e d to p a r t i c u l a r s p a t i a l locations. Only three types of s k e l e t a l layout are found i n the Late period: supine (2 8, 9 3 . 3 % ) , r i g h t side (1, 3.3%) and l e f t side (1, 3 . 3 % ) . Burials L24, L60 and L127 do not contain a skeleton. Thus, s k e l e t a l layout i s more homogeneous i n the Late period as well as orientation. The b u r i a l with the skeleton on the l e f t side, L117, i s i n a d i f f e r e n t s p a t i a l group than the b u r i a l on the r i g h t side, L25. Although both b u r i a l s have log tombs with second l e v e l platforms, other b u r i a l s with t h i s type of grave form have the supine s k e l e t a l layout type. As i n the Early period, there i s no co r r e l a t i o n between orientation, grave form, and body d i s p o s i t i o n . For the undatable b u r i a l s , the association of orienta-t i o n , grave form, and body d i s p o s i t i o n with s p a t i a l location cannot be properly assessed because t h e i r period and s p a t i a l groups are unknown. However, information in Figure 5-6 FIGURE 5-5. Correlation of grave form, grave orientation, body d i s p o s i t i o n , age and sex with s p a t i a l location of grave (as i n Fig. 5-2), Late period. location grave form grave orientation body disp o s i t i o n age sex SP ER LT TE NE E-W SE S LS RE P A SU -> U RM M FM FF F MU Cluster 1 (3 total) 2 1 3 3 3 1 2 Cluster 2 (15 t o t a l : 1 multiple 2 no ske l -eton) 12 2 1 1 12 2 13 1 14 4 4 2 1 4 1 DM, F Cluster 3 (3 t o t a l : 1 no skel-eton) 1 1 1 2 1 1 1 1 1 1 1 1 Cluster 4 10 1 1 10 2 12 12 4 1 3 2 2 KEY: Refer to F i g . 5-4 FIGURE 5-6. Correlation of grave form, grave orientation, body d i s p o s i t i o n , age and sex with s p a t i a l location of grave, Undatable b u r i a l s . KEY F = female RM = r e l i a b l y estimated male Unlabelled: simple p i t s , supine b u r i a l s , adults and unsexed b u r i a l s . Burials i n rough order from northern section of cemetery to southern section: 70 E-W 37 E-W 39 E-W 40 E-W 50 E-W 57 E-W 83 E-W 68 E-W 74 E-W 85 SE 95 SE 113 SE 133 SE (90 (86C (84( (90C (88C (87C (92C (84C (91* (110 (110C (125C (124C unsexed multiple c h i l d c h i l d RM, l e f t side c h i l d F c h i l d c h i l d F c h i l d flexed Isolated graves 92 SE (110°) , unsexed multiple 128 N-S (180 ) - 143 -regarding these b u r i a l s does not contradict the conclusion that the poten t i a l s o c i a l subgroup a f f i l i a t i o n variables are not associated with s p a t i a l location. Both southeast and east-west oriented b u r i a l s are located i n more than one s p a t i a l l o c a t i o n . However, the one north-south oriented grave (X12 8) i s s p a t i a l l y i s o l a t e d . Some s p a t i a l groups have more than one type of body di s p o s i t i o n represented as well. A l l of the undatable b u r i a l s have the same form of grave: the simple p i t . 5.4. Correlation of Ceramic Style with Spatial Location For both periods, there i s l i t t l e c o r r e l a t i o n of ceramic style with s p a t i a l location of grave. Figure A5-3 shows that of 62 ceramic styles that appear i n more than one b u r i a l i n the Early period, only f i v e (8.1%) are found exclusively in one s p a t i a l area (clusters 5, 4, 1). None of these ceramic styles i s found i n every b u r i a l i n these c l u s t e r s . Thus, ceramic st y l e does not appear to be a r e f l e c t i o n of s o c i a l a f f i l i a t i o n in the Early period. Figure A5-4 shows that of 66 ceramic styles that appear i n more than one b u r i a l i n the Late period, only 11 (16.7%) are exclusive to a p a r t i c u l a r s p a t i a l area. None of these styles is-, found i n s p a t i a l c l u s t e r 2, and two i n cl u s t e r 4. As i n the Early period, none of these styles i s found i n a l l the b u r i a l s within the p a r t i c u l a r s p a t i a l areas involved. Although the percentage of ceramic styles exclusive to a s p a t i a l area has doubled from the Early to the Late period, 16.7% i s probably not a high enough figure to propose that ceramic - 144 -style i s a r e f l e c t i o n of s o c i a l a f f i l i a t i o n i n the Late period, either. However, a trend towards the demarcation of s p a t i a l area by ceramic style may be developing. The Late period styles are associated with a variety of subtypes and functional s t y l e s . It i s not clear whether a trend towards the development of symbolism of s o c i a l a f f i l i a t i o n by ceramic s t y l e i s occur-ring i n the Late period and whether t h i s trend pertains to certai n s p a t i a l areas of the cemetery (e.g., Cluster 2) more than others. Due to time l i m i t a t i o n s , a study of the d i s t r i -bution of production tool and ornament styles i n the various s p a t i a l areas of the cemetery cannot be made here. Pearson (19 81:10 86) suggests that a study involving the styles of more than one type of a r t i f a c t could r e f l e c t d i f f e r e n t lineages within the cemetery. An additional d i s t r i b u t i o n a l study with t o o l types and ornaments could provide support for the possible trend indicated by ceramic s t y l e . Study of the s p a t i a l d i s t r i -bution of a r t i f a c t styles i n other Dawenkou culture s i t e s could indicate whether the trend i s a regional one. A study of the d i s t r i b u t i o n of ceramic design attributes in the Dawenkou s i t e and i n other Dawenkou Culture s i t e s i n Shandong could be useful (M. Blake, personal communication 1983). These design elements could symbolize s o c i a l a f f i l i a -t i o n within the region, as suggested i n Wobst's (1977) information exchange model (Braun 1977:119). A d i s t r i b u -t i o n a l study of ceramic design attributes could be useful - 145 -for understanding Dawenkou Culture regional s o c i a l organiza-t i o n . A study of design elements would necessitate v i s u a l inspection of the ceramics from the various Dawenkou Culture s i t e s . The design elements are not clear from published reports. The Dawenkou s i t e report indicates that there may be some differences i n painted design as well as in c i s e d and impressed design. 5.5. Interpretation of the Potential Social Subgroup A f f i l i a t i o n - Related Variables 5.5.1. Introduction The lack of c o r r e l a t i o n between orientation of grave, grave form, body d i s p o s i t i o n , and ceramic s t y l e with s p a t i a l location indicates no additional support for the p o s s i b i l i t y that s p a t i a l groups of b u r i a l s from the Early and Late periods r e f l e c t s o c i a l subgroup a f f i l i a t i o n . For reasons discussed below, i t i s also not l i k e l y that the v a r i a b i l i t y i n eithe r orientation, grave form, or body d i s p o s i t i o n alone r e f l e c t s s o c i a l subgroup a f f i l i a t i o n . I propose these variables r e f l e c t other aspects of mortuary r i t u a l . 5.5.2. Orientation Figure 2-3 indicates that o r i e n t a t i o n of grave may r e f l e c t sex as well as s o c i a l a f f i l i a t i o n . The known sexed and aged b u r i a l s from the Early and Late periods are shown i n Figures 5-4 and 5-5. In the Early period, three (.30%) of the 10 known females from single graves have a northeast orientation, f i v e (50%) have an east-west one, and two (20%) have a southeast one. One (1.1%) of the nine - 146 -known males from single graves has- a northeast orientation, four (4.4%) have an east-west one, and four (4.4%) have a southeast one. Inclusion of the r e l i a b l y and f a i r l y estimated males and two females from Figure 4-5 res u l t s i n the follow-ing d i s t r i b u t i o n : northeast: four males (13.3%), three females (25%), east-west: 17 males (56.7%), seven females (58.3%), and southeast: nine males (30%), two females (16.7%). I t i s apparent that sex and orientation are not associated. In the Late period, none of the fi v e known males are oriented northeast, four (80%) east-west, and one (20%) south-east. None of the four known females i s oriented northeast, three (75%) east-west, and one (25%) southeast. I f the r e l i a b l y and f a i r l y estimated males and females are included, the d i s t r i b u t i o n i s : northeast: male (none), female (1, 12.5%), east-west: male (12, 85.7%), female (5, 62.5%), and southeast: male (2, 14.3%), female (2, 14.3%). Thus, sex and orientation are not associated for the Late period b u r i a l s , e i t h e r . The l i m i t e d data on age of skeleton for both periods does not allow an assessment of whether differences i n orientation are a function of age. A l l that i s known i s that two of the four children have graves oriented south-east and two, east-west. The one subadult i n the Late period i s oriented east-west. - 147 -The pairs of in t r u s i v e b u r i a l s within the Early and Late periods can be u t i l i z e d to assess whether differences i n orientation are a function of time. Figure 5-4 shows that within the Early period, three of the seven pairs of in t r u s i v e b u r i a l s have the same orientation (east-west), and three have d i f f e r e n t types of orienta t i o n . In Figure 5-5, both b u r i a l s in the in t r u s i v e Late period pair have the same orientation. If differences i n orient a t i o n are a function of time, i t would be expected that the older b u r i a l s i n each pair of i n t r u s i v e b u r i a l s would have a d i f f e r e n t o r i e n t a t i o n than the younger b u r i a l s . A more l i k e l y explanation for differences i n orientation in the Early and Late periods i s season of interment. This proposition has been made for other mortuary s i t e s by Gruber (1971:67) and Saxe (19 71:49-50). Gruber (1971:67) proposes that the s l i g h t variations i n pattern of east-west orientation at the Mohr s i t e i n Pennsylvania are due to the pos i t i o n of the sun on the horizon at dawn. The mortuary r i t u a l c a l l e d for interment of graves i n an east-west d i r e c t i o n so that the heads of the deceased faced the r i s i n g sun i n the east. The differences i n orientation are due to the d i f f e r e n t positions of the r i s i n g sun at d i f f e r e n t seasons ( i b i d ) . When Gruber (.1971:6 7) calculated the position of the r i s i n g sun from season to season, he found that orientations of almost a l l graves i n the cemetery f e l l within the solar arc. Some ethnographic groups have been known to place b u r i a l s with respect to the r i s i n g sun (Gruber 1971:71). Saxe (1971:49-50) - 148 -concludes t h a t some b u r i a l s of the Wadi H a i f a s i t e i n Sudan are o r i e n t e d a c c o r d i n g to the p o s i t i o n of the r i s i n g sun from season to season and o t h e r s , a c c o r d i n g to the p o s i t i o n of the sun a t o t h e r times of day. Gruber's (.1971:73) c a l c u l a t i o n s f o r the Mohr s i t e i n d i c a t e t h a t summer interments range from 69° - 100° from n o r t h and w i n t e r interments range from 101° to 131°. I t i s p o s s i b l e t h a t a c a l c u l a t i o n of the seasonal p o s i t i o n of the sun a t Dawenkou may r e s u l t i n a s i m i l a r d i s t r i b u t i o n . The o r i e n t a t i o n s I i n t e r p r e t as n o r t h e a s t (43° - 80°) and e a s t -west (82° - 100°) may r e f l e c t summer interments and those which are southeast (102° - 125°) may r e f l e c t w i n t e r i n t e r -ments. I f the above i n t e r p r e t a t i o n i s accepted, graves were i n t e r r e d i n a l l l o c a t i o n s of the cemetery r e g a r d l e s s of the season. The east-west, n o r t h e a s t and southeast o r i e n t e d graves are found i n a l l s p a t i a l l o c a t i o n s d u r i n g the E a r l y and Late p e r i o d s (and f o r the undatable b u r i a l s as w e l l ) . The two e x c e p t i o n s are b u r i a l s E45 (2 70°) and X12 8 (180°). I f the season o f interment i s the e x p l a n a t i o n f o r v a r i a b i l i t y i n o r i e n t a t i o n , the g r e a t e r c o n s i s t e n c y i n the Late p e r i o d o r i e n t a t i o n s may be a f a c t o r of more b u r i a l i n the summer. However, one would expect a f a i r number of i n d i v i d u a l s to d i e d u r i n g the w i n t e r . Seasonal p o s i t i o n o f the r i s i n g sun may a l s o e x p l a i n the v a r i a b i l i t y i n grave o r i e n t a t i o n a t other Dawenkou C u l t u r e s i t e s . Pearson ( i n press:22-23) mentions three other Dawenkou C u l t u r e s i t e s i n Shandong wi t h predominantly east-west - 149 -orientations: Yedian (Late Dawenkou Culture, 90° - 116°), Dafanzhuang (Late Dawenkou Culture, 70° - 9 0°) and Wangyin (fourth layer, Early Dawenkou Culture, 70° - 128°). The bu r i a l s i n the late s i t e of Xixiahou have predominantly an east-west orientation (Shandong Archaeological Team 1964:58). The majority of graves from the Dadunzi s i t e i n Jiangsu also have an east-west orientation (Pearson i n press:22). However, there i s some v a r i a b i l i t y for s i t e s located in d i f f e r e n t regions and of varying time periods. Two s i t e s e a r l i e r i n time than Dawenkou (see Figure 1-3) have b u r i a l s oriented north-south: Songze i n Zhejiang and L i u l i n i n Jiangsu. The early s i t e of Beiyinyanying has a northeast pattern ( a l l from Pearson in press: 22-23). Thus, there are differences i n orientation pattern among the Dawenkou Culture s i t e s which have not been explained. At some s i t e s , b u r i a l s may have been oriented according to d i s t i n c t topographic features v i s i b l e from the cemetery. Kao (19 83:14) proposes that b u r i a l s at the Qianzhai s i t e i n Shandong (Late Dawenkou Culture) were oriented with respect to a mountain range located southeast from the s i t e . 5.5.3. Grave Form Figure 2-3 indicates that grave form may r e f l e c t condi-tion of death, age, and s o c i a l position as well as s o c i a l a f f i l i a t i o n . Condition of death can be proposed for f i v e b u r i a l s . B u r i a l E2 8 (female with infant) may represent a woman who died i n c h i l d b i r t h . As mentioned i n Chapter 4, the b u r i a l s with no skeleton (E54, L24, L60 and L127) may - 150 -represent individuals who died away from the community. The sparse data on s k e l e t a l age indicates no apparent association between age and grave form. As F i g . 5-4 and F i g . 5-5 ind i c a t e , both subadults and adults have simple p i t s , log tombs, and second l e v e l platforms. A l l of the undatable b u r i a l s have simple p i t s . Figure 5-4 indicates that three pairs of intru s i v e b u r i a l s i n the Early period have d i f f e r e n t grave forms: E54 and E58, E9 and E23, and E78 and E129. The one pair of intrusive b u r i a l s i n the Late period has the same grave form represented. Thus, i t appears that grave form i s not a r e f l e c t i o n of time differences among b u r i a l s . Figures 5-4 and 5-5 demonstrate that grave form i s not associated with sex, either. Of the known females i n the Early period, eight (80%) have simple p i t s and two (20%) have p i t s with second l e v e l platforms. Of the nine known males i n the Early period, four (44.4%) have simple p i t s , three (33.3%) have p i t s with second.level platforms, and two (22.2%) have log tombs. Inclusion of the f a i r l y and r e l i a b l y estimated males and two females results i n the d i s t r i b u t i o n : p i t s : male (20 or 66.7%), female (12 or 75%); p i t s with second l e v e l platforms: male (six, 20%), female (three, 25%); log tombs with second l e v e l platforms: (two males, 6.7%); and log tombs: (two males, 6.7%). For the Late period, four of the fi v e known males (80%) have simple p i t s and one has a log tomb with second l e v e l platform (20%). Inclusion of the f a i r l y and r e l i a b l y - 151 -estimated males indicates: 10 with p i t s (71.4%), one (7.1%) with a second l e v e l platform, two (14.3%) with a log tomb and second l e v e l platform, and one (7.1%) with a log tomb. Three (75%) of the known females have p i t s and one (25%) has a log tomb with second l e v e l platform. The incl u s i o n of the four f a i r l y estimated females gives: f i v e (62.5%) with p i t s , one (12.5%) with a log tomb and second l e v e l platform and two (25%) with a log tomb. I am only aware of one other Dawenkou Culture s i t e with the log tomb, a recently discovered s i t e i n Shanxi ( Zou, personal communication 19 83). Only one Dawenkou Culture s i t e appears to have the second l e v e l p l a t -form - the late s i t e of Qianzhai i n Shandong (Kao 19 83:14). 5.5.4. Body Disposition Body d i s p o s i t i o n may r e f l e c t condition of death, location of death, age, or s o c i a l p o s i t i o n as well as s o c i a l a f f i l i a -t i o n . Condition of death cannot be properly assessed since four of the f i v e b u r i a l s that may r e f l e c t condition of death do not contain a skeleton. A d i f f e r e n t location of death has been proposed for the b u r i a l s with no skeleton. Body di s p o s i t i o n cannot be assessed for association with age i n either the Early or Late periods due to the limited sample of aged b u r i a l s . However, both children and adults i n the Early period have supine positions, and both adults and the one youth i n the Late period are positioned on t h e i r sides. Child and adult undatable b u r i a l s have supine positions. The c u l t u r a l s ignificance of the one flexed burial, i n the cemetery CX133) and the one prone b u r i a l (E45) i s not - 152 -apparent. The Wangyin s i t e has a few prone b u r i a l s with heads to the west, l i k e E45. The Wangyin b u r i a l s may r e f l e c t cause of death or a special s o c i a l group (Shandong Archaeological Team 19 79).. Since each b u r i a l i n the intru s i v e pairs of b u r i a l s from the Early and Late periods has a supine p o s i t i o n , the i n t r u -sive b u r i a l s do not indicate that there i s no association between time differences within each period and body d i s p o s i -t i o n . Body d i s p o s i t i o n may be related to sex. For the Early period, none of the known females or f a i r l y estimated females have a body position other than supine. The same holds true for the known and f a i r l y estimated females i n the Late period. Of the eight b u r i a l s with a side body position i n the Early period (either l e f t or r i g h t ) , three (33.3%) are known males and two (20%) are f a i r l y estimated males. Three of the bu r i a l s with a side body position are of unknown sex. Of the two bu r i a l s with a side position in the Late period, one (.20%) i s a known male and one (16.7%) i s a r e l i a b l y estimated male. The one undatable b u r i a l with a side position i s a r e l i a b l y estimated male. Since the side position i s not a ch a r a c t e r i s t i c of a l l males i n the Early or Late period, some other aspect of mortuary r i t u a l must be r e f l e c t e d as well. F i n a l l y , the reason for the greater homogeneity i n sk e l e t a l layout i n the Late period than i n the Early period i s not apparent. The supine position i s most common to Qingliangang cultures (Pearson in press:18). At the Early Dawenkou - 153 -s i t e of Wangyin, the majority of b u r i a l s are supine and a few are on t h e i r sides such as at Dawenkou (Shandong Archaeological Team 19 79). The Wangyin s i t e exhibits v a r i a b i l i t y i n body d i s p o s i t i o n not seen elsewhere: a large secondary b u r i a l p i t (Pearson 1981:1985). The Nanjing Museum (1978) reports that b u r i a l s from the northern Yangtze r i v e r area, including the area of the Dadunzi s i t e are predominantly supine. In the southern area including the Shanghai region, b u r i a l s tend to have a prone position ( i b i d ) . Wu (19 73) reports that the following s i t e s have b u r i a l s with mainly a supine pos i t i o n : Dadunzi (the Qingliangang stratum, L i u l i n and Huating s t r a t a ) , the L i u l i n stage of the L i u l i n s i t e , and Xixiahou. Another Early Dawenkou Culture s i t e with mainly the supine position i s Beiyinyanying i n Jiangsu (Chang 19 79: 164). Another s i t e that exhibits v a r i a b i l i t y i n body d i s p o s i -t i o n i s Qianzhai, with over 50 ash-pits (Kao 1983:114). 5.6. Concluding Propositions 5.6.1 Argument for a Descent Group at Dawenkou I t was seen that each s p a t i a l area determined by v i s u a l inspection for the Early and Late periods i s not character-ized by d i s t i n c t grave orientation, grave form, body d i s p o s i -t i o n , and ceramic s t y l e . The extent of v a r i a b i l i t y i n the above four aspects of mortuary treatment i s exhibited i n each s p a t i a l group. Therefore, the cemetery as a whole can be considered as consistent i n the above aspects of mortuary treatment. This consistency i n b u r i a l treatment from s p a t i a l - 154 -area to s p a t i a l area suggests that the whole cemetery r e f l e c t s one type of s o c i a l subgroup. I propose the b u r i a l s from both periods represent a descent group. Due to the o v e r a l l consis-tency of mortuary r i t u a l i n terms of orientation, grave form, body d i s p o s i t i o n , and s p a t i a l location of grave, I propose that the same type of descent group i s represented throughout the duration of the cemetery's use. Since test implications for a descent group versus another type of s o c i a l subgroup have not been developed, the proposition i s tentative. My argument i s that a descent group i s possible, given the importance of descent to a g r i c u l t u r a l ethnographic s o c i e t i e s . Also descent systems were an important part of Shang s o c i o - p o l i t i c a l organization. H i s t o r i c a l data indicate the presence of ranked lineages (Chang 1980:78-79) and the importance of ancestor worship (Thorp 1979:152). The importance of descent i n mortuary r i t u a l may be another c u l t u r a l t r a i t that continues from the Dawenkou Culture to the Shang dynasty. Chang (1979: 110) claims that ancestor worship was established i n China by the Longshan period. Pearson (19 81:10 86) proposes that lineages may be r e f l e c t e d at Dawenkou on the basis of the furnished graves with no skeleton. Also, an increasing importance of lineage by the Late Dawenkou period may be indicated by the s p a t i a l segretation of some bu r i a l s composed of men, women and children ( i b i d ) . Ethnographic tests by Saxe (1970) and Goldstein (1976) indicate that cemeteries and other discrete disposal areas tend to r e f l e c t corporate descent groups (Goldstein 1980:7-8). - 155 -However, more ethnographic tests are necessary to substantiate t h i s r e l a t i o n s h i p (Chapman and Randsborg 1 9 8 1 : 2 3 ) . Also, the c r i t e r i a to define an organized and formal disposal area have not been c l a r i f i e d ( i b i d ) . This problem i s relevant to the Dawenkou cemetery. The possible use of the pottery k i l n during the Late period may indicate that the cemetery was not used exclusively for the disposal of the dead during the Late period. Whalen (19 8 3 : 3 5 ) maintains that almost complete uniformity of b u r i a l position and orientation indicates a formal, w e l l -organized disposal area. These c r i t e r i a are met i n the Dawenkou cemetery. Some ethnographic studies have indicated a r e l a t i o n s h i p between descent, the ancestors, and mortuary r i t u a l . Rights to land are j u s t i f i e d by descent for p a r t i c u l a r ancestors among the Nuba (Hodder 19 8 0 : 1 6 4 ) . Nuba mortuary r i t u a l r e f l e c t s m a t r i l i n e a l descent groups i n an i d e a l manner: In practice i n d a i l y l i f e , the m a t r i l i n e a l l i n e i s continually frustrated by male dominance and competing paternal r i g h t s . But i n death, the m a t r i l i n e a l group i s assembled 'pure', without the husband presence (Hodder 1 9 8 0 : 1 6 5 ) . The members of a m a t r i l i n e a l descent group buried together i n a common cemetery l i v e d i n d i f f e r e n t settlements during t h e i r l i f e t i m e s (Hodder 1 9 8 0 : 1 6 5 ) . Goody ( 1 9 6 2 ) describes the importance of descent i n LoDagaa mortuary r i t u a l (Chapman 1 9 7 7 : 2 2 ) . The LoDagaa are also buried with members of t h e i r descent groups instead of with people from the settlements i n which - 156 -they had l i v e d ( i b i d ) . Goody (1962:412) points out that mortuary r i t u a l i s regarded by descent groups as important for the continuation of t h e i r descent groups (Chapman and Randsborg 1977:22-23). Corporate descent groups that regulated the inheritance of land and other resources could have developed during the N e o l i t h i c period i n various world areas (Keesing 1975:17-18). Ethnographic data indicate that there i s great v a r i a t i o n i n types of descent groups (Allen and Richardson 19 71:4 7-48, Pader 1982:64). In many processual mortuary studies, t h i s known v a r i a t i o n i s not considered. Also, one should not assume a descent system continued in one l o c a l i t y through time (Allen and Richardson 1971:49). I argue that the general con-sistency of grave orientation, grave form, body d i s p o s i t i o n and s p a t i a l location of grave from the Early period to the Late at Dawenkou does not r e f l e c t a change i n type of descent group (or any other type of s o c i a l subgroup) through time. 5.6.2 Kinship or Residential Groups at Dawenkou The s p a t i a l groups of b u r i a l s for the Early and Late periods remain to be explained. Even though they are not characterized by d i s t i n c t orientation, grave form, or body di s p o s i t i o n , they could represent either a type of kingroiip or a type of r e s i d e n t i a l group. In Nuba cemeteries, s p a t i a l clusters of b u r i a l s r e f l e c t named community sections consisting of related i n d i v i d u a l s (Hodder 1980:165). Settled a g r i c u l t u r a l -i s t s tend to have v i l l a g e subdivisions (MacDonald 1980:44). Ethnographic groups in Southeast Asia are known to have two - 157 -types of i n t r a - v i l l a g e subdivisions: the family compound and the r e s i d e n t i a l ward or neighborhood (MacDonald 1980:36). Family compounds are distinguished s p a t i a l l y i n settlements ( i b i d ) . MacDonald (1980:36) proposes that s p a t i a l areas of bur i a l s at the Bang s i t e i n Thailand r e f l e c t one of these types of v i l l a g e subdivisions. Family plots in cemeteries are known from other ethno-graphic data (King 1969:49, Chapman and Randsbord 1981:15). Spatial groups of graves are said to r e f l e c t family groups i n the processual mortuary studies by King (1969:55), Petersen (1981:187) and Goldstein (1980:124). The multiple b u r i a l s at Dawenkou could represent family groups. The e a r l i e r Dawenkou Culture s i t e s of L i u l i n and Wangyin have d i s t i n c t s p a t i a l areas of graves. Investigators think the b u r i a l s from the second excavation of L i u l i n are located i n f i v e d i s t i n c t areas, each of which r e f l e c t s one or more families within a clan (Pearson 1981:1083). The four s p a t i a l groups at Wangyin are thought to represent a clan as well (Shandong Archaeological Team 19 79). The l a t e r Dawenkou Culture s i t e of Dafanzhuang appears to have s p a t i a l groups of graves s i m i l a r to those at Dawenkou (Archaeological Team of ^ i n g i , County 1975). An eastern and western sector of bu r i a l s i s reported for the l a t e r Dawenkou Culture s i t e of Qianzhai (Kao 1983:14). 