Of the Chinese Bronze

Home , Anyang, Changdao County, Dawenkou culture, Deng (state), Dos Cabezas, Dui (vessel)

READ ONLY/NO DOWNLOAD Ar chaeolo gy fthe of

Archaeology of the Chinese Bronze Age is a synthesis of recent Chinese archaeological work on the second millennium BCE—the period Ch associated with China’s ﬁrst dynasties and East Asia’s ﬁrst “states.” With a inese focus on early China’s great metropolitan centers in the Central Plains Archaeology and their hinterlands, this work attempts to contextualize them within Br their wider zones of interaction from the Yangtze to the edge of the onze of the Chinese Bronze Age Mongolian steppe, and from the Yellow Sea to the Tibetan plateau and the Gansu corridor. Analyzing the complexity of early Chinese culture Ag From Erlitou to Anyang history, and the variety and development of its urban formations, e Roderick Campbell explores East Asia’s divergent developmental paths and re-examines its deep past to contribute to a more nuanced understanding of China’s Early Bronze Age. Campbell

On the front cover: Zun in the shape of a water buffalo, Huadong Tomb 54 ( image courtesy of the Chinese Academy of Social Sciences, Institute for Archaeology).

MONOGRAPH 79 COTSEN INSTITUTE OF ARCHAEOLOGY PRESS Roderick B. Campbell READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

From Erlitou to Anyang

Roderick B. Campbell READ ONLY/NO DOWNLOAD

Cotsen Institute of Archaeology Press Monographs Contributions in Field Research and Current Issues in Archaeological Method and Theory

Monograph 78 Monograph 77 Monograph 76 Visions of Tiwanaku Advances in Titicaca Basin The Dead Tell Tales Alexei Vranich and Charles Archaeology–2 María Cecilia Lozada and Stanish (eds.) Alexei Vranich and Abigail R. Barra O’Donnabhain (eds.) Levine (eds.)

Monograph 73 Monograph 74 Monograph 75 The History of the Peoples of Rock Art at Little Lake: An The Stones of Tiahuanaco the Eastern Desert Ancient Crossroads in the Jean-Pierre Protzen and Stella Hans Barnard and California Desert Nair Kim Duistermaat (eds.) Jo Anne VanTilburg, Gordon E. Hull, and John C. Bretney READ ONLY/NO DOWNLOAD

To Jaline, Rui, and Kai

3 READ ONLY/NO DOWNLOAD

TheCotsenInstitute of Archaeology Press is the publishing unit of the Cotsen Institute of Archaeology at UCLA. The Cotsen Institute is apremier research organization dedicated to the creation, dissemination, and conservation of archaeological knowledge and heritage. It is home to both the Interdepartmental Archaeology Graduate Program and the UCLA/Getty Master’s Program in the Conservation of Archaeological and Ethnographic Materials. The Cotsen Institute provides aforum for innovative faculty research, graduate education, and public programs at UCLA in an effort to positively impact the academic, local and global communities. Established in 1973, the Cotsen Institute is at the forefront of archaeological research, education, conservation and publication and is an active contributor to interdisciplinary research at UCLA.

The Cotsen Institute Press specializes in producing high-quality academic volumes in several different series, including Monographs, World Heritage and Monuments, Cotsen Advanced Seminars, and Ideas, Debates, and Perspectives. The Press is committed to making the fruits of archaeological research accessible to professionals, scholars, students, and the general public. We are able to do this through the generosity of Lloyd E. Cotsen, longtime Institute volunteer and benefactor, who has provided an endowment that allows us to subsidize our publishing program and produce superb volumes at an affordable price. Publishing in nine different series, our award-winning archaeological publications receive critical acclaim in both the academic and popular communities.

The Cotsen Institute of Archaeology at UCLA Charles Stanish, Director WillekeWendrich, Editorial Director Randi Danforth, Publications Director

Editorial Board of the Cotsen Institute of Archaeology: Willeke Wendrich | Area Editor for Egypt, North, and East Africa Richard G. Lesure | Area Editor for South and Central America, and Mesoamerica Jeanne E. Arnold | Area Editor for North America Aaron Burke | Area Editor for Southwestern Asia Lothar VonFalkenhausen | Area Editor for East and South Asia and Archaeological Theory Sarah Morris | Area Editor for the Classical World John Papadopoulos | Area Editor for the Mediterranean Region

Ex-Officio Members: Charles Stanish and Randi Danforth External Members: Chapurukha Kusimba,Joyce Marcus, Colin Renfrew, and John Yellen

Copyediting by Ann Lucke |Cover design by Tracy Miller |Design by BLKVLD &Bos

Library of Congress Cataloging-in-Publication Data

Campbell, Roderick (Roderick B.) Archaeology of the Chinese bronze age :fromErlitou to Anyang /Roderick B. Campbell. pages cm. -- (Cotsen institute of archaeology monographs ;79) ISBN 978-1-931745-98-7 ISBN 978-1-938770-40-1(eBook) 1. Bronze age--China. 2. Excavations (Archaeology)--China. 3. China--Civilization. 4. China--Antiquities. I. Title.

GN778.32.C5C36 2014 931--dc23

2013043792

Second printing 2017 READ ONLY/NO DOWNLOAD

Contents

List of Figures 6

List of Tables 8

Author Biography 9

Acknowledgments 10

Chapter 1. Introduction 12

Chapter 2. The Erlitou Period 18

Chapter 3. The Erligang Period 68

Chapter 4. The Xiaoshuangqiao-Huanbei Period 106

Chapter 5. The Anyang Period 128

Chapter 6. Conclusion: The Central Plains Bronze Age from Erlitou to Anyang 184

Abbreviations /Bibliography 188

Index 213

5 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age Figures

Figure 2.1. Erlitou-period ceramic traditions (base map from Harvard Geospatial Library). Figure 2.2. Erlitou site map (after Xu et al. 2004:24, fig.1). Figure 2.3. Erlitou palace-temple area (Xuetal. 2005:14, fig. 1). Figure 2.4. Palace-temple 2atErlitou (after ZSKY 2003:67, fig. 2-4). Figure 2.5. Reconstruction of palace-temple 1(from Yang 2005:236, fig. 5). Figure 2.6. Erlitou elite artifacts (after ZSKY 2003:104–106, figs. 2-9, 2-10, 2-11, 2-12). Figure 2.7. Erlitou-variant ceramics (after ZSKY 1999:208, fig. 131; 211, fig. 133; 224, fig. 141; 133, fig. 79; ZSKY 2003:72–75, fig. 2-5). Figure 2.8. Dongxiafeng-variant ceramics (after ZSKY 2003:92, fig. 2-7, part A). Figure 2.9. Niujiaogang-variant ceramics (after ZSKY 2003: 92, fig. 2-7, part B). Figure 2.10. Yangzhuang-variant ceramics (after ZSKY 2003:92, fig. 2-7, part C). Figure 2.11. Xiawanggang-variant ceramics (after ZSKY 2003:92, fig. 2-7, part D). Figure 2.12. Map of Dongxiafeng site (after ZSKY et al. 1988:3, fig. 3). Figure 2.13. Jinzhong-tradition ceramics (Taigu Baiyan) (after ZSKY 2003:571, fig. 8-35). Figure 2.14. Zhukaigou-tradition ceramics (after ZSKY 2003: 580-81, fig. 8-39). Figure 2.15. Zhukaigou bronze artifacts (after ZSKY 2003:578, fig. 8-38). Figure 2.16. The Luwangfen-Songyao tradition and “influences” (ZSKY 2003:163, fig. 3-6). Figure 2.17. Xiaqiyuan-tradition variants (after ZSKY 2003:153, fig. 3-4). Figure 2.18. Lower Xiajiadian ceramics (after ZSKY 2003:601, fig. 8-47). Figure 2.19. Ceramics from Dadianzi Tomb 726(after ZSKY 2003:599, fig. 8-46). Figure 2.20. Datuotou-tradition ceramics (after ZSKY 2003:607, 8-48). Figure 2.21. Yueshi-tradition variant ceramics (after ZSKY 2003:450, fig. 8-2, A-B). Figure 2.22. Yueshi-tradition variant ceramics (after ZSKY 2003:451, fig. 8-2, C-D). Figure 2.23. Maqiao-tradition ceramics (after ZSKY 2003:465, fig. 8-5). Figure 2.24.Sanxingdui Erlitou–period ceramics (after ZSKY 2003:502, 503, fig. 8-17). Figure 2.25. Sanxingdui site (after ZSKY 2003:494, fig. 8-15). Figure 2.26. Jade blades from Sanxingdui (after ZSKY 2003:500, fig. 8-16). Figure 2.27. Qilidun-variant ceramics (after ZSKY 2003:542, fig. 8-27). Figure 2.28. Qijia bronze artifacts (after Li 2005:241, fig. 1). Figure 3.1. Erligang-period ceramic traditions (base map from Harvard geospatial library). Figure 3.2. Zhengzhou (after Yuan and Zeng 2004:60, fig. 1). Figure 3.3. Yanshi Shangcheng (redrawn from ZSKY 2003:206, fig. 4-5). Figure 3.4. Palace-temple complex at Yanshi (after Du 2005:198, fig. 4). Figure 3.5. Palace-Temple 4atYanshi (after ZSKY 2003:212, fig. 4-6). Figure 3.6. Erligang ceramic-tradition variants (after ZSKY 2003:172, 176–183, 193–195; figs. 4-1, 4-2, 4-4). Figure 3.7. Panlongcheng (after HWKY 2001:4, fig. 3). Figure 3.8. Panlongcheng Lijiazui M2 (after HWKY 2001:154, fig. 100). Figure 3.9. Bronze and jade artifacts from Lijiazui M2 (after HWKY 2001:176, 178, 180; figs. 117–119; Institute of Archaeology 2003:244, fig. 4-15).

6 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Figure 3.10. Panlongcheng Buildings 1and 2(after ZSKY 2003:233, fig. 4-11). Figure 3.11. Reconstruction of Buildings 1and 2(after HWKY 2001:644, fig. 13). Figure 3.12. Hushu-tradition ceramics (after ZSKY 2003:469; fig. 8-6). Figure 3.13. Jingnansi-tradition ceramics (after ZSKY 2003:475; fig. 8-7). Figure 3.14. Baota-tradition ceramics seriation (after ZSKY 2003:481; fig. 8-10). Figure 3.15. Wucheng-tradition ceramics (after ZSKY 2003:485; fig. 8-11). Figure 3.16. Wannian-tradition ceramics (after ZSKY 2003:490, fig 8–14). Figure 3.17. Sanxingdui-tradition phases III–IV ceramics (after ZSKY 2003:502, 503; fig 8–17). Figure 3.18. Sanxingdui-tradition Chaotianzui variant (after ZSKY 2003:510, 511; fig. 5-18). Figure 3.19. Siwa-tradition variant ceramics (after Zhao 1989:149; fig. 2). Figure 3.20. Xindian-tradition ceramics (after Nan 1989:76; fig. 2). Figure 4.1. Xiaoshuangqiao-Huanbei–period ceramic traditions (base map from Harvard Geospatial Library). Figure 4.2. Huanbei walled center (after ZSKYAG, ZJHQKDK 2010:10, fig. 1). Figure 4.3. Huanbei palace-temple Foundation 1(AT,IA, CASS 2004b:22, fig. 1). Figure 4.4. A) Baijiazhuang variant; B) Caoyanzhuang variant (after ZSKY 2003:256; fig. 5-3). Figure 4.5. C) Taixi variant; D) Daxinzhuang variant (after ZSKY 2003:256; fig. 5-3). Figure 4.6. E) Panmiao variant; F) Dachengdun variant (after ZSKY 2003:258; fig. 5-3). Figure 4.7. Panlongcheng variant (after ZSKY 2003:260; fig. 5-3). Figure 4.8. G) Beicun variant; H) Xiaoshen variant (after ZSKY 2003:259; fig. 5-3). Figure 4.9. Baoshan-tradition ceramics (after ZSKY 2003:522; fig. 8-21). Figure 4.10. Baoshan-tradition bronze artifacts (after ZSKY 2003:518, 519; fig. 8-20). Figure 5.1. Anyang-period ceramic traditions (base map from Harvard Geospatial Library). Figure 5.2. Anyang site map (redrawn from Campbell 2007). Figure 5.3. Duration of Anyang foundries (redrawn from Li 2003:311, fig. 7.9). Figure 5.4. Anyang palace-temple area (after Chang 1980:94, fig.23). Figure 5.5. Beixujiaqiaocun courtyard structure cluster (after Meng 2003:68, fig. 2). Figure 5.6. Xibeigang royal cemetery and sacrificial pits (after Tang 2004, fig. 7.5). Figure 5.7. Lineage cemeteries at Liujiazhuang (after Tang 2004, fig. 4.10). Figure 5.8. Anyang-variant everyday ceramics seriation (after ZSKY 2003:291, fig. 6-3). Figure 5.9. Anyang mortuary ceramic seriation (after ZSKY 2003:293, fig. 6-4). Figure 5.10. Subutun-variant ceramics (after ZSKY 2003:308, fig. 6-9). Figure 5.11. Anqiu-variant ceramics (after ZSKY 2003:307, fig. 6-9). Figure 5.12. Chang Zi Koutomb: Human remains (after HWKY,ZW2000:11, fig. 6). Figure 5.13. Chang Zi Koutomb: Grave Goods (after HWYK, ZW 2000:16, fig. 10). Figure 5.14. Qianzhangda-variant ceramics (after ZSKY 2003:307, fig. 6-9). Figure 5.15. Qianzhangda cemetery layout (after ZSKY 2005a:7, fig. 3). Figure 5.16. Bronze jue bearing insignia shi (after ZSKY 2005a:257, fig. 182). Figure 5.17. Tianhu-variant ceramics (after ZSKY 2003:309, fig. 6-9). Figure 5.18. Laoniupo-variant ceramics (after ZSKY 2003:308, fig. 6-9). Figure 5.19. Laoniupo burial 86XLIII1M41 (after Liu 2001:269, fig. 234). Figure 5.20. Laoniupo chariot pit 86XLIII1M27 (after Liu 2001:273, fig. 237).

7 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Figure 5.21. Laoniupo bronze burial goods: bronze mask from 86XLIII1M41 (after Liu 2001:296, fig. 256); gu-goblets from 86XLIII2M44 and 86XLIII1M33, respectively (after Liu 2001:284, fig. 245); and jue-vessels from 86XLIII2M44 and 86XLIII1M33, respectively (after Liu 2001:285, fig. 246). Figure 5.22. Weifang III ceramics (after ZSKY 2003:610; fig. 8-49). Figure 5.23. Weiyingzi-tradition ceramics (after ZSKY 2003:616; fig. 8-50). Figure 5.24. Zhenzhumen-tradition ceramics (after ZSKY 2003:462, fig. 8-4). Figure 5.25. Zhouliangyuqiao-tradition ceramics (after ZSKY 2003:476; fig. 8-8). Figure 5.26. Duimenshan-Feijiahe–tradition ceramics (after ZSKY 2003:479; fig. 8-9). Figure 5.27. Wucheng-tradition bronze artifacts from the Xin’gan tomb (after ZSKY 2003:488, 489, figs. 8-12,13). Figure 5.28. Lujiahe-tradition ceramics (after ZSKY 2003:515; fig. 8-19). Figure 5.29. Nianzipo-tradition ceramics and stone tools (ZSKY 2003:527, fig. 8-23). Figure 5.30. Liujia-type tomb, bronze bell,tubes, and ceramics (after ZSKY 2003:529; fig. 8-24). Figure 5.31. Zhengjiapo-type artifacts (after ZSKY 2003:532; fig. 8-25). Figure 5.32. Fenghao site pre-Zhou ceramics (after ZSKY 2003:534; fig. 8-26). Figure 5.33. Lingshi M1 (after Shanxisheng kaogu yanjiusuo 2006:16, 86, figs. 8, 93). Figure 5.34. Bronze artifacts showing Northern Complex influence or affiliation from the Lingshi Tombs(after Shanxisheng kaogu yanjiusuo 2006:134, 178, 84, 78 figs. 150, 203, 91, 85). Figure 5.35. CPBA-style bronzes from the Jin-Shaan Plateau area (after ZSKY:2003: 587, fig. 8-40). Figure 5.36. Northern Complex bronze and gold artifacts of the Late Shang Jin-Shaan Plateau (after ZSKY 2003:588–589; fig. 8-41). Figure 5.37. Lijiaya-tradition ceramics (after ZSKY 2003:590; fig. 8-42).

Tables

Table 2.1. Bronze artifacts discovered at Erlitou. Table 2.2. Comparison of Erlitou and Dongxiafeng cooking vessels. Table 3.1. Erligang relative chronology (adapted from ZSKY 2003:187). Table 3.2. Periodization of the Wucheng tradition (a comparison of Peng [2004:71), JBKY et al. [2005], and ZSKY [2003]). Table 4.1. Comparative chronology of important northern and southern sites. Table 5.1. Anyang periodization (based on ZSKY 2003:294, Keightley 1997:18, and Duandai 2000:88). Table 5.2. Anyang foundries and their mold fragments (after Li 2003:304, table 7.2). Table 5.3. Comparison of major sites, 1800–1050 BCE.

8 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Author Biography chaeological Initiative and Wenner-Gren Foundation funding for this project. Dur- Roderick Campbell is Assistant Professor of ing his time in China, Campbell had the East Asian Archaeology and History at the opportunity to participate in awide vari- Institute for the Study of the Ancient World, ety of ﬁeld projects and visit many of the NYU.Heobtained his MA in East Asian important sites mentioned in this book. Studies at the University of British Colum- The present work is asynthesis of seven bia (2001) and adual PhD in Anthropology years of research, interviews, and site visits. and East Asian Languages and Civilizations from Harvard University (2007). He was a Harvard-Yenching Fellowship visiting student at Peking University (2004–2006) and visiting student at the Chinese Academy of Social Sciences, Institute of Archaeology (2006–2007). Campbell has held postdoc- toral appointments at the Institute for the Study of the Ancient World (NYU), the Jouk- owsky Institute for Archaeology and the An- cient World (Brown University), and Merton College, Oxford University.Heiscurrently assistant professor of East Asian Archaeol- ogy and History at the Institute for the Study of the Ancient World. His specialty is East Asian Archaeology with afocus on Shang archaeology and epigraphy; his research interests span the rise and development of complex polities, violence, history, and production. Campbell’s published work includes an edited volume, Violence and Civilization: Interdisciplinary Sketches of a Deep History (Joukowsky Institute Publica- tions, 2011), apaper outlining a“networks and boundaries” approach to complex polities published in Current Anthropol- ogy (2009), and acoauthored Antiquity (De- cember 2011) article on his collaborative analysis of agigantic bone workshop at Anyang, China. He is currently codirector of the Sino-American Research Project on the Tiesanlu, Anyang Bone-Working Site in collaboration with the Chinese Academy of Social Sciences, Institute of Archaeol- ogy, and has received Luce East Asian Ar-

9 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Acknowledgments

Among the many people who wittingly University’s School of Archaeology and Mu- or unwittingly contributed to this book, I seology.Professor Liu’s seminars on Three would like first to acknowledge the original Dynasties archaeology taught me much, anonymous authors of The Xia, Shang Juan. and his generosity in bringing me to visit Tang Jigen, as one of those authors, as well the sites of Wucheng, Xingan, and Niutou as alongtime friend and mentor, needs in Jiangxi gave me invaluable firsthand ex- special mention. Many of Tang’s ideas perience of those sites, not to mention in- about Chinese archaeology have grown so troductions to local archaeologists. The intertwined with my own that it is difficult opportunity to participate in Peking Uni- for me to disentangle them. Although all versity’s field school at Zhougongmiao was of the archaeologists at the Anyang station also averyvaluable experience. Conversa- contributed to my knowledge of Chinese tions with the teachers and students there archaeology, He Yuling and YueZhanwei also enriched my knowledge of Chinese ar- were especially important for their generos- chaeology.Ofthese, Peking University pro- ity in sharing so much of their time during fessors Xu Tianjing, Dong Shan, Lei Xing- my many visits between 2004 and 2009. Pro- shan, Sun Qingwei, and Princeton graduate fessor Jing Zhicun of the University of Brit- student Kyle Steinke especially stand out. ish Columbia not only entertained me with Professor Zhao Chaohong’s course on jade many interesting hours of conversation on and Sun Hua’s course on bronze vessels Chinese archaeology, but also allowed me also influenced my understanding of these to participate in his and Tang Jigen’s survey artifact types, while Professor Sun’s invita- project while avisiting graduate student at tion to aconference on Sanxingdui allowed the Institute of Archaeology, CASS.Iwould me to visit the site, field station, and mu- also like to thank Professor Wang Wei, who seum and to talk to excavators. Professor took the time to supervise me as avisiting Li Shuicheng not only befriended my fam- student at CASS, IA, during 2006—2007 ily and me, but also helped me with sourc- despite his many duties as director of the es for the sections on the western areas. Institute of Archaeology.Jiang Bo was also Professor Wu Xiaohong indulged me with instrumental during the period Iwas astu- many interesting conversations about radi- dent in Beijing (2004—2007), especially ocarbon dating, expanding my understand- in helping me with various administra- ing of controversies in dating certain sites. tive issues throughout my year as avisiting student at CASS, IA, facilitating invita- Imust also thank my advisers at Harvard tions to conferences at Erlitou and other for whom aversion of this text was first places and generally being agood friend. written as an appendix to my dissertation. Especially significant in this regard was Another important influence on my think- Professor Rowan Flad’s careful reading and ing about Chinese archaeology was Profes- suggestions, not to mention his invitation sor Liu Xiu of Peking University, who kindly to participate in the 2005—2006 season of served as my adviser during my two years his and Gwen Bennett’s survey project in (2004—2006) as avisiting student at Peking Sichuan. The opportunity to meet Chengdu

10 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Archaeological Institute researchers (who also excavated the site of Jinsha) contributed to my knowledge of Sichuan archaeology.

Postgraduation, Imust acknowledge the New York University’s Institute for the Study of the Ancient World, Brown Uni- versity’s Joukowsky Institute for Archaeol- ogy and the Ancient Wworld, and Merton College, Oxford, for their generous post- doctoral (and now, in the case of ISAW, employment) support during the writing and researching of this and other projects. In particular, Iwould like to thank Profes- sors Sue Alcock and Jessica Rawson for their unﬂagging support during my time at Brown and Oxford respectively.Aspe- cial note of thanks should go out to Lothar von Falkenhausen, who encouraged me to work my dissertation appendix into abook and then waited patiently while Iprocras- tinated with other projects. Professor von Falkenhausen’s detailed editorial efforts enriched this text in innumerable ways and helped me avoid numerous errors. Iwould also like to thank the anonymous reviewer who made many useful suggestions. Finally Iwould like to thank Zhang Li, who as visiting scholar at ISAWspent many hours discussing ideas about Chinese archaeology and making suggestions for my manuscript. While the account Iampresenting here is almost entirely based on the work of other scholars, Ialone bear responsibil- ity for any errors of representation, ﬂaws of interpretation, or sins of omission.

Roderick B. Campbell

11 READ ONLY/NO DOWNLOAD

Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 1 Introduction

What is the “Chinese Bronze Age”? For deed, it is likely that many if not most, of many, “Chinese” or “China” refers to the those within the area of what is now the people and culture of the world’s longest People’s Republic of China did not speak continuous civilization. Yetlooked at his- any language ancestral to modern Chinese. torically, there are many “Chinas” despite In addition to archaic Chinese, there would theunifyingandprofoundlyahistoricmyths have been speakers of other Sino-Tibetan of the modern nation state that attempt to languages, as well as Altaic, Austroasiatic, project its boundaries and hard-won po- Hmong-Mien, Tai-Kadai, Austronesian, and litical self-consciousness back through the perhaps even Indo-European languages. mists of time. In the second millennium The geographic referent of this work is BCE, the period covered in this book, there also not that of the PRC(more or less the was no China. The people whose material nationalized boundaries of the Manchu culture is studied here did not yet, as far as conquests) but, rather, focuses on the Cen- we know, use the Eastern Zhou term Zhong- tral Plains region of mainland East Asia and guo,or“middle kingdoms,” noristhere any surrounding regions. The Chinese Bronze evidence that they considered themselves Age, then, is “Chinese” only in the weak and to have acommon collective identity.In- heuristic sense that what happened in the READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

second millennium BCE Central Plains has cessity of using familiar terms to allow po- been considered to be in the main current tential readers to identify the sort of book of dynastic history by subsequent regimes this is, “Bronze Age” is justified insofar as it and, thus, central to later elite historio- indexes aCentral Plains technological and graphical self-identifications (which were, cultural complex that was fundamental to in turn, foundational for both the Western contemporaneous political and religious construction of “China” and China’s self- life and continued to be so despite huge construction as anation state). socio-political changes over the course of the second and much of the first millennia The “Bronze Age,” like “China,” is both BCE as well. This work is thus not about the ubiquitous and, because of its superficial “Chinese Bronze Age” in either the sense of familiarity, treacherously problematic. the entirety of the area covered by the PRC Ever since Thompson periodized his Dan- or of the entirety of the period for which ish antiquities into Stone, Bronze, and Iron bronze vessels were central to the politi- ages in 1819, the advent of bronze has been cal economy.Rather, thiswork is about the seen as indicative of acertain level of so- second millennium BCE Central Plains and cial-technological development. Moreover, surrounding regions. Rather than trying to Childe’s (Childe 1963[1930]) association of archaeologically tell the story of achimeri- the Bronze Age with the first “states,” “cit- cal proto-China or monolithic Bronze Age ies,” and the rise of “civilization” is still part civilization, this work focuses on aseries of of the conceptual baggage accompanying major urban centers, their macroregional the term. Thus, despite recent work chal- contexts, and the changing forms of their lenging the idea that “cities” and “states” linked traditions. are necessarily developmentally connected (Blanton 1998; Smith 2003) (not to mention The rationale for writing this book is chiefly the difficulty in defining either term), there that the English-speaking (or, more accu- is still apernicious idea that the beginnings rately, -reading) world has fallen rather far of this one particular technology (bronze) behind the pace of archaeology in China caused the dawning of anew age. In fact, it since K. C. Chang published his last edition has been abundantly shown that the use of of The Archaeology of Ancient China in 1986 bronze has different socio-political entail- (Chang 1986). That is not to say that there ments around the world, meaning that it have been no English-language works on would be better to speak of “Bronze Ages” the Chinese Bronze Age since then but that than “The Bronze Age.” Moreover, in an either they too are now out of date, were in- area of the world where silk, lacquer, and tended as introductory textbooks, or were sophisticated, fast-wheel ceramics indus- narrowly focused on some specific thesis or tries, as well as some of the largest urban category of material. centers of their time, all predate metallurgy, we should, perhaps, reconsider our My own account, it should be noted, is chronological divisions and their evolu- largely based upon the Chinese Academy of tionary assumptions. The “Bronze Age” of Social Sciences, Institute of Archaeology’s the title, then, is not asocio-political evo- recent Chinese-language synthesis, Chinese lutionary epoch. Beyond the practical ne- Archaeology: The Xiaand Shang (Zhongguo

14 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 1

Shehuikexueyuan KaoguYanjiusuo hereaf- beyond these fundamentals. Even more ter ZSKY; ZSKY 2003), and would have not problematically, archaeological cultures been possible without it (or at least avastly are derived largely from formal ceramic ty- more difficult endeavor). While not without pologies, which are then assumed to cor- its limitations, The Xiaand Shang is easily respond with ethnic or political groups the most comprehensive and up-to-date ar- (Cohen 2001). While it is possible to avoid chaeological account of second-millenni- following these problematic ethno-polit- um BCE China published in any language. ical interpretations, the remaining narra- It is also important as amore or less repre- tive of ceramic or bronze typologies in time sentative Chinese archaeological state-of- and space is often less informative than the-field circa 2003. one might hope, especially when these culture-histories were constructed without What Iampresenting here, however, is consideration of issues of production, dis- not atranslation of The Xiaand Shang, but tribution, or consumption. Likewise, the rather an often-critical re-presentation of assumption that political and cultural his- material found in that work supplemented tory could be derived from material cultural with more recent publications; site reports; typology has meant that more nuanced ap- conversations with Chinese archaeologists; proaches to social and political archaeolo- observations based on fieldwork; and vis- gy, even basic work on settlement distribu- its to museums, sites, and archaeological tion, site structure, activity areas, or artifact stations. Although the scale of syntheses use, are still largely lacking, although work generally necessitates their being works at in the last decade has begun to address a distance, the Archaeology of the Chinese some of these issues. Bronze Age is even more so for its being an attempt to reinterpret alarge and uneven The traditional historiographic orienta- body of work based on more or less prob- tion of Chinese archaeology (Falkenhausen lematic premises. 1993) is evident in the very title of The Xia andShang.Focused on the archaeology of As Isee it, there are three main problems China’s traditional first two dynasties, The with The Xiaand Shang:its culture-histor- Xia and Shang devotes entire chapters to ical approach and assumptions, its tradi- debates concerning the origins of the Xia, tional historiographic orientation, and its Shang, and Zhou peoples, assuming not Central Plains-centric format. These are is- only the veracity of later chronicles con- sues inherited from the larger traditions of cerning these “dynasties,” but also that they Chinese archaeology (Falkenhausen 1993; corresponded to ethnic groups identifiable Liu Li 2004) and largely shared with the pri- through their ceramics. While it is easy mary work that The Xiaand Shang synthesis enough to avoid these debates and their was based on. problematic assumptions, the focus of The Xiaand Shang (symptomatic of Chinese ar- While material cultural classification, dis- chaeology in general) on the putative cent- tribution, and chronology are basic to ar- ers of Central Plains dynasties, both derives chaeological practice everywhere, Chinese from, and contributes to, arelativelyweak archaeology frequently does not get much coverage of “the periphery” and asense that

15 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

historical and cultural agency was the sole sity of ancient Mainland East Asia and any possession of the Central Plains polities. sense of history as authored by social actors This Central Plains bias is also on display (as opposed to faceless ethnic groups and in the placement of the non-Xia or Shang their pseudo-historic leaders) embedded in traditions into asection at the end of The diverse economies, societies, and techno- Xiaand Shang and in their absence on the cultural complexes. Lacking is any sense volume’s site distribution maps, making of social, cultural, or economic processes, the Central Plains traditions look curiously transformations, or eventful changes. One isolated. While Ihave attempted to work goal of this work then, is to recover what against this bias by discussing contempo- traces can be found of social change in the raneous regional traditions together and archaeological record of the second millen- placing them all on asingle map for each nium BCE. period, the present work is nevertheless unavoidably Central Plains-centric, where, My use of the phrase “ceramic tradition,” in- indeed, most of the archaeological work in stead of the more standard “archaeological China has been done. culture,” to translate the Chinese term kaogu wenhua,isintentional and based on the While the culture-historical, traditional fact that archaeological cultures in Chinese historiographic, and Central-Plains-centric archaeological practice are fundamentally biases are problematic enough from the based on formal ceramic typologies. As has perspective of achieving abalanced under- been repeatedly shown, however, the distri- standing of the peoples of the second-mil- bution of any single type of material culture lennium BCE Central Plains, their societies, does not necessarily correspond to ethnic, and their lifeways, there is yet another more linguistic, or political boundaries, much subtle and pernicious problem: namely, less all three. My use of the term “ceramic the way in which culture-historical identifi- tradition” instead of “culture,” then, aims cations of ethnic groups wedded to the tra- to de-link material cultural production and ditional historiographic tradition tend to consumption from ethnicity and social-po- produce anarrative of national ancestors, litical boundaries while foregrounding the even while the crypto-Marxist evolution- fact that ceramics are only one (if especially ary perspective and Central Plains–centric chronologically and geographically muta- bias supply both adirection and location ble) aspect of material culture. In Willey and for History.Inthis view, the stage of Chi- Phillips’s (1958:37) classic formulation, “An nese history was its Yellow River cradle, archaeological tradition is a(primarily) tem- where groups archaeologically identified poral continuity represented by persistent with the dynastic and predynastic ances- configurations in single technologies or other tors of the Chinese nation created civiliza- systems of related forms”(italics in original). tion. This centrally created Chinese culture By this definition, it would seem that what then magnetically drew in and sinified sur- Chinese archaeologists call “cultures” are, rounding, but backward, “ethnic minori- in fact, closer to pottery traditions. ties,” establishing the teleology that would lead to the modern Chinese nation. Lost in this narrative is both the cultural diver-

16 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 1

Layout of the Book mation for most of the questions to which they are normally put, they are nonethe- Like The Xiaand Shang,this work is essen- less information. Culture histories form tially aculture history of the Central Plains starting points from which more nuanced Bronze Age from Erlitou to Anyang. It is studies might begin. Moreover, changes in divided into four chronological chapters ceramic traditions over time, in addition to and asummary.While Ihave chosen to fol- changes in other traditions, the distribu- low the Chinese archaeological practice tion of sites, and other lines of evidence, of naming periods for the major center of can all contribute to abetter understand- that period, as well as The Xiaand Shang’s ing of the period in question. At minimum, well-motivated chronological divisions, I this material, for all its limitations (and es- have not followed its terminological prac- pecially if we are cognizant of these limita- tices. Thus, instead of “Xia,” “Early Shang,” tions), contributes apiece to alarger and “Middle Shang,” and “Late Shang,” Iuse more complex puzzle than has hitherto the less historiographically tendentious been adequately recognized in the Chinese periodization of “Erlitou,” “Erligang,” “Xi- archaeological literature. aoshuangqiao-Huanbei,” and “Anyang.” I have attempted to place the nonmetropoli- One ﬁnal note on bibliographic practice: tan traditions into the appropriate chrono- given that this book is aimed at an audience logical chapters, as opposed to being in that does not necessarily read Chinese, I their own section at the end of The Xiaand will attempt to be as comprehensive as pos- Shang.This integrated culture history, how- sible in citing English-language works, as ever, has the caveat that most of the periph- well as in citing Chinese-language works eral traditions are less ﬁrmly dated than the that are not included in The Xiaand Shang metropolitan traditions, and the former are (which is itself amassive Chinese-language largely dated through comparison with the bibliographic resource). latter.The utilityofthis comparative dating method, moreover, drops off the farther one gets from the metropolitan centers, not to mention the potential issues of time lag in transmission as well. In the future, it is hoped that more absolute dates will become available for corroboration or correc- tion.

Given all these issues, what is the utility of ahistory of ceramic and other material cultural traditions? Material cultural traditions are the remains of human practices, which are in acomplicated relationship with group afﬁliations and political and economic networks. Although ceramic traditions are not sufﬁcient sources of infor-

17 READ ONLY/NO DOWNLOAD

Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2 The Erlitou Period (ca. 1850–1600 BCE)1

Erlitou marks the beginning of the widely distributed across the Central Central Plains bronze tradition. This Plains and beyond, prompting some time and place have conventionally to speak of an “Erlitou expansion” beenunderstoodtousherintheChinese analogous to the “Uruk” expansion (Liu Bronze Age, but this is so only in a and Chen 2003; Allan 2007). Marked teleological sense, and then only if one differences in wealth are apparent believes that the mainstream of later from both residences and burials Chinese traditions have their wellspring at Erlitou, and there is evidence for intheCentralPlains. Erlitouisgenerally bronze, semiprecious stone, bone, and associated with China’s first dynasty ceramics industries. The agricultural (the Xia) in Chinese-language literature economy was built upon the advances of and, perhaps unsurprisingly, with the the previous millennium, with millets, first states (Liu and Chen 2001, 2003; rice, wheat, soybeans, cattle, sheep, Liu 2004, etc.) as well. Erlitou appears goats, pigs, and dogs providing both to be the largest East Asian center of agricultural diversity and the potential its time and shows evidence of long- for intensification. In terms of social, distance contacts. Erlitou ceramics are political, and economic life, however, READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

2.1. Erlitou-period ceramic traditions (base map from Harvard Geospatial Library).

beyond the broadest characterizations, tradition. Not surprisingly, the Western relatively little of substance is known. anthropology-derived evolutionary metanarratives of “state formation,” The Erlitou period is named for the site “Bronze Age,” and “the rise of of Yanshi Erlitou (Figure 2.2), asite of civilization” have become intertwined some 300 ha. Though originally thought with the traditional dynastic narrative to be an early Shang-dynasty site (Xu at Erlitou (see Liu and Chen 2003 for a 1959), it has since become associated prominent English-language example). with the capital of the Xia dynasty in Although Erlitou marks the beginning muchoftheChinese-languageliterature of the Central Plains bronze-casting (e.g., Zhongguo Shehuikexueyuan tradition and, thus, “The Chinese Kaogu Yanjiusuo, Erlitou Gongzuodui Bronze Age,” there are, in fact, earlier [hereafter, ZSKYEG] 1974; ZSKY bronze-using cultures in the territory 2003). The Xia was the ﬁrst of Chinese of the People’s Republic of China. historiography’s Three Dynasties, asort What makes Erlitou distinct is that it of political watershed in the textual marks the beginning of the compound

20 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2

2.2. Erlitou site map (after Xu et al. 2004:24, fig.1). mold casting of ritual bronze vessels— Although Erlitou was long thought to artifacts that were to remain culturally, mark asocio-political watershed, as politically, and ritually central to befitted its “first dynasty” status in Central Plains dynasts into the late first the Chinese-language literature, the millennium BCE. archaeology of the last fifteen years has brought to light agrowing flood of

21 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

2.3. Erlitou palace-temple area (Xuetal. 2005:14, ﬁg. 1).

large third-millennium centers, in the district; rammed earth monumental Central Plains and beyond, that rival or courtyard structures; and bronze ritual surpass Erlitou in size2.Erlitou, rather vessels, apparently made their ﬁrst than being adeparture, continued the appearance at Erlitou (Figures 2.2–2.6). third-millennium pattern of megasites, The Erlitou tradition, however, was centering an expansive sphere of not alone on the Mainland East Asian material cultural inﬂuence (Figure 2.1). stage. The land between the Yellow and At the same time, anumber of features Yangtze rivers was home to avariety of that were to become central to Central local and regional ceramic and other Plains Bronze Age elite traditions, such material cultural traditions beyond asarectangular,walled“palace-temple” those of Erlitou.

22 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2

2.4. Palace-temple 2atErlitou (after ZSKY 2003:67, ﬁg. 2-4).

23 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

2.5. Reconstruction of palace-temple 1(from Yang 2005:236, ﬁg. 5).

Erlitou was probably already a“central place.” During phase II Erlitou grew into a Discovered in 1959 by archaeologists, site of 300 ha with “palace-temples”; the site of Erlitou (Figure 2.2) has been agrid of four roads surrounding the divided into eight phases, only four of “palace” area; bronze-casting remains; them belonging to the Erlitou period and elite tombs3 containing bronze, (ZSKY 1999). The individual duration of jade, lacquer, turquoise, shell, proto- each of the phases is unclear, although porcelain, and white ceramics artifacts. theyareconventionallyassumedtobeof In phase III arammed earth wall was equal length. According to recent work, built enclosing the 10.8 ha “palace” duringphaseIoftheErlitouoccupation, area. New large-scale rammed earth Erlitou remains covered about 100 ha, platforms were built in the enclosed although it is unclear whether this was area, while the rammed earth asingle large settlement or acluster of foundations of the previous phase were several smaller sites (Xuetal. 2004). Xu leveled, and the area emptied of daily Hong and his collaborators also note, use features (such as wells, storage pits, however, that given its size and the etc.; Erlitou Fieldwork Team, Institute bronze, turquoise, and ivory artifacts of Archaeology, Chinese Academy of dating from this period, phase IErlitou Social Sciences [hereafter, EFT,IA,

24 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2

CASS] 2005a; Xu et al. 2004, 2005). The as loci of elite crafts production and piece-mold cast bronze vessels found hubs of resource extraction networks. at Erlitou also begin to appear in phase These, however, are not necessarily III.4 In phase IV,although the site center contradictory positions: large centers remained densely occupied, occupation like Erlitou were certainly connected to on the site periphery declined. larger political networks that extended Nevertheless, construction continued beyond the boundaries of the site, even within the walls of the “palace” area, while the mutual constitution of built bronze casting went on as before, and environment and human practice are the tombs of this phase on average crucial for understanding the nature exceeded those of phase III in quantity of Central Plains centers. Likewise, of jade and bronze artifacts (Xuetal. Erlitou may very well have been at 2004). Erlitou continued to be aplace of once apolitical capital, aceremonial importance into at least the beginning center, and anexus of elite production. of the Erligang period, after which it Unfortunately, information concerning shrank to about 30 ha concentrated in spatial practices at Erlitou and other the former “palace” area. By this time, Bronze Age Chinese sites is fragmentary where large-rammed earth platform at best, and characterizations are buildings once stood, only small house necessarily somewhat crude and foundations, middens, and tombs speculative on current evidence. remained: Erlitou had become avillage (see also Liu and Xu 2007). Palace-Temples

Interpreting Erlitou The large rammed-earth features near the center of the site, the so- Since at least Childe (1950), cities have called “palace-temples” (Keightley figured prominently in discussions of 1973; Thorp 1991), are perhaps the ancient polities. More recently, Yoffee earliest examples of what islater (2005) has argued that the cities acted unambiguously the standard form of as crucibles for new social and political Chinese palatial architecture (Figures relations, reconfiguring (indeed, 2.3–2.5). Indeed, wall-enclosed areas creating as such) the countryside, and of large, platform-built courtyard serving as centers of the first states. structures characterize later, imperial Smith(2003),ontheotherhand,cogently Chinese “forbidden cities.” writes that cities should be studied in their own right, de-linked from issues As noted above and described in more of complexity or state-formation. In detail in Xu et al. (2004; 2005) and Liu China, Chang (1985) argued that pre- and Xu (2007), the “palace-temple” Eastern Zhou cities were “king’s cities,” area appears to have undergone while Wheatley (1971) proposed that change throughout its use. These they were ceremonial centers. Liu and changes include, repeated leveling Chen (2003) more recently portrayed and rebuilding of courtyard structures; Early Bronze Age centers like Erlitou the apparent reduction of domestic

25 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

2.6. Erlitou elite artifacts (after ZSKY 2003:104–106, ﬁgs. 2-9, 2-10, 2-11, 2-12).

features, such as wells and middens, wood, thatch, and mud are eminently in phase III; and the construction of a perishable. The continual reordering surrounding wall in the same phase. of palace-temple space and if the Thesedevelopmentshavebeenlinkedto received tradition is to be believed, the narratives of wider (especially political) movement of capitals as well, moreover, change (see Liu Xu 2001), but given the suggest acultural logic of ceremonial limitations of the research that has been space that privileged aprocess of done thus far, caution is advisable (Liu constant reordering over rootedness or and Xu 2007). Nevertheless, the relative monumental permanence. ephemerality of Central Plains Bronze Age “palace-temple” structures appears The changes in the “palace-temple” to be afeature that outlived Erlitou. In area that took place in period III, from part, this might be explained in terms the building of arammed-earth wall of the building material: rammed-earth around the area, to the abandonment foundations with superstructures of of “palaces” 3and 5; the building of

26 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2

“palaces” 1(Figure 2.5),2(Figure 2.4),4, 7, 8, and 9ina“regulated pattern” (Xu Artifact Type Quantity et al. 2004, 2005); and the reduction of Knives 36 “domestic features” such as wells and Arrowheads 16 middens, may signal achange in the Jue-vessels 13 nature of the“palace-temple” area. They Chisels 7 suggest, perhaps, amore ceremonial, Bells 5 less residential function beginning in Awls 5 phase III (Xuetal. 2004, 2005). This, in Animal mask plaques 3 turn, may signal achange in the nature Round plaques 3 of the site or of elite spatial-political or Jia-vessels 3 religiouspractices.Unfortunately,inthe Fish hooks 3 absence of fine-grained archaeological Dagger axes 2 research aimed at recovering the Adzes 2 activities that may have taken place in Saws 1 the palace-temple areas before or after Ding-cauldron 1 phase III, nothing specific can be said He-vessel 1 about the wider significance of those Yue-axe 1 changes at present. Spindle whorl 1 Bead 1 In terms of later developments, Table 2.1. Bronze artifacts discovered at Erlitou the walled, rectangular “palace- temple” area (Figure 2.3) is certainly the predecessor for elite building literature, thereisverylittle evidence practices that continued in the Central otherthanlater traditions to suggest Plains through the second and first their function(Thorp 1988),and indeed, millennia BCE and on to later Imperial as Xu et al. (2004,2005) suggests, that times (Thorp 1988). The shift of elite functionmay havechanged over thelife architectural orientation, moreover, of the site, never mind the course of the from afew degrees west of north at secondmillenniumBCE. Do their large Erlitou, to afew degrees east of north courtyards suggestthe open spacesof in Erligang sites, such as Yanshi publicarchitectureand collective ritual, Shangcheng and Zhengzhou, is cited or does their limited access through a by some authors (e.g., Sun 2009) as single entrance suggest morerestricted evidence of dynastic change. Indeed, use (see Figures 2.3–2.5)? Erlitou similar changes in the orientation of courtyard structuresalsoappearto major architecture occurred again with differ from their descendants at nearby the Zhou conquest of the Shang. YanshiShangcheng, laterHuanbei andAnyang, in therelative paucity of It is important to note,however, sacrificial remains associated with that despite their characterization them. Indeed, of the 57 irregularburials as gongdian “palaces” in Chinese notedinthe sitereport(ZSKY 1999) and or “palace-temples” in the English cited in Huang (2004)asevidence of

27 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

sacrifice,onlytwo wereassociated with some of them yielding bronze and alarge courtyardstructure and only jade grave goods. One of the ash pits, nine were located in the central area dating from phase IV,apparently (Figure 2.2: V). Instead, and alsounlike had many irregular burials though later “palace-temples,” they share with no further information is presently smaller Erlitou structuresacustom of available (Zheng 2005). Irregular mortuary ritualwithintheir precincts burials, however, are not necessarily during or after their use (ZSKYEG1992; sacrificial victims, and deposition of Liu and Xu 2007). One of the “ash- human remains in middens is aburial pit”burials (VM62) notedabove,for practice that occurs at later sites as well, instance, appears to have grave goods, distinct from more obviously sacrificial but the clearest examplesofthis practice pits. Indeed, looking back over the arethe “severalrowsofmiddle-size 57 “irregular burials” cited by Huang burials” discovered in the central and (2004) as evidence for human sacrifice southerncourtyardsof“palace” 3and at Erlitou, only nine of them show dating from phase IIthatyieldedbronze, clear evidence of even anonstandard jade, lacquer,white pottery,stoneware, mortuary arrangement of the body, let turquoise,and shellartifacts,including alone perimortem violence. Moreover, the famous turquoise“dragon”(ZSKYEG some of the “irregular burials” are 2005; ET,IA, CASS2005b). By Erlitou irregular only in not having aproper standards, and despite their designation burial pit: they are otherwise laid out as “medium-sized”,these areclearly in accordance with normal Erlitou highelite burials. death ritual including grave goods. Like ordinary Erlitou tombs, these irregular Sacrificial Remains burials seem to be scattered around the site, occurring in areas II, IV,V,and VIII, Recent Erlitou site maps show a in addition to the unknown number in sacrificial area north of the palace- the “sacrificial area.” In other words, the temple zone stretching between areas spatial organization of ritual activities, VI and IX, giving the impression that including death ritual, at Erlitou is not Erlitou was neatly divided into different entirely clear on present published precincts and that ritual was mostly evidence. conducted there (Figure 2.2). However, the designation “sacrificial area” is Burials based on unpublished excavations done in 1995, when “ash trenches,” The layout of burials, their grave burials, and avariety of structures goods, and the comparative wealth of were found. Because the structures tombs within sites or regions have long apparently had no superstructures and been staples of Chinese archaeology no easily identifiable function they and, indeed, archaeology beyond were labeled “ritual.” As with many China as well. Based on the current other structures at Erlitou, however, state of excavation and publication, these were associated with burials, Erlitou burials display some apparent

28 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2 similarities and dissimilarities with is the relative insensitivity of grave latersites.Ontheonehand,thetradition size to tomb wealth, giving rise to the of burying drinking vessel sets that may appellation “middle-sized tombs” have begun with the Dawenkou culture for the largest and richest tombs (ca. 4500–3000 BCE) in Shandong found at the site (including the elite (Underhill 2002) and peaked at Late burials found within “palace” 3). The Shang Anyang is in evidence at Erlitou assumption behind this terminology with the signiﬁcant appearance of is that the tombs of the high elites or vessels cast in bronze. At the same time, royalty have not been found, with the the apparent lack of distinct cemeteries one large, looted “burial” in “palace set apart from residential areas 1” (Figure 2.5) now thought to have supposedly distinguishes Erlitou from been asacriﬁcial pit (ZSKY 2003:129). earlier sites, such as Taosi (Zhongguo An alternative hypothesis might be Shehuikexueyuan Kaogu Yanjiusuo that mortuary distinction was not as Shanxi Dui [hereafter ZSKYSD] 1980, marked at Erlitou as at later Central 1983), or later sites, such as Anyang Plains sites (or at earlier Taosi) and that (see chapter 5). Indeed, it has recently what distinctions were on display were been proposed that this lack of distinct more in the nature of the grave goods burial areas indicates the absence of than the size of the tomb. Indeed, this extended kin groups as asocio-political hypothesis seems to be borne out in organizing principle (Liu and Xu arecent statistical study of Central 2007). Instead, the site is said to have Plains mortuary distinction over drawn together unrelated people in a time (Campbell 2007), albeit with the process of urban ethnogenesis. While important caveat that the publication the centripetal demographic draw of of Erlitou tombs has been fragmentary alarge site like Erlitou and the forging and heavily biased toward elites. of new identities in the resulting urban crucible (Yoffee 2005) are surely part Workshops /Craft Production of the story, as will be argued below, Areas the distinction between burial and residential areas at such Late Shang Bronze-Casting sites as Anyang ismore apparent than Approximately 300 meters south of the real. In fact, the spatial and temporal palace-temple enclosure is a1-ha area contiguity between structures and that has been dubbed abronze foundry burials and the ordered clusters of the (Figure 2.2; ZSKY 2003; Xu et al.2005). latter might more readily suggest the Bronze casting remains, such as slag, opposite conclusion: that living and pieces of copper, tin, and lead, as well dead were closely connected through as mold and crucible fragments, have continuity of place and, perhaps, been excavated within the area (ZSKY kinship. 1999). Moreover, several structures associated with bronze casting were Asecond, although not so obvious, also excavated between 1982 and 1984 distinction of Erlitou burial practice (ZSKY 2003), although these remain

29 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

to be fully published. The main part workshop location or scale? Hopefully of the bronze-working area consisted future work and the publication of of anorth-south line of rectangular previous research will help advance our structures that were oriented east- knowledge of these issues. west. Three of these buildings were relatively well preserved. The first, In terms of the products of the bronze F9, was asemisubterranean structure workshops at Erlitou, the many ceramic constructed in phase II, renovated mold fragments appear to be mostly several times, and open on at least one vessel molds, while only one stone mold side. The phase III floor shows evidence has been discovered at the site (ZSKY of such casting activities as burned 2003). The majority of the published areas, slag, and crucible fragments. bronze artifacts discovered at Erlitou, South of F9 were two long shallow however, are weapons and tools (17 semisubterranean structures, Z1 (18 percent vessels, 18 percent weapons, x4m) andZ2(9 x5m), with postholes 55 percent tools, respectively). There is suggesting aroof of some kind and a also aclear increase in bronze artifacts burned surface embedded with slag, over time, with the vast majority dating crucible fragments, and small lumps to phases III and IV (2 percent phase I, 6 of bronze. Z2 was superimposed on the percentphaseII,42percentphaseIII,50 western half of Z1, indicating that Z2 percent phase IV; ZSKY 2003). Evidence was later.All threestructures had child for compound mold-cast bronzes also and adult burials under them as appears date from phase III onward. While to be typical of Erlitou in general. Akiln the full publication of more recent was also found in the bronze workshop excavations (such as the phase II area, but whether or not it was used for elite tombs found in the courtyard of firing molds is unknown (ZSKY 2003). “palace” 3) might change these figures somewhat, it nevertheless seems that More work aimed at reconstructing in terms of both the organization of the production processes and workshop palace-temple area and the expansion layout is needed before much can be of the bronze industry, phase III was a said concerning bronze production watershed for Erlitou. at Erlitou. How was the production process divided up? What was the scale Jade and Turquoise Working of production, and how much of the South of the palace-temple area, foundry area was actually occupied turquoise-working debris were found by workshops as opposed to debris scattered over an area of about 1,000 scatter? Indeed, crucible fragments and m2 beginning in phase III (Figure 2.2). other casting debris have been found in Apit containing over 1,000 pieces of areas III, V, and VI, in addition to the turquoise, showing signs of drilling, foundry in area IV (Figure 2.2). Does sawing, and grinding, was found in the this mean bronze casting took place area as well and dates to phase IV (Liu all over Erlitou, or simply that debris and Xu 2007). Since turquoise artifacts scatter is an unreliable indicator of (such as the famous “dragon” found

30 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2 in tomb 02VM3 [ET,IA, CASS 2005b]), included sawing, grinding (and in some dating from phase II, have been found cases carving), and, finally, polishing. at Erlitou, it seems likely that turquoise Technical studies of Erlitou bone- working was occurring at some part of working are currently underway. the site during phase II, if not earlier. Althoughnoevidencefortheproduction Although no jade working sites have of either lacquer objects or textiles has been published for Erlitou, many jade been found at Erlitou, there is mortuary artifacts have been excavated from evidence for both categories of object. tombs at the site. These prominently The remains or imprint of textiles is include large jade blades, such as axes most often found on jade blades, bronze and dagger-axes, but also shovels, plaques, or bronze bells in tombs at arrowheads, bracelets, and “handle- Erlitou. Based on the number of strands shaped objects”(Cao 2008), among per square centimeter (ranging from 8 other forms (ZSKY 2003) (Figure x8to 50 x50strands), the textiles are 2.6). Jade artifacts are found only in hypothesized to include both hemp burials at Erlitou, continuing Neolithic cloth and silk (ZSKY 2003). Although traditions of jade use in mortuary ritual. receiving far less attention in the Nevertheless, later tradition suggests literature on the Chinese Bronze Age, the use of jade in ritual, or as ornament, ceramic, stone, bone, wood, and textiles among the living. Systematic use-wear would have figured much larger in the and residue analyses might shed light lives of people living at the time than on the social life (Appadurai 1986) of bronze or jade and, it is to be hoped, jades before terminal deposition. Thus will form afocus of future research. far, such work is only just beginning Dozens of lacquer objects or their (Jing et al., 2007). remains have been found in Erlitou tombs. The majority are vessels, the Other Production mostnumerousofwhicharegu-drinking Ceramic kilns and bone production beakers. Interestingly, in tombs, debris have been found throughout lacquer gu-beakers are sometimes the site, and no specific areas of accompanied by bronze jue-tripods pottery or bone tool production have and ceramic he-pots (ZSKY 2003). In been identified, although amidden other words, drinking and feasting containing alarge quantity of worked sets that would become made entirely bone was excavated in area VI (ZSKY of bronze in future Central Plains 2003). Ceramic production techniques elite assemblages were composed of a at Erlitou are said to include hand- variety materials at Erlitou. In addition building, wheel-throwing, and mold to vessels, boxes, spoons, ladles, and use, sometimes used in combination on even drums and coffins made of lacquer asingle vessel (ZSKY 2003).5 Bone tools, have been found in Erlitou tombs (ZSKY which included spades, decorative pins, 2003:117). needles, awls, spoons, and arrowheads, were generally made in aprocess that