5.6.3. The Mortuary Population at Dawenkou If the cemetery was actually used for 600 years i n the early period and 400 years i n the Late, i t seems clear that - 158 -85 b u r i a l s (or 91 people, including the individuals from multiple burials) and 33 b u r i a l s (or 34 people, including the one multiple b u r i a l ) , respectively represent only a portion of the l i v i n g population for each period. It i s l i k e l y that the cemetery r e f l e c t s a portion of one or more communities. Thus, the mortuary population supports my proposition that a descent group i s represented at Dawenkou. Chang (1979:108) proposes that late N e o l i t h i c cemetery s i t e s i n the North China P l a i n region are composed of individuals from more than one v i l l a g e . This proposition i s plausible for Dawenkou and possibly other s i t e s in the Dawenkou Culture region. The numbers of b u r i a l s in Dawenkou Culture cemeteries vary from 11 at the l a t e r s i t e of Xixizhou (Shandong Archaeological Team 1964) to 885 at the e a r l i e r s i t e of Wangyin (the majority of which appear to be i n the same stratum) (Shandong Archaeological Team 19 79). Other e a r l i e r s i t e s have large numbers of b u r i a l s : 344 from the second excavation of Dadunzi i n Jiangsu (The Nanjing Museum 1981) and 197 from L i u l i n , also i n Jiangsu (Zhang 1979). Other l a t e r s i t e s contain r e l a t i v e l y fewer b u r i a l s : 15 at Xedian and 26 at Dafanzhuang, both i n Shandong (Pearson i n press:22), and 74 at Qianzhai (Kao 1983:14). I t i s not possible to • determine whether t h i s v a r i a t i o n i n b u r i a l population i s a function of time. The v a r i a t i o n may be a function of inten-sive land use i n certain areas or p a r t i a l excavation (Pearson, personal communication 1983). The large numbers of b u r i a l s at Wangyin in comparison to other Dawenkou Culture s i t e s may - 159 -be due to more complete excavation at that s i t e . I t appears that b u r i a l i n the Dawenkou cemetery was r e s t r i c t e d to a smaller number of people by the Late period. The mortuary population indicates a sharp decrease i n the number of bu r i a l s by the Late period (from 85 bu r i a l s i n the Early period to 33). I t i s not possible to ascertain whether t h i s i s also a difference i n the number of children buried i n both periods. Children may have been buried i n a separate location by the Late period, since the only subadult i n the Late period i s the youth, L117. A separate b u r i a l location for some subadults i s also conceivable for the Early period given the ..fact that subadults .are found i n only, s i x .of the 85 graves (including multiple burials).. However, the s i x undatable c h i l d b u r i a l s could be from the Early or Late period. The smaller mortuary population i n the Late period could be a function of the proposed shorter Late period (400 years, versus 600 years for the Early period). However, I propose that a more plausible explanation for the smaller mortuary population i n the Late period i s a change i n the system of status d i f f e r e n t i a t i o n from the Early period to the Late. - 160 -CHAPTER 6 ANALYSIS OF STATUS DIFFERENTIATION 6.1. Introduction 6.1.1. Procedure The purpose of t h i s analysis i s to understand 1) the nature of status d i f f e r e n t i a t i o n during the Early and Late periods, and 2) change through time i n status d i f f e r e n t i a t i o n at Dawenkou. The analysis of status for the Early period b u r i a l s i s described i n section two and the analysis for Late period b u r i a l s i n section three. Change through time i n status d i f f e r e n t i a t i o n at the cemetery i s discussed i n section three, as well. Section four i s comprised of conclusions and implications for the Ne o l i t h i c period of the eastern seaboard region. Comparison of my results with the published a r t i c l e s such as Pearson (19 81) regarding the nature of status d i f f e r e n t i a t i o n at Dawenkou i s included. I conclude the nature of status d i f f e r e n t i a t i o n i n each period i n terms of wealth (indicated by quantity and qua l i t y of grave goods) on the basis of the multidimensional scaling and cl u s t e r analyses r e s u l t s . Then, d i f f e r e n t i a t i o n among bu r i a l s i n terms of energy expenditure (indicated by grave form and grave size) i s compared with that by grave goods. V a r i a b i l i t y i n body d i s p o s i t i o n and grave location i s also assessed. Status d i f f e r e n t i a t i o n i n terms of grave goods, energy expenditure, - 161 -body d i s p o s i t i o n and grave location i s evaluated by means of test implications for achieved versus ascribed status. The multiple b u r i a l s from each period are not included i n the multi-variate analyses because i t i s d i f f i c u l t to determine the grave goods associated with p a r t i c u l a r i n d i v i d u a l s . The s o c i a l statuses of the multiple b u r i a l s are discussed separately. F i n a l l y , the in t r u s i v e pairs of b u r i a l s within each period are employed i n an attempt to assess change through time i n status d i f f e r e n t i a t i o n within each period. Status d i f f e r e n t i a t i o n among the undatable b u r i a l s i s also discussed. 6.1.2. Test implications Since status i s symbolized i n d i f f e r e n t manners i n d i f f e r e n t s o c i e t i e s , any number of the following attributes of mortuary treatment for achieved or ascribed s o c i a l status may be found to characterize Dawenkou i n either the Early or Late period. A society i n which s o c i a l status i s achieved (or an egal-i t a r i a n society) i s expected to exhibit d i f f e r e n t i a t i o n of bu r i a l s according to the age and sex of the deceased, on the basis of any number of the following c h a r a c t e r i s t i c s : 1) certa i n forms of grave goods (for example, certa i n forms may be regularly associated with certain age or sex classes), 2) r e l a t i v e quantities of grave goods, 3) elaborateness of grave form, 4) r e l a t i v e grave s i z e , 5) grave location (for example, children and infants buried i n separate location than adults), and 6) certain body d i s p o s i t i o n s ( s ) . - 162 -A society i n which s o c i a l status i s ascribed (or a society with hereditary ranking) i s expected to exhibit d i f f e r e n t i a t i o n of b u r i a l s which cross-cuts age and sex classes, on the basis of any number of the following c h a r a c t e r i s t i c s : 1) certain forms of grave goods (including "badges" which symbolize p a r t i c u l a r high ranks, imported goods, goods made from imported raw materials, goods which require much s k i l l i n manufacture, goods made from rare raw materials symbolizing high rank and goods made from l o c a l l y abundant materials and not requiring extraordinary s k i l l i n manufacture symbolizing low rank), 2) r e l a t i v e quantities of grave goods (greater quantities of a l l forms of a r t i f a c t s for high status b u r i a l s and small quantities for low status b u r i a l s ) , 3) elaborateness of grave form (elabor-ate for high status b u r i a l s , simple for low), 4) r e l a t i v e grave size (large for high status b u r i a l s , small for low status b u r i a l s ) , 5) grave location (separate location for certain status l e v e l s , e s p e c i a l l y individuals of high status), and 6) certain body d i s p o s i t i o n f o r certain status l e v e l s . A ranking pyramid i s also expected - or successively fewer b u r i a l s the higher the s o c i a l rank. The highest status category i s expected to have only one or.a few b u r i a l s . Children and adults of both sexes are expected to characterize each status l e v e l except the highest. Some d i f f e r e n t i a t i o n on the basis of age and sex i s expected i n addition to the above d i f f e r e n t i a t i o n which cross-cuts age and sex categories (the "subordinate dimension" from Peebles and Kus 1977:431). - 163 -Some d e v i a t i o n s to the standard t e s t i m p l i c a t i o n s d e s c r i b e d above t h a t have been u t i l i z e d i n p r o c e s s u a l mortuary analyses have been suggested. They are co n s i d e r e d here i n s e c t i o n s two and t h r e e . R o t h s c h i l d (1979:666) maintains t h a t h i g h r a n k i n g females may be symbolized i n a d i f f e r e n t manner than h i g h r a n k i n g males. Thus, not a l l mortuary a t t r i b u t e s which symbol-i z e h i g h s t a t u s are expected to be shared between the sexes. T h i s p o s s i b i l i t y i s a l s o noted by Chapman and Randsborg (1981: 9) and Pader (1982:59). Others comment t h a t wealthy c h i l d b u r i a l s do not necess-a r i l y i n d i c a t e a s c r i p t i v e r a n k i n g . Hodder (1980:163) and Braun (1979:68) ma i n t a i n t h a t wealthy c h i l d b u r i a l s may r e f l e c t the achieved s t a t u s of t h e i r parents i n s t e a d o f the a s c r i b e d s t a t u s of the c h i l d . Bayard (19 83:18) s t a t e s t h a t wealthy c h i l d b u r i a l s may r e f l e c t p a r e n t a l a f f e c t i o n i n s t e a d o f a s c r i b e d s t a t u s but t h e i r presence does allow the sugge s t i o n to be made t h a t a s c r i p t i v e r a n k i n g i s prese n t . 6.2. A n a l y s i s of Sta t u s , E a r l y P e r i o d 6.2.1. Data 6.2.1.1. The M u l t i v a r i a t e Analyses A t o t a l o f 23 v a r i a b l e s r e p r e s e n t i n g d i f f e r e n t aspects of mortuary treatment and 79 b u r i a l s are i n c l u d e d i n the m u l t i -v a r i a t e a n a l y s e s . The 79 b u r i a l s are l i s t e d i n F i g . A6-1. One b u r i a l , E27, i s d i s t u r b e d . The 23 v a r i a b l e s c o n s i d e r e d as - 164 -p o t e n t i a l l y status-related during the Early period are l i s t e d i n F i g . A6-2. The frequency of occurrence of each attribute among the b u r i a l sample i s shown i n F i g . A6-2 as well. The variables included i n the multivariate analyses for each period r e f l e c t grave goods i n form, quantity, and raw material. V a r i a b i l i t y i n energy expenditure i n terms of grave form (either log tomb, log tomb with second l e v e l platform, p i t with second l e v e l platform, or simple p i t ) and grave size i s del i b e r a t e l y excluded from the multivariate analyses for both periods i n order to compare status d i f f e r e n t i a t i o n based upon grave goods and that based upon energy expenditure. A good correspondence would strengthen the interpretation of r e l a t i v e status since each approach i s independent of the other and would indicate that both energy expenditure and grave goods symbolize status. Another check on the patterning from the multivariate analyses i s my expectations derived from my exploratory study of the cemetery. As much v a r i a b i l i t y i n grave goods as possible that poten-t i a l l y symbolizes status d i s t i n c t i o n s i s included i n the multi-variate analyses for each period. Each variable was coded on a presence/absence basis as most variables did not occur very frequently. Some poten t i a l status related variables could not be included because they occur only once or twice among the b u r i a l s . Three occurrences was chosen as the minimum number to make the analyses for each period viable because a minimum of - 165 -three occurrences among the 115 b u r i a l s was viable i n the chronological analysis of Chapter 3. I regard these excluded a r t i f a c t types as probable high status items. Therefore, the d i s t r i b u t i o n of these a r t i f a c t types i n the derived groupings representing equivalent status from the c l u s t e r i n g and scaling serves as s t i l l another check on the a n a l y t i c a l r e s u l t s . For each period, the variables are coded from the information i n the "grave goods" column (sui zang q i wu), pages 136-155 of the s i t e report In some cases, additional a r t i f a c t s are l i s t e d under the "remarks" (bei zliu) column. Two ceramic ware types from the Early period may r e f l e c t high status: painted and black wares. The painted wares from Dawenkou culture s i t e s are described as f i n e l y made by Thorp (1979:7) and as rare and l i t t l e understood by Shangraw (1978: 31). The black wares at Dawenkou culture s i t e s are considered f i n e l y made (with thin vessel walls and polished) by Shangraw (1978:35). The ceramic forms (functional types) thought to r e f l e c t high status at Dawenkou are the wine vessel (zun), serving stand (dou), and t a l l stemmed cup (bei gao). Bronze wine vessels are found i n high status contexts i n Shang tombs (Thorp 1980: 56 ). Since there are c u l t u r a l c o n t i n u i t i e s i n mortuary practices from the Dawenkou Culture period to the Shang, i t i s possible the wine vessel symbolizes high status at Dawenkou as well. This p o s s i b i l i t y i s also raised by Rawson (1980:29). The serving stand i s regarded by Pearson (personal - 166 -communication, 19 83) as possibly having a r i t u a l i s t i c function and as symbolic of high status. The serving stand and the high stemmed cup appear f i n e l y made and as not having purely u t i l -i t a r i a n functions i n the photographs provided i n the s i t e report. The high stemmed cups and serving stands are considered as having r i t u a l i s t i c functions by Shangraw (1978:34). Rawson (1980:29) describes the serving stands and high stemmed cups from several Dawenkou Culture s i t e s as f r a g i l e and suggests these forms had a ceremonial function. Three ornamental forms considered status-related are included i n t h i s analysis: r i n g (zhThuan), bracelet (bihuan), and hairpin (ji_) . Bone hairpins have been found i n high status contexts i n Shang s i t e s (Hay 1973:55). The functions of several a r t i f a c t types from both periods of the cemetery are unclear. Three a r t i f a c t forms included here may be ornamental or formerly part of an ornament: possible hair t i e s (shu fa qi', Gao 1978:31; Pearson 1981:1080), small round f l a t stone (bing), and small f l a t piece of bone (ban), (appearing worked i n photograph 97 of the s i t e report). The variable problable ornament parts includes forms which appear to be the same as those forms which are part of neck+. laces for the head or neck (tdu slfi or j l n g shi) i n known female b u r i a l s : f l a t , thin piece of stone (pian), stone or bone tubular bead (guan), stone c i r c u l a r bead (zhu), small stone annular shaped piece, possibly an earring (huan). Also - 167 -each of these forms appears ornamental from plate 9 7 of the report. These forms are grouped together because each i s present only once or twice across the b u r i a l sample. Other a r t i f a c t forms i n the analysis with a probable decorative as well as u t i l i t a r i a n function are elaborately carved bone or elephant ivory carved cylinder (diaotong), t u r t l e s h e l l (gui j i a ) , and spoon or spatula (b_i) . Only one Early period b u r i a l contains an elephant ivory cylinder. Elephant ivory i s one of two raw materials i n the Early period that may symbolize high status. On the basis of the discussion of the l o c a l environment i n Chapter 1 i t i s possible that elephant were obtained l o c a l l y . The photographs i n the s i t e report indicate that some cylinders from both periods are i n l a i d with stone, but the descriptions of the b u r i a l s do not mention i n l a i d stone (except for one Late period b u r i a l ) . It i s not clear from .photographs i n the s i t e report whether the t u r t l e s h e l l s are worked. Chang (1979:280) describes them as polished and states they were used for containers at several Dawenkou Culture s i t e s . Gao (1978:31) states the s h e l l s were strung together to form a pouch. Seven variables are coded i n an additive, presence/absence manner because the r e l a t i v e quantities of these variables are thought to r e f l e c t status d i s t i n c t i o n s . Frequency d i s t r i b u t i o n s showing r e l a t i v e quantities of these variables were drawn, and the natural breaks evident from the graphs were regarded as r e f l e c t i n g c u l t u r a l l y meaningful d i s t i n c t i o n s . These variables - 168 -are the serving stand, t o t a l number of ceramic vessels, stone t o o l t o t a l , bone t o o l t o t a l , pig s k u l l s (zhu_tou), deer teeth (zhang ya), and pieces of raw material ( l i a o , either bone, horn, or tusk). The associated frequency d i s t r i b u t i o n s are i n Figs. A6-3 to A6-9. Since only one Early b u r i a l i n the analysis has no ceramic vessels, i t was not possible to have a separate category for absence of vessels. The functions of the pig s k u l l s , deer teeth, raw material pieces, and oyster s h e l l pieces are not clear. The pig s k u l l s at Dawenkou Culture s i t e s may r e f l e c t wealth (Zhang 19 79, Luo and Zhang 19 79, Lu 19 76). A contemporary society from Hainan Island i n southern China uses pig jaws for t h i s purpose (Luo and Zhang 19 79). Since pigs may have been important i n sub-sistence, p i g s k u l l s and other p i g parts may r e f l e c t wealth i n terms of food supply. Some or a l l of the deer teeth may have been part of cutting tools for agriculture (Gao 19 78:31, Chang 1979:160). The raw material pieces may have been for manufacture of various goods. The variable pig parts includes front teeth (zhu men ya), various pieces of bone (zhu gu kuai), and lower jaw bone (xia he gu). The oyster s h e l l pieces (bang pian) are not shown i n a photograph i n the s i t e report. It i s not clear i f they are worked. The ceramic categories i n t h i s analysis and i n the Late period analysis are p a r t i a l l y redundant because ware color, form and quantity are a l l included. The impact of t h i s redundancy on the patterning of status d i s t i n c t i o n s i n these analyses i s judged to be minimal. - 169 -6.2.1.2. High Status Unique Items The a r t i f a c t forms or materials excluded from the multi-variate analyses that are considered probable high status items in the Early period are: elephant ivory disk ( b l ) , elephant ivory comb (shu), animal shaped ceramic vessel (tab shou xing qi) , red ochre (chi. t i e kuang s h i , "iron ore", which I interpret as red ochre), and jade (yu). The elephant ivory disk i s found i n two single b u r i a l s and i n one multiple b u r i a l . None are found i n Late period b u r i a l s . The authors of the s i t e report i d e n t i f y the object as a long hollow object with rectangular sides c a l l e d a corig. However, on the basis of photo plate 9 4 i n the s i t e report, Shangraw (personal communication 19 83) i d e n t i f i e s the object as the b i disk, maintaining the Chinese character i n the s i t e report i s incorrect. The b l form i n jade i s known from h i s t o r i c a l data to have r i t u a l i s t i c s ignificance during the Shang period (Rawson 1980:83). It i s c a l l e d the symbol of heaven i n the Zhou L i , a text from the late Zhou dynasty (Rawson 1980:83, Sull i v a n 1977:48). The Zhou L i also states that the jade b l symbolized the lower ranks of viscount and baron (Sullivan 1977:48). The elephant ivory b l may be a badge of high rank i n the Early period. Pearson (in press:29) suggests that some types of ornaments from Dawenkou resembling ornaments from the Shang period are badges and r e f l e c t inherited s o c i a l p o s i t i o n . Pearson (1981:1086) suggests that c e r t a i n jade forms from Late Dawenkou Culture s i t e s that do not resemble - 170 -u t i l i t a r i a n a r t i f a c t s are badges of high status. He has since indicated that these forms may be of other materials as well (personal communication 19 83). The elephant ivory comb i s found i n one Early b u r i a l and i n one Late. Only one animal shaped ceramic vessel i s found i n the cemetery. This vessel resembles a pig. The same b u r i a l (E9) contains the two pieces of red ochre. Ochre i s not present i n Late period b u r i a l s . It i s possible that the animal vessel and the ochre r e f l e c t r i t u a l rather than status d i s t i n c t i o n s . Other animal shaped vessels have been found i n late Dawenkou Culture s i t e s , i n high status contexts (Pearson 19 83:140). Only one Early b u r i a l contains jade while i t occurs .in four Late period.graves. B u r i a l E78 contains one jade huan (small annular object). This object i s i d e n t i f i e d as jade from page 96 of the s i t e report, not from the description for the b u r i a l . Jade was highly regarded i n Chinese h i s t o r i c a l periods for i t s beauty and r i t u a l significance (Rawson 1980:81). It was an important part of Shang mortuary r i t u a l (Chang 19 80:156, Rawson 19 80:32). Rawson (1980:32) maintains that nephrite i s the mineral which was worked i n east coast N e o l i t h i c cultures. Jade i s extremely d i f f i c u l t to work; abrasives must be used as even modern s t e e l tools cannot cut i t ( i b i d ) . The source for jade found i n east coast Neol i t h i c s i t e s i s presently unknown ( i b i d ) . However, at least one western investigator, J e f f r e y Kao, Depart-ment of Anthropology, Harvard University, i s currently research-ing the problem (Pearson, personal communication 19 83). - 171 -Wu (19 73) suggests some jade was available from Zhejiang and Anhui provinces. Lu (1976) maintains that jade was available from the Tai Shan region of Shandong. Chang (1980:156-157) suggests southwestern Henan and parts of Shaanxi were sources during the Shang period. Rawson (1980:32) suggests jade was imported over long distances from the north (Baikal region) or west (Central Asia) - the main source of jade i n h i s t o r i c a l times. The amount of labor involved i n procuring jade and working i t strongly suggests i t symbolized high status and i t s use was r e s t r i c t e d for mortuary r i t u a l or other types of r i t u a l ( i b i d ) . S u l l i v a n (1977:40) mentions that early h i s t o r i c texts state that jade was obtained from several locations i n China. Also, a source for a variety of jade other than the highly valued nephrite has been found i n Henan ( i b i d ) . The interpretation of the results from the multivariate analyses for both periods takes into account the known associa-tions of p a r t i c u l a r a r t i f a c t types with sex and age. As discuss-ed i n Chapter 4, the small f l a t stone (bing) i s not d e f i n i t e l y female-related. The elephant ivory disk ( b i ) , although possibly i n c o r r e c t l y grouped with the horn zhui i n Chapter 4, may be male-related. One of the two single b u r i a l s i n which i t i s found (E59) i s a known male and the other (E26) i s a r e l i a b l y estimated male. None of the 2 3 variables included i n the multi-variate analyses are exclusive to the Early period c h i l d b u r i a l s . In fact, none of the subadults i n the cemetery (including Late period and undatable burials) contain a r t i f a c t forms or raw - 172 -materials not found i n adult b u r i a l s . 6.2.1.3. Energy Expenditure Several researchers have pointed out the d i f f i c u l t y of i d e n t i f y i n g c u l t u r a l l y s i g n i f i c a n t differences i n energy expend-itu r e (Tainter 1978:128, Brown 1981:29, Goldstein 1981:55-56). The differences i n grave form for the b u r i a l s of both periods are s e l f evident: p i t , p i t with second l e v e l platform, log tomb, log tomb with second l e v e l platform. It was previously concluded that grave form does not seem to r e f l e c t s o c i a l sub-group a f f i l i a t i o n or age, i n either the Early or Late period. The authors of the s i t e report d i f f e r e n t i a t e between three types of log tombs that r e f l e c t varying degrees of elaboration and energy expenditure. These types were not discussed previously. These types may r e f l e c t status d i s t i n c t i o n s . As Pearson (19 81: 10 80) describes, type one i s a l i d of undressed logs and represents the lowest amount of energy expenditure. Type two has a l i d , four sides, and a bottom of loose logs. Type three, the most elaborate, i s a chamber made of trimmed logs f i t t e d together and placed i n the middle of a large p i t . The authors of the s i t e report (1974:5-7) do not indicate the type represented i n each b u r i a l with a log tomb. Only examples of each type are given. The example of a type one log tomb i s b u r i a l E5 3 and type two, E94. Only one Early b u r i a l and six Late b u r i a l s have log tombs of type three. The Early b u r i a l i s a multiple b u r i a l , E13. The map of the s i t e indicates great v a r i a b i l i t y i n grave - 173 -size for both periods. Grave size i s described i n the s i t e report i n terms of length, width, and depth i n meters. Grave volume i s not calculated due to my uncertainties over the depth measurements i n the report. It i s not clear whether depth was measured from the same point i n every grave. Therefore, area of grave i s employed as the measurement of grave si z e . The v a r i a b i l i t y i n grave form and grave size for the 79 Early b u r i a l s i s indicated i n F i g . A6-1. The d i s t r i b u t i o n of values for grave area i s shown i n Fig . A6-10. The sharp break i n the d i s t r i b u t i o n 2 of values aft e r 2.50m i s taken to r e f l e c t a c u l t u r a l l y s i g n i f -icant d i s t i n c t i o n i n grave s i z e . Seven graves (9.1%) have areas 2 2 over 2.50m (up to 3.82m ). These graves are considered large and the others, small. The areas for two graves, E4 3 and E51, could not be calculated due to lack of length and width measure-ments. In a few cases from both periods, grave size as indicated by my calculations does not seem to agree with r e l a t i v e grave size as indicated on the s i t e map. The calculations are consid-ered the more accurate indicator of grave s i z e . However, since the size of E43 and E51 appear c l e a r l y small on the s i t e map in r e l a t i o n to the seven large graves i d e n t i f i e d above, these two graves are interpreted as small for the purposes of t h i s analysis. 