31 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Stone Tool Production at Huizui Agriculture Huizui was an Erlitou-tradition, Erlitou- variant village roughly 10 km southeast The study of paleobotanical and of Erlitou (Ford 2004; Liu et al. 2007; zooarchaeological remains at Erlitou, Zhongguo shehuikexueyuan kaogu as in Chinese archaeology in general, yanjiusuo Henan diyi gongzuodui is in its infancy.Knowledge of the [hereafter, ZSKYHDG] 2010a, 2010b). It agricultural practices that supported was acenter for stone tool production, the presumably large population at taking advantage of local lithic Erlitou is, thus, minimal. Nevertheless, resources (Webb et al. 2007). There under the auspices of the Chinese were four other stone tool production government-sponsored “Origins sites within 10 km (ZSKYHDG 2010a), of Chinese Civilization Project,” suggesting the entire region was some preliminary paleobotanical, specialized in stone tool production. zooarchaeological, and stable-isotope Huizui had been producing stone research have been done at Erlitou tools for nonlocal consumption since (Yuan and Campbell 2009). Zhao (2007) Longshan times and continued to noted that in the centuries preceding produce the same artifacts in the same Erlitou, rice and soybeans were ways through Erlitou times, albeit on added to the earlier millet-based crop alarger scale (ZSKYHDG 2010b). This assemblage. During Erlitou times small suggests that the rise of nearby Erlitou amounts of wheat began to appear as did not qualitatively affect the stone tool well. Moreover, between the Erlitou and productionatHuizuioritsorganization. the Erligang period, wheat dramatically Given the general population increase increased in quantity (see also Lee in the Yiluo valley during Erlitou times, et al. 2007). With Erlitou civilization the increase in the scale of production then, the elite crafts and architecture or number of producers at Huizui could that has so preoccupied archaeologists be explained by increased demand, and art historians of China, was increased population at the site, or, built upon adeveloping multicrop most likely, both. Sites like Huizui and assemblage that included millets, rice, Guandimiao (see chapter 6) suggest wheat, and soybeans. The breadth of that certain forms of Central Plains this assemblage would have not only Bronze Age craft production occurred insured against the failure of any one in economically specialized villages, crop, but perhaps also presented the and that regional trade networks for a opportunity for multicropping and variety of goods were not necessarily agricultural intensiﬁcation. controlled by elites in the metropolitan centers, nor were centers like Erlitou Recent faunal analysis at Erlitou and necessarilytheonlynodesofproduction previous Longshan sites such as Taosi, for exchange. Wangchenggang, and Xinzhai (Yuan et al. 2007; Yuan and Campbell 2009) have shown that from the Longshan times to Erlitou, while pig continued to

32 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2 predominate, new domesticates, such (Zhongguo Shehuikexueyuan Kaogu as cattle (Bos sp.) and sheep (Ovis sp.), Yanjiusuo Xinzhai Dui,Zhengzhoushi began to form ever-greater percentages Wenwu KaoguYanjiuyuan [hereafter of the faunal assemblage. At the same ZSKYXD, ZWKY] 2009; ZSKY2003).6

2.7. Erlitou-variant ceramics (after ZSKY 1999:208, fig. 131; 211, fig. 133; 224, fig. 141; 133, fig. 79; ZSKY 2003:72–75, fig. 2-5). time, cattle and sheep played an At the sametime, and beginning from increasingly important role in ritual as phase II, Erlitou ceramic influencesand sacrificial victims and in marking status vessels began to showupinplacesasfar differences from Erlitou through Zhou awayasmodern LiaoningProvincein times (Yuan and Flad 2005). the north, Hubei in the south, Shandong inthe east, and Gansu in thewest.7 The Erlitou Ceramic Tradition TheZSKY(2003)lists five different regional variantsofthe Erlitou “culture,” The site of Erlitouformedthe center of or as Ihave termedit, theErlitou period ceramic traditionknown as the Erlitou CentralPlainsMetropolitanTradition: traditionand satwithinthe further Erlitou (Figure2.7); Dongxiafeng subcategorization of theErlitou variant, (Figure 2.8);Niujiaogang(Figure which is saidtohave evolved outoflocal 2.9);Yangzhuang(Figure 2.10); and Henan Longshan ceramic traditions, Xiawanggang (Figure 2.11).All ofthe suchasWangwan IIIand Xinzhai variants, with the exception of Erlitou,

33 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

2.8. Dongxiafeng-variant ceramics (after ZSKY 2003:92, ﬁg. 2-7, part A).

datefromErlitou II or later,and some or measuresofsome or all of these. What do not seem to be derived from previous, thisceramic tradition expansion says local, ceramic traditions.8 Thissuggests about political organization is even less an expansion of the Erlitou ceramic clear.Nevertheless,wewillattempt to traditionoutfromtheErlitouvariantcore piece together what evidencecurrently where it hadindigenously developed. exists concerning ceramicproduction, This wascoupledwithanexpansionin distribution, and consumption for the thesizeofthesiteandsocialstratiﬁcation Erlitou traditionand one of the better evident at Erlitou.Intermsofnetworks known variants: Dongxiafeng. ofceramicproduction,distribution, and consumption, however, unfortunately Erlitou and Dongxiafeng very little canbesaidonpresent Variants: Ceramics and Society evidence.Thatis, it is unclear whether the expansion of the Erlitou ceramic The best-known site of the Dongxiafeng traditionwas thespread of styles,the variant (Figure 2.8) is the Dongxiafeng spread of productiontechniques, site (Figure 2.12) in southeast Shanxi themovementofceramic producers, Province. This site has four phases theresultofdistribution networksof dated to the Erlitou period, and two ceramics produced in the Erlitou core, to the Erligang period (Zhongguo

34 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2

2.9. Niujiaogang-variant ceramics (after ZSKY 2003: 92, ﬁg. 2-7, part B).

2.10. Yangzhuang-variant ceramics (after ZSKY 2003:92, ﬁg. 2-7, part C).

Shehuikexueyuan Kaogu Yanjiusuo, ZLB, SKY 1988; ZSKY 2003). The actual Zhongguo Lishi Bowuguan, size of the Dongxiafeng site during the Shanxisheng Kaogu Yanjiusuo [ZSKY, Erlitou period is not known, although

35 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

2.11. Xiawanggang-variant ceramics (after ZSKY 2003:92, ﬁg. 2-7, part D).

there is aroughly 3haarea surrounded were used to inter the dead, although, by double ditches that appears to interestingly, there is only evidence of be the focus of phase III activities.9 this from period IV. Unfortunately, asystematic coring or Other than some “ash pits,” two burials, sampling of the site was never done and some artifacts from phase Iand II, beyond attempting to delimit the the majority of the evidence concerning surrounding ditches and excavating the Erlitou-period site occupation and use west corner of the enclosed area and a comes from phase III. It was in phase III few other locations. The nature of the (Erlitou phase III) that the double ditch- site as awhole is still not adequately es were built in the center of the site understood.10 and most of the houses discovered were dug. As noted above, alarge area of the The Dongxiafeng site has some unusual western corner of the double-ditched characteristics in comparison with area (locality 5) was excavated (approxi- other Central Plains sites of this period. mately 3,600 m2 of the total 6,723 m2). The surrounding ditches were used for Most of the remains found in this area the construction of cave-houses. Cave- also date to phase III, except the cave- houses are atype of residence that do house burials, which date to phase IV, not appear at Erlitou, where houses are suggesting that at least this portion of either semisubterranean or built on the the double-ditch was partially depopu- surface.11 Cave houses, are, however, lated during or before phase IV.12 Dat- found at the southern Shanxi site of ing from phase III, however, there are Yuanqu and are typical of late Neolithic 37 cave-houses, four pottery kilns, two settlements in southern Shanxi (ZSKY wells, ﬁve tombs, 13 “storage” caves, 20 2003). Dongxiafeng also has “cave” small pieces of slag, 6stone molds, and burials where abandoned cave-houses 21 ash pits in locality 5alone, suggest-

36 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2

2.12. Map of Dongxiafeng site (after ZSKY et al. 1988:3, ﬁg. 3). ing to some that this area was “a work- ential understandings of the site as an ing and residential area of craftsmen” elite-managed, specialized, and inten- (Liu and Chen 2001:17). Although inter- sive craft and resource transshipment pretations of this craft-working area and center within the larger framework of its wider context have motivated inﬂu- an expansionist Erlitou state (Liu and

37 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Chen 2003), the evidence is thin. Four tradition variant (Jiang 2008, however, small kilns (average firing chamber size considers it aseparate archaeological of 1.6 m3), 20 “small pieces” of slag, culture altogether), as we have and 6stone molds for casting small mentioned above. This categorization bronze artifacts do not necessarily sug- is based on stylistic (color, vessel form, gest large-scale or intensive production. vessel decoration, and the presence Furthermore, the “elite burial” in local- of diagnostic vessel types) rather than ity 4isnot elite by Erlitou standards,13 functional or production criteria. If we nor does acontrastbetween this burial compare the manner of production, and the five found in locality 5(not a Dongxiafeng (Figure 2.7) and Erlitou very large sample to begin with) show (Figure 2.8) show some interesting agreat disparity in wealth.14 Moreover, differences. The great majority of the claim that locality 4isasegregated Dongxiafeng phases Iand II ceramics elite residential area is predicated on a are hand-coiled (ZSKY,ZLB,SKY 1988). single burial. Asingle large cave-house AtErlitou,ontheotherhand,mostofthe (F2: 10 m2)dating from period III was ceramics are supposedly wheel-made also found in locality 4, although it was during this and later periods (Erlitou not as large as some of the cave-houses phases II–IV; ZSKY 1999, but see Zhang in locality 5(the “low-status craft pro- 2012). During Dongxiafeng phases III duction area”) and does not show any and IV,the occurrence of wheel-made obvious signs of being an “elite resi- ceramics increases but hand-coiled dence.” In short, amoresystematic in- pottery is still common. This suggests vestigation of the structure of the site is that the similarities between the Erlitou required for phase III before the nature and Dongxiafeng variants are not the of the area encircled by the ditches and result of the movement of ceramic its relationship to other parts of the site specialists from Erlitou to Dongxiafeng can be determined with any certainty (or at least cannot be explained entirely (not to mention we would gain abetter in these terms). understanding of the other three Er- litou period phases). As it stands, there Another interesting point of appears to have been alarge Erlitou comparison is the relative percentages period site at Dongxiafeng with inten- of ceramic types in the assemblages as sive occupation during at least phase awhole. According to the published III, along with ceramic production and figures of cooking vessel types found in some casting of simple bronze artifacts. the site reports for Erlitou (ZSKY 1999) Evidence for social stratification at the and Dongxiafeng (ZSKY,ZLB,SKY 1988) site, whether in mortuary treatment tabulated below (Table 2.2), there are or architecture, is slight, and there is significant differences in their relative no evidence of large-scale or intensive frequency, despite overlap in vessel bronze casting or smelting.15 forms. Looking at Table 2.2 we can see that unlike Erlitou, but similar to the In terms of ceramics, Dongxiafeng is contemporary Jinzhong tradition (see considered by most to be an Erlitou- below), yan-steamers are an important

38 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2 cooking vessel type accounting for that originated at Erlitou. The kilns an average of 25 percent of cooking discovered at the site suggest that much vessels during Dongxiafeng phases or all of the pottery found (for at least Ithrough IV.Another difference is phase III) could have been produced that ding-cauldrons, although afairly at the site. The distribution of vessels common cooking vessel at Erlitou, all produced at Dongxiafeng is simply but disappear from Dongxiafeng after unknown. The ceramic vessels types phase I. While it could be said that both used at Dongxiafeng appear to follow sites share the major cooking vessel asomewhat different pattern than at form of guan-pots, in fact, despite the Erlitou, while vessel decoration and common Chinese name, the deep guan shapes show similarities with Erlitou and round-bellied guan-pots (Figure ceramics. In sum, the site structure, 2.7: 15, 16; 3, 4) that predominate at nature, and extent of activities of Erlitou are quite dissimilar to the single- the Dongxiafeng site are still not eared guan-pots (or perhaps pitchers) of entirely clear during its Erlitou period Dongxiafeng (Figure 2.8: 4). occupation, as are the political, cultural, While it is well beyond the scope of and economic relationships between this study to perform an exhaustive this site and Erlitou, or even this site re-analysis of the Dongxiafeng variant and other sites of the Dongxiafeng of the Erlitou ceramic tradition, the variant. Given this, it is very difﬁcult differences in house type and major to say more than that there is some cooking vessel forms argue against the similarity in the ceramics, and thus hypothesis that this ceramic tradition evidence of interaction—the nature variant was the product of apopulation of which is unknowable on present

Phase Ding YanSEguan RB guan DB guan Jia Zeng Li Total Dongxiafeng I4 78 30022 26 15.4% 26.9% 30.8% 11.5% 0% 0% 7.7% 7.7% 100% Dongxiafeng II 415352 2221072 5.6% 20.8% 48.6% 2.8% 2.8% 2.8% 2.8% 13.9% 100.1% Erlitou II 15 24 19 9082 59 25.4% 3.4% 6.8% 32.2% 15.3% 0% 13.6% 3.4% 100.1% Dongxiafeng III 124576 9412 104 1% 23.1% 54.8% 5,8% 8,7% 3,8% 1% 2% 100.25 Erlitou III 13 10 17 29 179 77 16,9% 1,3% 0% 22,1% 37,7% 1,3% 9,1% 11,7% 100,1% Dongxiafeng IV 3523713186045174 1.7% 30.0% 21.3% 7.5% 10.3% 3.4% 0% 25.9% 100.1% Erlitou IV 12 10 29 16 1818 85 14.1% 1.2% 0% 34.1% 18.8% 1.2% 9.4% 21.2% 100%

Table 2.2. Comparison of Erlitou and Dongxiafeng cooking vessels

39 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

evidence—apparently did not involve surrounding ditch (ZSKY 2003), but no either wholesale population movement additional details are available as the or large-scale importation of Erlitou site remains unpublished. Rice paddy ceramics. agriculture was practiced at this site.

Niujiaogang Erlitou Variant Xiawanggang Erlitou Variant

Located in eastern Henan (Figure 2.1), This variant gets its name from the the best-known site of this variant Xiawanggang site in Xichuan county, is the Niujiaogang site located in Qi southwest Henan Province. This site county, eastern Henan. Its Erlitou- dates from Erlitou ItoIII,but because period ceramic variant (Figure 2.9) is the remains dating to Erlitou Ishow said to show obvious differences from relatively little Erlitou influence, only that of the Erlitou variant (ZSKY 2003) the Erlitou II and III period remains are and dates from the end of Erlitou II to considered to be avariant of the Erlitou ErlitouIV.Althoughitsceramictypesare tradition (Figure 2.11). The distribution basically the same as those of Erlitou, of this variant is also not completely its ceramic tradition is said to also show clear at present (see Figure 2.1). The influences from the adjacent Yueshi houses of this site are all round and andXiaqiyuan traditions. The houses semisubterranean. Both urn burials at the Niujiaogang site take the form and pit burials are known from this either of rectangular semisubterranean site, with urn burials generally used dwellings with asingle room and a for children.16 Although in most of the fire pit or stove or of small surface adult burials the bodies were extended structures. Beyond this, however, very and supine, there were also cases of little can be said about this variant on crouching posture. Also notable were present evidence. the burials, which, though poor in grave goods, occasionally had bovine skulls in them.17 Scapulamancy was also Yangzhuang Erlitou Variant practiced at the site with deer, pig, and sheep scapula (ZSKY 2003). Named for the Zhumadian Yangzhuang site in southern Henan, the distribution While the polity centered at the site of this variant is not entirely clear of Erlitou may very well have sought at present (see Figure 2.1). The to secure the resources of southern Yangzhuang-site Erlitou remains Shanxi and other areas (Liu and Chen date from Erlitou periods II and III 2001, 2003), the mechanisms by which only.Thisceramic tradition (Figure they may have done so and the political 2.10) shows many continuities with relationship between Erlitou and preceding upper– and middle–Huai the settlements of the Dongxiafeng River Longshan traditions (ZSKY and other variants can neither be 2003). A“ritual” area and remains were understood in terms of equations of reported found at this site, as well as a ceramic styles with political boundaries

40 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2 nor derived from general models of “early state” political economic evolution.18 To be sure, the late Erlitou bronze industry had to acquire its raw materials from somewhere, but the nature of this acquisition (trade, tribute, direct extraction, etc.) will only be understood when we have more robust evidence. Likewise, even if the extent of political control from the Erlitou center could be deduced simply from the ceramic tradition, the issues of the nature and mechanisms of political control would still need to be addressed (and on aregional basis).19 Much more archaeological work on regional centers and regional surveys needs to 2.13. Jinzhong-tradition ceramics (Taigu Bai- be done for the Erlitou period before yan) (after ZSKY 2003:571, ﬁg. 8-35). these issues can begin to be adequately addressed. and the Lower Xiajiadian traditions (ZSKY 2003)—in other words, the Beyond the Erlitou Ceramic archaeological traditions immediately Tradition adjacent to it. Of the burials known from this period, very few have grave The North goods and, to my knowledge, nothing The Jinzhong Tradition has been published on settlements. The Jinzhong20 tradition is distributed to the north of the Dongxiafeng Erlitou The Zhukaigou Tradition variant in central Shanxi, traditionally Distributed in the Ordos region of known as the Jinzhong area (Figure Inner Mongolia, the best-known site 2.1; ZSKY 2003). Developing out of of this tradition is Zhukaigou (see local Longshan ceramic traditions, Figure 2.1; Neimenggu Zizhiqu Wenwu this ceramic tradition (Figure 2.13) has Kaogu Yanjiusuo, Erduosi Bowuguan been called Guangshe culture, Baiyan [hereafter NZWKY,EB] 2000; ZSKY culture, Dongtaibao culture, Yicun 2003).Houses havebeenexcavatedfrom type, as well as Erlitou culture (Jinzhong many sites of this tradition and fall into type, Dongxiafeng type, Dongtaibao semisubterranean and surface types type, etc.). The major sites for the with the majority semisubterranean. Erlitou period include Taigu Baiyan Most of the semisubterranean houses phase four and Taiqu layer three. This are rectangular or square, and most are tradition is said to show inﬂuence from comprised of asingle room, although the Erlitou metropolitan tradition, as some have two or even three rooms. well as interaction with the Zhukaigou The Zhukaigou tradition (Figure

41 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

2.14. Zhukaigou-tradition ceramics (after ZSKY 2003: 580-81, ﬁg. 8-39).

2.14) is said to have developed out of Northeast local Longshan traditions while also showing influences from the Erlitou, The Luwangfen–Songyao Jinzhong, and Qijia traditions (ZSKY Tradition 2003). Its burial practices are very To the northeast of the Erlitou variant similar to those of the Qijia tradition and contiguous with it, is the Luwang- (NZWKY,EB2000; ZSKY 2003), and in fen–Songyao tradition22 (see Figure the Erlitou period, its bronze artifacts 2.1; hereafter LWF-SY). With the Er- also show similarities with Qijia (Figure litou tradition, Erlitou variant to the 2.15). There is evidence of mortuary southwest, Dongxiafeng variant to the differentiation both in terms of tomb west, Xiaqiyuan tradition to the north, size and contents, suggesting at least and Yueshi traditions to the east, the some degree of social hierarchy.21 Luwangfen-Songyao tradition (Fig- ure 2.16), not surprisingly, shows influences from all of these traditions, although it is said to have its unique characteristics as well (ZSKY 2003). Deep guan-pots are the main cooking vessels, while li-tripods are the second most common vessel type. Unlike at Erlitou or Xiaqiyuan, stone knives are more common than stone sickles. No detailed site or settlement data is available, and there is no information on craft production. Only one bronze artifact has been found associated with 2.15. Zhukaigou bronze artifacts (after ZSKY remains of this tradition: that is, a“rec- 2003:578, fig. 8-38). tangular” knife similar to those found

42 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2

2.16. The Luwangfen-Songyao tradition and “influences” (ZSKY 2003:163, fig. 3-6). in other areas of northern Henan and Hebei (Figure 2.1), the Xiaqiyuan southern Shanxi (ZSKY 2003:161). In tradition (Figure 2.17) is divided into short, there is not much evidence avail- three variants: the Zhanghe variant, able about this tradition apart from for- the Yuegezhuang variant, and the mal ceramic typology. Lutaigang variant (ZSKY 2003:152–156). Although the origins of this tradition The Xiaqiyuan Tradition are debated, most scholars believe its Distributed along astrip running most important influences are the local from eastern Henan to northern Longshan Hougang II tradition and the

43 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

2.17. Xiaqiyuan-tradition variants (after ZSKY 2003:153, ﬁg. 3-4).

Central Shanxi Longshan and Erlitou The Zhanghe variant is located period traditions. This tradition is in northern Henan and southern divided into four periods corresponding Hebei, and is said to be the center of roughly to Erlitou I–IV or slightly later.23 the Xiaqiyuan tradition in terms of geography, as well as in unique features

44 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2

(ZSKY 2003). Houses of this variant types (such as plain bowl-shaped dou take the form of either subterranean and zun-shaped vessels), which derive (“cave”-houses or “kiln”-houses) or fromYueshiinfluence.Consideringthat semisubterranean24 dwellings. In early that there are contemporaneous Yueshi periods, the main cooking vessels sites in the immediate area (see Figure were deep guan-pots, which in later 2.1), this is perhaps not surprising. periods were replaced by li-tripods. Egg-shaped weng-urns are also found In summary, with almost no in Zhanghe-variant assemblages—a information concerning anything but vessel type shared with traditions to formal ceramic typology and afew its west (such as the Zhukaigou and other aspects of material culture, it is Jinzhong traditions, as well as the very difficult to say what the Xiaqiyuan Dongxiafeng Erlitou variant). The only tradition represents socially, culturally, bronze artifacts discovered thus far are politically, or economically.Its variants arrowheads and knives, while the most (like most of the “cultures” described in common form of bone artifact is the bi- ZSKY 2003) tend to show influences of ladle, atrait shared with the Luwangfen- traditions nearest them, and although Songyao tradition to the south. the Xiaqiyuan tradition is said to have more salient common traits than The Yuegezhuangvariant is located in differences between variants, some of Northern Hebei Province. No houses or this may be due to the presuppositions burials have been found. Li-tripods and behind the practice of ceramic typology yan-steamers are the most common in China, including the tendency to see cooking vessels, and some of the ceramic traditions in rigid, bounded vessel forms are said to show obvious terms, and to identify these with similarities with northern ceramic different ethnic groups (see Cohen 2001 traditiontypes,suchasDatuotou(which foracritique).InthecaseoftheXiaqiyuan is immediately to the north). Bronze tradition, this is particularly acute, as artifacts include knives, earrings, this tradition is generally thought to arrowheads, and hairpins. Of these, belong to the predynastic Shang (Zhou the knives, earrings, and arrowheads 1980). Since the Xiaqiyuan tradition all show northern complex affiliations is said to have Erlitou influences, and (ZSKY 2003). since Erligang is said to combine both Xiaqiyuan and Erlitou traditions (not to The Lutaigang variant is located in mention the controversies surrounding eastern Henan and, so far, has only one the fuzzy boundary between early verified site.25 The one house that was Erligang and late Erlitou traditions), found at the site dating to this period the appellation of “pre-dynastic Shang” was around surface dwelling with to the Xiaqiyuan tradition seems “mud” walls. The main cooking vessels to be based only on rather subtle were deep guan-pots and li-tripods, with ceramic typological distinctions and the former being far more common. semihistorical geography.Moreover, Lutaigang is said to have several ceramic there is no reason to suppose that this

45 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

2.18. Lower Xiajiadian ceramics (after ZSKY 2003:601, ﬁg. 8-47).

formal ceramic typology delimits a settlements on hilltops and ridges common ethnic or political group even overlooking rivers (ZSKY 2003).27 These if the pre-Shang dynasts did originate settlements frequently made use of in an area where this ceramic tradition natural defenses in addition to walls, predominated. suggesting, perhaps, aprevalence of inter-community conﬂict in this area. The LowerXiajiadian Tradition While the authors of ZSKY (2003) feel Located in present-day Liaoning that the Lower Xiajiadian–tradition Province and Inner Mongolia, the Lower societies arose through amixing of Xiajiadian is best known from sites, Central Plains and Xiaoheyan cultures, such as Chifeng Xiajiadian (Zhongguo others believe the Xiajiadian tradition Shehuikexueyuan Kaogu Yanjiusuo (Figure 2.18) directly developed out Neimenggu Gongzuodui [hereafter of the indigenous Xiaoheyan culture ZSKYNG]1961,1974)andAohanbanner (which in turn evolved out of abranch Dadianzi (ZSKY 1996) (see Figure 2.1). of Hongshan culture). Settlements are Lower Xiajiadian-tradition sites feature said to show three levels of hierarchy, some unique characteristics, such and Xiajiadian social structure is as densely clustered,26 stonewalled hierarchical (ZSKY 2003; Shelach

46 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2

1994, 1996, 1999). Grave goods include ceramics, jade, bone, stone, and bronze artifacts, as well as cowrie shells (ZSKY 1996; Flad 2002). At the Dadianzi cemetery, 13 of the approximately 600 adult graves contained ceramic jue, he, or gui wine vessels similar totypes found at Erlitou (Figure 2.19). This has been interpreted (ZSKY 2003; Thorp 2006) as evidence for direct and powerful inﬂuence from Erlitou. However, while the presence of vessels typical of elite Erlitou mortuary ritual in 2percent of the burials is interesting, the overall picture of mortuary practice, bronze industry, ceramic tradition, and built environment is radically different from that of Erlitou, suggesting instead an independent center of social complexity with relatively minor and indirect interaction with Erlitou traditions (Shelach 1999).28

The Datuotou Tradition/Variant Located across northern Henan and 2.19. Ceramics from Dadianzi Tomb 726 (after southwest Liaoning Province (see ZSKY 2003:599, fig. 8-46). Figure 2.1), this tradition (Figure 2.20) shows many similarities with Lower trumpet-shaped earrings, and bronze Xiajiadian such that many scholars feel rings. it is aregional variant of the Xiajiadian tradition. The authors of ZSKY (2003), The East while noting aclose relationship with the Lower Xiajiadian tradition, feel The Yueshi Tradition there are also significant differences To the east of the Erlitou and and that Datuotou should be treated Xiaqiyuan traditions and located as adifferent “culture.” Nevertheless, in Shandong, eastern Henan, and it shares atradition of polychrome in northern Jiangsu Province is the painted pottery, the location of sites Yueshi tradition. The Yueshi tradition on ridges and hilltops, and similar is divided into five regional variants assemblages of burial ceramics. (ZSKY 2003):29 Zhaogezhuang (Figure Datuotou sites have also produced 2.21a), Haojiazhuang (Figure 2.21b), bronze artifacts of northern complex Yinjiacheng, Anqiugudui (Figure type, including ring-pommelled knives, 2.22c), and Wanbei (Figure 2.22d) and

47 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

2.20. Datuotou-tradition ceramics (after ZSKY 2003:607, 8-48).

is generally thought to have developed Shandong Gongzuodui [hereafter ZSKY, out of Shandong Longshan traditions SG] 1985). The houses at Dakou were in interaction with neighboring semisubterranean and rectangular with traditions.30 Yueshi-tradition material rounded corners and white ash spread culture is said to show very obvious and over the floor.Mosttools and weapons unique characteristics, such as using were made of stone and bone, and few planks and bundled rods in making artifacts of bronze have been found. rammed-earth structures, aprevalence The graves found at Dakou were small, of plain or polished ceramics, jue-hoes, rectangular, and uniformly lacking and half-moon-shaped stone knives in grave furniture or goods. The main (ZSKY 2003). cooking vessels were yan-steamers and deep guan-pots. The use of mica or talc The Zhaogezhuang variant is for temper is afeature of this variant. distributed across the east end of the Shandong peninsula, on islands off the The Haojiazhuang variant is coast, and in the southern part of the distributed north of theTaiyiMountains. Liaodong peninsula (see Figure 2.1). The houses are mainly rectangular Very little data on tombs or houses is surfacedwellingswithfiredfloors.Several available for this tradition, with almost sites of this variant are surrounded all such information coming from by rammed-earth walls,31 andsome, the Dakou site on Tuoji island, off the such as Chengziya, are fairly large(17 north coast of Shandong (Zhongguo ha;ZSKY2003).Afew of the rough- Shehuikexueyuan Kaogu Yanjiusuo, temperedvesselsaretemperedwithmica

48 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2

2.21. Yueshi-tradition variant ceramics (after ZSKY 2003:450, ﬁg. 8-2, A-B).

49 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

2.22. Yueshi-tradition variant ceramics (after ZSKY 2003:451, ﬁg. 8-2, C-D).

or talc,and blackceramicsare relatively are common. Multiedged stonehoes frequent.Mostvessels are undecorated, are common, as arehalf-moon-shaped and yan-steamers anddeep guan-pots stoneknivesand single-edgedadzes

50 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2

(ZSKY 2003). variant. This Yueshi variant issaid to TheYinjiachengvariant is distributed show Erlitou-, Xiaqiyuan-,and Doujitai- from the southern flanksofthe Taiyi tradition influences,perhaps not Mountains to thenorthernedgeof surprisingly, given thatthese traditions JiangsuProvince(Figure 2.1).The houses arecontiguous. discovered thus farare rectangular, above ground, and generallyhave TheWanbei variant isdistributed in rammed-earth floors.Somehouseswere Jiangsu Province northofthe Yangzi built on rammed-earth foundations. (Figure 2.1).The sitewith the most Bronze arrowheads, knives, chisels, and abundant materialisthe Gaoyou awlsoccurrelativelyfrequently,although Zhoubeisite(southernmost on map). the majority of tools are madeofstone, Stone tools include axes, knives, adzes, and the most common weapon types are arrowheads, and pendant-shaped stoneaxesand bone arrowheads. The artifacts.Coarse-tempered ceramicsare mostcommonstonetoolsarehalf-moon- more commonthanfine tempered,and shapedknives, rectangular single-edged somestampedstoneware occurs. The spades, andhoes. Ahigh frequencyof majority of the ceramics are undecorated shell tools is afeature of this variant. gray ware.The most common cooking Most of the ceramics are undecorated. vessel types are ding-cauldrons, yan- Ding-cauldrons, yan-steamers,and deep steamers,and li-tripods. Beyond guan-potsare common. ceramics andtooltypes,however, very little information is available forthis The Anqiugudui variant is distributed variant. across western Shandong and northeasternHenan. The housesofthis The Southeast variantare mostly surfacedwellings, sometimes withfragments of fired clay The Doujitai Tradition used to strengthen their foundations. This tradition is locatedbetween the Only afew small bronze weapons and Huai andYangzi riversinAnhui Province tools have beenfound,and most tools (Figure 2.1). Mostofthe ceramics of this are made of shell. The most common tradition arecoarse-tempered black- tools ofthisvariant are single-edged gray ware or coarse-tempered brown stoneshovels, half-moon-shapedstone ware. Mostvessels are plain,but basket- knives, shell shovels, and shell knives. marked andcord-marked decoration Fine-paste ceramicsare said tobemore is alsorelativelycommon. The main common than rough-paste ceramics. cookingvessels of this traditionare Mostpottery is undecorated,but rope- ding-cauldrons and deep guan-pots. marked pottery becomes increasingly Althoughthis tradition issaidtoshow common,eventuallyaccountingfor mostlyindigenous ceramic-tradition around one-quarter of allceramics bythe influences,Doujitaialso shows endofthe Erlitou period (ZSKY 2003). influences fromall of its contemporary Yan-steamers, ding-cauldrons, anddeep neighbors, includingErlitou,Yueshi, guan arecommonvessel typesofthis and lower Dianjiangtai (ZSKY 2003).

51 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

2.23. Maqiao-tradition ceramics (after ZSKY 2003:465, ﬁg. 8-5).

The Lower Dianjiangtai predates the Hushu tradition in the Tradition area, and probably does not last un- Located in the Nanjing and Zhenjiang til the end of the Erlitou period.32 No regions of Jiangsu and Anhui Provinces bronze artifacts have been found in the respectively (Figure 2.1), this tradition few sites of this tradition that have been

52 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2 excavated. The ceramics are said to orange-red or red-brown and are mainly show both Liangzhu- and Longshan-tra- cooking vessels, such as ding-cauldrons dition-derived characteristics. Accord- and yan-steamers. The main cooking ing to the ZSKY (2003), 38 percent of the vessels for this tradition are ding- ceramics are coarse-tempered red ware, cauldrons, yan-steamers, and fu-pots. 24 percent coarse-tempered gray ware, Fine-tempered black dou and gui33 are and 23 percent ﬁne-tempered black among the most ﬁnely made ceramics ware. Ding-shaped yan-steamers are the in the Maqiao repertoire. Stone artifacts main cooking vessels, with ding-caul- include fu-axes, adzes, chisels, knives, drons also appearing. However, unlike yue-axes, dagger-axes, spears, and the following Hushu tradition, there are arrowheads. Bronze artifacts generally no li-tripods. There is also no stamped seem to consist of small tools, such as stonewear or protoporcelain, although chisels and knives, but larger bronze there are some thin-walled black pol- weapons resembling Maqiao stone ished ceramics. No information con- weapon types have also been found cerning settlement structure or hierar- in the general Maqiao tradition area, chy, mortuary customs, or residences is though not securely associated with available for this tradition. archaeological deposits (ZSKY 2003).

The Maqiao Tradition The South The Maqiao tradition (Figure 2.23) is located in the general area around The Middle Reachesofthe Yangzi Lake Tai(seeFigure 2.1) and is said to In present-day Hubei Province, sites have evolved out of Liangzhu culture such as Panlongcheng, Jingnansi, under early Hushu, Yueshi, and Erlitou and Baimiaozhu (Figure 2.1) are all cultural influences (ZSKY 2003). The said to have Erlitou or Late Erlitou– type site of Maqiao is arelatively large Early Erligang–type ceramics (ZSKY (10 ha) residential site, but it and all 2003). Nevertheless, important Erlitou other known sites of this tradition have vessel types, such as deep or round- been badly disturbed by later activities, bellied guan-pots, are missing from and little can be said about settlement the assemblages, and other types show layout or architecture. marked local characteristics. Thus, although there seems to be some kind of Relatively few burials are known, and interaction between the Middle Yangzi these are generally pit burials. There and Central Plains regions during this are some cases in which remains show time, it is far from clear what social, evidence of having been bound hand political, and cultural practices might and foot before interment, as well as lie behind these ceramic “influences.”34 some secondary burial of children’s remains. Three-quarters of the ceramics The Southwest are fine-tempered vessels with colors ranging from orange-red to black, The Sanxingdui Tradition while the remainder are coarse-pasted, The Sanxingdui tradition was widely

53 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

2.24. Sanxingdui Erlitou–period ceramics (after ZSKY 2003:502, 503, ﬁg. 8-17).

distributed in the Sichuan basin (Figure that continue to surround the absolute 2.1),developingoutoftheearlierBaodun dating of Sanxingdui. At present, it tradition (Flad and Chen 2013; Xu 2008; appears that the Sanxingdui tradition ZSKY 2003). The Sanxingdui tradition and the huge walled site of Sanxingdui (Figure 2.24) is divided into six phases at least partially overlap with Erlitou. by the ZSKY (2003) and into four by Sun Moreover, during the late Baodun (2000) and Xu (2008).According to The period (the ﬁrst centuries of the second Xiaand Shang,phase I, andpossibly millennium BCE), either predating or part of phase II of the Sanxingdui site, contemporaneous with the early Erlitou date to the “Late Erlitou” period (ZSKY period, the Chengdu plain was dotted 2003:506). On the other hand, it is with anumber of walled sites, some of currently believed (Flad and Chen 2013; them, like Baodun, of prodigious size Xu 2008; Sun 2000) that the ﬁrst phase (220 ha; Flad and Chen 2013). Thus, of the Sanxingdui site belongs to the the rise of complex societies and urban Baodun period. Flad and Chen (2013) centers in the Sichuan basin seems to provide dates of 2700–1700 BCE for the be yet another example of local process Baodun culture and 1700–1150 BCE for within larger webs of interaction (see the Sanxingdui culture, while Xu (2008) below) and to be rooted in local third estimates Sanxingdui phase Itobefrom millennium BCE developments, not, the early third millennium to 2000 BCE as some have argued (see ZSKY 2003), and the Sanxingdui tradition to extend merely the result of early Central Plains over the entire second millennium Bronze Age migrations or conquest. BCE. Interestingly, both estimates are based on published radiocarbon dates Forthe early part of the Sanxingdui and demonstrate the uncertainties tradition, the major site appears to be

54 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2

2.25. Sanxingdui site (after ZSKY 2003:494, ﬁg. 8-15).

Sanxingdui. Surveys have suggested framed surface dwellings, 10–25 m2 in that the Sanxindui covers 12 km2, surface area (ZSKY 2003; Xu 2001a), with walls surrounding an area of 3.6 continuing Baodun building traditions km2 (ZSKY 2003:494; Xu 2008). It is (Flad and Chen 2013). Sanxingdui’s unclear at present, however, what part walls appear to continue atradition of the Sanxingdui site (Figure 2.25) of building begun in the preceding dates to which phase, and no precise Baodun period, one that was, perhaps, chronology of most of the other sites borrowed from, or inﬂuenced by, the of this tradition currently exists35 (see Middle Yangzi Shijiahe tradition (Xu Xu 2008). While some of the walls are 2001a; Faulkenhausen 2006). said to date from the early period of this site (phase II), others apparently Sanxingdui is famous for its “artifact date from the Anyang/Early Western pits,”widelythoughttoberitualremains Zhou period.36 At present, the size of (Xu2001a; So 2001). The most famous Sanxingdui during the various phases are the Anyang or Xiaoshuangqiao- of this tradition is unknown. The Huanbei–period artifact pits 1and 2, houses of the Sanxingdui tradition are with their giant bronze masks, statues, generally rectangular or square, wood- and other artifact types unseen in the

55 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

as the presence and form of he-and gu-vessels (ZSKY 2003). Nevertheless, the Sanxingdui tradition favors a tall, slender he-type unlike the short, squat Erlitou he.38 Interestingly, these are both vessel types associated with drinking and associated with higher- status burial contexts at Erlitou.39 Indeed, throughout the period of the Sanxingdui tradition, Central Plains inﬂuences generally take the form of ritual or status artifacts rather than those of daily use. Sanxingdui ceramics also show similarities to the late third-, early second -millennium BCE Shijiahe tradition of the Middle-Yangzi (Xu2008). And it is possible that the 2.26. Jade blades from Sanxingdui (after ZSKY Sanxingdui he-and gu-prototypes came 2003:500, ﬁg. 8-16). from this direction.

Central Plains (Xu2001b, 2006, 2008; Sanxindui ceramics were generally Ge and Linduff 1990; Bagley 1999; handmade in the early periods with Rawson 1996; Falkenhausen 2006; wheel-made ceramics becoming more etc.). Five other artifact pits have been common in later periods. Most of the found at Sanxingdui, and three more early ceramics are undecorated and at other sites, all apparently dating to coarse-tempered. Characteristic vessel the late phases of this tradition.37 Other types of this tradition include the above- than the two famous Anyang-period mentioned slender he,tall-stemmed pits, which contained alarge variety of dou, yan-shaped vessels, ping-vases, artifacts, including stone, bone, bronze, and bird-headed ladles. Flat-bottomed ivory, and ceramics, the other pits guan-pots and flat-bottomed pan-dishes appear to have contained mostly jade are also common vessel types.40 or stone artifacts (So 2001; ZSKY 2003). If these are, in fact, ritual remains, they Although there is not much information appear to be of atype different from concerning stone, bone, or other the sacrificial pits found in the Central artifact types in the early period, two Plains, suggesting different practices turquoise inlaid bronze plaques were (Bagley 1999; Xu 2001b; So 2001; found in alate-phase artifact pit in Falkenhausen 2003). 1988 (Sichuansheng Wenwu Kaogu Yanjiusuo [hereafter SWKY]1999) that Sanxingdui-tradition ceramics (Figure are similar to those found at Erlitou and 2.24) are said to show Erlitou-ceramic- some Qijia sites. Jade and stone zhang- tradition (Figure 2.7) influences, such blades (Figure 2.26) are also prevalent

56 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2

2.27. Qilidun-variant ceramics (after ZSKY 2003:542, fig. 8-27). in Anyang-period Sanxingdui artifact In summary, while the early stages of pits despite their relative scarcity in the the Sanxingdui tradition are somewhat Central Plains after the Erlitou period. murky, the emerging picture seems to Given these facts, it seems likely that be one of alocal development of social Erlitou-period Sanxingdui had similar complexity resulting in an urban site artifacts. Indeed, the forked zhang- thatrivaledErlitouinsizebyperhapsthe blades that appear so prominently eighteenth century BCE. Nevertheless, at Sanxingdui were also found in the Shijiahe, Qijia, and Erlitou (and numbers in the late third-, early second- perhaps Shimao) influences suggested millennium BCE site of Shimao, by the material culture discovered at Shenmu, in northern Shaanxi (So 2001; Sanxindui testify to its (and earlier Wang and Sun 2011)—a site that recent Baodun centers) inclusion in, as yet survey has revealed to be enormous unclear, webs of interaction between (400+ ha) and enclosed by astone wall large centers and culture areas. (Xinhua 2012). Lithic manufacture techniques and the styles of some jade The West and stone artifacts suggest possible Qijia (see below) influences as well (So Although there are Erlitou tradition-, 2001). At the same time, some authors Erlitou-variant sites on the eastern (Falkenhausen 2003; Xu 2008) argue edge of Shaanxi Province (Figure 2.1), on stylistic grounds that Sanxingdui the situation in the rest of Shaanxi is jade and bronze designs share many unclear for this period (ZSKY 2003). similarities with those found on Shijiahe artifacts. Not surprisingly, the The Qijia Tradition directions of interaction accord well The Qijia tradition is distributedtothe with the routes into the Sichuan basin: west of Erlitou, extending from western from the northwest of the Chengdu Shaanxi Provincethrough Gansu plain, the Hanzhong region, and along Provincetoeastern Qinghai(Figure2.1). the Yangzi through the Three Gorges This tradition is divided into several (Xu2008; Falkenhausen 2003). regional variants, and sites of this tradition are located in agreat variety

57 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

2.28. Qijia bronze artifacts (after Li 2005:241, ﬁg. 1).

of environments(ZSKY 2003). The Qijia on agriculture, although assessment has traditionlastedfrom ca. 2200to1500 been determinedlargelyonthe analysis BCE. This means thatonly the latter part of tooltypes (sickles, mortars and of the Qijia tradition overlappedwith the pestles,etc.).Animalhusbandrywasalso Erlitou period. The ZSKY (2003) divides part of the subsistence economy, and the tradition into ﬁve variants, only two pig, sheep, horse, donkey,cattle, and of which are showninthe western edge dogremainshaveallbeenfound,though of Figure 2.1,41 namely the Shizhaocun pig remains are the most common. Qijia and Qilidun variants.However, of potterywas mostlyhand-built, although the two, only Qilidun (Figure2.27), molds were used in somecases, as well beinglater, mayhave datedtoErlitou as the slow wheel.Characteristic vessel times. Settlements of thistradition are forms of the Qilidunvariant include usually foundonriver terraces close high-necked, double-eared guan-pots to the water.Settlementsize varies and double large-eared guan-pots. The from 12.5 ha to less than 1ha, with an jade- and stone-artifactindustry was average of 5–7 ha. The structure of Qijia also well developed, and unadorned, settlementsisalsohighly variable. Most smooth, polished disks and tubes houses are semisubterranean or surface of various sizes are characteristic. dwellings and have ahard and smooth Scapulamancy was widespread, usually white ash layeronthe living surface.42 using sheep scapulae, althoughless Subsistence appears to have been based commonly cattle and deer were also

58 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2 used. Burials are generally grouped and Conclusion orderedinrows,and Qijiatombs vary in sizeand richness. Most tombs consistof The Erlitou period is generally rectangular shafts, and coffins made of a considered to be the beginning of single log are common, althoughcoffins the Chinese Bronze Age. Forsome built of planks appear as well. Grave scholars it is also the period of the first goods are most commonlyceramics. Chinese state, centered at the large There are both single and paired burials, (300 ha) Erlitou site, and, perhaps not although the sex combinations vary, and unrelated,formany(especiallyChinese) some of these are secondary burials. scholars, Erlitou can unambiguously Nevertheless,some of the multiple be equated with the first dynasty of burialsare interpreted as evidence of traditional Chinese records: the Xia. death attendants. Archaeologically, the Erlitou period is characterized by numerous overlapping One of the most interesting aspects of regional ceramic, stone, metallurgical, the Qijia tradition is its relatively and other traditions. advanced bronze metallurgy and its possible role in cultural and The Erlitou period might be called the technological exchange between beginning of ahorizon insofar as it the Central Plainsand thebronze marks the beginning of the expansion traditions of southern Siberia and of an interconnected complex of CentralAsia (Fitzgerald-Huber 1995, traditions, practices, and styles of 2003; Bagley 1999; Li 2005).43 The Qijia widespread dispersal (Willey and bronzeartifact assemblage (Figure Phillips 1958). The problem, however, 2.28) consists of weapons, tools, and with dividing up Chinese prehistory ornaments,including ring-pommelled into horizons and intermediate periods knives, bronze mirrors, rings, and (modeled on Peruvian archaeology) is earrings showing affinity with Central the fact that in broad material cultural Asian and southern Siberian traditions. terms, the trend from the Longshan to In terms of affinities and influences, at least the Qin is one of increasingly the Zhukaigou tradition is saidto intensified interaction over larger share some common features with the and larger areas, with much evidence Qijia in terms of ceramics, metallurgy, of change but none of collapse. and burial traditions(ZSKY2003). In Upon making finer culture-historical addition,Erlitou-type artifacts, such as a divisions into particular traditions, the turquoise-inlaid animal face plaque and Erlitou period, though evidence is still closed-mouth he-vessels (Li 2005) have quite fragmentary, reveals alandscape been found in Qijia sites,although given of multiple overlapping networks of the earlier date of the Qijia tradition, we material cultural interaction. shouldtake seriously the possibility that turquoise inlaying was originally aQijia The elite crafts of this period can be tradition (see also Zhang 2012). divided into some traditions that seem to have been innovated at anumber

59 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

of sites and others, such as the jade practices that developed out of earlier industry, which were inherited from antecedents. earlier times and further developed (You 2002). In general, elite traditions The presentstate of archaeological that appear to have been innovated information makesitextremely difficult at Erlitou, such as bronze-vessel to do more than speculateaboutthe casting and monumental courtyard- socialpractices, attitudes, and networks style architecture show adistribution of exchange thatproduced the everyday limited to the site of Erlitou itself until ceramicswhichare thebasis for defining the end of the period, when they spread ceramic traditions.Nevertheless, on to other elite centers like Yanshi and presentevidence, itappearsthatkilns Zhengzhou. On the other hand, either were common features of residential bronze plaques were exchanged, or sites, kiln sizeswere small,and much the tradition of their manufacture of theeverydayceramics were probably was shared, between Qijia sites in locallymade.Ifthis is indeed the case, Gansu (Li 2005) and Sanxingdui in then large-scale distributionnetworks Sichuan. The Erlitou jade tradition, canbediscounted as amajorfactor moreover, emerged from preceding in producing similarities in everyday Longshan traditions, and Erlitou jades ceramics overlarge areas, except, correspondingly show more affinities to perhaps, in thecaseofelite ceramics jades found in other places (especially such as protoporcelains. In otherwords, such jade blades as those found at the Erlitou expansion(considered earlier Shimao and in even greater in ceramic terms) ought to have had numbers later in Sichuan).44 Ceramic somethingtodowiththe movement drinking vessels associated with higher- of people.Atthe same time, variants stratum Erlitou burials also have a of theErlitou traditionretain local relatively wide distribution yet also have characteristics,and similarity to Erlitou fairly widespread Neolithic antecedents tends to fall away with distance from (Underhill 2002). The presence of the center.Tothe extent this iscorrect, Erlitou-type drinking vessels in Gansu, itsuggests, rather thanpopulation Liaoning, and Sichuan suggests awide- replacement, an increased interaction ranging elite material cultural horizon within asphere centered on Erlitou (and perhaps elite interaction), but not, decreasinginintensity withdistance. despite some claims in the literature, Thesocialprocessesthatproduced amonocentered one and, even less, these effectsmay have included apolitically centralized one. Indeed, marriages, economic exchange, military given the widespread third-millennium alliance, conquest, colonization, or distribution of drinking vessels and combinations of the above. At the same their varied assemblages at different time, ceramic traditions andtheir sites, amore obvious conclusion variants tendtoshow associations with might be that Erlitou was one (major) their neighbors inall directions,and, in player in abroad complex of more or general, ceramictraditionboundaries less interrelated drinking and ritual appear to be soft,45 suggestingthat the

60 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2 socialinteractionmirrored in ceramic scale of bronze production at Erlitou traditions wasprobably local in scale and in comparison with that in evidence multidirectional. Althoughdatafrom at Anyang at the end of the second burials, houses, and lithic industries are millennium, it seems likely that even fragmentary at best, regional traditions Erlitou’s bronze industry could have appear to holdsway46—notalways been supplied through low-intensity, coextensive with ceramictraditions— noncentralized exchange, from elite and seem to have evenless to do with gifting or tribute, to trade or booty.The putative political entities. An even larger samecouldprobablybesaidofcinnabar, material cultural distribution sphere lacquer, turquoise, and cowrie shells, can be seenwiththe NorthernComplex although the former two are difﬁcult bronze tradition, its mirrors,knives, to quantify.Asfor other materials for axes, and earrings showingupfrom elite material cultural production Liaoning to Gansu,and from Zhukaigou that have not been preserved, such as to Erlitou. Thisphenomenon bespeaks silks, rare woods, costly foods, and so even wider-ranging contactsthanany on, we have little to go on either way. other materialculturaltradition, but Even less is known about the nonelite its extensive diffusion in avariety of economy, although based on work on cultural contextsisevenless likely to be stone tool production at Huizui, Chen explained by asinglepolitical master- (2005) suggests that the production and narrative,suggesting ratheramyriadof distribution of subsistence goods were interactionsover centuries. “more likely to have been organized in decentralized patterns” (p. 9) than as From the perspective of social and part of “centralized state economy.” economic exchange networks and their relationship to political organizations The political organization of the polity and the circulation of social power, the centered at Erlitou is unclear.Erlitou sat data presently available for Erlitou is far atop asettlement hierarchy of unknown from sufﬁcient to do much more than size,48 and for perhaps acentury, maybe speculate. Although small-scale bronze less (1700–1600 BCE), it was the largest metallurgy was likely taking place in urban center in East Asia for which we many places in northern and western have evidence (Shimao may overlap with China, including the Central Plains the early phases of Erlitou; Sanxingui, area, and all of these sites would have with at least the late phases; and Yanshi had to procure copper, lead, and tin and Zhengzhou were probably both from somewhere, we have neither data major centers by the Erlitou phase IV). on contemporaneous mining sites nor While the Erlitou ceramic tradition was evidence for the routes by which metals widespread, the mechanisms of this reached the various large- and small- expansion are probably only indirectly scale workshops.47 Muchless do we have related to political activity (if pots the social mechanisms of exchange or don’t equal people, they are even less how they were politically structured. representative of conquering armies or Indeed, given the relatively small “state” administrators). The degree of

61 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

centralization, mechanisms of political culture different from that at Erlitou50 control, and social organization can even while apparently in interaction. only be guessed at or extrapolated Inner Mongolia and Liaoning in the through comparison with Zhengzhou far northwest likewise were home and Anyang. This comparison can to very different societies from that justiﬁably be made insofar (and only found at Erlitou, displaying different insofar)asmanyofErlitou’selitecultural architectural techniques, settlement forms appear to be ancestral to those structures, cultural practices, and elite found at Zhengzhou and Anyang, from assemblages. North Central China architecture, to symbols of status and in the Erlitou period, then, far from implements of ritual.49 Nevertheless, being monocentric and culturally as will be discussed in more detail later homogeneous, was adiverse landscape on, there are considerable qualitative of peoples organized in amultitude and quantitative differences between of ways, with avariety of economies, Erlitou, Zhengzhou, and Anyang, producing and using different material making their comparison—and cultures and technologies, and if especially the derivation of the lesser historical China is any guide, likely known from the better known—a more speaking amultiplicityoflanguages, all complex problem than most have within multiple, overlapping, spheres credited. of inﬂuence, innovationand exchange.

Seen in regional context, if Erlitou has its civilizational sphere (Baines and Yoffee 1998; Allan 2007), it was not alone. In Sichuan near the end of this period, constellations of walled sites were undergoing astill poorly understood process of consolidation that culminated in the huge walled center at Sanxingdui, with its distinctive material culture and architectural styles. In the west and northwest quite differentsocietieswithmaterialcultural assemblages showing steppe and southern Siberian inﬂuences existed in theQijia-andZhukaigou-traditionareas of Gansu and Inner Mongolia. To the northeast, the still poorly understood traditions of Luwangfen-songyao and Xiaqiyuan were located in what would later be core areas of the polity at Anyang, while in the east, the Yueshi traditions continued to display material

62 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2

Endnotes phaseIIandIIIatErlitouhaslongbeenatthecenter 1These dates are based on both the of controversy.While most Chinese scholars now radiocarbon dates published in Xia Shang Zhou seem to be of the opinion that the Erlitou tradition Duandai Gongcheng Zhuanjiazu [hereafter is aceramic manifestation of the Xia dynasty with XSZDGZ] (2000) and ZSKY (2003). Previous work Erlitou as its capital (ZSKY 2003), there are still had suggested that Erlitou dated from 1900–1500 holdouts claiming that phases III and IV should be BCE for atotal of 400 years with each phase being considered Shang (Yin 1986). Given that this issue about 100 years long (Qiu, Cai, Xian, and Bo 1983; is framed by traditional historical concerns and ZSKY 1999). Recent radiocarbon work using based almost completely on ceramic typological “wiggle-matching” techniques have dated the site arguments, I will not attempt to deal with it here between 1750 and 1520 BCE, reducing Erlitou except to note that elite cultural remains from site occupation to amere 200 years, the last 50 or Erlitou and Erligang can basically be considered as so years claimed to be under Shang occupation developments of asingle metropolitan tradition, (Qiu, Cai and Zhang 2006). If this is really the case and it is surely significant that “Xia” and “Shang” then the “Erlitou expansion” during Erlitou II and material remains are so similar that debate still III was both rapid and short-lived (see discussion swirls around the question of which “dynasty” or below). “culture” aparticular artifact or stratigraphic level 2Insouthern Shanxi province, Taosi belongs to. and Zhoujiazhuang; in northern Shaanxi, the 5However, according to Zhang (2012) newly discovered Shimao; in southern Shandong, none of the supposedly wheel-thrown ceramics at Liangchengzhen and Yaowangcheng; in Zhejiang, the Erlitou station appeared to have been made on the Mojiaoshan Liangzhu site: all range between awheel. 200 and 500 ha. Given that most of these sites 6Infact, Xinzhai is supposed to be later have been discovered only in the last ten years— than Wangwan III and is seen as an intermediate the product of intensive, systematic survey—it is step between Wangwan III and Erlitou. In arecent unlikely that these discoveries will be the last. dissertation, Zhang (2012) argues that, in fact, 3These are Erlitou’s “medium-sized” Xinzhai and Erlitou are contemporaneous and (ZSKY 1999; ZSKY 2003; Liu and Chen 2003; etc.) belong to different traditions. tombs. Their “medium-size” appellation is based 7Nevertheless, as we will see below, there is on acomparison with large third-millennium BCE agreat rangeinboth the intensity of influence and tombs, such as the largest found at the Taosi site, variety of interpretations concerning what these as well as the royal tombs found at Anyang. This influences or material culture intrusions indicate. expectation, however, assumes ahomogeneity in 8This is nottosay that Erlitou variants, North Chinese burial practice (see also Cao 2004; such as Dongxiafeng, show no continuity with Campbell 2007). The “medium-sized” tombs are, the ceramic traditions previously dominant in moreover, the largest and richest tombs yet found the region (Taosi and Sanliqiao), but rather that at Erlitou. they are said to show more similarity with the 4Xuetal. (2004,2005) cite these and other Erlitou tradition, Erlitou variant. The replacement changes as evidence that the structure and nature of one ceramic tradition with another is widely of the “palace” area and the large-scale structures understood in China in terms of population themselves changed in this period, although they movement, but as we have argued above, and will do not venture an opinion concerning the nature further argue below, this is not necessarily the case. of the change. Aperception of difference between Moreover, as it is often easier to see similarities