6.2.2. Results: The Multidimensional Scaling The multidimensional sc a l i n g analysis for the Early period b u r i a l s resulted i n ten dimensions, the f i r s t two accounting for 51.2% of the v a r i a b i l i t y i n the data set. No tr i a n g l e i n e q u a l i t i e s i n the program were v i o l a t e d , i n d i c a t i n g t h i s data set i s metric. Dimension 1 accounted for 36.1% of the - 174 -v a r i a b i l i t y and Dimension 2, 15.1%. The t h i r d and fourth dimensions accounted for 10.2% and 8.8% of the v a r i a b i l i t y , respectively. When the f i r s t and second dimensions are plotted together, i t becomes evident that Dimension 1 can be interpreted as status d i f f e r e n t i a t i o n . The plot of the 7 9 b u r i a l s i s i n F i g . 6 - 1 . Clusters from the clus t e r analysis are superimposed on F i g . 6 - 1 . I interpret the four b u r i a l s on the extreme l e f t of the plot ( E 2 6 , E 9 , E 5 4 , E 5 9 ) as highest i n status, the b u r i a l s i n the mid-section of the plot as intermediate, and the 1 7 b u r i a l s i n the extreme ri g h t section as lowest i n status ( E 2 9 , E 3 6 , E 4 1 , E 4 3 , E 5 1 , E 7 1 , E 8 6 , E 9 0 , E 1 1 4 , E 1 2 0 , E 4 8 , E 2 7 , E 2 0 , E 4 5 , E 6 2 , E 8 8 , E 8 9 ) . The four b u r i a l s at the extreme l e f t end have the greatest occurrence of the attributes I expected to r e f l e c t high status, the b u r i a l s i n the midsection have the next greatest amount, and the 1 7 b u r i a l s at the extreme right end have the least. The bu r i a l s at the extreme l e f t and extreme ri g h t ends of the plot are s p a t i a l l y separated from other b u r i a l s and appear to r e f l e c t d i s t i n c t differences i n status. The b u r i a l s i n the midsection of the pl o t along Dimension 1 seem to represent a status con-tinuum because they do not seem to be c l e a r l y separated s p a t i a l l y from other b u r i a l s . The bu r i a l s located at the same point (for example, the ten b u r i a l s at the extreme ri g h t end) have the same attributes i n terms of grave goods. These 1 0 b u r i a l s appear to r e f l e c t a redundantly symbolized status l e v e l . My tentative interpretation for Dimension 2 i s also status - 175 -FIGURE 6-1. Multidimensional scaling plot of Early Period b u r i a l s i n the analysis of status, with clusters from Ward's Method. l a r g e in a r e a ( ? ) s m a l l in a r e a ( 7 2 ) log t o m b (6) s e c o n d l eve l p l a t f o r m (12) 4 3 d i s t u r b e d bur ia l (I) 4 3 c o n t r o v e r s i a l f o r m e r M i d d l e B u r i a l ( 3 ) /o no s k e l e t o n (I ) 43 high s t a t u s , unique ob jec ts (4) « ch i ld (4) N o t e : C l u s t e r s 1 - 7 f rom W o r d s seven group solution ore c i r c l e d . In the four group so lu t ion , grouped together are C l u s t e r s 1 8 2 , C l u s t e r s 4 8 5 , ond C l u s t e r s 6 8 7. C luster 3 is d is t ingu ished by both g r o u p s o l u t i o n s . - 176 -d i f f e r e n t i a t i o n . The b u r i a l s i n the lower midsection of the plot appear to be lower i n status than those i n the upper mid-section of the plot ( p a r t i c u l a r l y E7, E78, E129, E67, E49, E58). The l a t t e r s i x b u r i a l s contain more of the variables which represent ornamental or decorative types of grave goods than those b u r i a l s i n the lower midsection of the plo t . Status d i f f e r e n t i a t i o n may be represented as a diagonal through hyper-space i n a roughly southeast to northwest d i r e c t i o n . 6.2.3. The Three Types of Cluster Analysis The dendrogram judged to best represent the status r e l a t i o n -ships among the b u r i a l s i s from Ward's Method. This judgement was made af t e r studying the clusters of b u r i a l s i n each dendro-gram from each of the three techniques at several levels of s i m i l a r i t y to determine the best ordering, on the basis of 1) the occurrence of the mortuary attributes shared among the bu r i a l s i n each clu s t e r and 2) my expectations of the status relationships among the b u r i a l s from my former exploratory analysis of the cemetery. The dendrogram from Average Linkage compares f a i r l y well with that from Ward's Method, while the dendrogram from Complete Linkage does not compare well with eithe r that from Ward's or Average Linkage. As i n the chrono-l o g i c a l analysis, the Complete Linkage dendrogram shows small clusters of b u r i a l s at several distance levels and the Average Linkage dendrogram exhibits some chaining. The f a i r l y good comparison of the clusters from Ward's Method with those from Average Linkage and the good comparison with the ordering of bu r i a l s on the scaling p l o t also lends support to the acceptance of the Ward's dendrogram. Only the dendrogram from Ward's Method - 177 -i s presented here. The dendrogram from Ward's Method i s presented i n Fig.6-2. The seven c l u s t e r solution (at the 1.0022 distance l e v e l ) , the four c l u s t e r solution (1.3262 distance l e v e l ) , and the two c l u s t e r solution (2.2437 distance level) are judged to indicate status relationships among the b u r i a l s most c l e a r l y . The seven-four, and two c l u s t e r solutions are l a b e l l e d i n Fig.6-1. Each of the three solutions contains clusters composed of s i m i l a r mortuary attributes and compares f a i r l y well with the multi-dimensional scaling solution. At f i r s t , I thought the seven cl u s t e r solution best depicts status relationships among the b u r i a l s . The mortuary attributes i n common for each clu s t e r of t h i s solution are given i n F i g . A6-11. This solution seemed to f i t my a p r i o r i expectations of status relationships based upon my exploratory study. However, further appraisal of the r e s u l t s indicated that the four group solution better depicts status r e l a t i o n s h i p among the Early period b u r i a l s . The mortuary attributes c h a r a c t e r i s t i c of the four group solution can be seen i n F i g . A6-11 as well. The clusters of b u r i a l s from the seven group solution appear at f i r s t to indicate equivalent status l e v e l s (see F i g . A6-11). Clusters 1 and 2 appear to represent low status c l u s t e r s , with Cluster 2 of s l i g h t l y higher status. The b u r i a l s i n these two c l u s t e r s contain only a few u t i l i t a r i a n a r t i f a c t types. Clusters 4, 5, 6, and 7 appear intermediate i n status, with Clusters 5 and 6 s l i g h t l y higher i n status than Clusters 4 and 7. Clusters 4, 5, 6, and 7 contain greater quantities - 178 -FIGURE 6-2. Method. Dendogram of Early Period b u r i a l s from Ward' d i s t a n c e Two Croup Solution - 179 -of a r t i f a c t s , a greater variety of a r t i f a c t types (including some ornaments and items of probable decorative function) and black and painted wares. Clusters 5 and 6. contain b u r i a l s with s l i g h t l y greater quantities of a r t i f a c t s . Cluster 3 c l e a r l y represents the highest status l e v e l , with great quantities of a r t i f a c t s , several ornamental and decorative a r t i f a c t types and painted and black wares. I think the four group solution better depicts status r e l a -tionships among the b u r i a l s because i t groups the clusters from the seven group solution that are not 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 the pot e n t i a l status-related variables. Clusters 1 and 2 of the seven group solution are grouped together as well as the clusters judged as intermediate i n status. Clusters 4 and 5 of the seven group solution are grouped together, and Clusters 6 and 7 of the seven group solution. The c l e a r l y high status group of b u r i a l s i s set apart i n the four group solution as well. The two group solution i s c l e a r l y evident i n F i g . 6-2 as a "natural break" i n the dendrogram. I t distinguishes the d i s t i n c t l y low status group of b u r i a l s (Clusters 1 and 2 of the seven group solution) from the other Early period b u r i a l s . The clus t e r s from the seven group solution of Ward's Method were drawn on the multidimensional scaling p l o t (Fig. 6-1) because I o r i g i n a l l y thought that the seven group solution best depicted status relationships among the b u r i a l s . Information regarding the clusters from the four group solution i s included i n F i g . 6-1. F i g . 6-1 depicts a f a i r l y good agreement between the s c a l i n g and cl u s t e r i n g r e s u l t s (from the seven and four - 180 -group solutions). The high status c l u s t e r (3) i s located on the l e f t end of the p l o t , the clusters intermediate i n status (4, 5, 6, 7) i n the midsection, and the low status clusters (1,2) at the right end. Like the chronological analysis of Chapter 3, the r e s u l t s from Ward's Method and Torgerson"s Metric M u l t i -dimensional Scaling are i n s u f f i c i e n t agreement to allow an interpretation of status d i f f e r e n t i a t i o n to be made based upon the results from both methods. Although both methods c l e a r l y depict the high, intermediate, and low status b u r i a l s , there are some differences i n the order-ings from each method. The Ward's solution depicts b u r i a l s 63, 102, and 12 as high i n status along with 26, 9, 54, and 59. The scaling solution shows 63, 102, and 12 as s p a t i a l l y separated from the l a t t e r b u r i a l s . As F i g . A6-11 shows, b u r i a l s 63, 102, and 12 do not contain as many of the attributes I interpret as related to high status as the other four b u r i a l s . There i s also some disagreement over the placement of b u r i a l s i n the lowest status group. Ward's solution includes more b u r i a l s i n the low status group than the scaling p l o t : b u r i a l s 80, 82, 97, 21, 91, 119. The l a t t e r s i x b u r i a l s are s p a t i a l l y separated along Dimension 1 from the other 17 b u r i a l s interpreted as low i n status. Other disagreements i n the orderings from both methods include b u r i a l s 81, 6, and 101 (located along Dimension 1 with b u r i a l s from Ward's Cluster 6 yet included with the b u r i a l s from Cluster 7) and b u r i a l 30 (included i n Cluster 7 yet located with b u r i a l s from Cluster 2 along Dimension 1). - 181 -6.2.4. Interpretation of Status Distinctions i n the Early Period 6.2.4.1. D i s t r i b u t i o n of the Unique Items The correspondence between status d i f f e r e n t i a t i o n based upon grave goods included i n the multivariate analyses with that based upon the d i s t r i b u t i o n of the probable high status unique objects can be seen i n F i g . 6-1. The d i s t r i b u t i o n of the unique objects supports my interpretation that the order-ing of b u r i a l s from the c l u s t e r i n g and scaling techniques r e f l e c t s status d i f f e r e n t i a t i o n . Five of the six high status objects are found i n the high status group of b u r i a l s (burials 26, 9, and 59 i n F i g . 6-1). The two elephant ivory disks (bi) are i n b u r i a l s E59 and E26. E26 also contains the elephant ivory comb. E9 contains the red ochre and animal shaped pot. The jade huan i s found i n E78 (located i n Cluster 5 on Fig.6-1). I o r i g i n a l l y interpreted the b u r i a l s i n t h i s section of the scaling plot as being higher i n status than those i n the lower midsection of the p l o t . These b u r i a l s also seemed higher i n status than other b u r i a l s of intermediate status from the seven group solution of Ward's Method. Due to the good corres-pondence between the d i f f e r e n t i a t i o n of b u r i a l s from the cl u s t e r i n g and sc a l i n g and from the probable high status unique objects, I conclude that status d i f f e r e n t i a t i o n i s at least p a r t i a l l y symbolized by wealth i n terms of grave goods during the Early period. - 182 -6.2.4.2. Argument for Four Status Levels The differences i n the ordering of b u r i a l s between the r e s u l t s from Ward's clus t e r analysis, the scaling r e s u l t s , and the d i s t r i b u t i o n of the probable high status unique items preclude a d e f i n i t i v e conclusion on the approximate number of status levels based upon grave goods represented among the Early period b u r i a l s . I argue that four status levels are the most c l e a r l y represented. The difference i n composition of mortuary attributes among the clusters of Ward's four group solution r e f l e c t s status d i f f e r e n t i a t i o n i n a more obvious manner than i n the clusters of the seven group solution. Because the composition of these four clusters i n terms of b u r i a l s agrees well o v e r a l l with the multidimensional scaling r e s u l t s , these clusters are concluded as representing the major status d i s t i n c t i o n s of the Early period. I recognize that the placement of some b u r i a l s i s open to debate. According to the scaling r e s u l t s , the placement of three b u r i a l s i n the high' status c l u s t e r (63, 102, 12) on Fig. 6-1 may be incorrect. Likewise, the placement of a few other b u r i a l s i n the low status group (eg., 119, 91, 80, 82, 97, 21) may be incorrect. However, each method seemed to d i s t i n g u i s h b u r i a l s of high, high i n t e r -mediate, low intermediate and low status. Since the objective of t h i s analysis i s to i d e n t i f y the approximate, rather than the exact, number of. status- . d i s t i n c t i o n s . that existed i n the society of the deceased, the d i f f e r e n t i a l placement of a few b u r i a l s does not s i g n i f i c a n t l y change the i n t e r p r e t a t i o n - 183 -of status d i s t i n c t i o n s among the Early period b u r i a l s . A l l the c h a r a c t e r i s t i c s of the four status lev e l s are indicated i n F i g . 6-3. Additional unique items found i n Early period b u r i a l s but not considered as represent-ing high status are i n E103 (a b i r d claw bone, zhao, from photo plate 101), and E19 (a stone b a l l , qiu, photo plate 101). It i s not clear whether t h i s b a l l i s of the type that have been found i n the mouths of skeletons from other Dawenkou Culture s i t e s (Han and Pan 1 9 8 0 ) . B u r i a l E34 contains the only horn zhui from the Early period. 6.2.4.3. D i f f e r e n t i a t i o n by Energy Expenditure D i f f e r e n t i a t i o n of b u r i a l s based upon the four types of grave form and the two types of grave size does not coincide to a great degree with the d i f f e r e n t i a t i o n based upon grave goods. This lack of correspondence was expected on the basis of my exploratory study of status d i f f e r e n t i a t i o n . As shown i n F i g . 6-1, only three of the seven b u r i a l s i n the high status l e v e l have grave forms that represent more energy expenditure than the simple p i t . Burials E9, E59, and E12 have p i t s with second l e v e l platforms. None of the high status b u r i a l s contain the grave forms that represent the greatest amount of energy expenditure, the log tomb, or the log tomb with second l e v e l platforms. However, none of the low status b u r i a l s (Clusters 1 and 2 i n F i g . 6-1) have log tombs or second l e v e l platforms. Burials with the grave form representing the greatest amount of energy expenditure, log tomb with second l e v e l FIGURE 6-3. Characteristics of the Four Status Levels, Early period. high status high i n t e r - low i n t e r - low status mediate status mediate status (7 burials) (17 burials) (32 burials) (23 burials) ~ % of tot. + % of tot. + % of tot. + % of tot, black ware 6 23.1 15 57.7 5 19.2 0 0 painted ware 5 29.4 1 5.9 11 64.7 0 0 wine vessel 5 55.6 2 22.2 1 11.1 1 11.1 serving stand 0 0 0 0 0 14 41.2 20 58.8 serving stand 1-2 4 9.8 16 39.0 18 43.9 3 7.3 serving stand 3-9 3 75. 0 1 25. 0 0 0 0 0 t a l l stemmed cup 0 0 3 100.0 0 0 0 0 t o t a l pots 0-7 0 0 9 14.1 32 50.0 23 35.9 t o t a l pots 8-14 4 33. 3 8 66.7 0 0 0 0 t o t a l pots 19-30 3 100.0 0 0 0 0 0 0 carved cylinder 2 40.0 1 20.0 2 40.0 0 0 t u r t l e s h e l l 2 40.0 0 0 3 60.0 0 0 ring 2 25.0 1 12.5 5 62.5 0 0 bracelet 0 0 3 60.0 2 40.0 0 0 hairpin 4 50.0 4 50.0 0 0 0 0 spoon 4 30. 8 2 15.4 5 38.5 2 15.4 stone t o o l t o t a l 0 0 0 10 22.2 13 28.9 22 48.9 stone t o o l t o t a l 1 4 18.2 6 27.3 11 50.0 1 4.5 Cont'd. FIGURE 6-3 continued stone t o o l t o t a l 2-7 stone t o o l t o t a l 9-16 bone t o o l t o t a l 0 bone t o o l t o t a l 1-7 bone t o o l t o t a l 13-25 h a i r t i e ornament p a r t b l n g ban p i g s k u l l s 0 p i g s k u l l s 1-3 p i g s k u l l s 4-5 other p i g p a r t s deer t e e t h 0 deer t e e t h 1-3 deer t e e t h 4-12 raw m a t e r i a l p i e c e s 0 raw m a t e r i a l p i e c e s 1-10 raw m a t e r i a l p i e c e s 16-40 o y s t e r s h e l l high s t a t u s (7 b u r i a l s ) + % of t o t . 2 22.2 1 33. 3 0 0 5 20.8 2 50.0 2 25.0 0 0 0 0 4 57.1 3 6.1 4 16.0 0 0 0 0 1 3.7 5 10.9 1 16.7 3 4.4 2 25.00 2 66. 7 3 50.0 high i n t e r - low i n t e r - low s t a t u s mediate s t a t u s mediate s t a t u s (17 b u r i a l s ) (32 b u r i a l s ) (23 b u r i a l s ) + % of t o t . + % of t o t . + % of t o t . 1 11.1 0 0 12 23.5 4 16.7 1 25. 0 4 50.0 4 80.0 3 100.0 2 28.6 14 28.6 1 4.0 2 40.0 1 11.1 2 7.4 13 28.3 2 33.3 17 25.0 0 0 0 0 0 0 6 66.7 2 66.7 17 33. 3 14 58.3 1 25.0 2 25.0 0 0 0 0 1 14. 3 9 18.4 20 80.0 3 60.0 5 55.6 11 40. 7 18 39.1 3 50.0 25 36.8 6 75.0 1 33. 3 3 50.0 0 0 0 0 22 4 3.1 1 •4.2 0 0 0 0 1 20.0 0 0 0 0 23 46.9 0 0 0 0 3 33. 3 13 48.1 10 21.7 0 0 23 33. 8 0 0 0 0 0 0 - 186 -platform, are E94, E53, and E81. E94 i s i n the high i n t e r -mediate status group (Cluster 4 i n F i g . 6-1) and E53 and E81 are i n the low intermediate status group (Clusters 6 and 7). E5 3 has a type 1 log tomb and E9 4 has type 2, which represents greater energy expenditure. The other b u r i a l s with log tombs are E107 and E99 (in the high intermediate status group, Cluster 4 on Fig. 6-1) and E116 (low intermediate status, Cluster 7 on Fig . 6-1). The other b u r i a l s with second l e v e l platforms are E l l , E109, and E3 8. The f i r s t two b u r i a l s are i n the low intermediate status group (Cluster 6) and the second, i n the high intermediate (Cluster 4). D i f f e r e n t i a t i o n of b u r i a l s based upon grave size also does not coincide to a great degree with that based upon grave goods. Three of the seven large graves are located i n the high status group (E9, E59, and E12) . The other large graves are E78, E99, E l l , and E8. E78 i s i n the high intermediate status l e v e l (Cluster 5 i n F i g . 6-1). However, t h i s b u r i a l contains jade huan. E99 i s also i n the high Intermediate status group (Cluster 4),and the l a t t e r two b u r i a l s are i n the low i n t e r -mediate status l e v e l (Cluster 6). None of the large graves i s located i n the low status group. There i s a better c o r r e l a t i o n between the d i f f e r e n t i a t i o n based upon grave goods and that based upon both grave form and grave s i z e . Six of the seven b u r i a l s large i n size have grave forms other than the simple p i t : E9, E59, E12, E7 8, and E l l have second l e v e l platforms and E99 has a log tomb. - 187 -E8 i s the only grave that i s large i n size and has the simple p i t grave form. E9, E59, and E12 are i n the high status group. E78 and E99 are i n the high intermediate status group (Clusters 4 and 5). E l l i s i n the low intermediate status l e v e l (Cluster 6 on F i g . 6-1). It appears that energy expenditure i n terms of grave form and grave area eith e r does not symbolize status d i s t i n c t i o n s among the Early period b u r i a l s or that i t symbolizes a d i f f e r -ent aspect of status than grave goods. The grave forms that r e f l e c t great energy expenditure and large grave size are not exclusive to eithe r the highest status group or the second highest. On the basis of the comparison of status d i f f e r e n t i a t i o n based upon grave goods and energy expenditure, I conclude that high status i s characterized by exclusive access to only three forms of a r t i f a c t s : the elephant ivory disk (bji) , the elephant ivory comb, and the animal shaped ceramic vessel. High status i s distinguished by exclusive access to two types of raw material: elephant ivory (the comb and the one carved cylinder) and red ochre. The t h i r d c h a r a c t e r i s t i c exclusive to the high status b u r i a l s i s the highest quantity of pottery vessels. The high intermediate status l e v e l i s characterized by exclus-ive access to the high stemmed cup, the small round stone (bing) and the jade huan. The other mortuary attributes occur at least once i n more than one status l e v e l . - 188 -6.2.4.4. Body Disposition Data regarding age, sex and body d i s p o s i t i o n are included i n F i g . 6-4. F i g . 6-4 indicates that on the whole body di s p o s i t i o n i s not associated with status i n the Early period. The high status b u r i a l s a l l have the common supine position. Each of the intermediate groups contain b u r i a l s with side positions. However, the only prone b u r i a l (E45) i s i n the low status group. The low status position of E45 does not contradict my proposition i n Chapter 5 that E45 may represent a s o c i a l deviant or a person who died of an unusual cause. 6.2.4.5. Grave Location The locations of the b u r i a l s within each status group are shown i n F i g . 6-5. Multiple b u r i a l s (to be discussed shortly) are included i n t h i s figure. Five of the seven high status b u r i a l s are located i n the northern c l u s t e r of b u r i a l s . E9 i s located i n the middle section of the cemetery and E102 i n the southern end. The b u r i a l s of high intermediate and low intermediate status are located i n a l l areas of the cemetery. None of the low status b u r i a l s are located i n the northern s p a t i a l c l u s t e r , but they are located i n several other areas. I t appears that a trend towards the s p a t i a l segregation of high status b u r i a l s i s developing, as noted by Luo and Zhang (1979) and Pearson (1981:1086). - 189 -FIGURE 6-4. Age and sex composition of the status l e v e l s , Early Period b u r i a l s . (Body dispositions other than supine are marked below, as well as children.) High Status Group: .7 t o t a l 63, 59, 54, 12, 26, 102, 9 known male: 59, 9 known female: 102 r e l i a b l y estimated male: 63, 54, 12, 26 54 has no skeleton intr u s i v e b u r i a l s : 26, 9, 54 High Intermediate Status Group: 17 t o t a l 99, 34, 130, 118, 38, 32, 94, 132, 107, 76, 14, 67, 49, 129, 58, 78, 7 age: 1 c h i l d (94) known male: 99, 34, 10 7 known female: 130, 6 7, 7 f a i r estimated male: 129, 38, 118 f a i r estimated female: 49, 58 body d i s p o s i t i o n : 38 on ri g h t side, 34 on l e f t side intr u s i v e b u r i a l s : 58, 32, 78, 129, 132 Low Intermediate Status Group: 32 t o t a l 103, 87, 79, 131, 110, 109, 73, 106, 19, 23, 11, 53, 18, 65, 8, 115, 61, 55, 33, 30, 81, 66, 56, 112, 52, 108, 44, 42, 116, 101, 84, 6 known male: 109, 73, 112 r e l i a b l y estimated male: 103, 110, 106, 19, 11, 61, 66 f a i r estimated male: 81, 101, 79, 87, 53 known female: 131, 115, 55, 30 body d i s p o s i t i o n : 81, 52, 109 on l e f t side 73, 84, 110 on right side in t r u s i v e b u r i a l s : 23, 33, 30, 18, 44, 61 Low Status Group: 23 t o t a l 48, 120, 114, 90, 86, 71, 51, 43, 41, 36, 29, 97, 82, 80, 89, 88, 62, 45, 119, 27, 91, 21, 20 age: 3 children (114, 36, 89) known male: 91 known female: 82 f a i r estimated male: 119, 27 body d i s p o s i t i o n : 45 prone, 9 7 on l e f t side in t r u s i v e b u r i a l s : 62, 43, 71 - 190 -FIGURE 6-5. Location of the four status groups i n the Early Period, i n terms of the s p a t i a l areas derived i n Chapter 5. Multiple b u r i a l s are included. <8> 41 57 a 5 * 40 I I 46 t I 128 126 tZU E A R L Y P E R I O D B U R I A L E U L A T E P E R I O D B U R I A L G 3 U N D A T A B L E B U R I A L £k HIGH S T A T U S © HIGH I N T E R M E D I A T E S T A T U S si I N T E R M E D I A T E S T A T U S 6" LOW S T A T U S - 191 -6.2.4.6. Test Implications for Achieved versus Ascribed Status Status d i f f e r e n t i a t i o n among the Early period b u r i a l s seems to r e f l e c t aspects of both an achieved and an ascribed system. Unfortunately, a conclusive test of whether status d i f f e r e n t i a t i o n i s based upon age and sex or whether i t cross-cuts age and sex categories cannot be made. Even a f t e r estimating the sex of several b u r i a l s i n Chapter 4, a chi-square te s t to assess whether the derived status d i s t i n c t i o n s are di f f e r e n t from one another i n terms of age and sex (Rothschild 1979) i s hot possible due .to low sample size of sexed b u r i a l s . Thirty-four of the 75 single adult b u r i a l s i n the Early period remain unsexed. However, some insight into whether status d i f f e r e n t i a t i o n cross-cuts age and sex categories or not can be made by a v i s u a l inspection of the age and sex groups within each status group (see Fig . 