63 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

with well-known Erlitou or Erligang ceramics than period at Dongxiafeng. According to the ZSKY with poorly understood local ceramic traditions in (2003) categorization schema for Erlitou burials asituation where multiple influences are present, (including Erlitou tradition burials beyond the site there may well be abias toward identifying Erlitou of Erlitou), the “elite” burial at Dongxiafeng phase or Erligang influences. III (M401) would be of middling rank. 9The report delimits aroughly 1km2area 14 The contrast is between three ceramic for the site, but not all of it dates to the Erlitou vessels and eight small pieces of turquoise with no period and it is not clear how its various localities grave furniture or cinnabar as compared to burials relate. ZSKY (2003) states that it is a20ha+ site. with one or no ceramic vessels. Erlitou “medium” Site development during the four phases of Erlitou burials, such as 87VIM57, on the other hand, have period occupation is also unclear, except that wooden grave furniture; cinnabar, bronze vessels, phase III seems to show intensified occupation, weapons, and ornaments; jade weapons and with most of the houses and kilns dating from ornaments; lacquer vessels; ceramic drinking sets; this period, although more of the site should be strings of turquoise beads; turquoise pieces; and sampled before there can be any certainty. cowrie shells. 10 One possibility, argued in Liu and Chen 15 Compare the 20 small pieces of slag (2001, 2003) and based on the occurrence of found in asingle 8x8munit for Dongxiafeng phase copper deposits and asaltlake in the region, is that III with the 1haof bronze casting and smelting Dongxiafeng was aregional craft-production and remains at Erlitou (Xuetal. 2004). resource transshipment center in alarger Erlitou 16The use of urn burials for infants and “state” network, and that the enclosed area was a children had wide distribution in China and was craft-production area. already being practicedinNeolithictimes. 11 Phase I: no houses found; phase II: no 17 Both the squatting burials postures and houses found; phase III: 100 percent (41/41) cave- bovine-head grave offerings are interesting for houses; phase IV: 25 percent (3/12) cave-house, 17 their suggestion of burial practices different from percent (2/12) surface dwellings, 58 percent (7/12) those at Erlitou, although with the present state semisubterranean; phase V(Erligang period): of published information we are alongway from no houses found; phase VI (Erligang period): no being able to do acomparative study of burial houses found. This apparent shift in dwelling type practices. between phases III and IV of the site merits further 18 Shelach (1999:208), for instance, argues investigation. that the scarcity of cowrie shells, turquoise, jade, 12 Of the 12 houses found in phase IV levels, and lacquer artifacts points to a“lack of intensity” 6were found in the eastern localities 1and 4(3cave- in the interactions that brought them to Erlitou, houses and 3semisubterranean houses) and 6in indicating, perhaps, down-the-line trading. the central localities 5and 6(4semisubterranean 19 Operating within aculture-historical houses and 2surface houses). If the double ditches paradigm wedded to atraditional account of were originally dug for cave-houses, they were Chinese history, ZSKY (2003) notes that the apparently not used for this after phase III, at least dating of the non-Erlitou variants to the Erlitou in the areas that have been excavated. II period suggests an Erlitou expansion that is 13 The grave goods for this tomb were the archaeological reflection of the Xia kingdom. two ceramic wine vessels, acookingpot, and Nonetheless, the authors state that on present eight small pieces of turquoise, making it the evidence it is impossible to say what exactly the richest burial of the six burials found for this political relationship was between the Erlitou

64 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2 tradition and its regional variants (making the transmitted texts. assumption that ceramic boundaries coincide 23 Xiaqiyuan IV is said to belong to the Early with cultural/ethnic boundaries, which, in turn, Shang period, and period III corresponds to Erlitou coincide with political boundaries) and cannot Late phase III or Early phase IV (ZSKY 2003:147– rule out the possibility that they were satellite 152). polities of the Erlitou “kingdom.” 24 The Chinese term for cave-house is “kiln- 20 The use of the name Jinzhong, or “Central pit type” house. As noted above, this house type is Shanxi” tradition (ZSKY 2003), implies something found in Shanxi in Dongxiafeng variant sites, but of the ill-defined nature of this tradition and not in Erlitou variant sites. highlights the unevenness of the understanding of 25 The Lutaigang variant type-site is located different regions, even in purely ceramic terms. at Qi county, Lutaigang 杞县鹿台岗 (Zhengzhou 21 Social hierarchy cannot, of course, be DaxueWenboyuan, Kaifengshi Wenwu simply read off of burial practice, but difference Gongzuodui [hereafter ZDW,KWG]2000). in mortuary investment minimally represents 26 These sites are located as closely as 300 difference in the resources that the deceased’s kin mfromone another and form several clusters were able or willing to muster.The fact that roughly separated by 5ormorekilometers (ZSKY 2003) in 25 percent of the tombs were relatively large and the Chifeng area. equipped with tomb furniture, rich burial goods, 27 The size of these walled settlements is and “death attendants,” minimally suggests a variable, but they do not exceed 10 ha. differential status in death and/or differential 28 Shelach (1999) cogently argues that access to wealth. since the socio-political system(s) of the Lower 22 The nature of this tradition and its Xiajiadian tradition were already fully developed affiliation has been amatter of controversy with by Erlitou III, it is impossible to derive the genesis some scholars claiming it was an Erlitou variant of Lower Xiajiadian social developments from (Zhao Zhiquan 1986), while others argued that Central Plains influences. Shelach also contends it was Proto-Shang (along with Dongxiafeng; that Erlitou and Lower Xiajiadian polities were at Zheng Jiexiang 1988). The debate over whether asimilar level of socio-political complexity, but Luwangfen-Songyao should be considered if we wish to jettison neoevolutionary categories “Xia” culture or predynastic Shang and the and to compare populations, urban spaces, craft connections between this tradition and others, production, social differentiation, and political such as Xiaqiyuan and Dongxiafeng, show both practicesandinstitutions,theevidenceuponwhich the tendency of Chinese archaeologists to equate to form this comparison does not presently exist. ceramic traditions with historical or semihistorical Minimally, however, from Erlitou II onward, Lower political entities and the arbitrary nature of Xiajiadian has nothing to compare with Erlitou in ceramic tradition “boundaries.” That Luwangfen- terms of settlement size and nucleation, internal Songyao is now considered its own tradition is differentiation of sites, large-scale architecture, or also representative of the formal logic of typology complexity of craft production. in standard Chinese archaeological practice, in 29 The Yueshi “culture” was officially which “typology is being created formulaically for recognized in 1982 and has been divided into as the sake of typology” (Cohen 2001:101) without few as three to as many as seven regional variants. the guidance of any higher-level theoretical Forinstance, Ren(1996) divides Yueshi “culture” concerns or bridging arguments linking ceramics into three regional variants, Wang (1994) divides it to people other than the dubious influence of later- into five variants, and Luan (1997a), into seven.

65 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

30 According to the ZSKY (2003), while most may have occurred in anumber of ways, and Chinese archaeologists see it as an indigenous ceramic traditions need to be understood in terms development (e.g., Yan1986; Luan 1997b), Zhang of production and distribution practices, which, in (1989) believes that this tradition is the result of turn, articulate with other social, economic, and wholesale population movements from the north. political networks. 31 The Chengziya, Dinggong, and Shijia 35 This means that the Sanxingdui sites all have rammed-earth walls encircling tradition sites on Figure 2.1 may not have all been the sites that were built in Longshan times but contemporary. continued in use through the Yueshi occupation 36 The current head of the Sanxingdui (see Cohen 2001 for amore in-depth discussion work team, Lei Yu (personal communication of Yueshi walled settlements). In addition, a July 2006) suggested alateErligangdate for the large well-like pit containing over 350 artifacts encompassing wall, but noted that as the wall was found at the Shijia site. According to the and the moat around it contained ceramics from report (Ziboshi Wenwuju, Zhiboshi Bowuguan, phase I, the site must have been at least as large as Huantaixian Wenwuguanlisuo [hereafter ZW,ZB, the area contained by the walls. If this is correct, HW] 1997), most of the artifacts were pottery (most it would have rivaled or surpassed Erlitou in size. of these, various types of guan-pots), although Substantiation of this claim, however, awaits there were stone, bone, and shell tools, as well as further excavation and the publication of the site two sheep scapulae with some burn marks and report. some undeciphered symbols. Despite the fact that 37 ZSKY (2003) tabulates these pits, their it is widely interpreted as a“sacrificial pit,” it may contents, and where they were reported in Table also have served as astorage pit or something else. 8-1 on page 497. See also Xu 2008. 32 ZSKY (2003) claims that there was abrief 38 In fact, tall, slender vessel forms seem to Yueshi tradition interlude between the lower be amarked characteristic of this tradition. Dianjiangtai and Hushu periods, and the Hushu 39 This assessment is based on the division tradition may have begun as early as Erlitou IV. of Erlitou tombs into grades, only the lowest of 33 Dou and gui are generally presumed to be which do not contain drinking vessels (ZSKY for food or beverage presentation/consumption, 2003). Moreover, drinking vessels are the most although these assumptions have not been backed frequent vessel forms cast in bronze or made with use-wear or residue analyses. with white ceramics or lacquer, again suggesting 34 Liu and Chen (2003) make the suggestive arelationship between status and drinking (see argument that since Panlongcheng lies along Underhill 2002 for an argument about drinking possible transportation routes to the north and and status from the fourth through the second there are major copper sources in Hubei, “Erlitou millennia BCE). cultural assemblages in this region indicates 40 Unfortunately, the Sanxingdui site report the earliest attempt by the Erlitou polity to gain has yet to be published and thus many specifics access to copper resources in the Yangtze River await future publication. valley”(p.78). While Liu and Chen may well be 41 The other variants are Qinweijia in central correct about political intervention in the region Gansu, Huangniangniangtai in western Gansu, from the north, this needs to be confirmed by more and Liuwan in eastern Qinghai. than hypothetical transportation routes, modern 42 At Shizhaocun, 22 of the 26 houses were resource distribution, and ceramic assemblages surface dwellings 5to6m2in surface area, with the showing northern influences. Increased contact largest no more than 10 m2 (ZSKY 2003).

66 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 2

43 Indeed these authors have argued that societies in modern Shandong province, Eastern early Erlitou received important technological Henan, or Northern Jiangsu and Anhui were within impetus from the West, upon which foundations aCentral Plains cultural orbit during the Erlitou the Central Plains bronze-casting industry period. developed its characteristic ceramic piece-mold casting techniques. 44 Although it is frequently assumed, as in the Sanxingdui case, that Erlitou-type jades found in other places must have come from Erlitou, it may be more acase of shared tradition than exchange or direct emulation given that all known Erlitou jade forms have Longshan or earlier antecedents,. 45 This again complicates the simple model of an expansionist state, coextensive with the distribution of reiﬁed ceramic boundaries seen in Liu and Chen (2003) and many other Chinese- language treatments of the subject (e.g., ZSKY 2003). 46 Witness the “kiln-cave” houses at Dongxiafeng, absent from Erlitou, but common to Shanxi and Shaanxi from Neolithic times to the present. 47 Liu and Chen’s (2003) hypothesis concerning Erlitou’s control of mining sites in the Zhongtiao Mountains remains an intriguing possibility, but essentially without any direct evidence. 48 Only limited areas around Erlitou and adjacent drainages have undergone systematic survey.The settlement pattern beyond these areas can only constructed from unsystematic surveys at present. 49 As will be discussed below and in later chapters, however, the form and especially the scale of these elite practices undergo great changes over this period. 50 In fact, many Chinese scholars associate the Yueshi with the Dong Yi of Zhou and Han texts, supposedly implacable enemies of the Xia polity (assumed to be Erlitou). While Cohen (2001) has deconstructed this traditional historical equation of the Yueshi with Erlitou, neither material cultural nor received textual evidence suggests that

67 READ ONLY/NO DOWNLOAD

Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3 The Erligang Period Zhengzhou and the Metropolitan Tradition

The Erligang period, also termed the all other known sites for this or any ear- “Early Shang” period in the Chinese lit- lier period of Chinese prehistory (ﬁgure erature, gets its name from the site of 3.2). Its unprecedented size, combined Erligang, Zhengzhou. The Erligang site with the 37-ha “palace-temple” area, was discovered in 1950, and excavations with its rammed-earth foundations, have continued in the area ever since. some in excess of 2,000 m2 (ZSKY 2003), Associated from the beginning with the bronze-casting remains, and elite buri- material culture of Anyang and, thus, als, have led to the consensus view in the Shang dynasty, the Erligang tradi- the Chinese literature that the Zheng- tion is conventionally divided into up- zhou site was an early “capital” site of per and lower Erligang periods, each the Shang dynasty.Unfortunately, the with two phases. In the intervening site is buried under the modern city of years other sites have been discovered Zhengzhou, which has severely limited in the Zhengzhou area, including in- the scope of excavations, and, in many ner and outer walls delimiting areas of ways, the site is still not well under- about 290 ha and roughly 13 km2 (Yuan stood. and Zeng 2004) respectively,1 dwarﬁng READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

3.1. Erligang-period ceramic traditions (base map from Harvard geospatial library).

The elite Metropolitan traditions of the type or quantity can be dated to this pe- Erligang period, from bronze-casting riod (currently only some jue and fewof to the form of palace-temple struc- those). The vessels, moreover, tend to tures, developed directly out of Erlitou have thin walls, suggesting, perhaps, traditions, even while the ceramic as- that there was not much metal in circu- semblage, despite showing some dif- lation. Nevertheless, some of the “pal- ferences from that of Erlitou, has many ace-temple” structures and city walls obvious points of similarity (compare were constructed in this phase,3 while Figures 2.7 and 3.6). the bronze foundry at Nanguanwai began production as well. During Erligang phase Ithe ceramics show the inﬂuences of multiple tradi- During phase II the ceramic assem- tions2 with many different styles of the blage still shows the multiple inﬂuenc- diagnostic li-tripods (ZSKY 2003:171). es of the previous period, but bronze Fewbronze vessels either in terms of artifacts increase in type and quantity,

70 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3

3.2. Zhengzhou (after Yuan and Zeng 2004:60, ﬁg. 1). as well as in thickness. Both bronze During Erligang phase IV (Xiaoshuang- jue and jia are known from this period. qiao-Huanbei phase I), most of the Phase II is considered to be aperiodof large structures in the palace-temple growth and development for the site. area were built over with nonelite structures, and the bronze foundries went Phase III (the ﬁrst phase of the upper out of service by the end. Nevertheless, Erligang period) is Zhengzhou’s apo- two bronze hoards, as well as several gee. Bronzes from this period increase bronze-vessel-yielding tombs, have all in both numbers and type, and the been found at Zhengzhou dating to foundry at Zijingshan North went into phase IV.IfZhengzhou was indeed in production. The foundry site at Nan- decline, the tombs suggest it had not guangwai continued to produce as well, yet been completely abandoned by its meaning that in phase III, Zhengzhou elites, at least as aburial ground. The had at least two major bronze work- hoards, on the other hand, might sug- shops in simultaneous operation. gest ascenario of rapid abandonment

71 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Erligang Phase Zhengzhou Yanshi Yuanqu Dongxiafeng Panlongcheng Wangchenggang Beicun II(Lower I(sub-phase II I Erligang I) 1-2) II II (Lower II (3-4) II II III Erligang II) III III (Upper III (5-6) III III II III I Erligang I) Xiaoshuangqiao- IV (Upper III (7) IV IV III IV II Huanbei Erligang II)

Table 3.1. Erligang relative chronology (adapted from ZSKY 2003:187).

similar to the case of the late Western inner city, while the important work- Zhou bronze hoards. shops were placed between the inner and outer walls, and very few cultural In addition to foundry sites, aceramic remains occur outside the outer walls. workshop was discovered 1,300 mwest of the inner city.The site contained Zhengzhou’s walls themselves were kilns, foundations, tombs, middens, constructed on amassive scale, espe- and white ash surfaces. Many daily use cially the outer walls, which were ap- ceramics were discovered, as well as proximately 6kmlong, up to an esti- some unfired cups, misfired ceramic mated 40 meters wide at the base, and waste, and numerous ceramic produc- 20 meters tall (Yuan and Zeng 2004; tion tools (ZSKY 2003; HWKY 2001). HWKY 2001). The inner walls surround- At Zijingshan, roughly half akilometer ed an area of about 289 ha, while some north of the inner city, apit was discov- parts, if not all, of the inner and outer ered containing over 1,000 finished and walls were apparently surrounded by a half-finished bone artifacts, raw materi- moat up to 20 mwide. Comparing the als, production waste, and more than 10 inner and outer walls, the inner walls bone-grinding tools and small bronze are built on amore or less rectangular knives, indicating that there was abone- plan aligned roughly 20 degrees east working site in the area. Interestingly, of north5 and were built directly upon all of the production sites discovered so the ground surface. The outer wall, on far at Zhengzhou (including the bronze the other hand, was built according to foundries) are outside of the inner city the contours of the land, with afounda- but within the outer wall or natural bar- tion trench to strengthen it, and is cur- riers, such as the lake to the east of the rently 12–17 mthick at the base. These site (Yuan and Zeng 2004).4 Considering facts suggest to Yuan and Zeng (2004) what is currently known about the site that the walls served different defensive structure, it seems that all the large and functions, the inner wall protecting the medium structures are located in the “palaces,” and the outer wall, moat, and

72 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3

Wucheng Variant Phase Wucheng Site Shang phase Wucheng Tradition Phases Panlongcheng Phase (Peng 2004) (Peng 2004) (Peng 2004) (ZSKY 2003) (ZSKY 2003) ILongwangling IELG II Panlongcheng IV ELG II II Wucheng IELG III, XSQ-HB IELG III Panlongcheng VELG III III Wucheng II AY I, II XSQ-HB IAYI Panlongcheng VI, VII XSQ-HB I IV Wucheng III AY III, IV AY II, III, IV

Table 3.2. Periodization of the Wucheng Tradition (a comparison of Peng (2004: 71], JBKY et al. [2005] and ZSKY [2003]) lake defending the site as awhole. An- large rammed-earth foundations tend other, additional possibility, is that the to be distributed within the inner walls outer wall served as ﬂood protection. (Figure 3.2).

Little work has been done on residential No special burial areas are known for areas, and there are only 11 houses pub- the Erligang occupation at Zhengzhou, lished that have been completely exca- and no “large” tombs have been found vated. In general, the greatest number (ZSKY 2003; HWKY 2001). Burials are of houses is in the palace-temple area scattered among buildings and fre- in the northeastern part of the inner city quently intrude on earlier cultural de- and they become less frequent as one posits or are intruded on by later ones. proceeds to the south and west, while Where in some areas tombs are found in the entire southern half of the walled small clusters, these tend to have been site was said to be almost devoid of Er- used only for short periods of time. All ligang cultural deposits (ZSKY 2003).6 currently known burial areas are inside More recent work, however, has shown the inner walls or between the inner and that the southern portion of the inner outer walls of the site, with the latter be- city also contained large-scale struc- ing more numerous. In addition, there tures (Yuan and Zeng 2004). As for the are also cases of remains deposited in distribution and structure of residential middens and sacriﬁcial pits. The lack of areas, nothing can be said on present discrete burial areas and the absence of evidence, nor is the development of the any “large” burials mirror the situation site over time entirely clear.Mostofthe at Erlitou, but given the state of research smallhouses that have been discovered at Zhengzhou and the limitations im- are to the sides of the inner city or in the posed on archaeological work there, it is area between the inner and outer walls; difﬁcult to say whether there might, in on present evidence, little can be said fact, be discrete burial areas which have regarding the spatial relationships of not been found or whether the practice common and elite residences (an issue of scattering burials throughout the set- that we will return to in the discussion tlement or on its outskirts was simply of Anyang) other than the fact that the the dominant practice of the time.

73 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

ble 3.1). At its height, it was about 200 ha in size and surrounded by arammed- earth wall. While Yanshi has been at the center of controversies regarding its status as apossible early capital of the Shang dynasty, it seems likely, based on date (contemporary with Zhengzhou), size (200 ha versus 1,300 ha for Zheng- zhou), and location (6 km from Erlitou), that it was asecondary center, the early function of which may have been related to control of the Erlitou site.7

Yanshi was constructed during Erligang phase I(Yanshi phases 1–2), and “palace-temples” 1, 4, 7, 9, and 10 date from this period. The inner wall was also built during phase I, as were the buildings of group II. To the northeast and outside the wall was the bronze foundry that began operation in this period. At this time the site was over 81 ha in size. The ﬁrst part of this phase likely overlaps with Erlitou IV,during which (as 3.3. Yanshi Shangcheng (redrawn from ZSKY noted above) construction continued 2003:206, ﬁg. 4-5). in Erlitou’s palace-temple area, and the

Yanshi Shangcheng

The Erligang site of Yanshi Shangcheng or Shixianggou (Figure 3.3) was founded roughly contemporaneously with Zhengzhou Shangcheng, and its building and early occupation seem to have overlapped with the latter part of Er- litou phase IV (given the circa 1550 BCE radiocarbon date for the latter). Yanshi Shangcheng was located some 6km northeast of the Erlitou site and on the location of previous large Yangshao and Longshan sites, as well as asmall Er- litou village. It was discovered in 1986 3.4. Palace-temple complex at Yanshi (after Du and divided into seven phases (see Ta- 2005:198, ﬁg. 4).

74 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3 center of the site remained densely oc- palace-temple compound without a cupied (ZSKY 2003).8 wall in the south and the palace-temple area open to the rest of the site. Then, Erligang phase II (Yanshi phases 3–4) abruptly, at the end of the middle part was aperiod of expansion, during which of Erligang phase III (Xiaoshuangqiao- the outer wall was built and “palaces” Huanbei I), the site was largely aban- 4and 7continued to be used, while 10 doned, becoming avillage by the end and 1were abandoned. North of pal- of Erligang phase III (Wang 1999; ZSKY ace-temple 9anew “palace,” number 2003). 8, was built, and upon the foundation of palace-temple 9itself, the greatly ex- As with Erlitou and later Anyang, it ap- panded palace-temple 2was built. The pears that the palace-temple area un- western wall of the palace-temple area derwent continual renovation and re- had to be moved tens of meters to the newal with multiple episodes of razing west to accommodate the construction and building. Unlike in the Mayan case, of palace-temple 2. Palace-temple 6was where pyramids were built atop earlier also built during this period. The build- pyramids (preserving them inside), it ings of group II were completely rebuilt seems that in the Central Plains Bronze on the site of the previous buildings. At Age traditions the continuity of struc- this point the walls enclosed an area of tures and their location was not put at a roughly 200 ha, surrounded by amoat premium. This suggests adifferent ori- and accessed through ﬁve gates (two on entation toward the built environment the west wall, two on the east, and one of elite ritual/administrative space, per- on the north). haps not unrelated to the constraints and potentials of building in rammed The early and middle part of Erligang earth, wood, thatch, and tile, as op- period III was another period of ﬂores- posed to stone.9 This phenomenon of cence, but one that was short-lived. No the constant rebuilding of ritual/ad- major changes occurred in the site in ministrative space within sites may also terms of site structure, but the “storage be related to the frequent movement of area” (group II) underwent extensive capitals recorded in later transmitted renovation and rebuilding. The north- texts, something that will be discussed ern and eastern sections of Yanshi’s in more detail below.10 inner wall were leveled. “Palaces” 2, 4, and 8were abandoned. On the founda- The Yanshi Shangcheng palace-temple tions of “palace” 6was built the largest area was about 4.5 ha in size and sur- palace-temple, number 5, while palace- rounded by awallabout 2mthick (com- temple 7was abandoned and palace- pared to the inner wall’s base thick- temple 3was built on top of it (Figure ness of about 6–7 m(ZSKY 2003) and 3.4). The southern and western sections the large wall’s thickness of 16–18 mat of the phase II palace-temple area walls the base [ZSKY 2003]). As with Zheng- were leveled to allow for the building zhou, the palace-temple area contained of palace-temples 3and 5, leaving the apond. Many of the palace-temple

75 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

3.5. Palace-Temple 4atYanshi (after ZSKY 2003:212, ﬁg. 4-6).

structures were built in the “siheyuan” east-west. Moreover, despite renova- four-sided courtyard style and number tions and rebuilding, this area retained among them the largest contempora- its basic layout through the occupation neous buildings so far discovered in of the site (ZSKY 2003). Because there China (Figure 3.4).11 The scale of the are no remains of living activities in the palace-temples and the existence of area (neither hearths nor middens) and the palace-temple district from the very because of the enclosed nature and or- beginning of the site suggest that Yan- derly structure of the area and its loca- shiShangcheng was planned and built tion near the palace-temple compound, as an elite religious/political structure most Chinese archaeologist interpret from the start. it as astorage area of some sort (Du et al. 1999; Wang Xuerong 2000; Liu and Aside from the palace-temple area, Chen 2003; Wang Wei2005; etc.).12 A there was agroup of buildings in the similar enclosed area (group III) was southwest of the site surrounded by a built in phase III to the northeast of the wall occupying an area of about 4ha. In- palace-temple area (Figure 3.3). While side the wall were more than 100 large the function of these buildings remains structures arranged in six rows running amatter of speculation, they do attest to

76 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3 the careful planning and scale of con- In addition to bronze workshops, pot- struction at the Yanshi site, along with tery production sites have also been the palace-temple area. found,mostlocatedinthenorthernpart of the site and dating from Erligang pe- Unfortunately, with interest focused on riod phase II. Of those found, eight are elite activities and on finding ever-earli- relatively clustered and suggest at least er dates for the site, relatively little work one area of more concentrated ceramic has been done on common residences production. In addition, some bone ar- and workshops at Yanshi. Based on cur- tifacts were discovered to the west of rent understanding, houses come in the phase Ipalace-temple compound four types: 1) single-room surface dwell- wall—mostly consisting of arrowheads, ings (averaging 10 m2); 2) multiroom spatulas, or pins, with some production surface dwellings (20–30 m2)oriented waste and semifinished bone artifacts east of north, like the large rammed- as well—suggesting aboneworkshop earth structures; 3) semisubterranean in the area (ZSKY 2003). houses (generally about 10 m2); and 4) houses built at the bottoms of pits that As at Zhengzhou, no large tombs have tend to be small: <10m2 and are often as- been found at Yanshi. Most of those dis- sociated with workshops (i.e., may not covered to date are small tombs with a actually be residences; ZSKY 2003).13 No few medium-sized ones as well. All are information is available concerning the rectangular pits. Fewofthe burials have distribution of residences in the site, yaokeng (waist pits),15 and fewer, inner and it is thus impossible to say anything or outer coffins. Almost half have no about the general residential pattern.14 grave goods, and those with grave goods usually have daily-use ceramics, with Evidence for craft-working areas have jade or bronze artifacts occurring rarely. also been found at the site, including Aunique characteristic of the distribu- abronze-casting area in the northeast tion of mortuary remains at Yanshi is section of the larger site, where molds, that they tend to be placed near or con- charcoal, and slag have been found. tiguous to the city walls. Again similar These remains date to period I, and the to Zhengzhou and Erlitou, no special workshop was destroyed when the outer cemetery areas are known for the site wall was built. Some bronze-casting re- (ZSKY 2003). Whether this fact reflects a mains were also found near the middle real pattern of burial, or simply the bias of the eastern wall, suggesting that this, of excavation focused on the walls and too, may have been abronze-working palace-temple area, is impossible to say area, and that Yanshi may, like Zheng- on present evidence. zhou and Anyang, have had more than asingle bronze-casting area (Zhong- Erligang-Tradition Variants and guo shehuikexueyuan kagu yanjiusuo, the Erligang Expansion Henan dier gongzuodui [ZSKY,HEG] 1998). During Erligang-period phase I, Erligang ceramic traditions were concentrated

77 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

3.6. Erligang ceramic-tradition variants (after ZSKY 2003:172, 176–183, 193–195; ﬁgs. 4-1, 4-2, 4-4).

in central Henan, especially the area be- areas in which they were formerly found tween and around Luoyang and Zheng- (ZSKY 2003). In Erligang period II, met- zhou, as well as southern Shanxi, more ropolitan ceramic variants were found or less replacing Erlitou traditions in the in northern Henan and southern Hebei,

78 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3 while in the south, Panlongcheng, in The Erligang Tradition Metropoli- Hubei, formed another variant. In east- tan Variant ern Henan, another metropolitan-tradition variant began to replace Yueshi tra- Located in central Henan and most ditions. In Erligang-period phase III, the densely concentrated in the Zhengzhou distribution of metropolitan ceramic tra- area, with the Yellow River forming a dition variants became even more wide- northern boundary and the Tongbai spread, stretching from Xi’an in the west, Mountains asouthern limit, this vari- to Jinan in the east, reaching the Yangzi ant includes Zhengzhou and Yanshi. River from Anhui to northern Hunan in It formed the metropolitan tradition the south, and extending in the north to from which the regional variants were Beijing. Generally speaking, the north- derived (Figures 3.1, 3.6). The houses ern and western Erligang-period vari- of this tradition were mostly small ants resemble the metropolitan tradition semisubterranean or surface construc- more than the southern or eastern ones, tions; only medium or large structures which show more local characteristics were built on rammed-earth platforms (ZSKY 2003), suggesting, perhaps, differ- (ZSKY 2003). Bronze artifacts have been ent mechanisms of the ceramic-tradition found at several sites in this area (ZSKY spread, different levels of local incorpora- 2003) in addition to Yanshi and Zheng- tion, or both. zhou. In general, the bronze industry In summary, not only was Zhengzhou a of this tradition was based on that of site of unprecedented size, but by the end Erlitou but continued to develop with of the Erligang period it sat at the center new types and styles of bronze vessels— of aweb of ceramic influence, distribu- some derived from ceramic prototypes tion, and exchange that dwarfed the geo- associated with Erligang assemblages graphic spread of any previous ceramic (e.g., li-tripods). Oracle-bone divina- tradition within the area of modern Chi- tion was widely practiced, with sheep, na. Just as with Erlitou, however, this was pig, and cattle scapulae, as well as turtle not asingle uniform tradition but rather plastrons as the media (Flad 2008). took the form of ametropolitan tradition based on Zhengzhou and Yanshi In the culture-historical idiom in which and several regional variants. These are Chinese ceramic traditions are defined, the Erligang metropolitan variant (Fig- the interaction with neighboring tradi- ure 3.6 bottom right), the Erligang traditions is understood in terms of influ- tion (ELG) Liulige variant (Figure 3.6A), ences, essentially indexing asimilarity the ELG tradition Taixi variant (Figure of style or form deriving from some sort 3.6B), the ELG tradition Dongxiafeng of interaction, the nature of which is variant (Figure 3.6C), the ELG tradition unknown (though usually assumed to Panlongcheng variant (Figure 3.6D), the be migration). The most important in- ELG tradition Beicun variant, the ELG tra- fluence on the Erligang-tradition met- dition Dachengdun variant (Figure 3.6E), ropolitan variant is that of the Xiaqiyu- and the ELG tradition Daxinzhuang vari- an tradition, with the Erlitou tradition ant (Figure 3.6F) (ZSKY 2003). aclose second (Zou 1980; ZSKY 2003).

79 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Yueshi-tradition influences can also be The Erligang Tradition Liulige seen, and given the association of Er- Variant ligang with the Shang dynasty and the later tradition of the Shang’s eastern Located in Henan south of the Huan origins, some scholars speculate about River, in the foothills of the Taihang aformative role for the Yueshi tradi- Mountains and north of the Yellow tion in Erligang culture (for acritique River, this variant occupies much of the see Cohen 2001). The Erlitou-period same area as the Luwangfen-Songyao Luwangfen-Songyao tradition, perhaps tradition of the Erlitou period and ap- not surprisingly, given its neighboring pears to develop out of it under Erligang location, is said to be another source influences (Figure 3.6A).16 During Erli- of influence. Stamped stoneware and gang period Ithe tradition in this area protoporcelain from the south are also was still that of Luwangfen-Songyao, the present in Erligang metropolitan-var- Liulige variant forming only in Erligang iant assemblages and were, perhaps, period II (ZSKY 2003). Two of the better- emulated as well as imported (ZSKY known sites of this variant are Meng- 2003). As noted above, the early phases zhuang and Fucheng, only the latter of of the Erligang tradition appear to com- which has been published (Yuan and bine the forms, styles, and decorations Qin 2000). Erligang-type bronze vessels of anumber of neighboring traditions. have been found at sites of this variant, This cosmopolitanism and the enor- and aside from ceramics, which come mous size of the Zhengzhou site, with to look more like Erligang-tradition as- its inherent nucleation of population, semblages over time, other forms of suggest that, whatever the geographi- material culture appear to be more or cal origins of its rulers, the population less the same as those found at Zheng- they ruled over at Zhengzhou and be- zhou. Nevertheless, it should be noted yond was likely drawn from avariety of that the entire Central Plains area had sources. Considering the distribution seen increasingly intense interaction of the strongest influences, however, over the preceding millennium, and the the impression is given of amelding previous Luwangfen-Songyao tradition of the previous metropolitan tradition was originally thought to be an Erlitou (Erlitou) with traditions from the north- tradition variant. All this is to say that east (Xiaqiyuan, Luwangfen-Songyao) there were broad similarities between and east (Yueshi). Whether the pattern Central Plains ceramic traditions even really signifies population movements before the Erligang period, and the “Er- from the northeast and east into the ligang expansion” was really astrength- Zhengzhou-Louyang area or something ening of existing similarities indicating more complex will have to await further more intense interactions centered on research based on more than just for- Zhengzhou, rather than simply the re- mal ceramic typology. placement of previously unrelated local traditions by the migrating colonists of an expansionist “state.”

80 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3

Fucheng, located in Jiaozuo City’s pies the area formerly occupied by the southwest suburbs, 15 km from the Xiaqiyuan tradition, which probably Taihang Mountains, is asmall (9 ha) lasted until at least Erligang period Iif walled site showing Erligang charac- not later (ZSKY 2003). While Taixi-var- teristics, including walls and rammed- iant ceramics show some Xiaqiyuan- earth building foundations oriented tradition characteristics, there are also roughly 20 degrees east of north. There some obvious differences (Figure 3.6B). are four areas of rammed-earth founda- However, since neither Xiaqiyuan traditions in the site, and the largest of the tion nor the Taixi variant of the Erligang foundations is almost 4,000 m2,rank- tradition are well understood, more ing it among the larger Erligang-period work in these areas may give aclearer structures presently known. The site picture. Nevertheless, since the Xiaqi- itself dates from sometime during the yuan-tradition area is widely believed lower Erligang period (Erligang periods to be the Shang homeland in Chinese I–II), to the early part of the Xiaoshuang- archaeological circles, it is interesting qiao-Huanbei period, after which it was that there are nonetheless obvious dif- abandoned. From what is known about ferences between the Xiaqiyuan tradi- the structure and orientation of the site, tion and the Erligang metropolitan vari- not to mention its ceramic tradition af- ant on the one hand, and between both ﬁliation, it appears that the builders the former and the latter and the Taixi of this site were at least participants in variant on the other.Minimally it sug- Erligang-period elite culture, even if the gests that potsherds, politics, and eth- exact nature of the political relationship nic groups are not as easily associated between this site and the Zhengzhou- as Chinese archaeologists have tradi- Yanshi core remains unclear. tionally assumed.

The Erligang Tradition Taixi Variant

Beginning in Erligang period III and represented by the early layers of Gaocheng Taixi in Hebei Province (Hebeisheng wenwu yanjiusuo [HWY] 1985), this variant is located north of the Huan River (and the Liulige variant) and south of the Yishui, bounded on the west by the Taihang Mountains and by the old course of the Yellow River in the east (ZSKY 2003). Published sites for this variant are few, and since none but Taixi have full site reports, this variant is still not well understood. It occu-

81 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

3.7. Panlongcheng (after HWKY 2001: 4, ﬁg. 3).

The Erligang Tradition 2003), or in other words, local tradition Dongxiafeng Variant continuing to develop under the inﬂu- ence of the new metropole. Interesting- Located in the same area as the Dongxi- ly, however, none of the tombs found afeng variant of the Erlitou tradition, in this area have waist pits, suggesting this Erligang-period variant is said to be that this aspect of Erligang mortuary amixture of Erlitou Dongxiafeng and practice was not followed. Basic tools Erligang inﬂuences (Figure 3.6D; ZSKY and artifacts in the Dongxiafeng variant

82 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3

3.8. Panlongcheng Lijiazui M2 (after HWKY 2001:154, ﬁg. 100). area are more or less similar to those shaped” urn vessels, and arelative pre- of the metropolitan variant, although dominance of stone knives over sickles given the connection between the two in the stone tool assemblage—similar regions in the preceding Erlitou period, to the Beicun variant to the west but dif- this is not especially remarkable. None- ferent from the Erligang variant. theless, the Dongxiafeng variant still retains some features common to tradi- The Dongxiafeng site itself (ZSKY,ZLB, tions to its west and north, such as “egg- SKY1988) has only undergone limited

83 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

3.9. Bronze and jade artifacts from Lijiazui M2 (after HWKY 2001:176, 178, 180; ﬁgs. 117–119; Institute of Archaeology 2003:244, ﬁg. 4-15).

excavations (ZSKY 2003), and it is difﬁ- shaped depressions on the ﬂoor surface cult to estimate the site size in Erligang- and no doors. Many Chinese archaeolo- period times. It had awalland defensive gists speculated that these buildings ditch in Erligang period I, although only were granaries, but recent soil chemical the southwestern side has survived. At analysis has suggested they were used the site an area of round building foun- for salt storage (Chen et al. 2010). Some dations has been discovered forming at small bronze artifacts have been discov- least seven rows, with six or seven build- ered at the site, and there is some evi- ings per row, totaling between 40 and denceofbronze-castinginthenortheast 50 structures. These structures have section, where slag and stone molds for adiameter of about nine meters and producing arrowheads were excavated were about 30—50 cm above the con- from Erligang-period contexts. temporaneous surface. They have cross-

84 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3

riod. It dates from Erligang phase III through to the Anyang period. Despite some local characteristics, such as the prevalence of round-bodied guan- pots (an important cooking vessel in both Late Neolithic and Erlitou sites in eastern Shaanxi), and the absence of jue-vessels, the ceramic assemblage of this variant basically resembles that of Erligang (ZSKY 2003). However, like Dongxiafeng, stone knives appear more commonly than stone sickles—apparently atradition common to southeast- 3.10. Panlongcheng Buildings 1and 2(after ern Shanxi and east central Shaanxi ZSKY 2003:233, ﬁg. 4-11). during this period.

The Erligang Tradition Panlongcheng Variant

This Erligang-period variant is mostly based north of the Yangzi and south of the Tongbai Mountains, centered on the Panlongcheng site (Figure 3.7), although there are some scattered sites outside this area. This variant (Figure 3.11. Reconstruction of Buildings 1and 2(after 3.6D) dates from Erligang period II HWKY 2001:644, ﬁg. 13). and continues into the Xiaoshuang- qiao-Huanbei period. Due to the large tombs found at Lijiazui cemetery, many bronze vessels and weapons have been The Erligang Tradition Beicun found (Figures 3.8, 3.9). Indeed, the Variant Panlongcheng Lijiazui cemetery has the largest and richest tombs known for Distributed in eastern central Shaanxi, the Erligang period. Although no molds the Beicun variant is represented by have been discovered at the site, it has Yaoxian Beicun (Beijing Daxue Ka- been claimed that these bronzes have oguxi Shang Zhou Zu, Shaanxisheng local characteristics (ZSKY 2003), sug- Kaogu Yanjiusuo [BDKSZZ, SKY]1994) gesting local casting, though this is still and Xi’an Laoniupo (Liu Shi’e 2001). amatterofdebate. Weapons include This variant is distributed no farther ge-dagger-axes, yue-axes, and spears west than Xi’an in the Erligang pe- (something not seen in the Central

85 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Plains during the Erligang period), as the northeast part of the site, including well as arrowheads. Panlongcheng also three large buildings atop a6,000 m2 has akind of spade that has been found rammed-earth platform (Figures 3.10, atZhengzhou,suggestingthecontinued 3.11), which show obvious affinities use of local weapon and tool types. The with palace-temple structures of the ceramics include both vessel forms and Central Plains metropolitan tradition. styles that resemble Erligang ceramics and vessels that are obviously local in Mid- and small-sized houses were found origin. Erligang-type ceramics account at Panlongcheng as well, the middle- for about 60 percent of the vessels that sized ones often built on rammed-earth can be reconstructed. The ceramics are platforms, while the small ones are mostly hand-built, and red ceramics oc- semisubterranean. Craft production– cupy over half the assemblage (Erligang related remains have been discovered ceramics are dominated by gray-ware).17 both north and south of the walled site, Some Erligang types do not occur at Pan- although their nature is not entirely longcheng, other Erligang types have lo- clear.19 Panlongcheng’s architectural, cal peculiarities, and red, thick-walled, ritual, and mortuary remains show rough-paste gang-urns are particularly unambiguous affiliation with Central common. This is avessel type inherited Plains elite traditions, and, yet, the ce- from the late third–early second millen- ramic assemblage shows strong local nium BCE Shijiahe tradition and is one characteristics. The consensus view of the most common vessel types in this among Chinese scholars is that the Er- area during Erlitou times. The conclu- ligang intrusion into the middle Yangzi sion seems to be that the Panlongcheng was related to the control of rich copper variant’s ceramic tradition is mostly deposits in the south. While this argu- based on the metropolitan variant but ment seems plausible, the nature of the influenced by local traditions, as well as intrusion and the political relationship the Wucheng tradition to the southeast between Panlongcheng and Zhengzhou and the Hushu tradition of the east. both in the beginning, and over time, are by no means clear.Indeed, as will The Panlongcheng site, at over 75 ha,18 be argued in chapter 5, if the Erligang- is the third-largest Erligang-tradi- period polity is anything like the An- tion site after Zhengzhou and Yanshi yang polity in its political practices and (Hubeisheng wenwu kaogu yanjiusuo networks of discursive, practical, and [HbWKY 2001]). It is walled and orient- material resources, Erligang royal con- ed to the northeast like other Erligang trol over Panlongcheng, if ever actually centers, elite buildings, and tombs. Un- direct, may have been fluctuating and like Wucheng, and previous Neolithic contingent on regional politics and southern walled sites, Panlongcheng’s shifting alliances among high elites. walls were built with rammed earth (as opposed to simply heaped earth). There was also amoataround the site. Most of the important architecture was in

86 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3

The Erligang Tradition conjectural. The Daxinzhuang variant Dachengdun Variant dates from Erligang period III and is characterized by aceramic assemblage The Dachengdun variant of the Erli- that is mostly Erligang style or Erligang- gang tradition (Figure 3.6E) begins in derived, but also contains Yueshi or Erligang period III and is distributed in Yueshi-derived ceramics. Despite li- modern Anhui between the Yangzi and tripods being the most important cook- Huai Rivers and north of Chaohu. The ing vessel, yan-steamers, which are one type site is the Hanshan Dachengshou of the most important Yueshi cooking site. Erligang-type bronze vessels have vessels, are far more frequent at Dax- been found at Dachengshou sites. Li- inzhuang than they are at Erligang. In tripods are the most important cook- terms of stone tools, half-moon-shaped ing vessel in the ceramic assemblage two-holed knives carry on the Yueshi (as in all Erligang-tradition variants), tradition, but most other tools and arti- and though most of the assemblage re- facts look similar to Erligang types. As sembles Erligang ceramics, there are with other variants, Daxinzhuang shows also vessel types and styles that appear amixture of traditions rather than the to be continuations of earlier local tra- total replacement of one material cul- ditions. Ceramic-tradition influences ture tradition by another.Whether this include the previous Doujitai tradition mixture is due to overlapping distri- (the prior tradition in the area), as well bution networks of different ceramic- as possible Yueshi influences. Beyond production traditions, the coexistence formal ceramic and bronze typology, of potters of different traditions at a however, very little information is avail- single site, the incorporation of differ- able. On present evidence then, all ent styles and forms by ceramic produc- that can be said is that there seems to ers, or some combination is uncertain have been an intensification of Central with the present state of research. The Plains contact with this area in Erligang predominance of Erligang-type ceram- times, while the Erligang-style bronze ics, however, does indicate intensive vessels suggest the presence of metro- contacts between this area of Shandong politan-style elites.20 and Erligang sites to the west.

The Erligang Tradition Traditions Beyond the Erligang Daxinzhuang Variant Ceramic Horizon

Jinan Daxinzhuang is the type-site for The North this variant (Figure 3.6G), which is distributed south of the present course of During the Erligang period, the Jin- the Yellow River and west of the Taiyi zhong tradition was distributed in mountain range. However, the only site Shanxi north of the Dongxiafeng vari- of this variant in this period that has ac- ant. Sharing features with both Dongxi- tually been excavated is Daxinzhuang, afeng- and Erligang-tradition ceramics, so this distribution is to some extent this tradition is nonetheless considered

87 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

3.12. Hushu-tradition ceramics (after ZSKY 2003:469; ﬁg. 8-6).

to be an indigenous development with Houses took the form of semisubter- aclose relationship to Zhukaigou in ranean and subterranean cave houses, the northwest and Lower Xiajiadian in aphenomenon also seen in southern the northeast, sharing with them egg- Shanxi and Hebei. shaped urns and deep-bellied basins.

88 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3

To the northwest, as in the Erlitou pe- although the cultural, political and riod, the Zhukaigou tradition occupied economic nature of this zone and its the northern bend of the Yellow River relationship to Central Plains cultures through Inner Mongolia. Toward the probably changed over time, it was a end of the Erligang period, Central zone of contact of lasting importance Plains influences become more pro- throughout the Central Plains Bronze nounced in Zhukaigou assemblages Age and beyond. corresponding with the advent of the Erligang period III Beicun variant in The East central Shaanxi and perhaps the begin- During the Erligang period, the princi- ning of the Xiaoshuangqiao-Huanbei pal non-Central Plains Bronze Age ce- period expansion. ramic tradition in the east was still the Yueshi-tradition variants in Shandong. The Datuotou tradition, described As noted above, the Erligang-tradition above in the Erlitou section, continued variant of Daxinzhuang takes form at through the Erligang period, distribut- the end of this period, and the Erligang ed to the northeast of the Taixi variant tradition is said to extend into eastern of the Erligang tradition in the region of Henan (ZSKY 2003). Nevertheless, the modern Beijing and Tianjing. The Dat- extent to which the Erligang tradition uotou metallurgical tradition belongs displaces the Anqiugudui Yueshi vari- to the northern complex and shares ant or the Yueshi Zhaogezhuang vari- many common features with the other ant, is unclear at present. Even if the Er- traditions spread across the northern ligang variants in former Yueshi areas periphery of the Central Plains. As in represent metropolitan colonial intru- the Erlitou period, the Lower Xiajiadian sions as opposed to increased contact tradition was distributed to the north- and emulation of the Central Plains east of the Datuotou tradition, occupy- megacenters in areas that had long ing parts of eastern Inner Mongolia and been interacting, it would still not be western Liaoning (see Erlitou period clear whether such intrusions should section above). be seen in terms of conquest, economic activity, or some other possibility.In Generally speaking, there seems to be short, on present evidence, we can only adegree of cultural interaction and say that Erligang material cultural influ- exchange over abroad band stretch- ences spread eastward near the end of ing from western Shaanxi and Gansu, the Erligang period, replacing Yueshi through Inner Mongolia, middle and traditions in some areas. However, the northern Shanxi, and the northern half ethnic, political, social, and economic of Hubei and Liaoning. Evidence for details of this “influence” are anything this interaction includes elements of but clear. ceramic traditions, mortuary practices (stone coffins), and metallurgical tradi- The Southeast tions. This interaction zone was already As mentioned above, the area that was present from at least Erlitou times, and formerly occupied by the Doujitai ce-

89 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

3.13. Jingnansi-tradition ceramics (after ZSKY 2003:475; ﬁg. 8-7).

ramic tradition during Erlitou times political details of this transition are came to be occupied by the Erligang- unclear, and no major Erligang centers tradition variant Dachengdun at the have been discovered in this area. Mini- end of the Erligang period. The socio- mally, the evidence suggests amarked

90 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3 increase in northern material cultural Recently, asurvey and excavation of elements in an area that had already protoporcelain production sites was seen interaction with the Central Plains conducted in the Maqiao-tradition area in Erlitou times. (Zhejiangsheng Wenwu Kaogu Yanjiu- suo [ZWKY] et al. 2011). The fruits of The Hushu Tradition this investigation included the earli- The Hushu tradition (Figure 3.12) was est evidence of protoporcelain produc- distributed to the east of the Erligang tion dating to the Erligang period and Dachengdun variant in the Erlitou-pe- perhaps earlier and an early form of riod Dianjiangtai-tradition area. The the “dragon kiln.” The site of Nanshan Hushu tradition continued down to contained protoporcelain-production Anyang times. After the Anyang period, remains ranging from Erligang (or per- under strong Western Zhou influence, haps Erlitou) to Anyang times (ZWKY et “Wu culture”21 is said to emerge in this al. 2011). area (ZSKY 2003:467). The small bronze artifacts, such as knives and arrow- The South heads, so far discovered in Hushu strata bear close resemblance to Erligang As mentioned above, the Erligang-tra- types and are probably copies or im- dition site of Panlongcheng was built ports. In terms of ceramics, the assem- in Hubei Province by at least Erligang blages are said to show both Erligang period II, seemingly the culmination and Maqiao influences on atradition of several centuries of interactions be- that mostly derives from the Dianjiang- tween northern Hubei and the Central tai tradition (ZSKY 2003). Nevertheless, Plains. While there is evidence of Pan- being entirely an analysis of formal fea- longcheng-variant Erligang-tradition tures of atradition seen as an organic sites from as far south as Hunan Yuey- whole, these culture-historical conclu- ang and possibly Jiangxi Ruichang sions may in fact mask avariety of so- (both with early mining sites nearby), cial processes and contacts between Erligang ceramic traditions are not the this and adjacent areas (e.g., exchange, only ceramic traditions in Hubei, Hu- intermarriage, emulation, population nan, and Jiangxi during this period. movements, warfare, etc.). The Jingnansi Tradition The Maqiao Tradition Distributed near the modern city of Jin- South and east of the Hushu tradition, gzhou in southern Hubei, and known in the area around Taihu, the Maqiao mainly from the type-site of Jingnansi tradition was distributed as in Erlitou and the adjacent Zhangjiashan site, times. This tradition is poorly under- this tradition (Figure 3.13) shows strong stood, although, as noted above, there Panlongchenginfluences.Nevertheless, is some evidence for contact between the tradition mainly evolved from prior this tradition and the Hushu tradition local traditions, with some influences of Jiangsu Province. evident from the Chaotianzui variant of the Sanxingdui tradition located up the

91 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

3.14. Baota-tradition ceramics seriation (after ZSKY 2003:481; ﬁg. 8-10).

Yangzi (ZSKY 2003). The Jingnansi tra- Sanxingdui tradition and the southern dition lasted from the Erligang period extension of the Erligang tradition. to the beginning of the Anyang period. At present the distribution of Erligang- The Baota Tradition period sites in this area is unclear.Much Distributed north and east of Lake more work needs to be done before we Dongting in Hunan Province, the Baota understand the relationships between tradition (Figure 3.14) shows Central Panlongcheng and areas farther up the Plains inﬂuences in its ceramic tradi- Yangzi beyond the observations that the tion as well as that of Jingnansi. Central ceramic assemblage indicates extensive Plains–style bronzes have been discov- contact of some kind and that western ered in this area. While this area is also Hubei may have been azoneofcontact poorly understood archaeologically, we between the eastern extension of the know minimally that there seems to

92 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3 have been extensive and intensive contact between people living in Hubei and Hunan along the Yangzi and its tribu- taries.

The Wucheng Tradition The Wucheng tradition gets its name from the site of Wucheng in Jiangxi Province. Approximately 60 ha in size and surrounded by an earthen wall, Wucheng was also the site of bronze- casting and the production of stonewares and protoporcelains, as well as having what may be an undeciphered script.22 Wucheng sites were distributed mostly in the eastern half of Ji- angxi Province (ZSKY 2003; Peng 2004; Jiangxisheng bowuguan kaogu yanjiusuo, zhangshushi bowuguan [JBKY,ZB] 2005), and over 200 sites of this type have been discovered (Peng 2004).23 The Wucheng site is divided into three periods, only the first of which probably dates to the Erligang period (Table 3.15. Wucheng-tradition ceramics (after ZSKY 3.2).24 The wall of the site was begun dur- 2003:485; fig. 8-11). ing Erligang period phase III (Wucheng I), but the extent of the site during this ment, however, that Wucheng shows period is unclear.The Wucheng tradi- evidence, in terms of both ceramics tion (Figure 3.15) shows Erligang in- and bronze artifacts, of contact with the fluences most strongly in period I, and Central Plains metropolitan sphere and according to JBKY,ZB(2005), Central that this contact was strongest in phase Plains–type vessels form the core of the I. It is also widely believed that Wucheng assemblage (although this assertion may have been one of the sources of is not supported with any quantitative the stamped and glazed stoneware and data). Also uncertain is the origin of “protoporcelain” found in such central Wucheng’s Central Plains “influences,” plains sites as Zhengzhou and Yuanqu. since the vessel types are similar to, yet differ from, known Erligang variants.25 On present evidence then, it would The authors of the Xiaand Shang (ZSKY seem that during the formative stages of 2003), on the other hand, are of the the Wucheng tradition, there was acon- opinion that during phase Ithe main siderable amount of contact with Pan- tradition is local but under strong Erli- longcheng-variant areas and perhaps gang influence. There is general agree- other Erligang variant areas as well. The

93 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

3.16. Wannian-tradition ceramics (after ZSKY 2003:490, ﬁg 8–14).

speciﬁcs of this interaction remain un- teraction would seem to have involved at clear, but it is probably no coincidence least forms of exchange. However, since that it corresponds with the Central most of the remains of the Wucheng Plains intrusion into Hubei. From the site date to Xiaoshuangqiao-Huanbei mixed assemblage at the Tongling min- phase I(Wucheng II), Iwill save further ing site and the presence of Wucheng- discussion of the site to alater section. type hard ceramics in the north, this in-

94 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3

3.17. Sanxingdui-tradition phases III–IV ceramics (after ZSKY 2003:502, 503; ﬁg. 8–17).

The WannianTradition26 kind that left atrace in the ceramic as- Distributed in the western half of Ji- semblage) with communities in Jiang- angxi Province and showing ties with su than with their Wucheng-tradition the ceramic traditions of western Fu- neighbors. jian, this tradition (Figure 3.16) is notable for its high proportion of stamped The Southwest hard ceramics and protoporcelain (ZSKY 2003). According to recent work, In the Sichuan basin, the Sanxingdui this tradition can be divided into two tradition that arose in late Erlitou times phases, the first dating to roughly the continued to flourish, and, if all of the Erligang and Xiaoshuangqiao-Huanbei Sanxingdui site was in use at this point, periods, and the second, to the Anyang its size at 12 km2 (360 ha within the period (Peng 2004). Over 40 sites of this walls) would have approached that of type have been found to date, but they Zhengzhou and been the second larg- are widely scattered, and the largest are est site known from this period. Its only 1hainsize. walls were massive in size (over 40 min width at the base and covered with ado- With a yan-shaped vessel as its chief be brick finish at the top (ZSKY 2003)). cooking vessel instead of a li-tripod, the Sanxingdui material culture influences ceramic assemblage of this tradition is expanded in this period reaching the very different from the Wucheng tradi- three gorges where the Chaotianzui var- tion. Dividing up the assemblage into iant formed. four types, Peng (2004) notes that while local ceramics are said to make up 80 The Sanxingdui Tradition percent of the assemblage, Hushu-type Since asite report for Sanxingdui has ceramics are the most abundant non- yet to be published, there are numerous Wannian element. This would suggest questions that remain, not least, the that Wannian-tradition communities chronology of the site. While based on had more interaction (at least of the comparison with Erlitou ceramic types,

95 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

3.18. Sanxingdui-tradition Chaotianzui variant (after ZSKY 2003:510, 511; ﬁg. 5-18).

Sanxingdui Iand II have been dated in the artifact pits (ZSKY 2003). Bagley to the Erlitou period, Sanxingdui III (1999), however, claims that the bronz- (which includes artifact pit I) has been es at Sanxingdui show no Erligang influ- dated to phase IorIIofthe Anyang peri- ences, noting instead their similarity to od (Figure 3.17). This obviously creates vessels from the Middle Yangzi that date the problem of accounting for the 300 to the Xiaoshuangqiao-Huanbei period plus years between the end of Erlitou (or “transitional period” in Bagley’s ter- and the beginning of the Anyang period minology; see also Falkenhausen 2006). (ZSKY 2003). Amorecoherent solution is that of Xu (2006, 2008), which, as not- If this is correct, then, as suggested for ed in Chapter 2, places Sanxingdui Iin the Erlitou period, the Central Plains in- the Baodun period and dates Sanxing- fluence on the Sichuan basin is actually dui II to ca. 2000–1500 BCE. Sanxingdui an indirect influence via Panlongcheng III is dated to 1500–1200 BCE, and Sanx- and the bronze traditions that arose ingdui IV,to1200–1000 BCE. Under this on the edges of its influence and after chronological schema, late Sanxingdui its fall. In political terms, there is even II and early Sanxingdui III would date to less reason to postulate direct linkages the Erligang period. between the elites of Sanxingdui and those of Zhengzhou. Although according to ZSKY (2003) there is increased contact between Si- The Sanxingdui-Tradition chuan and the Central Plains in the Chaotianzui Variant Erligang period, the evidence for this The Chaotianzui variant of the Sanx- comes from some bronze vessels and ingdui tradition is distributed all along jades supposedly dating to the Erligang the Yangzi gorges area. Despite afair and Xiaoshuangqiao periods found amount of excavation on sites of this

96 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3

Figure 3.19. Siwa-tradition variant ceramics (after Zhao 1989:149; ﬁg. 2).