6-4). Including estimated sexed b u r i a l s , the high status group contains six males and one female. However, the four d i s t i n c -t i v e l y high status b u r i a l s are male (E26, 59, 54, 9). This fact may r e f l e c t an achieved system of status d i f f e r e n t i a t i o n because females may not have access to the highest status positions. B u r i a l 5 4 has no skeleton. On the other hand, i t i s possible that the ivory disk (bl) i s a badge of rank that symbolizes only male high status. It i s conceivable that the disk could be a marker of a p a r t i c u l a r achieved status. This p o s s i b i l i t y i s not considered i n the l i t e r a t u r e . The two bu r i a l s with the high status raw material of elephant ivory - 192 -are E26, a r e l i a b l y estimated male, and E59, a known male. The ochre and animal shaped pot i n E9 may r e f l e c t r i t u a l i s t i c status rather than status based on wealth. However, t h i s b u r i a l i s grouped with the other high status b u r i a l s i n terms of grave goods. The high intermediate status l e v e l contains six males and one female. The fact that t h i s female and the female i n the high status group (E102) are of higher status than other females i n the cemetery suggests that status d i f f e r e n t i a t i o n in the Early period i s ascribed. The b u r i a l i n t h i s status l e v e l with the jade huan i s of unknown sex. The low intermediate status group and the low status group contain both males and females, suggestive of ascribed status. The sex composition of the low status group i s e s p e c i a l l y unclear. The fact that three of the four children i n the Early period are located i n th i s c l u s t e r i s suggestive of achieved status. The fourth c h i l d , E94, i s i n the high intermediate status group. This factor i s suggestive of ascribed status, as well as the fact that the b u r i a l has a grave form that represents the greatest amount of energy expenditure, the log tomb with second l e v e l platform. Also, the log tomb i s type two, which represents more energy expenditure than the other log tombs among the single Early b u r i a l s (although the type of log tomb i s not described i n every case). The other c h i l d b u r i a l s are i n simple p i t s . The small number of c h i l d b u r i a l s i n the Early period - 193 -(4 out of 79) i n h i b i t s a comparison of mortuary treatment of children versus adults. The p o s i t i o n of E94 along Dimension 1 i n Fig. 6-1 i s quite far to the r i g h t , away from many high status b u r i a l s . Also, i f energy expenditure i n terms of grave form does symbolize ascribed status for E94, there should be a greater c o r r e l a t i o n of high status i n terms of grave goods and high status i n terms of grave form for the adult b u r i a l s . It i s also possible that the grave form of E94 r e f l e c t s the achieved status of his or her parents. The number of indivudual b u r i a l s i n each status l e v e l (including multiple burials) b a s i c a l l y indicates a ranking pyramid, expected by many processual mortuary studies to r e f l e c t ascribed ranking: eight b u r i a l s i n the high status group, 17 i n the high intermediate group, 36 i n the low i n t e r -mediate group, and 2 4 i n the low status group. It i s possible, though, that the presence of a ranking pyramid i s not a r e l i a b l e test implication for ascribed status i n every case. Suttles (1960:297) states that high status individuals i n Coastal S a l i s h communities comprise the majority of the population, not the minority. 6.2.4.7. Multiple Burials It i s d i f f i c u l t to compare the status of individuals i n the six Early period multiple b u r i a l s because the grave goods associated with each i n d i v i d u a l are not apparent. However, the status positions of the i n d i v i d u a l s i n the multiple b u r i a l s can be broadly assessed. My interpretation of status d i f f e r e n t i a -- 194 -t i o n among the multiple b u r i a l s also indicates that elements of both achieved and ascribed s o c i a l status are represented i n the Early period. In F i g . A6-12, the multiple b u r i a l s are described i n terms of the mortuary attributes included i n the analysis of the single b u r i a l s . Grave 13 seems c l e a r l y a high status b u r i a l , equivalent to the b u r i a l s i n the high status group. In terms of grave goods, i t has many of the t r a i t s exclusive to the high status group: the highest quantity of pottery, one elephant ivory carved cylinder and two elephant ivory disks. It also has 14 pig s k u l l s , where the greatest quantity among the single b u r i a l s i s f i v e . Another distinguishing feature i s i t s large s i z e . While the • 2 areas of the other multiple b u r i a l s range from 2.04 to 2.90m , 2 the area of grave 13 i s 6.46m . The largest area of the single 2 b u r i a l s i s 3.82m . Even though two bodies are i n E13, the grave area i s comparable to the large single graves. Grave 13 i s also the only Early period grave with the most elaborate log tomb type, three. S i g n i f i c a n t l y , grave 13 i s located i n the northern s p a t i a l c l u s t e r (see F i g . 6-5) with fiv e of the high status b u r i a l s . The body d i s p o s i t i o n of both individuals i n the grave i s supine. The r e l a t i v e statuses of the other multiple b u r i a l s can be t e n t a t i v e l y interpreted as follows: the individuals i n E31 can be considered low i n status;and E28, E69, E l l l , and E35 as intermediate i n status. The r e l a t i v e statuses of males and females i n multiple b u r i a l s from Dawenkou Culture s i t e s have received much comment in the Chinese archaeological l i t e r a t u r e (Pearson 1981:1084-- 195 -1085). Pearson (1981:1084) explains that E35 (and LI) are often c i t e d as i n d i c a t i n g low status of females i n comparison to males because the majority of grave goods within the grave are placed on the side of the male. The photos of E13 and E l l l i n the s i t e report also show the majority of grave goods near the male. This d i s p a r i t y i s more apparent i n E13 and E35 than E l l l . S i g n i f i c a n t l y , the male i n E13 i s associated with the two elephant ivory disks and the elephant ivory carved cylinder. The disks are placed d i r e c t l y on the male's upper chest. How-ever, the large quantities of ceramics are placed over the head of both the male and female. The pig s k u l l s are placed on the side of the female. It i s d i f f i c u l t to determine from the photographs whether the males and females i n each grave were buried at separate times or at one time. The association of the high status items i n E13 i s support for the p o s s i b i l i t y that the Early period represents a system of achieved status d i f f e r -e n t i a t i o n . 6.4.2.8. Conclusions It i s not possible to ascertain conclusively whether status d i f f e r e n t i a t i o n i s achieved or ascribed, i n terms of the test implications that have been u t i l i z e d i n processual mortuary studies. The Early period appears to have elements of both types of status systems. Rothschild (19 79:672) has si m i l a r r e s u l t s at an Archaic period and a Mississippian period s i t e . Regardless of the descriptive term applied to the Early period system of status d i f f e r e n t i a t i o n ("ascribed" or "achieved") the b u r i a l s are d i f f e r e n t i a t e d into four broad - 196 -status levels on the basis of grave goods. Even though the placement of some b u r i a l s into certain levels i s debatable, the four levels are indicated by the results from both the multidimensional scaling and Ward's Method of cluster analysis. However, the lowest and highest status levels on the basis of grave goods are the most c l e a r l y distinguished. The i n d i v i d u a l b u r i a l s i n six of the seven pairs of int r u s i v e b u r i a l s are not i n the same status l e v e l (see Figs. A3-2 and 6-4). Since each status l e v e l includes some intrusive b u r i a l s within the Early period, i t i s possible that none of the four status l e v e l s i s r e s t r i c t e d to one part of the Early period. It i s not clear why the d i f f e r e n t i a t i o n of b u r i a l s i n terms of grave form and grave size d i f f e r s from that based upon grave goods. Because some of the high or high intermediate status b u r i a l s i n terms of grave goods also have log tombs, second l e v e l platforms, or large graves and none of the low status b u r i a l s have these t r a i t s , i t seems that energy expend-itu r e r e f l e c t s status d i f f e r e n t i a t i o n at least i n part. Although elements of both an ascribed and achieved system of status d i f f e r e n t i a t i o n are present i n the Early, period, I propose that an ascribed system i s more l i k e l y . The d i s p a r i t y i n the quantity and quality of grave goods i n the high status group i n comparison to the low status group i s suggestive of ascribed status. However, the high status group only has exclusive access to a few a r t i f a c t forms or raw materials. - 197 -Pearson (1981:1086) proposes that the presence of stone orna-ments i n male and female graves i s suggestive of ranking. I think the presence of hairpins, rings, and bracelets i n male and female Early graves could r e f l e c t ranking. I think that the quantity and qu a l i t y of grave goods i n the high and high intermediate status groups by i t s e l f i s suggestive of ascribed status. My analysis also depicts a p a r t i a l segregation of wealthy graves. Pearson (ibid) states that lack of a complete segregation may indicate lack of a hereditary r u l i n g class. Or, lack of a complete segregation may indicate a c h i e f l y o f f i c e (Pearson 1983:140). Although evolutionary typologies such as those by Fr i e d (1967) and Service (1975) are controversial (see Chapter 2), they have u t i l i t y for descriptive purposes. The Early period b u r i a l s may r e f l e c t a "ranked" society i n Fried's (196 7) sense. In Chapter 5, I argued that a l l the b u r i a l s from both periods could r e f l e c t a descent group. Fried (1967:116) states that one of the major c h a r a c t e r i s t i c s of a ranked society i s organ-i z a t i o n by kinship i n terms of descent. Members of a ranked society are related by descent usually i n terms of a lineage or clan (Fried 1967:125-126). However, Fr i e d does not discuss the nature of a ranked.society i n terms of i t s mortuary r i t u a l . One p o s s i b i l i t y i s that the Early period b u r i a l s represent a conical clan i n Kirchhoff's (1955) sense. I argue that Neolithic communities could have kept a record of relations of individuals to p a r t i c u l a r ancestors of concern by location of graves i n the same cemetery i n which the ancestors of concern are buried. - 198 -The Early period b u r i a l s could r e f l e c t a chiefdom i n Service's (1967) sense. Service (1967:79) includes the con-cept of conical clan after Kirchhoff (1955) i n his d e f i n i t i o n of chiefdom. Peebles and Kus (1977:422) point out that the term "chiefdom" includes much v a r i a t i o n i n degree of socio-c u l t u r a l complexity. Due to the lack of extreme d i f f e r e n t i a -t i o n among the majority of b u r i a l s , the Early period may r e f l e c t a simple chiefdom. Individuals of high r i t u a l i s t i c standing are expected i n both the "ranked" (Fried 1967:137) and "chiefdom" (Peebles and Kus 1977:422) types of s o c i e t i e s . As discussed previously, E9 may r e f l e c t r i t u a l i s t i c standing. However, I r e a l i z e now that E9 i s probably dated to the Late period instead of the Early (see Chapter 3). Even though t h i s b u r i a l may have been i n c o r r e c t l y c l a s s i f i e d as Early, I do not think my i n t e r -pretation of the high status group of b u r i a l s changes s i g n i f i c -antly. I f E9 i s a Late period b u r i a l , high status during the Early period i s not characterized by the animal shaped vessel and red ochre r but by the elephant ivory disk, comb and carved cylinder, and the greatest quantity of ceramic vessels. 6.3. Analysis of Status, Late Period 6.3.1. Data 6.3.1.1. The Multivariate Analyses Twenty-eight variables r e f l e c t i n g grave goods and 32 single b u r i a l s are included i n the multivariate analyses for the Late period. The 32 b u r i a l s are l i s t e d i n F i g . A6-13. Three of - 199 -these b u r i a l s are disturbed: L3, L46, and L77. The 28 v a r i -ables and t h e i r attributes are l i s t e d i n Fig . A6-14. As i n the analysis of status for the Early period, black-ware and painted ware are included. A t h i r d type of ware appears in the Late period that i s l i k e l y to r e f l e c t high status, white ware. The white ware vessels from Dawenkou are extremely f i n e l y made and :. were f i r e d , at. high.temperatures (Shangraw 1977:386). They are extremely b r i t t l e and were probably not used for u t i l i t a r i a n purposes (Shangraw, personal communication 1983). White ware i s found i n apparent high status Shang dynasty graves (Li Chi 1977:202). The color white may have symbolized the Shang state (Chang 19 80:57) or the soul during the Han Dynasty (Yu 1981:83). The frequency of occurrence of mortuary attributes i s given i n Fig . A6-14. Four forms of ceramic vessels are included i n the analysis. The wine vessel, serving stand and t a l l stemmed cup, but not the bottle (ping), occur i n Early b u r i a l s as well. The appear-ance of the bo t t l e s from the photographs of the s i t e report, as well as the fact that a few b u r i a l s contain great quantities, suggest they are high status items. The b o t t l e i s not coded in an additive manner because i t does not occur i n many b u r i a l s . The serving stand and t a l l stemmed cup are coded i n an addi-t i v e manner. The frequency d i s t r i b u t i o n s for a l l variables coded i n an additive manner are i n Figs. A6-15 to A6-21. By the Late period there i s a s l i g h t increase i n the frequency of the serving stand i n b u r i a l s and a great increase i n the - 200 -frequency and quantity of the t a l l stemmed cup. The t o t a l s of ceramic vessels increase a great deal by the Late period as well. The ornamental forms included i n t h i s analysis which are i n the Early period analysis are the r i n g , bracelet, h a i r p i n , h a i r -t i e , , small round stone (bong), and probable ornament parts. There i s an increase i n the frequency of occurrence of the brace-l e t by the Late period. No huan (the small f l a t annular piece of stone) occur i n the Late period. Since the small f l a t piece of bone (ban) i s i n only one Late b u r i a l , i t must be excluded from the multivariate analyses. Additional ornamental categories i n the Late period analysis are the neck or head ornament and the horn zhui (as discussed i n Chapter 4, either an ornament or a net weight). In Chapter 5, I concluded the head and neck ornaments are most l i k e l y female a r t i f a c t s and the zhui, a male a r t i f a c t . The a r t i f a c t forms with a decorative function which were in the Early period analysis are t u r t l e s h e l l , bone or horn spoon, and the carved cylinder. The l a t t e r a r t i f a c t type occurs with greater frequency by the Late period, i n both elephant ivory and bone. Jade occurs more frequently i n the Late period but s t i l l appears to be a r e s t r i c t e d material. It occurs i n four b u r i a l s i n f i v e forms: h a i r p i n , bracelet, r i n g , c h i s e l , and spade. The two jade spades (chan) are unused and were apparently made for symbolic purposes (Gao 1978:33). This form may be a badge of rank. The remaining variables were also i n the Early period analysis. Pig s k u l l s do not occur as frequently and there i s - 201 -l i t t l e difference i n the quantities among b u r i a l s . The deer teeth (additively coded) are present i n varying quantities. Oyster s h e l l pieces and other pig parts are included. Other ad d i t i v e l y coded variables are stone t o o l t o t a l , bone tool t o t a l , and raw material pieces. As i n the Early period status analysis, some of the included variables are s l i g h t l y redundant. The three ware colors are s l i g h t l y redundant with the four ceramic forms, and the two add i t i v e l y coded ceramic forms with the t o t a l amount of vessels. Upon retrospect, I believe i n c l u s i o n of the category "jade" created more than a s l i g h t redundancy with the hairpin, bracelet, and ring categories. The b u r i a l s with both jade "present" and either of the three ornamental categories above are: L10 (bracelet, r i n g ) , L25 (hairpin), and L117 (hairpin). The p o s s i b i l i t y that the status of these b u r i a l s i s misrepresented by the multivariate analysis i s considered i n the interpretation, section. Since the c h i s e l was included i n the stone t o o l t o t a l s , the status of the b u r i a l with both the c h i s e l and jade present (L4) should not be misrepresented. The unused jade spades were not included i n the stone t o o l t o t a l s . 6.3.1.2. Energy Expenditure The Late period b u r i a l s that have a grave form other than simple p i t are marked on Fi g . A6-13. Six of the seven log tombs are type three, the type which r e f l e c t s the greatest amount of energy expenditure. The authors of the s i t e report do not state the type of the seventh log tomb, L10 4. It i s implied t h i s tomb i s either type one or two. Only one of the - 202 -seven Early period log tombs was a type three. The d i s t r i b u t i o n of grave size for the Late period single b u r i a l s i s i n Fig . A6-22. While two d i s t i n c t i o n s i n area of grave were noted for the Early period, three are apparent for the Late. Three b u r i a l s are of large s i z e , nine of medium si z e , and 19 small. The area of one grave, L12 7, could not be c a l -culated due to lack of information provided i n the s i t e report. Due to i t s small r e l a t i v e size on the s i t e map, i t i s tenta-t i v e l y c a l l e d a small b u r i a l for the purposes of t h i s analysis. The large size i s much larger than the largest single b u r i a l 2 2 i n the Early period (13-15.00m versus 3.82m i n the Early period). Also the medium size range i n the Late period encompasses the large size of the Early period b u r i a l s (2.60-7.99m2). 6.3.1.3. High Status Unique Items The four a r t i f a c t forms excluded from the multivariate analyses that are considered probable unique, high status items are: one elephant ivory comb (in L10), two semi-circular stone pendants (in L72, huarig) , 84 pieces of a l l i g a t o r hide (in L10), and turquoise i n the form of a necklace (in L10) and i n l a i d i n a bone carved cylinder (in L4, described on pages 101-102 of the report). In the western and Chinese archaeological l i t e r a -ture regarding Dawenkou, the a l l i g a t o r hide i n L10 i s referred to as crocodile hide. However, the term A l l i g a t o r sinensis (page 15 7 of the report) refers to a l l i g a t o r , not crocodile. This difference i n genus and species i s s i g n i f i c a n t because the Chinese crocodile probably ranged only as far north as the - 203 -southern t i p of China, while the Chinese a l l i g a t o r probably was present i n the lower Yangtze River v a l l e y region (Campbell 1981: 37). The source area for turquoise is unknown. I n l a i d turquoise items have been found i n Shang Dynasty graves (Chang 19 80:345). Huang pendants of jade have been found at Shang and Zhou period s i t e s (the Zu Hai dictionary 1978:315). The huang form may be a badge of rank during the Late period. 6.3.2. Results: The Multidimensional Scaling Nine dimensions resulted from the multidimensional scaling analysis. The f i r s t two accounted for 43.04% of the v a r i a b i l i t y i n the data set - with the f i r s t dimension accounting for 28.25% and the second, 14.79%. The t h i r d and fourth dimensions accounted for 12.60% and 8.84% of the v a r i a b i l i t y , respectively. Fi g . 6-6 depicts the relationship of the Late period b u r i a l s when the f i r s t two dimensions are plotted together. Like the Early period scaling r e s u l t s , Dimension 1 on F i g . 6-6 can be interpreted as c l e a r l y r e f l e c t i n g status d i f f e r e n t i a t i o n . Burials on the f a r l e f t side (L10, L25, L117, L47) are i n t e r -preted as highest i n status, those i n the midsection (the majority) as intermediate i n status, and those at the ri g h t end (L123, L93, L16, L122, L96, L15, L100, L46) r e l a t i v e l y low i n status. The in f e r r e d high, intermediate and low status b u r i a l s contain the mortuary attributes from F i g . A6-14 that I expected from my exploratory study of status d i f f e r e n t i a t i o n to r e f l e c t high, intermediate and low status, respectively. F i g . 6-6 does not depict any sharp s p a t i a l breaks between b u r i a l s along Dimension 1. The high status b u r i a l s (L25, L10, L117, L47) - 204 -FIGURE 6-6. Multidimensional scaling plot of Late Period b u r i a l s i n the analysis of status, with clusters from Ward*s Method. l a rge in area (3) medium in area (9) s m a l l in a rea (20) log tomb (7) s e c o n d leve l p l a t f o r m (5) 4 3 d isturbed burial (3) 4 3 controversial former Middle B u r i a l (I) 7 y o u t h (I) la no skeleton (3) 4 3 high s ta tus , unique objects (3) Note: Clusters 1-8 from Ward s Method eight group solution are circled. In the three group solution Clusler I is the same; grouped together are Clusters 2 - 5 ond C lus te rs 6 - 8 . In the two group solution grouped logether are C lus te rs 1-5 and C lus te rs 6 - 8 . - 205 -are most c l e a r l y distinguished. It appears that status d i s -t i n c t i o n s among most b u r i a l s are gradual and not abrupt. Unlike the Early period r e s u l t s , a group of low status b u r i a l s i s not distinguished. The low percentage of v a r i a b i l i t y explained by Dimension 1 may be due to the low number of attributes i n common among the Late period b u r i a l s , as i n the Early period. The dimension of v a r i a b i l i t y explained by Dimension 2 i s not apparent. Assessment of the mortuary attributes of b u r i a l s along Dimension 2 does not indicate that Dimension 2 r e f l e c t s status d i s t i n c t i o n s . In the analysis of status d i f f e r e n t i a t i o n for both the Early and Late periods, I do not attempt to interp-ret dimensions three and four due to time l i m i t a t i o n s , the fact that Dimension 1 seems to c l e a r l y r e f l e c t status d i s t i n c t i o n s , and the fact that there i s f a i r l y good agreement i n the status d i s t i n c t i o n s exhibited by the multidimensional s c a l i n g and cl u s t e r analysis. 6.3.3. The Three Types of Cluster Analysis Once again, the dendrogram from Ward's Method best r e f l e c t s status d i s t i n c t i o n s among the b u r i a l s . This judgement i s based upon assessment of the composition of each b u r i a l i n terms of the mortuary attributes within the clusters at varying distance levels and from my a p r i o r i expectations based upon my explor-atory assessment of status d i f f e r e n t i a t i o n . The dendrograms from the Complete Linkage and Average Linkage techniques exhibit some of the properties that were encountered i n the dendrograms of the Early period b u r i a l s . Again, the average - 206 -Linkage dendrogram compares better with the Ward's dendrogram. The Ward's dendrogram i s presented i n F i g . 6-7. The clusters from four distance levels i n the Ward's dendrogram can be interpreted as e x h i b i t i n g status relationships among b u r i a l s : the eight group solution at the 0.6209 distance l e v e l , the three group solution at 0.9813, and the two group solution at 1.0018. The attributes c h a r a c t e r i s t i c of the eight clusters are shown i n F i g . A6-23. At f i r s t , the eight clusters appeared d i s t i n c t i n terms of composition of mortuary attributes and seemed to meet my expectations from my exploratory study. However, the three and two group solutions suggest that some of the status d i s t i n c t i o n s indicated by the eight group solution are not s i g n i f i c a n t . Cluster one contains four b u r i a l s that are c l e a r l y high i n status and d i s t i n c t from other c l u s t e r s . These b u r i a l s contain great quantities of ceramics, t a l l stemmed cups, serving stands, and bone and stone too l s . They also contain jade, white ware, and a variety of ornamental types. The three b u r i a l s that may have been misrepresented are i n t h i s c l u s t e r : L10, L25, and L117. However, the fact that these b u r i a l s are distinguished by several attributes as well as jade and jade ornaments indicates the status of these b u r i a l s may not have been over-represented i n the multivariate analyses. L10 and L117 are also distinguished by the presence of high status unique items. - 207 -FIGURE 6-7. Dendrogram of Late Period b u r i a l s , Ward's Method. I i i i i i i i i I i o I -L2 -L5 -L64 -L77 -L127 -L104 -L105 Cluster 7 -L121 L15 L100 -L46 -L96 -L16 -L122 -L93 -L123 -L124 -L3 -L72 -L24 -L60 -L22 -L75 -L98 -L4 -L125 -L17 -L10 -L25 -L47 -L117 l l I I l I I I l l c o r * - u " i n c s O o o r ^ m *y c s o rH i H rH i-t rH rH O O O O O O Cluster 8 Cluster 6 Cluster 5 Cluster 4 Cluster 3 Cluster 2 Cluster 1 d i s t a n c e - 208 -Cluster two i s characterized by the largest quantity of bone t o o l s , stone t o o l s , and raw material. Cluster 4 i s characterized by white ware, great quantities of serving stands, and a few ornament types. Cluster 5 i s characterized by white ware, the medium quantity of ceramics, and the necklace and other ornaments. Although clu s t e r s 2, 4, and 5 are d i s t i n c t i n terms of mortuary a t t r i b u t e s , i t i s d i f f i c u l t to judge whether one c l u s t e r represents higher status than another. Clusters 3, 7, and 8 appear d i s t i n c t i n terms of mortuary attributes but a l l appear t h i r d highest i n status. Burials i n these clusters contain a few ornament types and decorative items. Cluster 6 appears lowest i n status, with the lowest quantities of ceramics and bone or stone t o o l s , and few other a r t i f a c t s , most of which are u t i l i t a r i a n . In the three group solution (see F i g . 6-7), the highest status c l u s t e r (1) of the eight group solution i s distinguished. The c l u s t e r s I had interpreted as intermediate and low i n status from the eight group solution are grouped s l i g h t l y d i f f e r e n t l y : clusters 2, 3, 4, 5, and clusters 6, 7, 8, are grouped together. The two group solution i s a clear natural break i n the dendro-gram i n F i g . 6-7. Clusters 1, 2, 3, 4, 5, of the eight group solution are grouped together and clusters 6, 7, and 8. The clusters from Ward's eight group solution are plotted on F i g . 6-6. There i s a f a i r l y good agreement i n the ordering of b u r i a l s by both the scaling and c l u s t e r i n g . The high status cluster (for the three group solutions as well) i s located at - 209 -the l e f t end of the p l o t . Cluster 2, one of the clusters interpreted as next highest i n status, i s located adjacent to cluster 1 on the p l o t . The low status c l u s t e r , 6, i s located at the r i g h t end of the p l o t . The clusters also previously interpreted as next highest i n status (4 and 5) are located i n the same area along Dimension 1 as Cluster 2. The clusters previously interpreted to represent another low status l e v e l , 3, 7, and 8, are located between clusters 2, 4, and 5 and cluster 6 along Dimension 1, as expected. However, there i s less agreement between the three and two group solutions of Ward's Method and the scaling plot i n Fig.6-6. The high status c l u s t e r i n the three group solution i s d i s t i n -guished on the scaling p l o t . However, the other two clusters do not match well with the scaling r e s u l t s . The clusters from the two group solution do not agree well with the scaling results either. 6.3.4. Interpretation of Status Di s t i n c t i o n s In the Late period and Assessment of Change through Time 6.3.4.1. D i s t r i b u t i o n of the Unique Items The correspondence between status d i f f e r e n t i a t i o n as r e f l e c t e d by grave goods in the multivariate analyses and the probable high status, unique objects i s not as good as that among the Early period b u r i a l s . Some of the objects are found i n clusters other than the highest status grouping. The c l e a r l y i d e n t i f i e d highest status group (Cluster 1) i n F i g . 6-6 contains - 210 -only one of the three b u r i a l s with the probable high status unique objects. However, the b u r i a l i n question, L10, contains three of the high status objects. This b u r i a l i s shown on the scaling plot as the b u r i a l of highest status i n Cluster 1. The status of L10 may be exaggerated due to i t s presence of jade, a jade bracelet, and a jade r i n g . The status of L25 and L117 may not be as exaggerated because they each contain two redundant categories instead of three (jade and the h a i r p i n ) . L10 contains the elephant ivory comb, the a l l i g a t o r hide, and the turquoise necklace. L4, located i n Cluster 2 (Fig. 6-6) contains the bone cylinder i n l a i d with turquoise. L72, with the two huang pendants, i s i n Cluster 5. Ward's Clusters 2 and 5 were i n t e r -preted as next highest i n status. 6.3.4.2. Argument f>6r Three or Two Status Levels It i s even more d i f f i c u l t to i n f e r an approximate number of status d i s t i n c t i o n s based upon wealth i n terms of grave goods for the Later period b u r i a l s than among the Early. The d i f f e r -ences i n status d i s t i n c t i o n s as indicated by Ward's eight, three and two group solutions, the multidimensional scaling r e s u l t s , and the d i s t r i b u t i o n of the probable high status unique objects suggest that many bu r i a l s are not d i s t i n c t i v e l y d i f f e r e n t i n status i n terms of grave goods from one another. I i n f e r that either three or two status levels based upon wealth i n terms of grave goods may be argued as being represented among the Late period b u r i a l s . However, the composition of these status levels i n terms of b u r i a l s i s open to some debate. - 211 -The b u r i a l s i n Ward's C l u s t e r 1 (L10, L25, L117, L47) should be c o n s i d e r e d the h i g h e s t s t a t u s l e v e l because they are d i s t i n g u i s h e d on the s c a l i n g p l o t by the e i g h t and three group s o l u t i o n s from the Ward's dendrogram and by the probable hi g h s t a t u s o b j e c t s i n L10. I argue t h a t C l u s t e r s 2, 4, 5, on F i g . 6-6 should be c o n s i d e r e d the next h i g h e s t s t a t u s l e v e l . I o r i g i n a l l y i n t e r p r e t e d these c l u s t e r s as d i s t i n c t , but roughly e q u i v a l e n t i n s t a t u s on the b a s i s of t h e i r mortuary a t t r i b u t e s . Ward's three group s o l u t i o n d e p i c t s C l u s t e r s 2, 3, 4, and 5 tog e t h e r . The m u l t i d i m e n s i o n a l s c a l i n g p l o t d e p i c t s C l u s t e r s 2, 4, and 5 as i n the same area along Dimension 1 wi t h C l u s t e r 3 c l e a r l y separated from them. I conclude t h a t the m u l t i d i m e n s i o n a l s c a l i n g p l o t d e p i c t s the s t a t u s r e l a t i o n s h i p s among the Late p e r i o d b u r i a l s more r e l i a b l y than Ward's Method. C l u s t e r s 2 and 5 c o n t a i n the other probable h i g h s t a t u s , unique o b j e c t s . C l u s t e r s 2, 4, and 5 are l o c a t e d adjacent t o C l u s t e r 1 on the s c a l i n g p l o t . I i n t e r p r e t the b u r i a l s i n C l u s t e r s 8, 3, 7, and 6 i n F i g . 6-6 as r e p r e s e n t i n g a low s t a t u s l e v e l . C l u s t e r 6 was o r i g i n a l l y judged as d i s t i n c t i n terms of mortuary a t t r i b u t e s . However, many o f the b u r i a l s i n C l u s t e r 6 are l o c a t e d i n the same area along Dimension 1 as b u r i a l s i n C l u s t e r 7 and 3. A l s o , C l u s t e r s 6, 7, and 8 are grouped t o g e t h e r i n Ward's three and two group s o l u t i o n s . I t i s p o s s i b l e t h a t C l u s t e r 1 should be c o n s i d e r e d grouped w i t h C l u s t e r s 2, 4, and 5 to form one s t a t u s l e v e l and C l u s t e r s - 212 -8, 7, 3, and 6 to form another. Three of the four b u r i a l s i n Cluster 1 contain the redundant categories (L10, L25, L117). Also, the c l e a r l y depicted two group solution i n the Ward's dendrogram groups Cluster 1 with 2, 4, and 5. The mortuary attributes that characterize the three status lev e l s are depicted i n F i g . 6-8. In the Early period, high status i s characterized by exclusive access to only a few a r t i f a c t forms and raw materials. The differences between the status levels are mainly quantitative, not q u a l i t a t i v e . In the high status group, b u r i a l 10 i s distinguished i n terms of i t s quantity of bottles (38 versus 4 or 5 for the other b u r i a l s i n t h i s status level) and quantity of ceramics (the highest i n the cemetery: 9 3). The next highest quantity of ceramics i s 71, i n L126. Burials L10 and L117 each contain the unused jade chan spade, the possible badge of status. As mentioned previously, L10 also contains the elephant ivory comb, turquoise necklace and a l l i g a t o r hide. B u r i a l L126 i s distinguished among the b u r i a l s of i n t e r -mediate status. The position of L126 i n the scaling plot i s very close to the b u r i a l s i n the highest status group. L126 contains the second highest quantity of b o t t l e s : 26. Bur i a l L72 with the stone pendants (huang) and L4 with the i n l a i d turquoise are i n t h i s status group. The disturbed b u r i a l s i n the low status group are L77 and L46. L77 contains an item found only i n one other b u r i a l , multiple b u r i a l LI, six end pieces of arrows. It i s possible the arrows symbolize a s p e c i f i c status and that L77 i s a high status i n d i v i d u a l as the authors - 213 -FIGURE 6-8. Characteristics of the Three Status Levels, Late period. intermediate high status status low status (4 burials) (8 burials) (20 burials) + % of t o t . + % of t o t . + % of tot. white ware 4 40.0 4 40. 0 2 20.0 black ware 4 26. 7 4 26. 7 7 46.7 painted ware 2 33. 3 3 50.0 1 16. 7 wine vessel 1 10.0 2 20.0 7 70.0 serving stand 0 1 8.3 2 16.7 9 75. 0 serving stand 1 0 0 4 28.6 10 71.4 serving stand 2-12 3 50.0 2 33.3 1 16.7 bottle 3 42.9 2 28.6 2 28.6 t a l l stemmed cup 0 0 0 1 7.1 13 92.9 t a l l stemmed cup 1-6 1 6.7 7 46.7 7 46. 7 t a l l stemmed cup 14-16 3 100.0 0 0 0 0 t o t a l pots 1-13 0 0 2 10.0 18 90.0 t o t a l pots 16-21 0 0 4 66. 7 2 33.3 jade 3 75. 0 1 25.0 0 0 bone cylinder 3 50.0 2 33.3 1 16. 7 ivory cylinder 2 40.0 2 40.0 1 20.0 t u r t l e s h e l l 1 33. 3 0 0 2 66.7 ring 2 40.0 2 40.0 1 20.0 bracelet 4 33.3 2 16.7 6 50.0 hairpi n 4 50.0 2 25.0 2 25.0 spoon 1 20.0 1 20.0 3 60.0 stone t o o l t o t a l 0 1 7.1 3 21.4 10 71.4 stone t o o l t o t a l 1-3 2 15. 4 2 15.4 9 69.2 stone t o o l t o t a l 5-19 1 20.0 3 60.0 1 20.0 Cont'd.. . - 214 -FIGURE 6-8 continued intermediate high status status low status (4 burials) (8 burials) (20 burials) + % of t o t . + .%. of t o t . + % of tot bone t o o l t o t a l 0 3 17.6 3 17.6 11 64. 7 bone tool t o t a l 1-6 0 0 2 18.2 9 81. 8 bone t o o l t o t a l 10-27 1 25. 0 3 75.0 0 0 hair t i e 2 20. 0 1 10. 0 7 70. 0 neck or head ornament 2 40.0 2 40.0 1 20.0 zhu"i 1 25.0 2 50.0 1 25.0 ornament part 1 33. 3 1 33. 3 1 33. 3 bing 0 0 2 66. 7 1 33. 3 pig s k u l l s 3 42.9 2 28.6 2 28.6 other pig parts 2 22.2 4 44. 4 3 33.3 deer teeth 0 1 16.7 2 33. 3 3 50.0 deer teeth 1-2 1 5.0 5 25.0 14 70.0 deer teeth 3-5 2 33.3 1 16.7 3 50.0 raw material 0 4 15.4 2 7.7 20 76.9 raw material 1 0 0 3 100.0 0 0 raw material 11-•26 0 0 3 100.0 0 0 oyster s h e l l 0 0 2 40.0 3 60.0 - 215 -of the s i t e report suggest. Burials L125, L17, L4 and L72 also contain items not included i n the multivariate analyses. B u r i a l L17 contains three jaw bones of the genus F e l i s , L125 contains two deer leg bones , L4 contains some grains of sand (sha li.) , (from page 27 of the report), and L72 has the small f l a t piece of bone (ban). The high status l e v e l i s not characterized by exclusive access to any forms of material except the unused jade spade and the elephant ivory comb. The only raw material exclusive to t h i s l e v e l i s a l l i g a t o r hide. The only form of a r t i f a c t present i n the greatest quantity i s the t a l l stemmed cup. Even i f the two status l e v e l i nterpretation i s accepted (combining the high and intermediate status l e v e l s , above), there are s t i l l only a few a r t i f a c t forms, raw materials, or quantities of a r t i f a c t s exclusive to the high status l e v e l : the greatest quantity of ceramics, t a l l stemmed cup, and raw material pieces; the presence of jade, turquoise, a l l i g a t o r hide; the elephant ivory comb, the unused jade spade, and the stone pendant (huang). Many of the p o t e n t i a l high status forms, materials, or great quantities of a r t i f a c t s are found i n b u r i a l s of the low status cl u s t e r , such as white ware, black ware, painted ware, great quantities of the serving stand, hairpins, head or neck ornament, and the elephant ivory carved cylinder. The low status b u r i a l s i n the Early period were much more c l e a r l y set apart from other b u r i a l s i n terms of grave goods. - 216 -6.3.4.3. D i f f e r e n t i a t i o n by Energy Expenditure There i s a change from the Early period to the Late i n terms of correspondence i n status d i f f e r e n t i a t i o n based upon grave goods with that based upon energy expenditure. There i s a better correspondence i n the Late period than i n the Early, as expected from my exploratory study of the cemetery and by Pearson (1981:1082). In the high status l e v e l , three of the four b u r i a l s contain log tombs with second l e v e l platforms. The fourth b u r i a l (L47) has the log tomb but no platform. The multidimensional scaling solution depicts t h i s b u r i a l as being s l i g h t l y lower i n status than the other three b u r i a l s of t h i s status l e v e l . In the intermediate status l e v e l , there are two more b u r i a l s with log tombs. A l l the log tombs i n the high and intermediate status l e v e l s are type three, the most elaborate. The l a s t log tomb i s i n L104. This log tomb i s probably type one or two. L10 4 i s i n the low status l e v e l . However, L10 4 also has a second l e v e l platform unlike b u r i a l s L47, L60, and L126 which are located i n higher status lev e l s i n terms of grave goods. The l a s t b u r i a l with a second l e v e l platform i s L9 8, i n the low status l e v e l . Thus, two b u r i a l s i d e n t i f i e d as low i n status i n terms of grave goods have grave forms other than the simple p i t . None of the Early period b u r i a l s i d e n t i f i e d as low status have log tombs or second l e v e l platforms. As i n the Early period, the c o r r e l a t i o n between status i n terms of grave size and status i n terms of grave goods i s not strong. The three large graves (L10, L60, L126) are i n the high and intermediate l e v e l s . Three of the nine medium-sized graves - 217 -are l o c a t e d i n the high s t a t u s group and two i n the i n t e r m e d i a t e group. Four are i n the low s t a t u s l e v e l . Thus, every s t a t u s l e v e l has a t l e a s t one b u r i a l w i t h an area t h a t r e p r e s e n t s more energy expenditure than the norm. The three b u r i a l s t h a t are l a r g e i n s i z e (L10, L60, and L126) a l s o have high-energy grave forms. Grave L10 i s d i s t i n c t because i t has the grave form t h a t r e p r e s e n t s the g r e a t e s t amount of energy expenditure (the type three l o g tomb with second l e v e l p l a t f o r m ) . F i v e b u r i a l s of medium s i z e have h i g h -energy grave forms: L25, L117, and L104 have the l o g tomb and second l e v e l p l a t f o r m ; L47 has the l o g tomb; and L98 has a second l e v e l p l a t f o r m . The h i g h and low s t a t u s l e v e l s are r e p r e -sented by . the above b u r i a l s . In sum, although the correspond-ence between s t a t u s d i f f e r e n t i a t i o n based upon grave goods and energy expenditure i s b e t t e r among the Late p e r i o d b u r i a l s than the E a r l y , each c o u l d symbolize a d i f f e r e n t aspect of s t a t u s d i f f e r e n t i a t i o n . 6.3.4.4. Body D i s p o s i t i o n and Grave L o c a t i o n The body d i s p o s i t i o n s o t h e r than the common supine p o s i t i o n are i n d i c a t e d on F i g . 6-9, as w e l l as age and sex data f o r each s t a t u s group. The o n l y two b u r i a l s w i t h s k e l e t o n s l y i n g on t h e i r s i d e s are i n the h i g h s t a t u s l e v e l . These b u r i a l s are L117, the male youth, and L25, a r e l i a b l y e s t imated male. The h i g h s t a t u s females i n t h i s s t a t u s l e v e l (L10 and L47, a r e l i a b l y e s t i mated female) have the supine body p o s i t i o n . I t seems t h a t by the Late p e r i o d , the s i d e body p o s i t i o n was r e s e r v e d f o r h i g h - 218 -FIGURE 6-9. Age and sex composition of the three status l e v e l Late Period. (Body di s p o s i t i o n other than supine i s marked below, as well as the one subadult burial.) Characteristics of the one Late Period multiple b u r i a l are included. High Status Group: 4 t o t a l 25, 10, 117, 47 1 youth: 117 known male: 117 known female: 10 r e l i a b l y estimated male: 25 f a i r estimated female: 4 7 body d i s p o s i t i o n : 25 on ri g h t side 117 on l e f t side Intermediate Status Group: 8 t o t a l 126, 60, 24, 17, 125, 4, 3, 72 no skeleton: 60, 124 known male: 125 known female: 72 r e l i a b l y estimated male: 17, 4, 126, 24 f a i r estimated female: 3 Low Status Group: 20 t o t a l 127, 77, 5, 2, 64, 98, 75, 22, 121, 105, 104, 12 16, 122, 123, 93, 15, 100, 96, 46 no skeleton: 12 7 known male: 12 3, 122, 15 known female: 121, 105 f a i r estimated male: 124, 75, 22, 98, 77 f a i r estimated female: 5 intrus i v e b u r i a l s : 123, 124 Multiple b u r i a l LI: (1 male, 1 female) p i t , medium-sized white ware, blackware, 1-13 pots, 1-6 t a l l stemmed cups 1 serving stand, t u r t l e s h e l l , h a i r p i n , arrow remnants as i n L77, 5-19 stone t o o l s , 10-27 bone t o o l s , 8 deer teeth, probable ornament parts (including one jade tubular bead), 1 bone ban, 5 raw material pieces Judged as intermediate i n status. - 219 -ranking males. In the Early period, the b u r i a l s with the side body position are i n a variety of status l e v e l s . The s p a t i a l location of bu r i a l s i n each of the three status lev e l s i s shown i n . F i g . 6-10. Unlike the s p a t i a l patterning i n the Early period, the high status b u r i a l s i n the Late period (status l e v e l one) are located i n a l l areas of the cemetery. The northern area only contains one b u r i a l of the high status l e v e l . I f the two status l e v e l i n t e r p r e t a t i o n i s accepted, the majority of high status b u r i a l s are located i n a l l areas of the cemetery as well. 6.3.4.5. Test Implications As i n the analysis of status i n the Early period, i t i s d i f f i c u l t to ascertain conclusively whether status d i f f e r e n t i a -t i o n cross-cuts age and sex categories to indicate ascribed status. There i s only one subadult i n the Late period and a minority of b u r i a l s for which sex i s known or estimated. The age and sex composition of the b u r i a l s i n the three status levels i s i n F i g . 6-9. However, the clear high status of the male youth, L117, allows the proposition to be made that status d i f f e r e n t i a t i o n i n the Late period i s ascribed. L117 i s more c l e a r l y of high status than the c h i l d i n the Early period, E9 4. L117 i s i n the high status group and contains the possible badge of status, the unused jade chan. The jade spade i s also i n the female adult b u r i a l L10. L117 i s also distinguished from some adult b u r i a l s i n terms of grave form (the type three log tomb - 220 -FIGURE 6-10. Location of the three status groups i n the cemetery, Late Period, i n terms of the s p a t i a l areas discussed i n Chapter 5, the one multiple b u r i a l included. CD E A R L Y P E R I O D B U R I A L EU L A T E P E R I O D B U R I A L £3 U N D A T A B L E B U R I A L A\ HIGH S T A T U S ( 7 2 ) I N T E R M E D I A T E S T A T U S 6 4 LOW S T A T U S - 221 -and second l e v e l platform) and a medium grave s i z e . It i s possible that the status of L117 i s a r e f l e c t i o n of his parents' pos i t i o n . I do not think t h i s p o s s i b i l i t y i s l i k e l y given the distinctiveness of L117's status position among the other Late period b u r i a l s . It i s also clear that status d i f f e r e n t i a t i o n crosscuts sex categories to a greater degree i n the Late period than i n the Early. Both male and female b u r i a l s are i n the high status l e v e l . L10, a known female, was described as consistently d i s -tinguished from the other b u r i a l s i n the high status l e v e l by both grave goods and energy expenditure. It may be the b u r i a l of highest status i n the Late period. Male and female b u r i a l s are i n each of the other two status l e v e l s . The known female L72 has the stone pendant, another possible badge of status. Both male and female b u r i a l s have grave forms and grave sizes that represent a greater than average amount of energy expend-i t u r e . F i g . 6-9 suggests that a ranking pyramid i s represented among the Late period b u r i a l s . The difference between the rank l e v e l s , however, i s not great. The grave goods, grave form and grave size of the one multiple b u r i a l i n the Late period, LI, i s indicated i n Fig.6-9. Due to the presence of jade i n the b u r i a l , L l can probably be characterized as similar i n status to the b u r i a l s i n the i n t e r -mediate status l e v e l . As i n the multiple b u r i a l s from the Early period, i t appears the status of the female i s lower than that of the male because the majority of grave goods are placed by - 222 -the male. However, other females i n the Late period are clearly-higher i n status than some males. There i s only one intr u s i v e p a i r of b u r i a l s within the Late period, L123 and L124. Both b u r i a l s are i n the low status group. Due to the lim i t e d information regarding time relationships within the Late period, i t i s not possible to judge whether the three status l e v e l s existed throughout the duration of the period. 6.3.4.6. Conclusions In the Late period as well as the Early, high status i s characterized by exclusive access to only a few a r t i f a c t forms, raw materials, or great quantities of objects. The elephant ivory comb symbolizes high status i n both periods. However, other forms symbolic of high status had changed by the Late period, from the b i disk and animal shaped vessel to the unused jade chan (and the stone pendant, i f the two status l e v e l i n t e r -pretation i s accepted). The raw material types that had changed are elephant ivory and red ochre i n the Early period to a l l i -gator hide i n the Late (and jade and turquoise i f two status levels are accepted). Since jade was not found i n the highest status l e v e l i n the Early period, access to i t may have become r e s t r i c t e d by the Late period. Elephant ivory was exclusive to high status individuals i n the Early period but not i n the Late. Great quantities of ceramics characterized the high status Early and Late period b u r i a l s , i f the two status levels are accepted. Another change i s that the t a l l , stemmed cup i s exclusive to the high intermediate status group i n the Early - 223 -period while i n the Late i t occurs i n the b u r i a l s of a l l status l e v e l s . However, by the Late period, the highest status l e v e l i s characterized by the greatest quantity of t a l l stemmed cup. If E9 (with the animal shaped vessel and red ochre) i s a Late period b u r i a l , I estimate i t to be equivalent i n status to the intermediate status l e v e l (see F i g . A6-11 for the grave goods i n b u r i a l 9 and Fig . 6-8). I f t h i s estimation i s accurate, then the animal shaped vessel and red ochre are not c h a r a c t e r i s t i c of high status i n eithe r the Early or Late periods. It appears that the nature of status d i f f e r e n t i a t i o n at Dawenkou changed from a system with elements of both an achieved and an ascribed system i n the Early period to a system that c l e a r l y represents ascribed ranking i n the Late. Thus, an increase i n c u l t u r a l complexity from the Early period to the Late i s indicated. This increase i n c u l t u r a l complexity i s also suggested by the general increase i n quantity and variety of grave goods or raw materials among the b u r i a l s from the Early period to the Late. There i s also an increase i n the complexity of grave form by the Late period (the prevalence of the type three log tomb) and an increase i n grave s i z e . As i n the Early period, few types of grave goods or raw materials are exclusive to the Late period high status b u r i a l s . However, the Late period high status b u r i a l s are characterized by high energy expenditure i n terms of grave form or grave size more than the Early period high status b u r i a l s . It may be that high status had become more redundantly expressed by the Late period, such that both grave goods and grave form or size were symbolic of - 224 -high status. The apparent increase i n c u l t u r a l complexity by the Late period i s accompanied by few types of grave goods exclusively associated with high status and by a lack of extreme d i f f e r e n t i -ation between the high and low status b u r i a l s . Low status b u r i a l s i n the Early period are d i s t i n c t i v e l y d i f f e r e n t from the intermediate and high status l e v e l s . In contrast, the Late period low status b u r i a l s are not as d i s t i n c t from other b u r i a l s . The decrease i n the number of status levels represented i n the cemetery from approximately four i n the Early period to three or two i n the Late i s unexpected i n l i g h t of the apparent increase i n c u l t u r a l complexity by the Late period. I argue that these apparent contradictions can be explained i f a l l or most of the Late period b u r i a l s are regarded as high i n status. I propose that b u r i a l i n the cemetery became r e s t r i c t e d to r e l a t i v e l y high status i n d i v i d u a l s by the Late period. The cemetery either became r e s t r i c t e d to individuals of one high status l e v e l or of two to three high status l e v e l s . It i s possible that some differences i n status among the Late period b u r i a l s are a factor of the "subordinate" dimension of Peebles and Kus (1977:431). Also the ranking system represented among the Late period b u r i a l s may have been part of a regional ranking system and cemeteries i n the Dawenkou Culture region (perhaps Shandong) may have r e f l e c t e d t h i s regional system. The individuals i n the Late period at Dawenkou may represent one status l e v e l i n the regional ranking system. Goldstein (1980: 136) points out that i n d i v i d u a l Mississippian cemeteries appear - 225 -eg a l i t a r i a n because each as a whole represents one status l e v e l i n a regional ranking system. I propose that the trend towards s p a t i a l segregation of high status b u r i a l s noted for the Early period had developed by the Late period such that the entire cemetery was r e s t r i c t e d to r e l a t i v e l y high status i n d i v i d u a l s . The Late period b u r i a l s may represent a " s t r a t i f i e d " society i n Fried's (1967) sense. According to Frie d , status d i f f e r e n t i a t i o n had increased from the l e v e l of ranked society such that there i s unequal access to basic resources (1967:186). The s t r a t i f i e d society i s organized by a mechanism that trans-cends kinship systems ( i b i d ) . In Chapter 5, I propose that the same descent group persisted i n the cemetery from the Early period to the Late. It i s possible that the whole descent group represented at the cemetery had become ranked i n r e l a t i o n to other descent groups i n the Dawenkou Culture region by the Late period. Other Dawenkou Culture cemeteries roughly contemporan-eous with Dawenkou may be ranked higher or lower than Dawenkou. The Late period b u r i a l s could also r e f l e c t a f a i r l y complex chiefdom. Whether t h i s chiefdom was as complex as that repre-sented by Moundville and other Mississippian s i t e s i s unclear. I do not have an explanation for the few c h i l d b u r i a l s i n the Early and Late periods, e s p e c i a l l y i n the Late period. A separate location for c h i l d b u r i a l s i s expected i n achieved status systems. Pearson (1981:1084) notes that c h i l d b u r i a l s at Dawenkou tend to contain fewer ceramics i n comparison to adults than at the e a r l i e r Dawenkou Culture s i t e s of L i u l i n . - 226 -It i s possible that children from the Late period were not buried i n the cemetery u n t i l they reached a certain age. There i s a c h i l d i n multiple b u r i a l 35, which probably belongs to the Late period (see Chapter 3). Including multiple b u r i a l s , a change from 85 b u r i a l s i n the Early period to 33 i n the Late i n i t s e l f suggests a change i n use of the cemetery, provided the length of time and popula-ti o n base remained roughly the same. The changes through time i n status d i f f e r e n t i a t i o n just described would probably hold regardless of the period to which the 15 undatable b u r i a l s belong. The r e l a t i v e status of each undatable b u r i a l i n comparison to Early and Late period b u r i a l s appears low. The age, sex, grave goods and energy expenditure information for the undatable b u r i a l s i s shown i n F i g . A6-24. Seven of the eight b u r i a l s lacking grave goods i n the cemetery are undatable b u r i a l s . Five of the b u r i a l s lacking grave goods are children. Like the high status graves i n both periods, the empty graves i n terms of grave goods (including multiple b u r i a l E31) are not r e s t r i c t e d to one s p a t i a l area of the cemetery. The undatable b u r i a l s appear r e l a t i v e l y low i n status i n terms of grave goods and energy expenditure. A l l of the b u r i a l s are p i t s and the largest area 2 i s 2.49m , smaller than most b u r i a l s of both periods. There are no photographs of the two multiple b u r i a l s to allow a comparison of the status of males versus females. - 227 -6.4. Implications Two of the trends through time i n status d i f f e r e n t i a t i o n noted by Pearson (1981) are supported i n t h i s analysis (see Chapter 1): an increase i n the mean number of ceramics i n buri a l s by the Late period and an increase i n the percentage of b u r i a l s with ceramics. The multivariate analyses did not depict greater v a r i a t i o n i n quantities of ceramics, t o o l s , and ornaments among b u r i a l s i n the Late period compared to the Early as expected by Pearson (1981:1082, 1085). On the basis of my exploratory study of status d i f f e r e n t i a t i o n i n the cemetery, I also had expected greater v a r i a t i o n among the Late period b u r i a l s . Pearson (19 81:10 86) suggests that the Late period does not r e f l e c t a highly ranked society and that c r a f t s p e c i a l -i z a t i o n was developing. Also, the status of men i n r e l a t i o n to women and children was increasing ( i b i d ) . I conclude that the Late period does r e f l e c t a highly ranked society. Also, both men, women and children had high status positions (eg., the female b u r i a l L10 and the youth, L117). The Wen Wu Correspondent (1978) states that Dawenkou represents the development of private ownership and s o c i a l classes. I suggest the Late period represents a highly ranked society i n which s o c i a l classes may have developed. My analysis shows the increase i n status d i f f e r e n t i a t i o n among b u r i a l s from the Early period to the Late noted by the Shandong P r o v i n c i a l Museum (1978) and the Kao Gu E d i t o r i a l Staff. - 228 -If my proposition i s correct that the Early period at Dawenkou represents a ranked society with a trend towards s p a t i a l separation of high status graves, and that by the Late period the cemetery may represent one or more high status lev e l s i n a regional ranking system, Dawenkou probably does not r e f l e c t an i n c i p i e n t system of ranking as Chang (19 79:161) maintains. It i s l i k e l y that ascribed ranking developed i n the Dawenkou Culture region at an e a r l i e r date than was previously thought. - 229 -CHAPTER . 