97 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

variant, reports have not been system- ern Shaanxi from Erligang to Western atic, and it is difficult to fix the date of Zhou times (Nan 1989). It is known al- this culture. It is said to roughly corre- most exclusively from burials, and vir- spond to the middle period of the Sanx- tually nothing is known about the socie- ingdui tradition, suggesting, perhaps, ties that produced the material culture dates between the Erligang and Xia- of this tradition. The Siwa was located oshuangqiao-Huanbei periods. This mostly to the east and south of the Xin- variant (Figure 3.18) can be divided into dian tradition, and though they are said three phases, but, because there have to be clearly distinct traditions, Xindian been few comprehensive reports, the and Siwa show some similarities, such periodization is still not settled. The as saddle-mouthed guan-jars and dou- ceramic assemblages of this tradition ble-eared, bag-footed li-tripods (Nan include vessel types commonly seen in 1989). Sanxingdui sites, some types rarely seen in Sanxingdui sites, and types that are Generally speaking, most Siwa tombs purely local. The Sanxingdui types are are rectangular pits, and treatment of the most common, but by phase III, the the body includes: intact straight-limb proportion of Sanxingdui ceramics de- burial, partial dismemberment, and crease in relation to local forms. Some secondary burial (Zhao 1989). Burial Central Plains–type bronze vessels have goods generally take the form of ce- been discovered in the area in addi- ramic vessels (anywhere from several tion to local style bronze arrowheads, to several dozen), with saddle-mouthed knives, axes, and small tools. Most tools guan-jars being the most common ves- and weapons excavated from Chaotian- sel type. Small bronze artifacts, such zui sites, however, were made of stone as weapons, tools, and ornaments, are or bone. occasionally also found in Siwa tombs. Many Siwa burials have niches used for The West placing burial goods and, sometimes, sacrificial victims or death attendants. Aside from the Beicun Erligang–period At one Siwa cemetery, some of the buri- variant, which dates from Erligang peri- als had coffins and, more rarely, even od phase III, the archaeology of Shaanxi outer coffins (Zhao 1989). The Siwa tra- is poorly understood for the Erligang dition, following the historiographic period. On present evidence, all that assumptions of Chinese archaeological can be said is that Central Plains mate- practice, has been assigned anumber rial cultural elements were spread far- of ethnic labels derived from tradition- ther into Shanxi during Erligang period al historical texts, including Di Qiang III and continued to spread during the (Xia 1949), Rong Di (Hu 1979), and even Xiaoshuangqiao-Huanbei period. Proto-Zhou (Li 1981). As Wang (1992) has argued however, the appellations The Siwa Tradition that Zhou texts gave to their non-Zhou The Siwa tradition was distributed in neighbors were not so much stable enti- eastern Gansu and on the edges of west- ties as discursive reflections of shifting

98 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3 political and ethnic boundaries. From amaterial culture perspective, with the current state of evidence, all that can be said is that the Siwa tradition shares some features with some of the traditions found in Shaanxi during the An- yang period.

The Xindian Tradition (The Xindian tradition was distributed to the west of the Siwa tradition, around the conﬂuence of the Yellow and Tao Rivers at the eastern end of the Gansu corridor and eastern Qinghai. This tradition (Figure 3.20) is generally believed to have developed out of the Qijia tradition (Zhang et al. 1993; Nan 1989) and dated from the end of the Erlitou period to the Western Zhou. During its later 3.20. Xindian-tradition ceramics (after Nan phases the Xindian tradition expanded 1989:76; ﬁg. 2). westward into Qinghai. Very little information exists concerning the Xindian tradition aside from formal typological ents at Erlitou, the Erligang period saw material and brief tomb descriptions. their development. From settlement Tombs take the form of rectangular size to distribution of sites and in terms pits, irregular pit burials, or “niche”- of the quantity, quality, and distribu- burials. Ceramics include divided- tion of elite craft production, as well as crotch, bag-footed li vessels, but the evidence of long-distance exchange, the most common vessel type appears to Erligang horizon was unprecedented. be saddle-mouthed guan-pots. Xindian The transition between the Erlitou and ceramics are also frequently painted Erligang periods, moreover, appears to (unlike Siwa ceramics). Compared to have been culturally seamless. Already, Qijia bronze-working, the Xindian tra- in Erlitou IV and increasingly toward dition seems relatively impoverished. the end of the phase, Erligang-type ce- What bronze artifacts are found in Xin- ramics were found in the Erlitou assem- dian sites mostly take the form of small blage. Six kilometers away and roughly weapons, tools, and ornaments (Li Shu- contemporary with the second half of icheng 2005). Erlitou IV,the large Shang walled site of Yanshi was built, and, in the Zheng- Conclusion zhou area, an even larger walled site was being constructed. The “Erlitou expan- If many of the material cultural aspects sion” was not followed by aretraction of Shang elite culture had their anteced- of Central Plains material culture, but

99 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

rather it was incorporated into an even scholars (e.g., Sun 2009; Liu and Chen larger“Erligangexpansion.”Thebronze 2003) have noted that the presence of vessel–casting tradition, likely invented walled Erligang sites on the periphery at Erlitou, was taken up and expanded tend to correspond to areas of strategic on in Zhengzhou and Yanshi and likely resources (such as copper and tin: Pan- such other sites as Panlongcheng. The longcheng, Dongxiafeng; salt: Yuanqu), bronze and stone weapon types found but military and political objectives at Erlitou continued to form the core may have been just as important (Tong of the Central Plains arsenal. The use 2003). These scholars also see the pres- of bronze and jade weapons or artifacts ence of these walled sites and the Erli- derived from weapons as symbols of gang expansion in general as evidence status or tools of ritual continued to be for the direct control of ahuge territory elaborated on. The Erlitou jade assem- by acentralized state (Bagley 1999). blage in general (with the exception of While the presence of walled Erligang- zhang-blades) remained the core of type sites over alarge area suggests a the Erligang-period jade assemblage, common elite cultural sphere, even with some types, such as the “handle- while the unparalleled scale of Zheng- shaped” artifact, increasing in quantity zhou suggests acultural and political and variety through to the eighth cen- core, the actual relationships between tury BCE (Cao 2008). Architecturally, sites and the mechanisms of putative the Erlitou enclosed courtyard style political control remain unknown. If rammed-earth platform palace-temple, the analogy to the Anyang period can be and circumscribed palace-temple area made, lacking the infrastructure of later at the center of the settlement formed Qin-Han–type imperial control, politi- the basis of Erligang and, indeed, later cal relationships even within the Cen- imperial Chinese architecture (Thorp tral Plains cultural sphere were likely 1991). indirect, mutable, and based on ritually reinforced kinship hierarchy, alliance, The Erligang period is also character- and sporadic, rather than routine, ized by some new developments, how- mechanisms of coercion. It is also likely ever, such as the appearance of aseries that—as with the Zhou dynasts, who set of Central Plains metropolitan ceramic up statelets in strategic areas after the tradition sites surrounded by rectan- conquest of the Shang—the political, gular rammed-earth walls oriented to economic, and cultural relationships 10–20 degrees east of north. These sites between sites changed over the course are distributed from Hubei to southern of their occupation, each site and each Shanxi.27 In fact, the change in the orien- region having its own local historical tation of major architecture from about trajectory related to, but not necessarily 5degrees west of north at Erlitou, to the determined by, the fate of the cultural 10–20 degrees east of north in the Erli- and political core. gang period, is viewed by many scholars as evidence conﬁrming the received tex- Looking at the Erligang expansion tual record of dynastic change.28 Some from the point of view of the ceramic

100 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3 traditions on which it is largely based, The civilizational sphere of influence several issues immediately present centered on the Zhengzhou polity, themselves. The first is the relationship though expansive, was neither homoge- between ceramic production and dis- neous nor unique within the area cur- tribution. If, as we tentatively proposed rently occupied by the PRC. In Sichuan, for Erlitou times,29 production is mostly the Sanxingdui tradition flourished, small-scale and distribution limited, while lower down the Yangzi, local so- then ceramic tradition changes can be cieties responded to Central Plains cul- more closely linked with the movement tural and perhaps political intrusion of people. However, given that the Erli- with avariety of responses, exporting gang-period ceramic tradition and vari- northward their characteristic stone- ant boundaries tend to be soft, whatev- ware and protoporcelains even as they er social or demographic changes were absorbed bronze-casting techniques reflected in the Erligang expansion, and perhaps other cultural forms. To they generally do not appear to have the west, north, and northwest in west- involved wholesale population move- ern Shaanxi and Gansu, Inner Mon- ments and certainly not replacements. golia, and northern Shanxi provinces, Rather, as in the case of Panlongcheng, local traditions, though showing in- where there is amixing of Erligang and creasing interaction with the Erligang local ceramic traditions, both Erligang period cultural world, nevertheless and local populations may have coexist- preserved material cultural traditions ed,30 despite it apparently being anorth- (including bronze-casting) and likely ern colony.The second issue is the na- social organization and lifeways differ- ture and developmental history of the ent from those of the Zhengzhou core. Erligang-period ceramic traditions and In Liaoning and Inner Mongolia in the the issue of ethnogenesis. If Zhengzhou northeast, the Lower Xiajiadian–tradi- ceramics show evidence of multiple tra- tion area was still home to societies liv- ditions whose variety gradually and syn- ing in networks of stone fortified sites, thetically changed into uniformity over who shared certain cultural forms with the generations of the site’s occupa- other sites across abroad expanse of tion, its seems more likely that political the north and northwest, while in the identity and culture continued to evolve east the Yueshi areas, while showing within and in interaction with the great increased contact with the Erligang-pe- settlement at Zhengzhou (Wang 2009) riod cultural sphere, was nonetheless than that they were the unchanging distinct. Although in its time, the Erli- property of a“Shang” ethnic group who gang-period urban site at Zhengzhou occupied the Xiaqiyuan-tradition area may have stood at the center of the larg- during Erlitou times. Rather, as the est and most influential sphere of civi- great center brought together people in lization in contemporary East Asia, the hitherto unprecedented numbers and elites at Zhengzhou nevertheless inhab- densities, new identities, social forms, ited acomplexculturallandscape, with and ways of life (Yoffee 2005) were multiple and multidirectional networks forged.31 of resources, culture, and knowledge.

101 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Endnotes the steps through which the authors arrived at 1Henansheng wenwu kaogu yanjiusu their absolute dating and thus difficult to assess its [HWKY] (2001) states that the Erligang-period site rigor.Moreover, it should be noted that there has was 25 km2.This estimate takes into account the been along-standing rivalry between the Yanshi distribution of Erligang-period cultural deposits Shangcheng and Zhengzhou Shangcheng archaeo- outside of the walls. logical teams revolving around which site is earlier, 2These are Xiaqiyuan, Erlitou, Yueshi, and and there remains some controversy concerning Luwangfen-Songyao. Indeed, at least according to the dating of both these sites, which, on present ZSKY (2003:164–168), the early occupation of the evidence, appear to be roughly contemporaneous site included at least three separate but overlap- at around 1600 BCE or perhaps abit earlier ping ceramic traditions: Erligang, Nanguanwai, 4Whether or not the wall actually sur- and Luodamiao. Erligang is said to be “Shang” (i.e., rounded the entire site is in doubt, and it may have mostly Xiaqiyuan in origin), Nanguanwai is said to been only an outer defense-works and/or levee on have heavy Yueshi influences, and Luodamiao is the most vulnerable side of the site. The possible apparently adevelopment of Erlitou ceramic tradi- defensive nature of the wall is further suggested tions. With this the authors of the XiaShang Juan by the discovery of apit just inside the outer wall stress the cosmopolitan origins of the Zhengzhou with acache of over 30 stone ge-dagger-axes (ZSKY center and, by implication, the “Shang” polity in 2003). general. 5Thisisgenerally true of all important ar- 3Recent work at the site has indicated that chitecture (large foundations, tombs, and most of there were rammed-earth structures being built in the walls in the Erligang to Anyang periods, unlike the “palace-temple” area and that the inner walls Erlitou, where buildings are oriented slightly west were built in the Luodamiao period before Erli- of north. In the Erligang period this generaliza- gang I(i.e., during the end of the Erlitou period). tion is true of Zhengzhou, Yanshi, Panlongcheng, As for the absolute date of Lower Erligang, radio- Dongxiafeng, and Fufeng (see below). carbon dating puts Zhengzhou’s Luodamiao site 6Thismay in part be due to the fact that the at 1670–1640 (BCE) and Lower Erligang phase I southern half of the site was intensely occupied in at 1600 (B.C.). “Now we have discovered that con- Han and Tang periods destroying any earlier cul- struction on the Shang center began before Lower tural deposits ther.(ZSKY 2003). Erligang and thus the Zhengzhou Shang center was 7This, of course, does not rule out the pos- constructed between 1640 and 1600 B.C.” (Yuan sibility that it was also aroyal center and, indeed, and Zeng 2004:66; Yuan and Zeng 2005 [an Eng- the large palace-temple structures suggest that lish version of the same article]). Liu Xu,however, it was an important ritual, administrative, and/ notes that there is some conflation of “Luodami- or economic site. Unfortunately, without clearer ao” and “Erligang I” stratigraphic levels involved understanding of what sorts of activities went on in this assessment (lecture notes Nov.2005, Peking in which types of buildings and who had access to University), while the C-14 dates, some with error them and who did not, it is difficult to say much for margins as small as ±17years were apparently de- certain about the precise nature of the site except rived through “wiggle-matching,” which brought that it has the same type and scale of elite struc- assumptions about stratigraphic relationships of tures seen at Anyang, which can be unambiguously units from different parts of the site into calcula- associated with the Shang kings. It may be that in tions of absolute dating. Moreover, without the the Erligang period, as in the Western Zhou case, uncalibrated C-14 dates, it is hard to reconstruct there was more than one capital. There is avolumi-

102 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3 nous Chinese-language literature on the subject, find discrete, kin-based living areas with amixture the bibliography for some of which can be found in of elite and nonelite residences. Unfortunately, ZSKY (2003). this, too, remains an untested hypothesis. 8Atthe beginning of this phase, it is like- 15 These are pits dug below the waist of the ly that Erlitou was larger than the Erligang site at principal interee, usually containing adog, or in Yanshi, asituation that reversed by the end of Erli- the royal tombs at Anyang, an armed person. The gang phase II. “waist pit” is widely believed to be aparticular 9Earth and wood architecture would, on characteristic of “Shang” tombs. The absence of the one hand, require constant maintenance and, waist pits in the Dongxiafeng variant perhaps sug- on the other, require less effort to build in the first gests different mortuary practices in this area. place than architecture of stone. 16 It is named for the Liulige site, Hui Coun- 10 Or, in other words, the intrasite constant ty, Henan (ZSKY 1956). refashioning of the ritual/administrative/econom- 17 This might simply be due to adifference ic built environment may be related to the frequent in local clays. While color is frequently used as a intersite rebuilding of the same. factor in the determination of archaeological cul- 11 Palace-temple number 5was almost 1ha ture in China, no attempt is made to understand in size, the largest Erligang-period building and the the practices behind the difference. The problem second largest Central Plains Bronze Age building with this approach is that it potentially lumps to- discovered thus far (the Xiaoshuangqiao-Huanbei gether different practices producing similar ar- period Huanbei F1 is around 1.6 ha in size). tifacts and separates different superficial effects 12 Nevertheless, there is no direct positive that may have been produced by similar practices. evidence concerning the use of these buildings, 18 Panlongcheng would be considerably and to my knowledge, neither flotation nor soil- larger if the locations outside the walls were taken chemical analyses were performed (nor, given into account. Liu and Chen 2012, citing an obscure Chinese excavation practices during the 90’s, even study claiming to have found outer walls (Liu Sen- sieving). Interpretations of the function of these miao(?) 2002), suggest the real site size is 290 ha. buildings range from royal treasuries/storehouses This is probably ahigh estimate. (Wang 2000), to armories (Underhill 2006). 19 Hubeisheng wenwu kaogu yanjiusuo 13 This (ZSKY 2003) account is apparently [HWKY](2001:84–87, 97–100) describes 50-m- based on firsthand, unpublished information, as long “kilns” dating from late Erlitou and Erligang no citations are given, and asearch through the period Iand II south of the walls. ZSKY (2003) various preliminary reports on Yanshi have failed notes more cautiously that it was asite related to to turn up any mention of ordinary residences. ceramic production (based on charcoal deposits 14 This makes it difficult to agree with Liu and high densities of sherds). If these were single and Chen’s (2003) claim that residences were strat- kilns, they resemble no other kilns known from ified and spatially discrete, with the elites living in the period, north or south. Trenches with charcoal, the “palaces,” and the commoners living outside large-mouthed gang-urns (缸), labeled “crucibles” the palace zone. In fact, based on the current state by some authors (e.g. Liu and Chen 2003) and slag of evidence, we know that there are large structures were also discovered at the site, indicating that that may have functioned as “palaces” but lack any some sort of metallurgy occurred at the site, but direct evidence concerning where different eco- whether this included casting vessels or simply re- nomic classes lived. If social organization at Yan- fining is presently unknown. shi was anything like Anyang, we might expect to

103 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

20 Whether these elites originated in Zheng- that the mixed assemblage of Wucheng Ihad its or- zhou or were local leaders to some degree incorpo- igin in “a branch of Central Plains Shang culture” rated into the Central Plains Bronze Age cultural and that when these people arrived in Wucheng, orbit is unknown. their culture came into contact with local cultures, 21 So named for the Eastern Zhou polity and in the mixture of traditions Wucheng culture known from transmitted texts and believed to be was born (JBKY,ZB2005:413). located in that area. 26 The Wannian tradition has also been 22 Inscriptions are all found on stone or ce- called the Xiaojiashan-Jiaoshan type (ZSKY 2003). ramic artifacts and are short (the longest is 12 char- 27 There were, of course, walled Long- acters). Despite being undeciphered, it is often shanoid and Erlitou-tradition sites prior to the claimed that they are the same system as the Shang Erligang period, but never before adhering to a oracle-bones. Some characters do indeed look like common architectural program over such alarge oracle-bone graphs, but there are not enough long region. strings of graphs to provide context for decipher- 28 Theorientation of major architecture ment, or determination of afﬁliation, if, indeed, it again changes with the Zhou. is ascript. 29 While there is some evidence for more 23 Including the Ruichang Tongling min- specialized ceramic production and possibly great- ing site. Both ZSKY (2003) and JBKY,ZB(2005) er distribution, the general picture (albeit based on give somewhat conﬂicted accounts of the “cul- scant evidence) seems to be that of local produc- tural identity” of the deposits at Tongling. On tion and consumption. the one hand, ZSKY (2003) in its section on the 30 This might have taken the form of a Panlongcheng variant lists Tongling as a“Middle Zhengzhou elite ruling over alocal population Shang” Panlongcheng-variant site even while in its (Bagley 1999) and/or close economic, social, and Wucheng section it lists it as aWucheng-tradition political ties to sites in the north that saw avariety site. Jiangxi Institute et al.[JBKY,ZB] (2005),while of forms of interaction from marriage to trade and claiming that Tongling is a Wucheng site, notes possibly war. that the northern Wucheng sites actually should “belong to the Panlongcheng Shang culture variant”(419). Part of the confusion seems to stem from the practice of trying to construct organic cultural wholes out of mixed assemblages of potentially varied origin. 24 There is actually some discrepancy between ZSKY (2003) on the one hand and Peng (2004) and JBKY,ZB(2005) on the other.The latter include an earlier phase than Wucheng I, based on limited excavations at Longwangling and do not take into consideration the now widely accepted Middle Shang period (Xiaoshuangqiao-Huanbei period) in their comparative ceramic dating. 25 Following the pottery-equals-people assumptions of most Chinese archaeological practice, the authors of the Wucheng report suggest

104 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 3

105 READ ONLY/NO DOWNLOAD

Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 4 Xiaoshuangqiao- Huanbei Period (ca. 1400–1250 BCE)

Xiaoshuangqiao-Huanbei work done in the 1990s on the Xiaosh- Period Metropolitan Traditions uangqiao site near Zhengzhou (Henan- sheng kaogu yanjiusuo [HKY] 1993; The Xiaoshuangqiao-Huangbei pe- Henansheng wenwu kaogu yanjiusuo riod is usually referred to as the “Mid- [HWKY] 1996) and the discovery of the dle Shang” period in the Chinese lit- walled center at Huanbei in 1999.3 As erature (ZSKY 2003). The term “Middle such, it is less well understood than the Shang,” however, originally referred preceding Erligang or following Anyang to the Erligang period in older perio- periods. Important Central Plains met- dization schemes.1 The use of “Middle ropolitan sites of this period in addition Shang” for the period between Erligang to the above-mentioned Huanbei and (and including upper Erligang II) and Xiaoshuangqiao sites are Gaocheng Anyang was ﬁrst put forward by Tang Taixi, Xingtai Caoyanzhuang, Anyang Jigen (Yang and Tang 1999)2 and has Sanjiazhuang, Xiaotun, and Jinan Dax- only received wide recognition in Chi- inzhuang (ZSKY 2003). Perhaps the two nese archaeological circles in the last most striking features of this period are ten years, in large measure due to the an apparent “shift” of the metropolitan READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

4.1. Xiaoshuangqiao-Huanbei–period ceramic traditions (base map from Harvard Geospatial Library).

center from the Zhengzhou-Luoyang re- Xiaoshuangqiao-Huanbei gion to the Anyang area in phase II and Phase I the maximum distribution of metropolitan-variant ceramic traditions and met- During this period, which also corre- ropolitan-style bronze vessels (see Fig- sponds to Erligang IV,the bronze found- ure 4.1). This period can be divided into ries at Nanguanwai and Zijingshan three phases: Xiaoshuangqiao-Huan- were still in operation, and Zhengzhou bei I, represented by Zhengzhou Baiji- was not yet abandoned, although the azhuang and Xiaoshuangqiao; Xiaosh- palace-temple area was no longer in uangqiao-Huangbei II, represented by use. The Baijiazhuang site, however, theearlyphasesofHuanbeiShangcheng just outside the northeast corner of and Gaocheng Taixi; and Xiaoshuang- Zhengzhou’s inner walls, has yielded qiao-Huanbei III, represented by the houses, tombs, and kilns all dating to later phases of Huanbei Shangcheng this period. Despite the apparent de- and Gaocheng Taixi (ZSKY 2003). cline in the Zhengzhou site as awhole,

108 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 4

Period Zhengzhou Yanshi Huanbei Panlongcheng Wucheng Panlongcheng Variant Distribution ES IX XX ES II XX Xx ES III XX Xxx MS IX XXX MS II XX XX MS III XXx LS IXx LS II-IV x X=period of flourishing; x=period of development or decline Table 4.1. Comparative chronology of important northern and southern sites. there are more large, bronze-vessel- has seen large-scale excavations since bearing tombs from this period than 1995 (HWKY 1996; ZSKY 2003). The previous ones (ZSKY 2003), and two of site contains large rammed-earth struc- the famous bronze “hoards” date to this tures (up to 500 m2),5 abronze-cast- period.4 The contents of these “hoards” ing site, many sacrificial pits, storage were assemblages of such unprecedent- pits, and amoat. In addition to bronze edly large vessels that they are widely weapons, tools, and vessels, abronze believed to have belonged to kings artifact described as an “architectural (Henansheng wenwu kaogu yanjiusuo; component” was discovered during ex- Zhengzhoushi wenwu kaogu yanjiusuo cavations (ZSKY 2003:275). This object 1999). Perhaps related to the decline of may have been used to cap abeam and Zhengzhou, Yanshi was largely aban- it had a taotie animal-face design. The doned during this period. Nevertheless, discovery of the putative beam fitting whatever the cause of the rapid demise is the first time bronze architectural of Yanshi Shangcheng at the begin- components have been discovered at a ning of Xiaoshuangqiao-Huanbei I, as Central Plains Bronze Age site and may well as the the decline of Zhengzhou indicate the high status of at least some throughout this period, the Zhengzhou of the occupants. Another interesting area remained the densest concentra- feature of the Xiaoshuangqiao site is tion of metropolitan tradition sites and the presence of Yueshi-type axes among was still apparently the locus of high the stone artifact assemblage, prompt- elite activities during Xiaoshuangqiao- ing some scholars to speculate about Huanbei I. Yueshi influences (ZSKY 2003); while this usually implies “ethnic” presence Xiaoshuangqiao in Chinese archaeological discourse, the axes could just have easily been ob- Located 20 km northeast of Zhengzhou, tained through trade. the Xiaoshuangqiao site is the largest known Central Plains site of this phase Speculations concerning the nature of at 150 ha. It was discovered in 1989 and this site range from capital, to second-

109 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

4.2. Huanbei walled center (after ZSKYAG,ZJHQKDK 2010:10, ﬁg. 1).

ary palace area, to royal ritual precinct, Xiaoshuangqiao-Huanbei but given the preliminary state of ex- Phases II and III cavations at the site, it is probably too Huanbei Shangcheng early to be sure. Whatever the nature of the site and whatever its relationship to During this period the partial abandon- metropolitan elites still in the Zheng- ment of sites, such as Zhengzhou6 and zhou area, this site was abandoned, Yanshi Shangcheng, was accompanied along with much of the Zhengzhou core, by the building of the large walled site at the end of Xiaoshuangqiao-Huanbei at Huanbei (470 ha)7 justacross the I, when the center of metropolitan cul- Huan River from the later metropoli- ture apparently shifted north to the An- tan center of Yinxu at Anyang (Figure yang area (HWKY1996). 4.2). Discovered only in 1999, this site lies buried under sediments up to 3m deep. The wall that surrounds the site appears to have afoundation trench approximately 9minwidth.8 While relatively little of the site has been excavated to date, awell-ordered area of

110 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 4

4.3. Huanbei palace-temple Foundation 1(AT,IA, CASS 2004b:22, fig. 1). large-scale rammed-earth structures dui, Zhong Jia Huanheliuyu quyu kaogu was discovered, including the largest diaocha ketizu [ZSKYAG, ZJHQKDK] Central Plains Bronze Age building (F 2010). Recent work on another build- 1) yet known at 1.6 ha in area (AWT,IA, ing, Foundation 2, south of Foundation CASS 2004b; Figure 4.3). Although work 1, has revealed some interesting con- at Huanbei is in arelativelyearly stage trasts. Foundation 1has anumber of compared with other major Central sacrificial deposits and is thought to be Plains Bronze Age centers, the results aroyal lineage temple (AWT,IA, CASS of what work has been done so far are 2004b; Figure 4.3). Foundation 2lacks extremely interesting. The large-scale sacrificial remains, uses layers of differ- rammed-earth structures, like those at ent-colored earth, and is thought to be other metropolitan centers since Er- residential rather than ritual in nature ligang, are oriented approximately 13 (Tang et al. 2010). degrees east of north. They occur in a walled 41-ha palace-temple precinct In contrast to the “palace-temple” pre- like those found at Erlitou, Zhengzhou, cinct, an intensively cored 32-ha area and Yanshi (Zhongguo shehuikexuey- in the east of the site appeared to be uan kaogu yanjiusuo Anyang gongzuo- mostly habitation and living-activity

111 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

4.4. A) Baijiazhuang variant; B) Caoyanzhuang variant (after ZSKY 2003:256; ﬁg. 5-3).

112 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 4 areas. According to the excavators, world during Xiaoshuangqiao-Huanbei “habitation areas, dominated with the I, even while Yanshi and Zhengzhou cultural deposits of buildings, are usu- went into terminal decline. Neverthe- ally surrounded by zones rich in refuse less, according to ZSKY (2003:255), deposits,” and “each habitation area is “Middle Shang phase two and three separated from another by zones with [i.e., Xiaoshuangqiao-Huangbei II–III] little or no cultural deposit”(AT, IA, remains for this area are still few and CASS 2004a:12). If this is correct, then it their characteristics not easy to general- suggests that Huanbei may have had a ize.” clustered residential pattern, afeature that, as we will see below, it shares with Xiaoshuangqiao-Huanbei Anyang. Rammed-earth foundations Tradition, Caoyanzhuang of relatively large-scale buildings were Variant also occasionally discovered both within and between habitation clusters, al- Developing out of the Liulige variant of though their function can only be spec- the Erligang tradition, and taking Xing- ulated at present. Sacrificial remains tai Caoyanzhuang as its type-site, the have also been discovered associated core area of this variant is in northern with both Huanbei’s walls (AT, IA, CASS Henan and southern Hebei. It is distrib- 2004a) and palace-temple Foundation uted in the area formerly occupied by 1(Du 2005). Buildingsatthe site gener- the Liulige variant, as well as the south- ally have an upper layer of burned de- ern part of the Erligang Taixi variant bris, and flood deposits across the site (Figure 4.4B). As mentioned above, in suggest it may have been flooded before Xiaoshuangqiao-Huanbei II and III, in it was abandoned (AT, IA, CASS 2004a). tandem with the building of Huanbei, this area appears to become the core of Xiaoshuangqiao-Huanbei the Central Plains metropolitan tradi- Period Metropolitan Tradition tion (ZSKY 2003). It is generally argued Variants that the Caoyanzhuang variant kept es- Xiaoshuangqiao-Huanbei sentially the same assemblage as that Tradition, Baijiazhuang Variant found at Baijiazhuang during Xiaosh- uangqiao-Huanbei Iand, thus, shows The type-site of this variant is Baiji- continuity (ZSKY 2003), even while cer- azhuang. The Baijiazhuang variant is tain ceramic types show local charac- primarily distributed in the Luoyang- teristics (ZSKY 2003; Figure 4.4). More Zhengzhou area, particularly the Zheng- recent, unpublished work, however, has zhou area. This variant develops direct- found that Huanbei Shangcheng shows ly from Erligang ceramic traditions, and marked local characteristics, and the sites of this type are most plentiful dur- supposed continuity with Baijiazhuang ing Xiaoshuangqiao-Huanbei I(Figure is problematic (Tang Jigen, personal 4.4A). As mentioned above, this area communication). It is to be hoped that appears to still have been the center of future work, based on more than formal the Central Plains cultural and political ceramic typology, will resolve the issue

113 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

have “waist-pits” with dogs in them, and 10 percent of the tombs have death attendants. Taixi-variant sites are most plentiful in Xiaoshuangqiao-Huanbei periods II and III, and the ceramics of at least one site is said to show “northern cultural inﬂuences” (ZSKY 2003:263).