7 CONCLUSIONS AND SUGGESTIONS FOR FUTURE RESEARCH 7.1. Conclusions regarding Dawenkou The primary goal of t h i s mortuary analysis has been to understand the nature of status d i f f e r e n t i a t i o n at Dawenkou i n the Early and Late periods, as well as the nature of change through time i n status d i f f e r e n t i a t i o n . Dawenkou has been regarded i n the western and Chinese archaeological l i t e r a t u r e as c r i t i c a l for understanding the development of ranking i n the eastern seaboard region. This study, u t i l i z i n g current arch-aeological method i n mortuary analysis, could help resolve the controversy i n the archaeological l i t e r a t u r e regarding the nature of status d i f f e r e n t i a t i o n at the s i t e . The methodology by which to accomplish the primary goal was outlined i n Chapter 2. I argued that since status d i f f e r -e ntiation i s expressed in a variety of manners and to d i f f e r e n t degrees i n ethnographic s o c i e t i e s , an investigation should regard a l l variables i n an analysis of status as only p o t e n t i a l l y status-related. I found that quantity and qu a l i t y of grave goods distinguished the Early and Late period b u r i a l s at Dawenkou. However, energy expenditure i n terms of grave size and grave form was not c l e a r l y status-related i n either period. The. relati o n s h i p between status and energy expenditure was more evident with the Late b u r i a l s . - 230 -In Chapter 2 I stated that cemeteries must be regarded as r e f l e c t i n g more than one s o c i a l system through time. Derived status d i s t i n c t i o n s should not be interpreted as the exact number of status levels that existed i n the community of the deceased. Another t h e o r e t i c a l reason for not interpreting the exact number of status levels i s the p a r t i a l d i s t o r t i o n of s o c i a l organization r e f l e c t e d i n mortuary chains, noted by Hodder (1980) and others. A methodological reason for not interpreting the exact number of status levels i s the s u b j e c t i v i t y and d i f f i c u l t y i n choosing optimum cl u s t e r i n g and scaling solutions that r e f l e c t status d i s t i n c t i o n s . Mortuary analyses should attempt to assess change through time i n status d i f f e r e n t i a t i o n . In t h i s study, change from the Early period to the Late i s emphasized rather than change within each period. The i n t r u s i v e pairs of b u r i a l s within each period are the only source of chronological v a r i a -t i o n of b u r i a l s within each period. I argued that a r e l a t i v e assessment of change through time i n the degree of status d i f f e r e n t i a t i o n at a s i t e may be more feasi b l e than discussing change i n terms of evolutionary typologies. Although i t was d i f f i c u l t to characterize the nature of status d i f f e r e n t i a t i o n among the Early b u r i a l s , I concluded an increase i n the degree of status d i f f e r e n t i a t i o n had occurred by the Late period. A system of ranking seemed present i n both periods, but there appeared to be a q u a l i t a t i v e change i n the nature of the ranking system by the Late period. If evolutionary typologies are considered, I conclude there may be a change from a "ranked" to a " s t r a t i f i e d " society i n Fried's (1967) sense or from a simple to a complex chiefdom. - 231 -My understanding of status d i f f e r e n t i a t i o n at Dawenkou would not have been as complete without the three preliminary analyses. In Chapter 3, the ordering of b u r i a l s on the basis of changes i n ceramic s t y l e s , subtypes, and functional types by the authors of the s i t e report was clos e l y r e p l i c a t e d by the c l u s t e r and scaling analyses. Recent a r t i c l e s i n the Chinese archaeological l i t e r a t u r e conclude that the Middle b u r i a l s are closer i n ceramic st y l e to the Early b u r i a l s . My analysis showed some Middle b u r i a l s grouped with the Early and some with the Late. The resultant grouping of b u r i a l s as Early and Late allowed me to accept two chronological periods for the duration of the study. Since the same typeuof ceramic c l a s s i f i -cation system appears to be u t i l i z e d at some other Dawenkou Culture s i t e s , chronological analyses of b u r i a l s u t i l i z i n g c l u s t e r i n g and scaling techniques could be conducted when radio-carbon or stra t i g r a p h i c data are limited. Addition of the estimated sexed b u r i a l s from Chapter 4 to the sample of known sexed b u r i a l s enhanced the inter p r e t a t i o n of status d i f f e r e n t i a t i o n i n Chapter 6 and the assessment of s o c i a l subgroup a f f i l i a t i o n i n Chapter 5. Temporal and geograph-i c v a r i a t i o n i n sexed-linked a r t i f a c t types could be better understood i f the simple inspection method and discriminant analysis were applied to other Dawenkou Culture s i t e s . It would be useful to determine whether males are exclusively associated with a greater number of a r t i f a c t types than females at other Dawenkou Culture s i t e s . I t i s possible that t h i s r a t i o of male to female a r t i f a c t s r e f l e c t s increasing status - 232 -of males compared to females i n the eastern seaboard region as proposed by Pearson (1981:1086). The analysis of status d i f f e r e n t i a t i o n i n Chapter 6 indicated females with equal status to males. In Chapter 2, I argued that an analysis of status d i f f e r -e n t iation i s more complete with an assessment of s o c i a l subgroup a f f i l i a t i o n represented at a s i t e . The assessment i n Chapter 5, a co r r e l a t i o n of s p a t i a l location of grave with grave orienta-t i o n , grave form, body d i s p o s i t i o n and ceramic style indicated much consistency within the cemetery i n terms of these variables. There were some changes from the Early period to the Late such as the greater consistency i n orientation by the Late period. I proposed that the consistency during both periods i n grave form, grave orientation and body d i s p o s i t i o n from s p a t i a l area to s p a t i a l area r e f l e c t s one type of s o c i a l group. I proposed that a descent group i s l i k e l y , although test implica-tions for a descent group have not been developed. I also proposed that the s p a t i a l areas i n the Early and Late periods could r e f l e c t k in groups or r e s i d e n t i a l groups. The analysis also indicated an increase i n number of ceramic forms that are exclusive to a s p a t i a l area by the Late period. It would be useful to determine whether t h i s trend through time i s a regional one. The analysis of status provided some insight regarding the controversy over the degree of ranking represented at Dawenkou. I concluded the degree of ranking had increased from the Early period to the Late. The system of ranking may have changed - 233 -from one i n which members within the descent group at the cemetery were ranked to one i n which whole kin groups i n the region were ranked. My re s u l t s t e n t a t i v e l y support Pearson's (1981:1086) proposition that lineages were becoming increasingly important by the Dawenkou Culture period. Also, the greater cor r e l a t i o n between s p a t i a l area and ceramic style i n the Late period (see Chapter 5) may indicate that symbolism of kin groups was increasing. 7.2. Future Research The process of understanding development of c u l t u r a l complexity i n the eastern seaboard region must involve assess-ment of status d i f f e r e n t i a t i o n at the regional l e v e l . An analysis of status at other Dawenkou Culture s i t e s roughly contemporaneous with the Early and Late periods at Dawenkou i n Shandong or Jiangsu i n p a r t i c u l a r , could indicate whether the same trends noted here are apparent at other s i t e s . The develop-ment of a regional system of status d i f f e r e n t i a t i o n i n Shandong or Jiangsu should be r e f l e c t e d i n other Dawenkou Culture s i t e s . At s i t e s contemporary with the Late period at Dawenkou, one might expect to f i n d the same aspects of mortuary treatment that characterize high status at Dawenkou. The jade chan, a possible badge of status i n the Late period, may symbolize high status throughout the region. Possible supralocal symbols (Peebles 1971: 69) include the chan and ceramic design attr i b u t e s . One could determine whether the other attributes c h a r a c t e r i s t i c of high status during the Late period are associated with high status - 234 -items at other Late Dawenkou period s i t e s : white ware, a l l i g a t o r hide, huang pendant, jade, turquoise, great quantities of ceramics (of the t a l l stemmed cup, i n p a r t i c u l a r ) . Also one could determine whether b u r i a l s from l a t e r s i t e s show l i t t l e d i f f e r e n t i a t i o n i n terms of grave goods — i n d i c a t i v e of the cemetery being used for one status l e v e l . For s i t e s roughly contemporary with the Early period of Dawenkou, one could assess whether the trend towards s p a t i a l segregation of high status graves i s evident. The L i u l i n s i t e shows such a trend. Pearson (1981:1083) concludes that the s p a t i a l areas at L i u l i n were a strong determinant of v a r i a b i l i t y i n terms of quantities of ceramics. A few mortuary attributes c h a r a c t e r i s t i c of high status i n the Early period found i n roughly contemporaneous Dawenkou Culture s i t e s (such as the b i disk, elephant ivory objects, and great quantities of ceramics) may indicate the i n i t i a l development of a regional system of status d i f f e r e n t i a t i o n . Pearson (1981:1085) notes an increase i n the degree of status d i f f e r e n t i a t i o n through time i n the eastern seaboard region. However, some of the a r t i f a c t forms or materials I interpreted as c h a r a c t e r i s t i c of high status i n either period at Dawenkou have been found i n s i t e s e a r l i e r than Dawenkou: white ware and painted ware . from the f i r s t excavation at Dadunzi i n Jiangsu (The Nanjing Museum 1964); the stone huang pendant, jade ax, jade knife and painted ware from the second season at Dadunzi (The Nanjing Museum 19 81); and the jade spade (chan), white ware, turquoise and painted ware at Wangyin - 235 -(Shandong Archaeological Team 1979). A r t i f a c t forms or materials found at l a t e r Dawenkou Culture s i t e s are: white ware, black ware, turquoise, ceramic b o t t l e , f u l l stemmed cup at the Dafanzhuang s i t e (Archaeological Teamof L i n y i County 1975); a jade huan, black ware, white ware at Yedian (Shandong P r o v i n c i a l Museum 1972); and white ware, painted ware, and the second l e v e l p l a t -form at Xixiahou (The Shandong Archaeological Team 1964). The presence of these a r t i f a c t forms and materials i n early s i t e s may indicate that access to these forms and materials became r e s t r i c t e d only l a t e r i n time. Status d i f f e r e n t i a t i o n at these s i t e s may be of a greater degree than currently considered. Instead, the v a r i a t i o n i n status d i f f e r e n t i a t i o n among Dawenkou Culture s i t e s may be a factor of geographical differences i n resource d i s t r i b u t i o n , for example, or i n demography. Mortuary analyses of other cemeteries from varying time periods i n the Dawenkou Culture region must be completed so that insight into the processes that may have influenced the development of c u l t u r a l complexity from the Early to the Late period at Dawenkou i s gained. Possible processes include population pressure (Fried 1967:196), a g r i c u l t u r a l i n t e n s i f i c a -tion (Pearson 1981:1086), warfare or trade. Warfare i s poss-i b l e , given that a b u r i a l at Dadunzi has a stone arrowhead embedded i n a thigh bone (Yang 1982:58). However, as Pearson(ibid) has pointed out, there i s no evidence for v i o l e n t b u r i a l of low status i n d i v i d u a l s from s i t e s i n the eastern seaboard region as at some Longshan s i t e s . Trade of ceramics, jade or elephant - 236 -ivory i n either period i s possible. If the k i l n i n Dawenkou cemetery was used during the Late period (see Chapter 1), i t could have been used for production of trade wares. Shangraw (personal communication 19 83) suggests that the cemetery had a sacred function by the Late period. The k i l n may have been used for production of white ware for purposes of mortuary r i t u a l . The b r i t t l e nature of t h i s ware suggests that i t was used for r i t u a l i s t i c purposes. My r e s u l t s support the notion of the evolutionary r e l a t i o n -ship of Dawenkou with the Shang Dynasty mentioned by Chang (1980:345-346, 19 83:509-510), Thorp (1980:51) and Fried (1983: 488). Many t r a i t s c h a r a c t e r i s t i c of high status i n the Shang period appear to r e f l e c t high status at Dawenkou: the log tomb, second l e v e l platform, wine vessels (although ceramic, not bronze), jade, white ware and ornaments such as hairpins. I also think that the concern for lineage possibly r e f l e c t e d at Dawenkou i s another c u l t u r a l continuity with the Shang period. Ranked lineages were an important part of the Shang system of s o c i o p o l i t i c a l organization (Chang 1980:163). Communities consisted of members of common lineages (Chang 1980:161). Line-age membership may have been symbolized i n decorative styles of bronzes (Chang 1980:165). At t h e i r deaths, individuals may have been buried with members of t h e i r lineage. At a recent excavation at Anyang described by Chang (1980:369), eight s p a t i a l areas of graves were found, each with a d i s t i n c t grave orienta-t i o n , b u r i a l method and pottery assemblage. The investigators maintain that each of these s p a t i a l areas represents a lineage ( i b i d ) . They also found that there was much v a r i a t i o n i n grave - 237 -goods within each s p a t i a l area, i n d i c a t i n g that ranking of individuals existed within lineages. For both periods at Dawenkou I found that the s p a t i a l areas within the cemetery contained more than one status l e v e l . However, I te n t a t i v e l y interpreted these s p a t i a l areas as representing kin groups such as families or r e s i d e n t i a l groups. Chang (1980:235) notes that records of economic a c t i v i t i e s are lacking i n the Shang state and that writing was more concerned with s o c i a l i d e n t i f i c a t i o n (1980:247). Writing for t h i s purpose may have begun during the Neo l i t h i c , at s i t e s i n several regional areas of China, i n c l u d -ing Dawenkou Culture s i t e s (1980:243). According to Chang (1980:161) lineage members symbolized relat i o n s h i p of the lineage to the ancestral l i n e by the Cong r i t u a l chamber. The cong i s not present at Dawenkou but i t has been found i n Ne o l i t h i c s i t e s from the lower Yangtze River area (Rawson 1980:36). Thus, the concern for lineage may have been well developed during the Dawenkou Culture period. In conclusion, t h i s mortuary analysis of Dawenkou suggests that the Dawenkou Culture region (Shandong province, i n p a r t i -cular) should be regarded as a dynamic, independent region of c u l t u r a l evolution during the Ne o l i t h i c period. My reasoning i s based upon the increase i n c u l t u r a l complexity noted for the Early period to the Late, the increase i n c u l t u r a l complexity through time noted by Pearson (19 81) for N e o l i t h i c 'sites i n Shandong and other provinces of the eastern seaboard region, and the c u l t u r a l c o n t i n u i t i e s from the Dawenkou s i t e to the Shang period. I maintain the results from the analysis of status - 238 -indicate that Dawenkou and other Dawenkou Culture s i t e s do not represent i n c i p i e n t ranking. I t i s l i k e l y that that ranking was f i r s t developed i n the eastern seaboard region at a date e a r l i e r than Chang (19 79:160) asserts. My results suggest that by the Late period, the Dawenkou s i t e r e f l e c t s the "Level of V i l l a g e Aggregates" defined by Chang (1980:361) as c h a r a c t e r i s t i c of Longshan cultures i n Shandong, Shaanxi, and Henan. It i s l i k e l y the Late .period b u r i a l s r e f l e c t i n t e r n a l d i f f e r e n t i a t i o n into poor and r i c h groups, spec i a l i z e d handicrafts and a socio-p o l i t i c a l organziation with i n t e r v i l l a g e leaders (as described by Chang 19 79:361). When the nature of regional s o c i a l organ-i z a t i o n i n the Dawenkou Culture region i s better understood, attempts can be made to assess the nature of in t e r a c t i o n between c u l t u r a l regions expected by Chang (1981:155) and An (1979-80: 45) for late N e o l i t h i c Lungshanoid cultures. 7.3. Methodological Conclusions The secondary goal of t h i s study was to make a methodolog-i c a l contribution to mortuary analysis. My methodology was derived from both the processual and symbolist approaches to mortuary studies. In both the chronological and status analyses, I found that Ward's Method of clus t e r analysis yielded r e s u l t s that best met my a p r i o r i expectations based upon my i n t u i t i v e study of the cemetery i n comparison to Complete Linkage and Average Link-age. The res u l t s from Ward's Method also compared reasonably well with those from Torgerson's Metric Multidimensional Scaling. In the Late period analysis of status, I judged the scaling - 239 -results to better depict the relationships i n the data set than Ward's Method. I recommend the use of Ward's Method and multidimensional scaling i n conjunction with each other for mortuary analyses of status d i f f e r e n t i a t i o n . I conclude that e f f o r t s to i d e n t i f y the sex of unsexed bu r i a l s by means of grave goods and the v i s u a l inspection method of Chapter 4 are worthwhile. Discriminant analysis may produce better results with data sets exh i b i t i n g more continuous v a r i a -t i o n . Social a f f i l i a t i o n i n Chapter 5 could only be assessed i n an exploratory manner, due to the lack of t e s t implications that would i d e n t i f y one type of s o c i a l subgroup versus another. It became evident i n the status analysis of Chapter 6 that mortuary analyses of status d i f f e r e n t i a t i o n must have a region-a l approach, e s p e c i a l l y when ranking i s expected. Also, the te s t implications which e x i s t for achieved versus ascribed status are probably too s i m p l i s t i c . As Rothschild (1979:659) points out, investigators have assumed a p o l a r i t y between e g a l i t a r i a n and ranked s o c i e t i e s . The Early period at Dawenkou did not appear c l e a r l y ranked or e g a l i t a r i a n on the basis of the t e s t implications i n the l i t e r a t u r e . As mentioned i n Chapter 2, systems of status d i f f e r e n t i a t i o n are quite complex i n most s o c i e t i e s . It was not clear whether the lack of a correspondence between status d i f f e r e n t i a t i o n based upon energy expenditure and grave goods meant that d i f f e r e n t aspects of status were being symbolized. It i s possible that the correspondence between status d i f f e r e n t i a t i o n based upon energy expenditure and upon - 240 -grave goods i s better the more complex the society. F i n a l l y , the multivariate techniques proved extremely useful for the analysis of status d i f f e r e n t i a t i o n . Thomas (19 78:236) claims that numerical taxonomy does not provide any insight i n relationships among items i n a data set that v i s u a l inspection cannot i d e n t i f y . The ordering of b u r i a l s that resulted from the cl u s t e r i n g and scaling techniques i n Chapter 6 indicated a general lack of s i g n i f i c a n t status d i s t i n c t i o n s among the b u r i a l s , e s p e c i a l l y i n the Late period. On the basis of my exploratory study, I had expected more extreme status d i s t i n c t i o n s among the Late period b u r i a l s . I conclude with Matson (1980:342) that multivariate techniques can be more e f f e c t i v e than v i s u a l inspection. - 241 -BIBLIOGRAPHY Alekshin, V.A. 19 83 B u r i a l customs as an archaeological source. Current Anthropology 24(2):137-149. Al l e n , William and James B. Richardson III 1971 The reconstruction of kinship from archaeological data: the concept, the methods, and the f e a s i b i l i t y . American Antiquity 36(l):41-53. AnZhimin 1979- The Neol i t h i c archaeology of China: a b r i e f survey 80 of the l a s t t h i r t y years. Translated by K.C. Chang, Early China 5:35-45. 1981 China's N e o l i t h i c period. Kao Gu 3:252-260 (in Chinese). 1982 China's Neol i t h i c period. China Reconstructs XXXI (6):57-62. Archaeological Team of L i n y i County 1975 Excavated b u r i a l s from the N e o l i t h i c s i t e of Dafanzhuang i n Shandong. Kao Gu 1:13-22 (in Chinese). Ba r t e l , Brad 1973 A multivariate analysis of European death r i t u a l . Ethnologia Europaea 7:111-128. 1982 A h i s t o r i c a l review of ethnological and archaeo-l o g i c a l analyses of mortuary practices. Journal of Anthropological Archaeology 1:32-58. Bayard, Donn 1983 Rank and wealth at Non Nok Tha: the mortuary evidence. Paper presented at symposium KIE, XV, P a c i f i c Science Congress, Dunedin, New Zealand. Beij i n g University Department of Archaeology 19 83 C u l t u r a l r e l i c s from Shandong province. Kao Gu 3:193-198 (in Chinese). Bender, Susan J. 19 79 Paleodemographic analysis of a Late Woodland s i t e i n southeastern Michigan. Mid-Continental Journal of Archaeology 4(2):183-208. Bettinger, Robert L. 1979 Multivariate s t a t i s t i c a l analysis of a regional subsistence-settlement model for Owens Valley. American Antiquity 44 (3): 455-470. - 242 -Binford, Lewis R. 1971 Mortuary practices: t h e i r study and t h e i r p o t e n t i a l . In Approaches to the s o c i a l dimensions of mortuary practices, edited by James A. Brown. Memoirs of the Society for American Archaeology 25:6-29. 1972 Galley Pond' Mound. In An archaeological perspective, edited by L.R. Binford, pp.390-420. Academic Press, New York. Blackmore, Charles, Mary Braithwaite and Ian Hodder 1979 Social and c u l t u r a l patterning i n the Late Iron Age i n southern England. In Space, hierarchy and society, edited by Barry C. Burnham and John Kingsbury. BAR International Series 59:93-111. Braun, David P. 1977 Middle Woodland-early Late Woodland s o c i a l change i n the p r e h i s t o r i c central midwestern U.S. Unpub. Ph.D. d i s s e r t a t i o n , Department of Anthropology, University of Michigan. 19 79 I l l i n o i s Hopewell b u r i a l practices and s o c i a l organization: a reexamination of the Klunk-Gibson Mound group. In Hopewell archaeology: the C h i l l i - cothe Conference, edited by David S. Brose, N'omi Greber, pp. 66-79. The Kent State University Press, Kent, Ohio. 1981 A c r i t i q u e of some recent North American mortuary studies; American Antiquity 46 (2):398-416. Brown, James A. 19 71 The dimensions of status i n the b u r i a l s at Spiro. In Approaches to the s o c i a l dimensions of mortuary, practices, edited by James A. Brown. Memoirs of the Society for American Archaeology 25:92-112. 1981 The search for rank i n p r e h i s t o r i c b u r i a l s . In The archaeology of death, edited by Robert Chapman, Ian Kinnes, Kla s Randsborg, pp. 25-37. Cambridge University Press. Buikstra, Jane E. 19 76 Hopewell i n the Lower I l l i n o i s Valley: a regional study of human b i o l o g i c a l v a r i a b i l i t y and p r e h i s t o r i c mortuary behavior. Northwestern University Archaeological Program S c i e n t i f i c Papers No. 2. Campbell, George R. 19 81 Jaws, too. The story of Sanibel's a l l i g a t o r s and other crocodilians. Sanibel-Captiva Conservation Foundation, Sanibel, F l o r i d a . Chang, K.C. 19 79 The archaeology of ancient China. Third e d i t i o n . Yale University Press, New Haven. 19 80 Shang c i v i l i z a t i o n . Yale University Press, New Haven. 1981a Archaeology and Chinese historiography. World Archaeology 13(2):156-169. 1981b In search of China's beginnings: new l i g h t on an old c i v i l i z a t i o n . American S c i e n t i s t (69):148-160. 19 83a Sandai archaeology and the formation of states i n ancient China: processual aspects of the origins of Chinese c i v i l i z a t i o n . In The ori g i n s of Chinese c i v i l i z a t i o n , edited by David N. Keightley, pp. 495-521. University of C a l i f o r n i a Press, Berkeley. 19 83b Concluding remarks. In The origins of Chinese c i v i l i z a t i o n , edited by David N. Keightley, pp. 565-581. University of C a l i f o r n i a Press, Berkeley. Chapman, R.W. 1977 B u r i a l p ractices: an area of mutual i n t e r e s t . In Archaeology and anthropology: areas of mutual i n t e r e s t , edited by Matthew Spriggs. BAR Supple- mentary Series 19:19-33. 1981 Archaeological theory and communal b u r i a l i n p r e h i s t o r i c Europe. In Patterns of the past: studies i n honour of David Clarke, edited by Ian Hodder, Glynn Isaac, Norman Hammond, pp. 387-411. Cambridge University Press. Chapman,Robert and Klaus Randsborg 19 81 Approaches to the archaeology of death. In The archaeology of death, edited by Robert Chapman, Ian Kinnes, K l a v s Randsborg, pp. 1-2 4. Cambridge University Press. Cheng, Te-k'un 19 82 Studies i n Chinese archaeology. The Chinese University Press, Hong Kong. Coles, J. and A. Harding 19 79 The Bronze Age i n Europe. Methuen, London. - 244 -Cook, Delia C. 1981 Mortality, age structure and status i n the i n t e r -pretation of stress indicators i n p r e h i s t o r i c skeletons: a dental example from the Lower I l l i n o i s Valley. In The archaeology of death, edited by R. Chapman, I. Kinnes, K. Randsborg, pp. 133-134. Cambridge University Press. Decker, Dean A. 1969 Early archaeology on Catalina Island: p o t e n t i a l and problems. Archaeological Survey Annual Report 11 :69-84. Department of Anthropology, University of C a l i f o r n i a , Los Angeles. Donnan, Christopher B. and Carol J. MacKay 1978 Ancient b u r i a l patterns of the Mbche Valley, Peru. University of Texas Press, Austin. Doran, J.E. and F. R. Hodson 19 75 Mathematics and computers i n archaeology. Edinburgh University Press. Doran, James 19 71 Computer analysis of data from the l a Tene cemetery at Munsingen-Rain. In Mathematics i n the archaeo- l o g i c a l and h i s t o r i c a l sciences, edited by F.R. Hodson, D.G. Kendall and P. Tautu, pp. 422-431. Edinburgh University Press. 1973 Explanation i n archaeology: computer experiment. In Explanation of culture change, edited by Colin Renfrew, pp. 149-153. Gerald Duckworth, Ltd., London. Drennan, Robert D. 19 76a A refinement of chronological s e r i a t i o n using nonmetric multidimensional scaling. American Antiquity 41 (3)-.290-302 . 19 76b Fabrica San Jose and Middle Formative society i n the v a l l e y of Oaxaca, edited by Kent Flannery. Memoirs of the Museum of Anthropology, University of Michigan 8. Fr i e d , Morton H. 1967 The evolution of p o l i t i c a l society. Random House, New York. 19 8 3 Tribe to state or state to t r i b e i n ancient China? In The origins of Chinese c i v i l i z a t i o n , edited by David N. Keightley, pp. 467-493. University of C a l i f o r n i a Press, Berkeley. - 245 -Gao Guangren 1978 Dawenkou: Neoli t h i c culture f i n d . China Reconstructs XXVII(8):31-35. 1980 A statement proposed for the stages of the Dawenkou Culture. Wen Wu J i Kan 1:47-65 (in Chinese). Goldstein, Lynne 1976 Spatial structure and s o c i a l organization:. regional manifestations of Mississippian society. Unpub. Ph.D. di s s e r t a t i o n , Department of Anthropology, Northwestern University. 1980 Mississippian mortuary practices. A case study of two cemeteries i n the Lower I l l i n o i s Valley. Northwestern University Archaeological Program S c i e n t i f i c Papers No. 4. 1981 One-dimensional archaeology and multi-dimensional people: spatial.organisation and mortuary analysis. In The archaeology of death, edited by Robert Chapman, Ian Kinnes, Klavs Randsborg, pp. 53-69. Cambridge University Press. Goodrich, David W. 1983 Recent archaeological publications i n China. China Exchange News 11(1):15-18. Goody, Jack 1962 Death, property and the ancestors. Tavistock, London. Gowe r, J.C. 1966 Some distance properties of latent root and vector methods used i n multivariate analysis. Biometrika 53:325-338. Greber, N'omi 1979a Variation i n s o c i a l structure of Ohio Hopewell peoples. Mid-Continental Journal of Archaeology 4(1):35-78. 19 79b A comparative study of s i t e morphology and b u r i a l patterns of Edwin Harness Mound and Seip Mounds 1 and 2. In Hopewell archaeology, edited by David S. Brose and N'omi Greber, pp. 2 7-38. The Kent State University Press. Gruber, Jacob W. 1971 Patterning i n death i n a late p r e h i s t o r i c v i l l a g e i n Pennsylvania. American Antiquity 36(l):64-76. - 246 -Han Gangxin and Ban Jifeng 19 80 A r t i f i c i a l deformation of the jawbones of Ne o l i t h i c people at Dadunzi and Wangyin. Kao Gu 2:185-191 (in Chinese, translated by Kazue Pearson). Hatch, James W i l l i s 19 76 Status i n death: p r i n c i p l e s of ranking i n Dallas Culture mortuary remains. Unpub. Ph.D. d i s s e r t a t i o n , Department of Anthropology, Pennsylvania State University. Haviland, William A. 1967 Stature at T i k a l , Guatemala: implications for ancient Maya demography and s o c i a l orgnaization. American Antiquity 32 (3):316-325. Hay, John 19 73 Ancient China. The Bodley Head, London. Hodder, Ian 19 80 Social structure and cemeteries: a c r i t i c a l appraisal. In Anglo-Saxon cemeteries, edited by P. Rahtz, T. Dickinson and L. Watts. BAR B r i t i s h Series 82:161-169. 19 82a Symbols i n action. Cambridge University Press. 19 82b Sequences of s t r u c t u r a l change i n the Dutch Neo l i t h i c . In Symbolic and s t r u c t u r a l archaeology, edited by Ian Hodder, pp. 162-177. Cambridge University Press. Hodson, F.R. 1977 Quantifying H a l l s t a t t : some i n i t i a l r e s u l t s . American Antiquity 42 (3):394-412. 19 79 Inferring status from b u r i a l s i n Iron Age Europe: some recent attempts. In Space, hierarchy and society, edited by B.C. Burnham and J. Kingsbury. BAR International Series 59:23-30. Hubers, Louisa G.F. 19 83 The relationship of the Painted Pottery and Longshan Cultures. In The origins of Chinese c i v i l i z a t i o n , edited by David N. Keightley, pp. 177-216. University of C a l i f o r n i a Press, Berkeley. 1981 The t r a d i t i o n s of Chinese Neol i t h i c pottery. B u l l e t i n of the Museum of Far Eastern A n t i q u i t i e s 53:1-253. Hull, C. Hadlai and Norman H. Nie 1981 SPSS update 7-9. McGraw-Hill Book Company, San Francisco. - 247 _ Huntington, Richard and Peter Metcalf 19 80 Celebrations of death. The anthropology of mortuary r i t u a l . Cambridge University Press. Jacobsen, T.W. and Tracey Cullen 19 81 A consideration of mortuary practices i n Neol i t h i c Greece: b u r i a l s from Franchthi Cave. In Mortality and immortality. The anthropology and archaeology of death, edited by S.C. Humphreys and Helen King, pp. 79-101. Academic Press, San Francisco. Johnson, L.,.Jr. 19 72 Introduction to Imaginary models for archaeological scaling and c l u s t e r i n g . In Models i n archaeology, edited by D.L. Clarke, pp. 309-379. Methuen, London. Jones, Richard 1980 Computers and cemeteries: opportunities and l i m i t -ations. In Anglo-Saxon cemeteries, edited by P h i l i p Rahtz, T. Dickinson, L. Watts. BAR B r i t i s h Series 82:179-195. Kao, Je f f r e y Y.T. 1983 Notes from an archaeological year i n China. China Exchange News 11(1):12-15. Kao Gu E d i t o r i a l Staff 19 79 Discussion on the nature of the society of Dawenkou, and related problems. Kao Gu 1:33-36 (in Chinese, translated by Kazue Pearson). Keesing, Roger M. 1975 Kin groups and s o c i a l structure. Holt, Rinehart and Winston, Inc.,New York. Kendall, David G. 1971 Seriation from abundance matricies. In Mathematics i n the archaeological and h i s t o r i c a l sciences, edited by F.R. Hodson, D.G. Kendall and P. Tautu, pp. 215-252. Edinburgh University Press. King, Linda 1969 The Medea Creek cemetery (LAn-243): an investigation of s o c i a l organization from mortuary practices. Archaeological Survey Annual Report 11:23-68. Department of Anthropology, University of C a l i f o r n i a , Los Angeles. Kirch, Patrick V. 19 80 B u r i a l structure and s o c i e t a l ranking i n Vavalu, Tonga. The Journal of the Polynesian Society 89(3): 291-308. - 248 -Kirchhoff, Paul 1955 The p r i n c i p l e s of clanship i n human society. Davidson Anthropological Journal 1:1-10. Klecka, William R. 1975 Discriminant analysis. In S t a t i s t i c a l package for the s o c i a l sciences. Second Edition. Edited by N. Nie, CH. H u l l , J. Jenkins, K. Steinbrenner, D. Bert, pp. 437-467. McGraw-Hill, San Francisco. Kruskal, J.B. 1964 Multidimensional scaling by optimizing goodness to a nonmetric hypothesis. Psychometrika 29:1-27. Larson, Lewis, J r . 19 71 Archaeological implications of s o c i a l s t r a t i f i c a t i o n at the Etowah S i t e , Georgia. In Approaches to the s o c i a l dimensions of mortuary practices, edited by James A. Brown. Memoirs of the Society for American Archaeology 25:58-67. Leach, Edmund 19 77 A view from the bridge. In Archaeology and anthro-pology: areas of mutual i n t e r e s t , edited by Matthew Spriggs, pp. 161-176. BAR Supplementary Series 19. Levine, H.G. 1977 I n t r a - c u l t u r a l v a r i a b i l i t y and ethnographic descrip- t i o n : a decision-making analysis of funerary behavior among the New Guinea Kafe. Unpub. Ph.D. d i s s e r t a t i o n , University of Pennsylvania. L i Chi 19 79 Anyang. University of Washington Press, Seattle. Lo Shyh-Charng 19 77 A review of Lungshanoid s i t e s using c l u s t e r analysis and multidimensional scaling. Unpub. M.A. t h e s i s , Department of Anthropology and Sociology, University of B r i t i s h Columbia. Longacre, William 1970 Archaeology as anthropology: a case study. University of Arizona Anthropological Papers No. 17. Lu Zhong 19 76 The emergence of private ownership as indicated by the archaeological finds of the late primitive society from Jiangsu province. Kao Gu 3:165-167, 152,(in Chinese, translated by Kazue Pearson). Luo Gun and Zhang Yongshan 19 79 Change of clan system seen through the Dawenkou Culture. Zhongguo Shi Yanjiu 2:122-131 (in Chinese, translated by Kazue Pearson). - 249 -Ma Chengyuan 19 80 The splendor of ancient Chinese bronzes. In The great Bronze Age of China, edited by Wen Fong, pp. 1-19. The Metropolitan Museum of Art, New York. MacDonald, William K. 1980 The Bang s i t e , Thailand. An alternative analysis. Asian Perspectives XXI(1):30-51. Mainfort, Robert C. 1977 The Fletcher s i t e cemetery (20 BY 28) Bay County, Michigan: a study i n the S o c i a l dynamics Of the contact period. Unpub. Ph.D. d i s s e r t a t i o n , Department of Anthropology, Michigan State University. Marquardt, William H. 1978 Advances i n archaeological s e r i a t i o n . In Advances i n archaeological method and theory, Vol. 1, edited by Michael S c h i f f e r , pp. 257-314. Academic Press, New York. Matson, R.G. 19 74 Clustering and scal i n g of Gulf of Georgia s i t e s . Syesis 7:101-114. 19 75 Metric multidimensional scaling computer program. Department of Anthropology and Sociology, University of B r i t i s h Columbia. 1980 The proper place of multivariate techniques i n archaeology. American Antiquity 45(2):340-344. Matson, R.G. and W.D. Lipe 19 77 Seriation of Pueblo ceramic assemblages from Cedar Mesa, southeast Utah. Manuscript i n the possession of the authors. Matson, R.G. and D.L. True 1974 Site relationships at Quebrada Tarapaca, Chile: a comparison of cl u s t e r i n g and scaling techniques. American Antiquity 39 (1) :51-75. Milisauskas, S. 19 78 European prehistory. Academic Press, New York. Morrison, Donald G. 1969 Discriminant analysis. Journal of Marketing Research 6:156-163. Nanjing Museum 196 4 T r i a l diggings at Dadunzi i n Si Hu Zhen, Bei Xian, Jiangsu province. Kao Gu Xue Bao (2):9-56 (in Chinese). - 250 -Nanjing Museum (cont'd) 19 78 Inquiry into some problems concerning the Neol i t h i c culture i n the Yangtze River area. Wen Wu 4:46-57 (in Chinese, translated by Kazue Pearson). 19 81 Second season of excavations from the Dadunzi s i t e . Kao Gu Xue Jikan (1):27-81 (in Chinese). Orme, Bryony 1981 Anthropology for archaeologists. Cornell University Press, Ithaca, N.Y. Orton, Clive 19 80 Mathematics i n archaeology. Cambridge University Press. Orton, Clive and F.R. Hodson 1981 Rank and c l a s s : i n t e r p r e t i n g the evidence from p r e h i s t o r i c cemeteries. In Mortality and immortality. The anthropology and archaeology of death, edited by S.C. Humphreys and Helen King, pp. 10 3-115. Academic Press, San Francisco. O'Shea, John 19 81 Social configurations and the archaeological study of mortuary practices: a case study. In The archaeology of death, edited by Robert Chapman, I. Kinnes, K. Randsborg, pp. 39-52. Cambridge University Press. Pader, Ellen-Jane 1980 Material symbolism and s o c i a l r e l a t i o n s i n mortuary studies. In Anglo-Saxon cemeteries, edited by P. Rahtz, T. Dickinson, L. Wa11s. BAR B r i t i s h Series 82:143-159. 19 82 Symbolism, s o c i a l r e l a t i o n s and the interpretation of mortuary remains. BAR International Series 130. Peacock, William 19 76 Cluster analysis and multidimensional scaling of c u l t u r a l components at Glenrose. In The Glenrose Cannery s*ite, edited by R.G. Matson. Archaeologica 1 Survey of Canada Paper No. 52:214-2 30. Pearson, Michael P. 19 82 Mortuary practices, society and ideology: an ethno-archaeological study. In Symbolic and s t r u c t u r a l archaeology, edited by Ian Hodder, pp.99-113. Cambridge University Press. - 251 -Pearson, R i c h a r d 1978 The N e o l i t h i c c u l t u r e s of the lower Yangtze and c o a s t a l China. Paper submitted to the Scandinavian I n s t i t u t e f o r A s i a n S t u d i e s f o r i n c l u s i o n i n the p u b l i c a t i o n o f the symposium on The o r i g i n o f a g r i c u l t u r e and technology: west or e a s t A s i a . Moesgard, Denmark, i n p r e s s . 1981 S o c i a l complexity i n Chinese c o a s t a l N e o l i t h i c s i t e s . Science 213:1078-1086. 1983 The Qingliangang C u l t u r e and the Chinese N e o l i t h i c , ( w r i t t e n w i t h the a s s i s t a n c e o f Shyh-Charng L o ) . In The o r i g i n s of Chinese c i v i l i z a t i o n , e d i t e d by David N. K e i g h t l e y , pp. 119-145. U n i v e r s i t y o f C a l i f o r n i a P r e s s , B e r k e l e y . Peebles, C h r i s t o p h e r 1971 M o u n d v i l l e and surrounding s i t e s : some s t r u c t u r a l c o n s i d e r a t i o n s of mortuary p r a c t i c e s I I . In • Approaches t o the s o c i a l dimensions o f mortuary p r a c t i c e s , e d i t e d by James A. Brown. Memoirs of the S o c i e t y f o r American Archaeology 25:68-91. 19 72 M o n o t h e t i c - d i v i s i v e a n a l y s i s of the Mound v i l l e b u r i a l s : an i n i t i a l r e p o r t . Newsletter o f Computer Archaeology VIII(2):1-13. 19 74 M o u n d v i l l e : the o r g a n i z a t i o n o f a p r e h i s t o r i c community and c u l t u r e . Unpub. Ph.D. d i s s e r t a t i o n , Department of Anthropology, U n i v e r s i t y of C a l i f o r n i a at Santa Barbara. Peebles, C h r i s t o p h e r and Susan M. Kus 1977 Some a r c h a e o l o g i c a l c o r r e l a t e s o f ranked s o c i e t i e s . American A n t i q u i t y 42 (3): 421-448. Peterson, F r e d r i c F. 19 81 The e x c a v a t i o n o f a Bronze Age cemetery on Knighton Heath, Dorset. BAR B r i t i s h S e r i e s 98. Pokot y l o , David 19 81 Numerical taxonomy and mollusc remains: an i n t e r -p r e t i v e study from the Ryukyu I s l a n d s . In Subsistence and sett l e m e n t i n OkinaWan p r e h i s t o r y — Kume and Iriomote, Ryukyu A r c h a e o l o g i c a l Research Team, pp.56-71. Laboratory o f Archaeology, Depart-ment of Anthropology and S o c i o l o g y , U n i v e r s i t y o f B r i t i s h Columbia. 1982 A m u l t i v a r i a t e a n a l y s i s o f interassemblage v a r i a -b i l i t y i n the S h i e l d A r c h a i c T r a d i t i o n . In Approaches t o Algonquian archaeology, e d i t e d by M.G. Hanna and B. Kooyman, pp. 213-231. U n i v e r s i t y of C a l g a r y A r c h a e o l o g i c a l A s s o c i a t i o n . - 252 -Randsborg, Klavs 19 81 B u r i a l , succession and early state formation i n Denmark. In The archaeology of death, edited by R. Chapman, I. Kinnes and K. Randsborg, pp.105-121. Cambridge University Press. Rathje, W. L. 1970 S o c i o - p o l i t i c a l implications of Lowland Maya b u r i a l s : methodology and tentative hypotheses. World Archaeology 1 (3):359-374. 1973 Models for mobile Maya: a variety of constraints. In The explanation of culture change, edited by Colin Renfrew, pp.731-759. Duckworth, London. Rawson, Jessica 19 80 Ancient China. Harper and Row, New York. Rothschild, Nan A. 1979 Mortuary behavior and s o c i a l organization at Indian Knoll and Dickson Mounds. American Antiquity 44 (4) :658-675. Saxe, Arthur A. 19 70 Social dimensions of mortuary practices. Unpub. Ph.D. dissertaion, Department of Anthropology, University of Michigan. 1971 Social dimensions of mortuary practices i n a Meso-l i t h i c population from Wadi Haifa, Sudan. In Approaches to the s o c i a l dimensions of mortuary practices, edited by James A. Brown. Memoirs of the Society for American Archaeology 25:39-57. Seeman, Mark F. 1979 Feasting with the dead: Ohio Hopewell charnel house r i t u a l as a context for r e d i s t r i b u t i o n . In Hopewell Archaeology, edited by David Brose, N'omi Greber, pp. 39-46. Kent State University Press. Service, Elman R. 1975 Origins of the state and c i v i l i z a t i o n . W.W.Norton and Co., Inc., New York. Shangdong Archaeological Team, Institute of Archaeology, CASS 19 79 Excavation of the N e o l i t h i c s i t e at Wangyin i n Yanzhou, Shandong province. Kao Gu 1:5-14 (in Chinese, translated by Kazue Pearson). Shandong Archaeological Team, IACC 1964 Excavations ( f i r s t season) at Xixiahou i n Ju Fu County, Shandong province. Kao Gu Xue Bao 2:57-106 (in Chinese). - 253 -Shandong P r o v i n c i a l Cultural Properties Commission, and the Jinan City Museum 19 74 Dawenkou, report of the excavation of a N e o l i t h i c cemetery. Wen Wu Press, Bei j i n g (in Chinese). Shandong P r o v i n c i a l Museum 19 72 Preliminary report on the t r i a l diggings at the Ne o l i t h i c b u r i a l s i t e at Yedian (Zou xian), southern Shandong. Wen Wu 2:25-30 (in Chinese). 1978 Notes on the Dawenkou Culture. Wen Wu 4:58-67 (in Chinese, translated by Kazue Pearson). Shangraw, Clarence F. 19 77 Early Chinese ceramics and k i l n s . Archaeology 30 (6):382-393. 1978 Origins of Chinese ceramics. China Institute i n America, New York. 1983 Treasures from Shanghai. Archaeology 36(3):60-63,79. Shanks, Michael and Christopher T i l l e y 19 82 Ideology, symbolic power and r i u t a l communication: a reinterpretation of N e o l i t h i c mortuary practices. In Symbolic and s t r u c t u r a l archaeology, edited by Ian Hodder, pp. 129-154. Cambridge University Press. Shennan, Stephen 1982 Ideology, change and the European Early Bronze Age. In Symbolic and s t r u c t u r a l archaeology, edited by Ian Hodder, pp. 155-161. Cambridge University Press. Shennan, Susan 1975 The s o c i a l organization at Branc. Antiquity XLIX (193):279-288. Shepard, John 1979 The s o c i a l i d e n t i t y of the i n d i v i d u a l i n i s o l a t e d barrows and barrow cemeteries i n Anglo-Saxon England. In Space, hierarchy and society, edited by Barry C. Burnham and John Kingsbury. BAR International Series 59:47-79. Sneath, Peter H.A. and Robert R.1 Sokal 19 73 Numerical taxonomy. The p r i n c i p l e s and practices of numerical c l a s s i f i c a t i o n . W.H. Freeman, San Francisco. Sokal, Robert and Peter Sneath 196 3 Pr i n c i p l e s of numerical taxonomy. W.H. Freeman, San Francisco. - 254 -Spencer, Charles S. 1982 The Cuicatlan Canada and Monte Alban. A study of primary state formation. Academic Press, New York. Sullivan, Michael 1977 The arts of China. Revised Edition. University of C a l i f o r n i a Press, Berkeley. Suttles, Wayne 1960 A f f i n a l t i e s , subsistence, and prestige among the Coast Salish. American Anthropologist 62(2):296-305. Tainter, Joseph 19 73 The s o c i a l correlates of mortuary patterning at Kaloko, North Kona, Hawaii. Archaeology and Physical Anthropology i n Oceania VIII(l):1-11. 19 75a Social inference and mortuary practices: an experi-ment i n numerical c l a s s i f i c a t i o n . World Archaeology 7(1): 1-15. 19 75b The archaeological study of s o c i a l change: Woodland systems i n west-central I l l i n o i s . Unpub. Ph.D. di s s e r t a t i o n , Northwestern University. 19 76 Spatial organisation and s o c i a l patterning i n the Kaloko cemetery, North Kona, Hawaii. Archaeology and Physical Anthropology i n Oceania XI(2):91-95. 1977a Woodland s o c i a l change i n west-central I l l i n o i s . Mid-Continental Journal of Archaeology 2(l):67-98. 19 77b Modelling change i n p r e h i s t o r i c s o c i a l systems. In For theory building i n archaeology, edited by Lewis Binford, pp. 327-351. Academic Press, San Francisco. 19 78 Mortuary practices and the study of p r e h i s t o r i c s o c i a l systems. In Advances i n archaeological method and theory, Vol. 1, edited by M. S c h i f f e r , pp. 105-141. Academic Press, New York. 19 80 Behavior and status i n a Middle Woodland mortuary population from the I l l i n o i s Valley. American Antiquity 45 (2):308-313. 19 81 Reply to "A c r i t i q u e of some recent North American mortuary studies". American Antiquity 46 (2): 416-420. Tainter, Joseph and Ross H. Cordy 1977 An archaeological analysis of Social ranking and residence groups i n p r e h i s t o r i c Hawaii. World Archaeology 9(1):95-112. - 255 -Thomas, David Hurst 1978 The awful truth about s t a t i s t i c s i n archaeology. American Antiquity 43 (2):231-244. 19 80 The gruesome truth about s t a t i s t i c s i n archaeology. American Antiquity 45 (2):344-345. Thorp, Robert 19 79 Mortuary art and architecture of Early Imperial China. Unpub. Ph.D. d i s s e r t a t i o n , University of Kansas. 1980 Bu r i a l practices of Bronze Age China. In The great Bronze Age of China, edited by Wen Fong, pp. 35-50. A l f r e d A. Knopf, New York. Torgerson, W.S. 195 8 Theory and methods of scaling. John Wiley and Sons, New York. True, D.L. and R.G. Matson 1970 Cluster analysis and multidimensional scaling of archaeological s i t e s i n northern Chile. Science 169:1201-1203. Ucko, Peter J. 1969 Ethnography and archaeological in t e r p r e t a t i o n of funerary remains. World Archaeology 1(2):262-280. Van de Velde, Pieter 19 79 The s o c i a l anthropology of a N e o l i t h i c cemetery i n the Netherlands. Current Anthropology 20(1):37-58. Wei J i n 19 76 The e a r l i e s t class d i f f e r e n t i a t i o n i n ancient China as indicated by the archaeological finds of the primitive society. Wen Wu 8:1-5 (in Chinese, translated by Kazue Pearson). Weiss, Kenneth M. 19 73 Demographic models for anthropology. Memoirs of the Society for American Archaeology 27. Wen Wu Correspondent 1978 Summary of the symposium on Neol i t h i c culture on the lower reaches of the Yangtze River. Wen Wu 3:35-39 (in Chinese, translated by Kazue Pearson). Whalen, Michael E. 1983 Reconstructing early formative v i l l a g e orgnaization i n Oaxaca, Mexico. American Antiquity 48(1):17-43. Whittlesey, Stephanie 19 78 Status and death at Grasshopper Pueblo: experiments toward an archaeological theory. Unpub. Ph.D. d i s s e r t a t i o n , Department of Anthropology, University of Arizona. - 256 -Winters, Howard 196 8 Value systems and trade cycles of the Late Archaic i n the Midwest. In New perspectives i n archaeology, edited by S.R. Binford and L.R. Binford, pp. 175-221. University of Chicago Press. Wobst, H.M. 19 77 S t y l i s t i c behavior and information exchange. University of Michigan Museum of Anthropology, Anthropological Paper 61;317-342. Wood, John J. 19 74 A computer program for h i e r a r c h i c a l c l u s t e r analysis. Newsletter of Computer Archaeology IX(4):l-9. Wright, Gary A. 1978 Social d i f f e r e n t i a t i o n i n the Early Natufian. In Social archaeology: beyond subsistence and dating, edited by C.L. Redman, M.J. Berman, E.V. Curtin, W.T. Langhorne, J r . , N.M. Versaggi, J.C. Wanser, pp. 201-223. Academic Press, New York. Wu Ren 1982 The developing order of p r e h i s t o r i c cultures i n Shandong. Wen Wu 10:44-56 (in Chinese). Wu Ruzuo 1982 On types and periodization of the Dawenkou Culture. Kao Gu Xue Bao 3:261-282 (in Chinese). Wu Shanjing 19 73 A b r i e f discussion of the Qingliangang Culture. Wen Wu 6:45-61 (in Chinese, translated by Kazue Pearson). Xia Nai 1979 Carbon-14 dates and the Dawenkou Culture. In A c o l l e c t i o n of studies that discuss the Dawenkou Culture, pp.1-12. Wen Wu E d i t o r i a l Committee (in Chinese). 19 77 Carbon-14, authenticated chronology and the pre-h i s t o r i c archaeology of China. Kao Gu 4:217-227 (in Chinese, translated by Kazue Pearson). Yang Hong 1982 China's ancient weapons. China Reconstructs XXXI (1) :58-62. Yu Yingshi 19 81 New evidence on the early Chinese conception of a f t e r l i f e — a review a r t i c l e . The Journal of Asian Studies XLI(1):81-85. - 257 -Zhang Zhongbei 19 79 Analysis of the remains of the L i u l i n period of the Dawenkou Culture. J i l i n Daxue Xuebao (l):87-95 (in Chinese, translated by Kazue Pearson). FIGURE A3-1. The fu n c t i Functional Types b§i (cup) bo (bowl) ding (tripod food vessel) dou (footed serving stand for food) gai (potlid) guan (jar) 1 types, subtypes and styles of pottery at the Dawendou Site. Subtypes Style Numbers cai (painted) - - -gi ta (other) I, II t6ng xing (tubular shaped) I - III gao bing ( t a l l stem) I - VIII dan ba (simple handle) I - V cai (painted) I, II zhe fu (bent body with ridge) I - X qi ta (other) I - XI cai (painted) If II 1 yu'an f u (round body) I - V to guan shi pan (guan-like dish) I - VII shuang ceng pan (double layered dish) I, II x i bing (thin stem) I - V da ldu kong (large cut out holes) I j i a ( f i r s t ) , I y i (second), CO tong xmg (tubular shaped) cai (painted) II j i a , II y i , I I I , IV I - IX cai (painted) I - III qi ta (other) I - VI yuan fu (round body) I - V zhe fu. (bent body with ridge) I - V cont 1d. FIGURE A3-1 continued Functional Types guan (cont/'d) gui he hu (tripod pitcher) (spouted vessel, possibly for wine) (storage vessel, possibly for wine) Subtypes shen fu (deep body) wu zu (no legs) shi zu (solid legs) kong zu (hollow legs) ping di ( f l a t , l e v e l base) Style Numbers cai (painted) san zu (3 legs) cai (painted) cai bei (painted with spout at back) shuang b i (2 lugs) wu b i (no lugs) kuan j i a n (wide shoulders) bdi (1 spout at back of vessel) kui xing qi - -(helmet shaped vessel) pen (basin for water) Ping (bottle) shou xing qi. - -(animal shaped vessel) wan I, I I , IV j i a , I - IV I - III I - VI 1/ II I - VI If II I - V I - IX I - VII I - V I, II I, II I - III I, II <_n (bowl) cont'd. FIGURE A3-1 continued Functional Types y i (ladle) zun (wine vessel) zuo (stand, pedestal) Subtypes ping d l ( f l a t or l e v e l base) quan zu (ring footed) Style Numbers I, II I, II I - IV Other: these types are given i n the b u r i a l descriptions, but they are not discussed in Chapter 5 of the s i t e report nor are they included i n the photographs. dou zuo (in burials E105, E51) (Functional type - dou ?) ddu pan (in b u r i a l L9 8) (Functional type - dou ?) Note: The translations for the above terms are from The Chinese-English Dictionary (The Commercial Press Ltd., Hong Kong, 1979); t r a n s l a t i o n terms compiled by Stanford University, Chang (1981a) and Ma (1980). - 261 -FIGURE A3-2. The 15 p a i r s o f i n t r u s i v e b u r i a l s . ; L = L a t e p e r i o d ; M = M i d d l e p e r i o d ; E = E a r l y p e r i o d ; X = U n d a t a b l e , (dapo = ' c u t s i n t o 1 ) . Y o u n g e r O l d e r L10 E26 M9 E23 E33 E62 L15 E33 L24 E30 E18 E 3 1 M44 E43 E54 E58 L123 L124 E32 E 6 1 M16 E 6 1 E78 E129 M121 E132 M121 X133 X70 E 7 1 N o t e : T h e r e i s a d i s c r e p a n c y i n t h e b u r i a l d e s c r i p t i o n s f o r E3 2 , E 6 1 , M16. The d e s c r i p t i o n f o r E61 s a y s E36 c u t s i n t o i t . T h i s d e s c r i p t i o n seems t o be i n e r r o r . E32 i s d e s c r i b e d as c u t t i n g i n t o E61 and M16 a s c u t t i n g i n t o E 6 1 . A l s o , C h a p t e r 2 o f t h e s i t e r e p o r t a n d t h e b u r i a l d e s c r i p -t i o n s f o r X70 a n d E71 s t a t e t h a t X70 c u t s i n t o E 7 1 . The V s i t e map i n t h e s i t e r e p o r t (page 4) shows X70 as c u t t i n g i n t o E l l . I i n t e r p r e t t h e b u r i a l d e s c r i p t i o n s a s b e i n g c o r r e c t a n d t h e map a s l a b e l l e d i n c o r r e c t l y ( w i t h b u r i a l s 11 and 71 s w i t c h e d ) . A l l t h e maps i n t h i s s t u d y show E l l i n t h e s o u t h e r n p a r t o f t h e c e m e t e r y a n d E l l i n t h e n o r t h -e r n m o s t , c o n t r a r y t o t h e s i t e r e p o r t , p a ge 4. - 262 -FIGURE A3-3. Burials included i n the chronological analysis. Early b u r i a l s from the s i t e report: 6, 7, 8, 11, 12, 13, 14, 18, 19, 20, 23, 26, 27, 28, 29, 30, 32, 33, 34, 38, 41, 43, 45, 48, 51, 52, 53, 54, 55, 56, 58, 59, 61, 63, 65, 66, 71, 73, 76, 78, 79, 80, 81, 82, 84, 86, 87, 88, 89, 90, 91, 94, 99, 101, 102, 103, 106, 107, 109, 110, 111, 112, 114, 115, 116, 119, 120, 129, 130, 131, 132. Middle period b u r i a l s from the s i t e report: 9, 16, 21, 22, 35, 36, 42, 44, 46, 49, 67, 69, 75, 93, 96, 97, 98, 118, 121. Late period b u r i a l s from the s i t e report: 1, 2, 3, 4, 5, 10, 15, 17, 24, 25, 47, 60, 64, 72, 77, 100, 104, 105, 117, 122, 123, 124, 125, 126, 127. - 263 -FIGURE A3-4. The 83 C e r a m i c Forms i n t h e c h r o n o l o g i c a l a n a l y s i s . 1 b e i t d n g x i n g I , I I 2 b e i t d n g x i n g I I I 3 b e i gao b i n g gao b i n g I , I I , I I I 4 b e i I V 5 b e i g ao b i n g V 6 b e i gao b i n g V I 7 b e i g a o b i n g V I I , V I I I 8 b e i d an b a I I I 9 b e i d a n b a I V 10 b e i d a n b"a V 11 • k° I 12 d i n g z h e f u I 13 d i n g zh'e f u I I 14 d i n g d i n g z h e f u I I I 15 z h e f u I V , V 16 d i n g z he f u V I 17 d i n g z he f u V I I 18 .1 d i n g z he f u V I I I , I X 19 d i n g z he f u X 20 d i n g q i t a I 21 d i n g q i t a I I I 22 d i n g qT t a I V 23 d i n g q i t a V I , V I I 24 d i n g y u a n f u I 25 d i n g y u a n f u I I 26 d i n g y u a n f u I I I , I V , V 27 dou guan s h i p a n I 28 ddu guan s h i pa'n I I , I I I 29 ddu guan s h i pa'n I V , V 30 ddu s h u a n g c e h g p a n I I 31 ddu x i b i n g I 32 ddu x i b i n g I I , I I I 33 ddu x i b i n g I V 34 ddu d a l d u k o n g I j i a 35 ddu d a l d u k o n g d a l d u k o n g 1 £ 1 36 ddu I I j i a 37 ddu d a l d u k o n g I I y i , I I I 38 ddu da l d u k o n g I V 39 ddu t d n g x i n g 40 g a i I 41 g a i I I 42 g u a n q i t a I 43 guan q i t a yucin f u V 44 guan I , I I 45 guan y u a n f u I I I , I V , V c o n t 1 d . . . - 264 -FIGURE A3-4 continued 46 guan zhe fu I 47 guan zhe fu I I , III 48 guan shen fu I, II 49 guan shen fu III j i a 50 guan shen fil III y i 51 guan shen fu IV j i a 52 giian shen fu IV y i 5 3 guan shen fu V 54 g\li shi zu I, II 55 gai k5ng zu II 56 gai k5ng zu III 57 h£ ping d l I, II 58 he ping d i III 59 he' s5n zu II 60 hti shuang b i I 61 hd shuang b i II 62 ha shuang b i III 63 hu wti bi" I, II 64 hu wu bi" III 65 hu wu b l IV, V 66 hu wd b f VI, VII 67 hu kuan~jfan II 68 hu kuan jfan III 6 9 ha kuan j i a n IV, V 70 ha kuan jIan VI 71 hd bei I 72 hu bei II 73 hu bei III 74 hd bei IV 75 hd bei V 76 kui xing q l I, II 77 ping I I , III 78 w&n I 79 y T ~ ^ I, II 80 zun ping d l I, II 81 zun guan zu I, II 82 zun guan zd III 8 3 zun guan zu IV - 265 -FIGURE A3-5. .Distribution of ceramic categories among the b u r i a l sample. 70 -' i i i i i i i i i i i i i t u 01 - 34.7 60 H u o oo cu u co o 01 •§ C 50 4 40 30 A 20 H io ^ 18.2 11.4 7.4 7.4 4.0 3.4 1.2 1.1 1.1 i i i J _ J • t i l l I I l l l I i l l 1 0.6 o CM 0.6 -number of times a ceramic category i s present among the b u r i a l sample - 266 -FIGURE A3-6. Frequency of occurrence of ceramic categories after lumping. 36 i i i i i i i i 2% § CO CO •rl a cfl ri 60 o CO 0) CU u 60 cu T3 O CO u 60 CO 0) CU T3 U O 60 -H cn cn cn cn cn Oj pj r t r t r t r t r t H r t M M H M M CD CD CD CD CD CD CTl CO 00 M - J l-i Hi Hi Hi Hi W & - J o OJ OJ H-1 H 1 r— 1 t o OJ tubular shaped cup I x x x tubular shaped cup V x x bowl (bo) I x x x bent body tr i p o d I x x x x x 1 bent body tr i p o d II x x x CTl bent body tripod III x x x x x x x x , other tripod I x x x other tripod II x x other tripod III x x other tripod IV x x other tripod VI x x other tripod VII x x painted tripod I x x round bodied tripod I x x x x x round bodied tripod II x x x x x x serving stand with x x x x guan dish I Cont'd... - 277 -C l u s t e r 5 C l u s t e r 4 C l u s t e r 3 Ks* KA K/t K/l KA KA Kj* K/t * rS A KN A >A • N A K N TJ" 4-1 A o c_> C l u s t e r 2 C l u s t e r 1 •V S** M M rN t/N rS rS •V* ly* -V KN rN KN E66, E73 E130,E131 E81, E82 E:79/ E80 E49, E67 X X X X E48 E36 TJ cu fl •rH 4-> a o o I s o l a t e d b u r i a l E14 ro I IT) < D U H &H . f l 4-> •H H & H H T J C .fl rrj to 4-> -H cn TJ tn fl fl JO -H * 3 > M CU CD * tn! . f l 4-) •H & TJ C rrj 4-> CO T J tn fl fi ,rd - r l ^ > tn H OJ CO . f l 4-1 •H £ H H TJ fl g rrj CU 4-1 4-> CO CO tn C •H > U CU CO fl; •si X! X! — 4-> 4-> •H CO S M •H TJ m fl — (0 -P H 10 CO tn CU fl H TJ fl O o CU CO o .fl Xi 4-> -H •5 T J fl — rtJ 4-> H CO CO tn CU fl rH •H O > Xi H CU CO Xi -P •H T J C rd H 4-> H CO CO tn CU C 4-1 CO H •H m o .fl X! 4J •H £ T J c (TJ 4-1 CO CO tn CU fl T J C O O CU CO H o . f l 4-> •H T J fl rrj 4-> H CO to tn CU fl <—I •r l > H CU CO H o X! X! 4-) •H & T J C rrj > 4-1 H CO CO tn cu fl •rH > H CU CO o Xi T J •H rH 4-1 o a> H T J •H rH 4-> O 04 FIGURE A5-3 continued tr H C tn n o H- H CU fu I—1 rt CD M a, h-1 it-painted j a r jar with round body I jar with round body II jar with round body III jar with bent body I jar with bent body II jar with deep body II jar with deep body III ( f i r s t ) jar with deep body IV ( f i r s t ) jar with deep body IV (second) jar with deep body V tripod pitcher with s o l i d legs spouted vessel with f l a t base I *painted spouted vessel painted storage vessel II M M H M H O o o o n -0 CO h-1 r— 1 t- 1 CO CD M U) CTi c O CO w en cn ^ r t f t r t r t r t H W H M H (D n> ro (D CD CTi 00 CO H - J r< H r-i h o t o 00 U ) h- 1 H " t o U l X X X X X X X X X X X X X X X X X X X X X X X X t o 00 X X X X X X X X X X X X X X X X Cont'd. FIGURE A5-3 continued cr H C co M 1-1 0 u> H- M CTi CD fu fu H rt CD M Cb painted storage vessel with spout I storage vessel with two lugs I storage vessel with two lugs II storage vessel with two lugs III storage vessel with no lugs I storage vessel with no lugs II storage vessel with no lugs III storage vessel with no lugs IV storage vessel with spout I storage vessel with spout II w M a M w o o o o o 00 h-1 CTi h-1 h-1 H M h-1 co CO r- 1 co CTi 0 d N o CO CO CO CO CO rt rt f t rt rt M M M H H CD CD CD CD CD CTi 00 00 r—' l-i r-i H o Ni C O CJO h-1 M to U) *>. Ul X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X Cont'd... FIGURE A5-3 continued tr H M W W H C CD H M - J 00 I—1 n 0 H H- r— 1 CO •* o pj pj M r r H M M H CD CTi oo CO M H Ch o to U) I—1 \-> *» storage vessel with spout III x helmet shaped vessel II basin I x bowl (wan) I ladle II *wine vessel with f l a t base I wine vessel with ring foot I wine vessel with ring foot II x wine vessel with ring foot III stand, pedestal x Note: Styles that occur i n only one b u r i a l are excluded x = present * = style exclusive to one s p a t i a l area - 281 -FIGURE A5-4. The d i s t r i b u t i o n of ceramic styles among the s p a t i a l locations derived i n Chapter 5 for the Late Period b u r i a l s (see F i g . 5-5). rH CN Ceramic s t y l e : Cluster Cluster Cluster Cluster *other cup II X tubular shaped cup I X X X tubular shaped cup II X X tubular shaped cup III X X X t a l l stemmed cup III X X t a l l stemmed cup IV X X X t a l l stemmed cup V X X t a l l stemmed cup VI X X X X * t a l l stemmed cup VII X t a l l stemmed cup VIII X X tripod with bent body IV X X tripod with bent body VI X X tripod with bent body VII X X X t r i p o d with bent body VIII X X t r i p o d with bent body X X X X t r i p o d other IV X X t r i p o d other V X X *tripod with round body IV X *serving stand with layered dish I serving stand with layered dish II X X X serving stand with th i n stem I X X serving stand with thin stem II X X serving stand with thin stem III X X Cont 1d... - 282 -FIGURE A5-4 continued serving stand with thin stem IV *serving stand with thin stem V tubular shaped serving sta p o t l i d I p o t l i d VII p o t l i d VIII p o t l i d IX other j a r I other j a r V jar with round body IV jar with bent body II jar with deep body III ( f i r s t ) j a r with deep body III (second) jar with deep body IV (second) jar with deep body V *tripod pitcher with hollow legs I tri p o d pitcher with hollow legs II tr i p o d pitcher with hollow legs III *spouted vessel with l e v e l base III spouted vessel with three legs II painted storage vessel VI •H CN ro H H u H CD CD CD CD -P 4-> 4-1 4-> CO (0 CO CO 3 3 3 •H rH rH rH CJ CJ CJ CJ X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X Cont'd... FIGURE A5-4 continued storage vessel with two lugs IV storage vessel with no lugs IV *storage vessel with no lugs V storage vessel with no lugs VII storage vessel with wide shoulders II storage vessel with wide shoulders III storage vessel with wide shoulders IV storage vessel with wide shoulders V storage vessel with wide shoulders VI storage vessel with spout III storage vessel with spout IV storage vessel with spout V bott l e III bowl (wan) I *ladle I *wine vessel with f l a t base II *wine vessel with ring foot II - 284 -FIGURE A5-4 continued wine vessel with ri n g foot III wine vessel with ring foot IV cup with simple handle III cup with simple handle IV cup with simple handle V x x CN ro u u H u OJ 'OJ OJ CD 4-> 4-> •P 4-> co CO CO CO rH rH rH rH u P u P X X X X X X X X X X X X Note: Styles that occur i n only one b u r i a l are excluded. x = present * = st y l e exclusive to one s p a t i a l area. FIGURE A6-1. The 79 burials in the Early Period, analysis of status: age and sex, grave form, grave size. 6 7 8 9 11 12 14 18 19 20 21 23 26 27 29 30 32 33 34 36 38 41 42 43 44 45 48 49 F LA M, ER, LA RM, ER, LA RM, ER, LA RM RM D, FM M C ER, FM KEY t o t a l (4) C — c h i l d (adults unlabelled) (9) F = known female (9) M = known male (1) NS - no skeleton (11) RM = r e l i a b l y estimated male (10) FM = f a i r estimated male (2) FF = f a i r estimated female (1) D = disturbed (6) LT — log tomb (12) ER — second l e v e l platform (7) LA = large i n size Simple p i t s and small graves undesignated. FF Cont 1d. FIGURE A6-1 .continued 51 90 52 91 M 53 LT, ER, FM 94 C, LT, ER 54 NS, RM 97 55 F 99 M, LT, LA 56 101 FM 58 ER, FF 102 F 59 M, ER, LA 103 61 RM 106 RM 62 107 M, LT 63 RM 108 65 109 M, ER 66 RM 110 RM 67 F, ER 112 M I 71 114 C NJ 73 M 115 F » 76 116 LT 78 ER, LA 118 FM 1 79 FM 119 FM 80 120 81 LT, ER, FM 129 FM 82 F 130 F 84 131 F 86 132 87 FM 88 89 C FIGURE A6-2. The 2 3 variables and the i r attributes i n the Early Period analysis of status. (Coded presence/absence or by additive coding - see text f o r explanation.) Variables Presence/Absence black ware 26 (32. 9%) / 53 (67. 1 & ^ o / painted ware 17(21. 5%) / 62 (78. 5%) wine vessel 9 (11. 4%) / 70 (88. 6%) t a l l stemmed cup 3 ( 3. 8%) / 76(96. 2%) bone or elephant ivory carved cylinder 5 ( 6 . 3%) / 74 (93. 7%) t u r t l e s h e l l 5 ( 3. 3%) / 74 (93. 7%) bone or stone ring 8 (10. 1%) / 71 (89. 9%) stone, bone or ceramic bracelet 5( 6. 3%) / 74 (93. 7%) i stone hairpin 8 (10. 1%) / 71 (89. 9%) bone or horn spoon, spatula 13 (16. 5%) / 66 (83. 5%) 00 -~J hair t i e (pair or one half) 8 (10. 1%) / 71(89. 9%) probable ornament parts: 5 ( 6. 3%) / 74(93. 7%) 1 bone or stone tubular bead - guan thin, f l a t piece of stone - p i an stone annular object - huan stone c i r c u l a r bead - zhu small, round f l a t stone (bfng) small, f l a t piece of bone (ban) 3( 3. 8%) / 76 (96. 2%) 7( 8. 9%) / 72 (91. 1%) other pig parts 9 (11. 4%) / 70(88. 6%) oyster s h e l l pieces 6( 7. 6%) / 73 (92. 4%) Cont 1d... FIGURE A6-2 continued Additive Coding serving stand t o t a l number of ceramic vessels stone tool t o t a l bone tool t o t a l pig skulls deer teeth pieces of raw material: bone, horn, or tusk 0/1-2/3-9 : 34(43.0%)/41(51.9%)/4(5.1%) 0-7/8-14/19-30 : 6.4.(81. 0%)/12 (15. 2%)/3 (3. 8%) 0/1/2-7/9-16 : 45(57.0%j/22(27.8%)/9(11.4%)/3(3.8%) 0/1-7/13-25 : 51(64.6%)/24(30.4%)/4(5.1%) 0/1-3/4-5 : 49 (62.0%)/25 (31. 6'%)/5 (6 . 3%) 0/1-3/4-12 : 27(34.2%)/46(58.2%)/6(7.6%) 0/1-10/16-40 : 68(86.l%)/8(10.l%)/3(3.8%) - 289 -FIGURE A6-3. D i s t r i b u t i o n of serving stands, Early Period. 30 -' 20 H number o f s e r v i n g s t a n d s (dou) - 290 -FIGURE A6-4. D i s t r i b u t i o n of pottery vessels, Early Period, j i i i i i i i i i i i i i i i i i I I '-D 25 5 20 M PM u W to 8 15 CO 01 > u 01 4-1 o 14-1 o u CU ! 10 I 1 i i i i l I I l i i l i I i I i i I i ' a in o m o CN number of b u r i a l s - 291 -FIGURE A6-5. D i s t r i b u t i o n of stone t o o l s , Early Period. 20 10 H i i i i i s t o n e t o o l t o t a l s , E a r l y P e r i o d - 292 -FIGURE A6-6. D i s t r i b u t i o n of bone to o l s , Early Period. t i i i i i i i 3 0 crj 4J O 25 T3 O •H U CU 20 u CO W CO r H o o cu c o r Q o r J o 4 u a) • i 3 a i i i i i i i number of p i g s k u l l s - 294 -FIGURE A6-8. D i s t r i b u t i o n of deer teeth, Early Period. _i i I i i i i i i i I I _ !_ 20 -r 15 -CO rt 10 •H U 3 rO o u CU • § 3 5 -I I I O number of d e e r t e e t h , E a r l y P e r i o d - 295 -FIGURE A6-9. D i s t r i b u t i o n of raw material pieces, Early Period. T3 o •H u cu Hi 40 u Cu w 20 o cfl •H C H O rC c o rQ 15 to 3 rt •H U -H r-l U O) CU CO cw 0) O 4-1 cO U HJ • i c I I I I I I I I I I I I I I I . D 10 i \ 1 1 1 1 1 1 1 i i i r o LO o number of b u r i a l s - 307 -FIGURE A6-16. D i s t r i b u t i o n of the t a l l stemmed cup, Late Period. •s 60 .•3' I I I I L 15 CO o< a -a t3 cu o I u at FN CO cu & 4J 60 CJ •rl- i-J 4-1 o CU • i -3 (3 10 i i — i — i — r ~ O CO vO number o f b u r i a l s - 308 -FIGURE A6-17. D i s t r i b u t i o n of pottery vessels, Late Period. 93 V 71 o 0 u a cu 55 al 0 1 38 20 o 15 e c 10 i o n u m b e r o f b u r i a l s - 309 -FIGURE A6-18. D i s t r i b u t i o n of stone tools, Late Period. 10 i i i i i i i i i i i i i i i i 1 1 cd •H M <4-4 o u CD 1 a •5 0 -f I 1 • I I I I I I I I I o CN s t o n e t o o l t o t a l s , L a t e P e r i o d - 310 -FIGURE A6-19. D i s t r i b u t i o n of bone tools, Late Period. o •H U CU PH cfl O O CU C O I I I I I I I I 30 25 20 n 15 u cu rO Cfl 4J O 10 1 — i — i — r - T" O CN -J-number o f b u r i a l s - 311 -FIGURE A6-20. D i s t r i b u t i o n of deer teeth, Late Period. J i i i i i_ 10 -m o u Qi 42 o -I—I—I—I—I number o f d e e r t e e t h , L a t e P e r i o d - 312 -FIGURE A6-21. D i s t r i b u t i o n of raw material pieces, Late Period. 01 G o l_ 26 CO 3 4-1 rH CO •H U cu CO e TJ !5 o cO •H U l-l 0) 4-1 P4 o 01 CO 4-1 QI CO O 01 •iH P. s o 4-1 /CO O •rl rH — 01 42 c 6 M 3 O (3 43 10 number o f b u r i a l s - 313 -FIGURE A6-22. Grave area of Late Period b u r i a l s . i i i i i i i i i i i i i CO u cu 4J CU e cu u CTJ 3 cr CO c •H CO CU u cu > cfl u M 1 a r g e m e d i u m s m a 1 1 14.00-13.00-12.00-11.00-10.00-9.00-8.00-7.00-6.00-5.00-4.00-3.00-2.00-1.00-0.90-0.80-•14. •13. 12. 11. 10. 9. 8. 7. 6. 5. 4. 3. 2. 1. 0. 0. 99 99 99 99 99 99 99 99 99 99 99 99 99 99 99 89 i I o |-number o f b u r i a l s - 314 -FIGURE A6-2 3. The mortuary attributes c h a r a c t e r i s t i c of the cluste r s from the eight group solution, Ward's Method, Late period b u r i a l s . (Underlined add i t i v e l y coded attributes represent the greatest quantity of that a r t i f a c t type.) Cluster 1 Bu r i a l 117 white ware, black ware, painted ware, bot t l e , 14-16 t a l l stemmed cups, 38-93 pots, jade, bone carved cylinder, elephant ivory carved cylinder, bracelet, h a i r p i n , TO-* 27 bone tools, hair t i e , horn(zhui) 47 white ware, black ware, wine vessel, 2-12 serving stands, 14-16 t a l l stemmed cups, 38-93 pots, t u r t l e s h e l l , bracelet, hairpin, 1-3 stone to o l s , hair t i e , head or neck ornament, pig s k u l l s , 3-5 deer teeth 25 white ware, black ware, 2-12 serving stands, b o t t l e , 14-16 t a l l stemmed cups, 3 8-93 pots, jade, bone carved cylinder, r i n g , bracelet, hairpin, spoon, 5-19 stone tools, 1-6 bone tools, pig s k u l l s , other pig parts, 1-2 deer teeth 10 white ware, black ware, painted ware, 2-12 serv-ing stands, b o t t l e , 1-6 t a l l stemmed cups, 38-93 pots, jade, bone carved cylinder, elephant ivory carved cylinder, r i n g , bracelet, hairpin, 1-3 stone tools, head or neck orna-ments, probable ornament parts, pig s k u l l s , other pig parts, 3-5 deer teeth Cluster 1 i s the same as Cluster 1 of the Three Group Solution. Cluster 2 B u r i a l 17 10-27 bone to o l s , other pig parts, 1-2 deer teeth, 11-26 raw material pieces, black ware, 1 serv-ing stand, 1-13 pots, elephant ivory carved cylinder, r i n g , bracelet, 5-19 stone tools 125 black ware, wine vessel, 1 serving stand, 1-6 t a l l stemmed cups, 16-21 pots, r i n g , spoon, 5-19 stone tools, 10-27 bone tools, hair t i e , (zhui), pig s k u l l s , 3-5 deer teeth, 11-26 raw material pieces, oyster s h e l l 4 painted ware, wine vessel, 1 serving stand, 1-6 t a l l stemmed cups, 1-13 pots, jade, bone carved cylinder, t u r t l e s h e l l , 5-19 stone tools, 10-2 7 bone t o o l s , 1-2 deer teeth, pig s k u l l s , 11-26 raw material pieces, oyster s h e l l Cont'd... FIGURE A6-2 3 continued Cluster 3 B u r i a l 98 75 22 1-6 t a l l stemmed cups, 16-21 pots, spoon, 1-3 stone t o o l s , 1-6 bone tools, 1-2 deer teeth, oyster s h e l l black ware, wine vessel, 1 serving stand, 16-21 pots, spoon, 1-6 bone to o l s , 1-2 deer teeth, oyster s h e l l 1 serving stand, 1-6 t a l l stemmed cups, 1-13 pots, ring, spoon, other pig parts, 1-2 deer teeth Cluster 4 Bu r i a l 60 126 white ware, black ware, 2-12 serving stands, b o t t l e , 1-6 t a l l stemmed cups, 3 8-9 3 pots, probable ornament parts, (bing), other pig parts white ware, 2-12 serving stands, b o t t l e , 1-6 t a l l stemmed cups, 38-9 3 pots, bone carved cylinder, 1-3 stone tools, 1-6 bone to o l s , other pig parts, 1-2 deer teeth, 1 raw material piece Cluster 5 B u r i a l 72 white ware, wine vessel, 1-6 t a l l stemmed cups, 16-21 pots, h a i r p i n , head or neck ornament, other pig parts, 1-2 deer teeth 3 white ware, black ware, painted ware, wine vessel, 1-6 t a l l stemmed cups, 16-21 pots, bracelet, head or neck ornament, (bing), pig s k u l l , 1-2 deer teeth, 1 raw material piece Cluster 2, Cluster 3, Cluster 4, Cluster 5 i s the same as Cluster 2 of the Three Group Solution. Cluster 6 Bu r i a l 124 wine vessel, 2-12 serving stands, 1-6 t a l l stemmed cups, 1-13 pots, bracelet, 1-3 stone t o o l s , 1-2 deer teeth 123 painted ware, wine vessel, 1 serving stand, 1-13 pots, 1-3 stone t o o l s , 1-2 deer teeth 93 1 serving stand, 1-13 pots, 1-3 stone t o o l s , 1-6 bone to o l s , h a i r t i e , 1-2 deer teeth Cont'd. - 316 -FIGURE A6-23 continued Cluster 6 (cont'd.) Bu r i a l 122 1 serving stand, 1-13 pots, 1-3 stone t o o l s , other pig parts 16 black ware, 1 serving stand, 1—13 pots, 1-3 stone tools 96 1-13 pots, hair t i e , 1-2 deer teeth 46 1-13 pots 100 1 serving stand, 1-13 pots, 1-2 deer teeth 15 1 serving stand, 1-13 pots, (zhui), 1-2 deer teeth Cluster 7 B u r i a l 121 wine vessel, 1-13 pots, bracelet, 1-6 bone tools, hair t i e , pig s k u l l s , 3-5 deer teeth, ovster s h e l l 105 1-6 t a l l stemmed cups, 1-13 pots, bracelet, hair t i e , head or neck ornament, 1-2 deer teeth 104 wine vessel, 1-13 pots, bone carved cylinder, bracelet, hair t i e , 1-2 deer teeth Cluster 8 Burial 127 white ware, black ware, 1-2 serving stands, b o t t l e , 1-13 pots, 1-3 stone t o o l s , 1-6 bone tools 77 black ware, b o t t l e , 1-6 t a l l stemmed cups, 1-13 pots, elephant ivory carved cylinder, 1-3 stone t o o l s , 1-6 bone to o l s , 1-2 deer teeth 6 4 white ware, black ware, 1-6 t a l l stemmed cups, 1-13 pots, t u r t l e s h e l l , 1-6 bone to o l s , hair t i e , 3-5 deer teeth 5 black ware, 1 serving stand, 1-6 t a l l stemmed cups, 1-13 pots, bracelet, hairpin, 5-19 stone t o o l s , hair t i e , (bing), 1-2 deer teeth 2 black ware, 1-13 pots, bracelet, h a i r p i n , 1-3 stone tools, probable ornament parts, 3-5 deer teeth Cluster 6, Cluster 7, Cluster 8 i s the same as Cluster 3 of the Three Group Solution. - 317 -FIGURE A6-24. Data regarding the undatable b u r i a l s : age, sex, grave goods, energy expenditure, body d i s p o s i t i o n . ( A l l single p i t s , small i n size.) 1 37 c h i l d . no grave goods. 2 39 c h i l d . no grave goods. disturbed. 3 40 r e l i a b l y estimated male. skeleton on l e f t side. 2 stone tools, 4 bone tools, 1 t u r t l e s h e l l , 1 pig s k u l l , 2 pieces of raw material. 4 5 0 c h i l d . no grave goods. 5 5 7 known female. 1 stone t o o l , 1 deer tooth, 1 pig s k u l l . 6 68 c h i l d . 1 bracelet, 1 deer tooth, 2 pig s k u l l s . 7 70 multiple, unsexed b u r i a l . 1 stone t o o l . 8 74 c h i l d . no grave goods. 9 83 3 deer teeth. 10 85 known female. 1 stone t o o l , bone t o o l , oyster s h e l l . 11 92 multiple, unsexed. no grave goods. 12 95 c h i l d . no grave goods. 13 113 1 piece of raw material, 1 deer tooth. 14 128 no grave goods. 15 133 flexed. 1 raw material piece, 3 deer bones. Note: unsexed adults not designated. - 318 -GLOSSARY ban: small f l a t piece (of bone), possibly part of JFHL an ornament bang plan: oyster s h e l l , f l a t thin piece bei : cup ben: adze *-fr b i : spoon, spatula b i : round f l a t disk with hole i n center biao: d a r t - l i k e weapon, harpoon ^ bihuan: bracelet ^ bing: small f l a t c i r c u l a r piece of stone (function uncertain, possibly part of an ornament) bo: bowl chan: spade, shovel chikou: wide-mouth chi t i e kuang s h i : iron ore, red ochre? chui: hammer fk a & i t %Y To cong: long hollow object with rectangular sides J - r j x dab: knife dapo: in t r u s i v e (eg. burial) diao tong: carved cylinder •n ding: t r i p o d , food vessel ft dou: footed stand, serving stand for food 3. er: handle ercengtai: second l e v e l platform, or ledge on which =- £ grave goods were placed /7 aa a ix% eyu;lin: a l l i g a t o r hide T fanglun: spindle whorl fangwei: orientation - 319 hua wen: in c i s e d huan: small round piece (of stone), possibly an earring huang: semi-circular pendant j l : chicken X L fu: prostrate, prone fu: axe, hatchet gai: pot l i d gou: f i s h hook gu: bone R gu j i a : skeleton H ~yf ^ _ A"">r guan: tube, probably part of an ornament guan: jar gui: tripod pitcher ^jc 1 a gui j i a : t u r t l e s h e l l T he: crane he: spouted vessel, possibly for wine hu: storage vessel, possibly for wine ^ hua wen: motif ^ j i : h a i r p i n on which hair i s bound at back of head ^ j i a sha: sand tempered, sand paste ^ ]ian wei: arrow remnant ]iao : horn, antler ift j i a o q i ba: horn t o o l handle j i n g s h i : neck ornament; types: shi zhu (stone bead), shu yao zing zhu (waist b e l t shaped beads?), shi pian (small f l a t pieces of stone), guan zhuang shi zhu (tubular shaped stone beads) - 320 -ju ch! wen: sawtooth design kui xing q1: helmet shaped vessel M_ / h% l i a n : s i c k l e l i a n kou: constricted mouth l i a o : pieces of raw material l i s h i : whetstone l i u : spout - - - n u t 4> l u t u i gu: deer thigh bone b K 7 J k - n lusongshi: turquoise ^ 0 raao: spearhead k mo bang: ground stone s t i c k mu zang: grave, b u r i a l n i , n i z h i : smooth clay, f i n e l y levigated 'jb ^ niao zhao: b i r d claw pen: basin, for water /7t plan: f l a t t h i n piece (of stone) probably part of j^ j an ornament ping: bottle qu: bent (flexed) qiu: b a l l sha l i : grains of sand & if sheng wen: cord decoration ""^ s h i : stone D s Q3lT£ D,° shou xing q i : animal e f f i g y vessel ~FT~"/ A oo shu: comb jft shuf aqi: possible pair of hair t i e s it? 7^v 00 suo xing q i . : weaving shuttle shaped object ^ ~fi 3^ tao pian: ceramic sherd t^^| til % - 321 -tao q i : pottery, ceramic tao yao: pottery k i l n 120 tdu s h i : head ornament; types: chang fang shi pian ^ ( f l a t rectangular stone pieces), guan zhuang shi zhu (tubular shpaed stone beads), huan xing ( c i r -cular) , chang fang xing (rectangular), bu gui ze xing (irregular) wan: bowl 7B xiang ya: elephant tooth, ivory jpc. ^ xie fang ge wen: cross hatched design ^ ^ yan: rim, l i p yang: supine i^^ y i : ladle 2) yu: jade yu gu bao: f i s h bone bundle or lump zang ju : b u r i a l apparatus, (log tomb) ^-p-zad: c h i s e l , punch zhang: r i v e r deer ^ % zhang ya gdu xing q i : deer tooth f i s h hook shaped ^h. ^ ^£/^'3S t o o l zhen: needle f t F zhen guan: needle case zhi htfan: r i n g (for' finger) zhu: c i r c u l a r bead, probably part of an ornament ^ E T J C zhu menya: pig i n c i s o r s j-j zhu gu kuai: various pieces of pig bone ^ j£-zhu t i gu: pig hoof bones 3^' ^ zhu tdu: pig s k u l l ^ zhu x i a he gu: lower jaw bone of pig ^ ~f 6 ^ ,3, zhui: awl zhui: a hanging object, a plummet (possibly a net [Ms weight or pendant) ^ - 322 -zu: arrowhead zun: wine vessel zuo: stand, pedestal Sources: 1) The Chinese-English Dictionary, 1979, The Commercial Press, Hongkong 2) t r a n s l a t i o n terms compiled by Stanford University 3) Chang (19 81a) and Ma (19 80) for some ceramic terms