Xiaoshuangqiao-Huanbei Tradition, Daxinzhuang Variant

The Xiaoshuangqiao-Huanbei period Daxinzhuang variant is adevelopment of the Erligang tradition Daxinzhuang variant (Figure 4.5D). This variant is distributed in Shandong Province north of the Taiyi mountains. In this period, the Daxinzhuang site was 30 ha in size, and houses, tombs, and middens have been found. The tombs all have waist- pits, and some have bronze vessels and jades. This variant still show Yueshi-tradition inﬂuences in some of its ceramic assemblage (ZSKY 2003).

4.5. C) Taixi variant; D) Daxinzhuang variant Xiaoshuangqiao-Huanbei (after ZSKY 2003:256; ﬁg. 5-3). Tradition, Panmiao Variant

of continuity and change between Xia- Distributed in southeast Shandong oshuangqiao and Huanbei. Province, this variant developed out of Xiaoshuangqiao-Huanbei Iand is an ex- Xiaoshuangqiao-Huanbei pression of the continual eastward ex- Tradition, Taixi Variant pansion of Central Plains metropolitan ceramic traditions from the Erligang This variant is adevelopment of the period onward (Figure 4.6E). Like other Erligang tradition Taixi variant (Figure regional variants, Panmiao-variant sites 4.5C). The Gaocheng Taixi site that is show differing degrees of “local ﬂavor” the type-site of this variant was occupied in their ceramic assemblages, perhaps from Erligang III to Xiaoshuangqiao- indicating interaction with non-Central Huanbei III (ZSKY 2003). Both houses Plains traditions or the retention of and tombs were excavated at this site, some aspects of older indigenous forms although no large-scale structures have and styles. In addition to the Panmiao been found to date. Some of the tombs site, Anqiugucheng is another site of

114 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 4 this variant, with strata spanning Xia- oshuangqiao-Huanbei to Anyang periods.

Xiaoshuangqiao-Huanbei Tradition, Dachengdun Variant

This Xiaoshuangqiao-Huanbei tradition Dachengdun variant developed out of the Erligang-tradition variant of the same name and is distributed in the same area (Figure 4.6F). Although the ceramics during this period are said to show strong local characteristics (while still having enough common Central Plains metropolitan characteristics to qualify as avariant), the bronze artifacts found in this area dating to this period are identical to those of the metropolitan centers (ZSKY 2003).9 No contemporaneous large sites have been found, and, at present, relatively little archaeological work has been done in this area.

Xiaoshuangqiao-Huanbei 4.6. E) Panmiao variant; F) Dachengdun variant Tradition, Panlongcheng (after ZSKY 2003:258; ﬁg. 5-3). Variant

The Panlongcheng site remained han, has deposits that date from Xia- the center of the Central Plains met- oshuangqiao-Huanbei II to perhaps An- ropolitan tradition Panlongcheng yang I(ZSKY 2003) as well. variant. The distribution of this variant greatly expanded during the Xi- South of the Yangzi, Panlongcheng- aoshuangqiao-Huanbei period, and variant sites take the form of scattered new Panlongcheng-variant sites ap- outposts (ZSKY 2003). These sites in- peared (Figure 4.7). Nevertheless, Pan- clude the copper mine site of Ruichang, longcheng itself was apparently not which dates from late Erligang or early occupied beyond Xiaoshuangqiao- Xiaoshuangqiao-Huanbei, and the Huanbei II (ZSKY 2003), while bronze nearby Jiujiang Shendun, which has hoards at Yingcheng Wuci and Qunli- strata dating between Xiaoshuangqiao- cun date to Xiaoshuangqiao-Huanbei Huanbei II and III. II and, perhaps, Anyang I, respectively. Xianglushan in Xinzhouxian, near Wu-

115 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

nomic, and political networks may have played an important role in facilitating this exchange. Moreover, the ceramic assemblages at Panlongcheng and other sites of this variant continued to display local characteristics right through to the end of their occupations (ZSKY 2003). In addition, it would seem that rather than being completely bound to the fate of the Zhengzhou metropolitan center, the Panlongcheng site, especially its ceramic-tradition variant, reached its maximum distribution and influence during and after the period that Zhengzhou was being largely abandoned (Table 4.1). Nor, as lead isotope 4.7. Panlongcheng variant (after ZSKY studies(Jin et al. 1990, 1994, 1995, 1998) 2003:260; fig. 5-3). and aprotoporcelain sourcing study (Chen et al. 1999) have suggested, did The Tonggushan site near Yueyang, extensive contacts within the Yangzi occupied since Erligang II, has non- area cease after the retreat of Central Central Plains–tradition strata overlay- Plains ceramic traditions in the Anyang ing Xiaoshuangqiao-Huanbei II strata. period. Indeed, even the Panlongcheng-variant strata of the site are “relatively compli- Xiaoshuangqiao-Huanbei cated” with Central Plains–dominated Tradition, Beicun Variant assemblages showing “obvious local influences” (ZSKY 2003:268). These “in- Originating in Erligang III, this Shaanxi fluences” are said to include Jingnansi, Province–variant of Central Plains met- and possibly Wucheng and Baota tra- ropolitan tradition ceramics spread ditions—doubly significant in light of west during the Xiaoshuangqiao-Huan- Bagley’s contention that large bronze bei period, reaching the Fufeng-Qishan bells found in the Wucheng-tradition area (Figure 4.8G). In addition to the tomb at Xin’gan “connect Wucheng Beicun site, Xiaoshuangqiao-Huanbei with abroad cultural province that II or III ceramics were found at Fufeng seems to have embraced the lower Yang- Baijiayao, while Fufeng Yijiabu has Xi- zi region and, perhaps somewhat later, aoshuangqiao-Huanbei III strata over- the middle Yangzi as well” (1999:173). lain by three strata of Zhengjiapo- and It would seem that even in Erligang and Liujia-tradition ceramics—traditions Xiaoshuangqiao-Huanbei times, there frequently associated with “Proto-Zhou was intense interaction in the middle culture” (see chapter 5). Xiaoshuang- and lower Yangzi regions and that Pan- qiao-Huanbei– period Central Plains– longcheng and its putative cultural, eco- style bronze vessels have also been

116 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 4

4.8. G) Beicun variant; H) Xiaoshen variant (after ZSKY 2003:259; ﬁg. 5-3).

117 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

found in this area. At the same time, shift of the metropolis to the Anyang at the Qishan Wangjiazui site Beicun area, south and southwestern Shanxi variant, Xiaoshuangqiao-Huanbei I–III underwent adramatic reduction in strata are overlain by non-Central Plains sites (ZSKY 2003). This claim, however, traditions in the Anyang period (ZSKY is with the important caveat that when 2003). Like other Central Plains metro- the survey upon which this assessment politan tradition variants, the Beicun rests was made (ZSKYSD 1989), the variant had “many vessels with substan- Xiaoshuangqiao-Huanbei period had tial local flavor” (ZSKY 2003:270), coex- not been put forward, and it is possible isting with more typical Central-Plains that some of the sites said to be of ei- vessels, indicating some form of inter- ther late Erligang or Early Anyang date action with, or continuation of, non- “may include Xiaoshuangqiao-Huanbei Central Plains traditions of ceramic cultural remains that at that time were production or use. not distinguished” (ZSKY 2003:270). At the least, recent research has suggested Xiaoshuangqiao-Huanbei that the Dongxiafeng site has Xiaosh- Tradition, Xiaoshen Variant uangqiao-Huanbei Iremains (Wang 1998), and Yuanqu IV is also said to be The Xiaoshen variant is mostly distrib- within the Xiaoshuangqiao-Huanbei uted in southeast and middle Shanxi. period. Thus, while information is still The type-site of Zhangzhi Xiaoshen is somewhat sketchy, it would seem that 18 ha in size and has fourth-to-late-first- based on present (and admittedly un- millennium BCE occupations (Figure systematic) evidence, the major Erli- 4.8H). The Central Plains metropolitan gang sites in the Dongxiafeng–variant tradition remains from this site date to area were abandoned during Zheng- either Xiaoshuangqiao-Huanbei II or III zhou and Yanshi’s decline, and in gen- (ZSKY 2003). Xiaoshuangqiao-Huanbei eral there appears to have been adepop- II and III bronze vessels have been dis- ulation of the area. covered in this area (Wang 1982). Al- though at present the Xiaoshuangqiao- Beyond the Xiaoshuangqiao- Huanbei–period culture history of this Huanbei –Tradition Areas area is poorly understood, minimally, it would appear that metropolitan-style The North elites were active in the region, along with some form of intensive interaction During the Xiaoshuangqiao-Huanbei with the metropolitan core region cen- period, the Jinzhong tradition dis- tered on Huanbei. played increasing Central Plains influences in its ceramic tradition over time Xiaoshuangqiao-Huanbei (ZSKY 2003). The shift of the metropoli- Tradition, Dongxiafeng Variant tan center to Northern Henan may have been related to this phenomenon, not During the Xiaoshuangqiao-Huanbei to mention the appearance of the Xia- period, especially after phase Iand the oshuangqiao-Huanbei Xiaoshen vari-

118 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 4 ant in southeastern and central Shanxi, contiguous with the distribution of the Jinzhong tradition.

The Xiaoshuangqiao-Huanbei period corresponds with phase IV of the Zhu- kaigou tradition and shows strong Cen- tral Plains inﬂuence. Indeed, one tomb from this phase (M1052) had an entirely Central Plains metropolitan burial assemblage (ZSKY 2003). Zhukaigou IV is said to span the Xiaoshuangqiao-Huan- bei period and perhaps even include some early Anyang period remains. In general, it appears that the Xiaoshuang- qiao-Huanbei period saw an increased interaction with the Zhukaigou-Jin- zhong tradition area, perhaps related to the northward movement of the metropolitan center and the processes and events that resulted in the northward and westward spread of Central Plains 4.9. Baoshan-tradition ceramics (after ZSKY metropolitan ceramic variants at the 2003:522; ﬁg. 8-21). time.

In the northeast, the Datuotou tradition may still have been distributed north and east of the Xiaoshuangqiao-Huan- bei Taixi variant, but the chronology of this tradition is uncertain. The Datuo- tou tradition may already have been replaced by the Weifang III tradition.10 It is not entirely clear what tradition was distributed in the former area of the Xiajiadian tradition during this period.

The East

The Xiaoshuangqiao-Huanbei period saw the continual eastward creep of Central Plains ceramic traditions south of the Taiyi Mountains, as well as in the north. Based on ZSKY (2003), aclear

119 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

judgment cannot be made as to which was the period of Wucheng’s zenith. Yueshi variants remain during this pe- During this period (Wucheng II), the riod or whether their distribution goes wall around the site was ﬁnished, and beyond ascenario of gradual replace- the site had reached 60 ha in size, mak- ment. As noted in Cohen (2001), the ing it comparable to the only other Yueshi tradition and its many variants known contemporary site of its size in are still not well understood, nor was the area, Panlongcheng. However, de- the eastward spread of Central Plains spite ten seasons of excavations begin- metropolitan ceramic tradition vari- ning in 1973, the overall structure of the ants ablanket replacement of previous site is still not well understood.11 Kilns, Yueshi traditions (Cohen 2001; Fang bronze-casting-related middens, hous- Hui, personal communication Oct. 19, es, and tombs have all been discovered 2005). at the site, which also includes what the excavators have called a“ritual area” The South (JBKY,ZB2005).12

As in theErligang period, Anhui was oc- As mentioned earlier,there is some cupied by the Dachengdun variant of controversy over the “cultural”affilia- the Xiaoshuangqiao-Huanbei tradition, tion of Wucheng,with the writersofthe while the neighboring Nanjing area was Wucheng report (JBKY,ZB2005) taking occupied by the Hushu tradition. The up the idea that survivors of the Pan- Lake Taiand Shanghai region is unclear longcheng collapse fled to Wucheng, at present beyond the observation that which, according to this story, had been the Maqiao tradition apparently dates originally founded by an intrepid group only to the Erlitou and Erligang periods of adventurers from theCentral Plains. (ZSKY 2003). Although at Wucheng there are definite- ly artifacts of Central Plains influence In Jiangxi province the Wucheng tradi- or origin, the Wucheng site, unlike Pan- tion flourished as the Panlongcheng longcheng, does not haveany northern variant of Xiaoshuangqiao-Huanbei architectural features. The site’s walls tradition reached its apogee, and the are not made of rammed earth,they Hubei, northern Hunan, and western arenot rectangular and oriented 10 to Jiangxi regions became linked in net- 20 degrees east of north, no rammed- works of interaction. Indeed, with the earthstructures have been found at heterogeneous assemblages, number the site at all, and none of the 23 buri- of traditions involved, and the variety als excavated at the site have waist-pits of interactions implied, this entire area or other of the Central Plains features might tentatively be termed the “Pan- of this period.13 The moldsfound at the longcheng interaction sphere.” site so far are all stone-molds for casting weapons and tools, and the kilns are of atype that is not found in the north.14 The Wucheng Tradition Moreover, it seems that the hypothesis The Xiaoshuangqiao-Huanbei period that Wucheng’s zenithcame only after

120 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 4

4.10. Baoshan-tradition bronze artifacts (after ZSKY 2003:518, 519; ﬁg. 8-20).

Panlongcheng’s eclipse is predicated on it can be said that Sanxingdui mate- acomparative chronology thatdoes not rial culture and influences can be seen recognize the Xiaoshuangqiao-Huan- in both the Yangzi gorges area and in bei15 period. If the ZSKY(2003)synthe- southern Shaanxi (Baoshan tradition). sis is correct, then Wucheng and Pan- If the Xiaoshuangqiao-Huanbei was the longcheng flourished together for overa period of the maximum distribution of hundredyears before Panlongcheng de- Central Plains metropolitan tradition clined and was abandoned (Table 4.1). material culture, it also seems to be the Moreover, the first half of Wucheng’s apogee of Sanxingdui influence.16 phase II corresponds to the period of the Panlongchengvariant’smaximum dis- In the Hanzhong region of southern tribution,atime when the middle and Shaanxi, the Baoshan tradition appears lower Yangzi were connected in aweb at this time (Figure 4.9). Named for the of intense contactand interaction. Nev- Chenggu Baoshan site (Xibeidaxue wen- ertheless, much more workneeds to be bo xueyuan [XWX] 2002), the Hanzhong done in Jiangxi and the Yangziareaand area has seen multiple discoveries of agreater use of absolute dating meth- the artifacts of astriking local bronze ods might help clear up some of the un- industry17since the fifties (ZSKY 2003; certainties concerning chronology. Luo 2010). These finds generally took the form of hoards buried in mounds, The Southwest often overlooking rivers. These mounds have no burials associated with them On the Chengdu plain, Sanxingdui sat (ZSKY 2003). According to ZSKY (2003) at the center of aweb of connections the bronze artifacts of this area can be and interactions the nature and extent divided into three groups, ranging from of which are still unclear.Minimally arelatively few Central Plains metropol-

121 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

itan vessels,18 to more frequent copies The West of metropolitan bronze types with local characteristics,19 and to artifact types Although there were certainly non- that are particular to the Hanzhong area Central Plains traditions in central (Figure 4.10). Among the latter artifact Shaanxi and farther west during the types are supposedly bronze masks that Xiaoshuangqiao-Huanbei period, the also occur in the Anyang period Central culture history and chronology of this Plains metropolitan variant site of Lao- area is unclear aside from the western niupo near Xi’an and the Xin’gan tomb expansion of the Beicun variant. This in Jiangxi, suggesting contacts between westward expansion was then followed these areas (see chapter 5). Atriangular- by replacement with non-Central Plains shaped ge dagger-axe, or kui 戣,typical traditions west of Xi’an toward the end of this area is also found in Sanxingdui of the Xiaoshuangqiao-Huanbei period, tradition sites and later in sites as far or perhaps alittle later. away as Anyang. Scepter and sickle like artifacts, on the other hand, are only Further west, the eastern Gansu areawas found in Hanzhong and are most likely stilloccupied by Siwaand Xindiantradi- to be local products (Luo 2010), acon- tions in the Xiaoshuangqiao-Huanbei clusion backed by metallurgical stud- period.What interaction,ifany, existed ies (Mei et al. 2009; Chen et al. 2009). In between societiesinthese areas andso- fact, rather than seeing the Hanzhong cietiesfurther east inShaanxiisunclear bronzes in terms of ametropolitan ver- on presentevidence. sus local dichotomy on stylistic (Luo 2010) and metallurgical grounds (Chen Conclusion et al. 2009; Mei et al. 2009), it would be more accurate to characterize them as The Xiaoshuangqiao-Huanbei period is having avariety of origins, including the least well-known period of the Cen- central Shaanxi (perhaps the masks), tral Plains BronzeAge.Understood by Sichuan, the middle Yangzi area, the some scholars as atransitionalorin- Hanzhong area, and the Central Plains termediary period, it saw the demiseof (Luo 2010). theZhengzhou centerand therise of a new central siteatAnyang(Huanbei). Stone cofﬁn burials have been found at In thesense that there was ashift in the some sites, and most burials appear to CentralPlains metropolitanarea, the be secondary, multiple burials. Accord- Xiaoshuangqiao-Huanbei period is in- ing to XWX (2002), there are strong sim- deedaperiod of transition.From the ilarities between the Baoshan tradition, perspective of materialcultural distri- Sanxingdui, the Three Gorges Area, and bution andcontinuity from theErligang western Hubei traditions, while Central period, however, it cannotbeconsidered Plains and western Shaanxi elements an intermediate periodinthe sense ofa can also be seen. collapse or disintegration of the Central Plains metropolitan horizon.20 Rather, theXiaoshuangqiao-Huanbei is thepe-

122 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 4 riod in which CentralPlainsmetropoli- tion continued to expandand flourish tan material culture reached its greatest through the Xiaoshuangqiao-Huanbei distribution. It is unclear exactly where period. The MiddleYangzi areaingen- the expected mega-center was in Xiaosh- eralseemstohave experienced increas- uangqiao-HuanbeiI(or indeedifthere ingly intense interactions duringthis pe- was one). In Xiaoshuangqiao-Huanbei riod, andthe Wucheng site in Jiangxi was II, however, the site of Huanbei appears at itsapogee. Sanxingdui wasalsoatthe to be, in bothits size and its monumen- height of its materialculturalinfluence, tal architecture,acenter of royal propor- and there seems to havebeen interaction tions.21 Nor do suchelite traditions as betweenthe Upperand Middle Yangzi bronze-castingseem to have declined regionsduring this period. Inthe north duringthis period,instead showing and west, from the Zhouyuan in Shaanxi awider distributionthan ever before. Province,tocentral Shanxiand Hebei, Whileevidencefor diverging Central Central Plains metropolitan material Plainsmetropolitan-originatedlocal culture continued to expand its distribu- bronze-casting traditions appears in a tion, although themechanisms through number of regions duringthis period which thistook place are farfromclear (such as the Sichuan basin,the upper and likelyvariedfromregion to region.23 Han RiverValley, and the middlereaches The east also shows expanded influence of theYangzi),theyare associatedwith as Central Plainsmetropolitanceram- societieswhose material culture and ics and evidence for metropolitan-style practices had never been part theCen- elitesincreasinglyappear in Shandong. tral Plains cultural sphere.Rather than Rather thanunderstanding theErligang evidence forthe collapse of acentralized political landscapeinterms of awell- system of bronze vesselproductionand integrated,centralized “state,” the col- distribution(as perLiu and Chen 2003; lapseofwhich precipitated thefall of all Bagley 1999),the trajectory of bronze of its secondary centers, thecontinued vessel castingfromErlitou through to prosperity of mostpartsofthe Central the Xiaoshuangqiao-Huanbei might in- Plains culturalworld in the Xiaoshuang- stead be seenasanexpanding horizon qiao-Huanbei bespeaks their relative of specialized metallurgical knowledge independence of the fate of Zhengzhou. and elite artifacts.22 Moreover, giventhe paucityofsystemat- ic regional survey,the ongoing destruc- If theexact nature of the political land- tion of exposedsites,and thedifficulty scapeisunclear for the Erligangperiod, of finding those buriedunder theYellow it is doubly unclear for Xiaoshuangqiao- River floodplain, we shouldremind our- Huanbei times.The southern Shanxi selves thatweactually know very little area seems to havesuffered depopula- aboutregional settlement structure for tionalong with theZhengzhou-Luoy- any part of the secondmillennium BCE. angregion. Panlongcheng flourished a It maywell be that importantXiaosh- few generations beyondthe demiseof uangqiao-Huanbeisitesawait discovery. Zhengzhou,and then it,too,was largely abandoned, although its ceramictradi-

123 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Endnotes not have been complete when abandoned (AT, IA, 1See Chang (1980) for an English discus- CASS 2004a) sion of the early attempts to fit Erligang and Erlitou 9Bagley (1999:175–176) remarks on the into achronology that included Anyang. large size and fine quality of “transition-period” 2Bagley’s (1999) transitional period, based bronzes found in this area, stating that the “assem- on bronze vessel analysis, corresponds to the post- blages have the composition of standard northern Erligang parts of the Middle Shang period (MS II– burial sets, and bronzes of this size and quality can III). only have belonged to avery high stratum of soci- 3The discovery of the site was first reported ety.” in the Guangming Daily,January 8, 2001, in astory 10 ZSKY (2003) is somewhat contradictory titled, “Henan Anyang xin faxian Shangdai cheng- on this score. In the section on Datuotou, it notes zhi” (Newly Discovered Shang Period WalledSite at that the Datuotou strata at the Yuxian site are over- Anyang, Henan). lain by Erligang III remains, while at other sites 4These are the hoards at Zhangzhai nan Datuotou remains are overlain by Weifang III re- jie and Chengdonglu huimin shipinchang. The in- mains. Nevertheless C-14 dates taken from several terpretation of the these pits is controversial, with Datuotou sites (including Yuxian) put Datuotou be- some scholars claiming that they are sacrificial pits tween 1530 and 1300 BCE, straddling the Erligang (e.g., An 1993, 1997), while most believe that they and Xiaoshuangqiao-Huanbei periods. Despite were stashes buried during political turmoil by this, based essentially on the determination of the owners who could not take them with them in anal- Yuxian site as having Erligang III remains (Zhangji- ogy to Western Zhou bronze hoards (ZSKY 2003). akou Archaeological Team 1984), Datuotou is 5Although these foundations are relatively claimed to date from Erlitou to Erligang times. Five small by comparison with the largest rammed- pages later, however, in adiscussion of the dates earth structures at Zhengzhou, Yanshi, and Huan- of Weifang III, which directly overlays Datuotou at bei during their apogees, one of the foundations anumber of sites, the earliest Weifang III remains had ahigher-preserved height than any other sec- are said to date from Anyang I(ZSKY 2003). Provid- ond millennium BCE foundation found to date. ed that the judgment of Yuxian’s ceramic tradition 6Despite this picture of site abandonment affiliations are correct, one explanation might be at Zhengzhou, the bronze hoard at Nanshuncheng- that the Datuotou tradition did not disappear in a jie might date to as late as Middle Shang II or III blanket fashion, as is apparently assumed, and the (based on stylistic indicators) indicating possible intrusion of Erligang elements in the area did not elite activity at the site after Middle Shang I(ZSKY spell the end of the tradition. 2003). 11 As it stands, 5,189 m2 have been excavat- 7Itis, in fact, the largest site completely en- ed of the 62-ha site (less than 1percent). The site, closed within awall known to that point from Chi- moreover, has never been systematically surveyed na. While the Zhengzhou site is larger, the outer or cored, suggesting that claims about the site’s wall did not entirely encircle the site. The Huanbei division into production areas, ritual areas, living occupation, moreover, may have exceeded the area areas, and so on (JBKY,ZB2005), are premature. surrounded by the wall as well, so the true site size 12 The mound, building on top of it, road, may well have been larger (Anyang Team,Institute and area of postholes seem to have been given the of Archaeology, Chinese Academy of Social Scienc- label “ritual” based on their not having any obvi- es [AT, IA, CASS] 2004a). ous function and on the presumed necessity of the 8The excavators note that the site may large sites of early civilizations to have ritual cent-

124 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 4 ers. Moreover, whatever the nature of the so-called blades, an artifact type common at Erlitou and “ritual area,” it does not resemble anything found Sanxingdui, but always manufactured of jade or in the Central Plains. If these features are, indeed, stone in the Central Plains. related to ceremonies of some kind, they do not 20 In Willey and Phillip’s (1958:33) classic seem to be of the same sort conducted in the north. formulation, ahorizon is “aprimarily spatial conti- 13 The burials include afew that the exca- nuity represented by cultural traits and assemblages vators feel are secondary burials and at least one whose nature and mode of occurrence permit the as- urn-burial. While urn-burials occur in the Central sumption of abroad and rapid spread”(italics in Plains, secondary burial is not aCentral Plains original). While Bagley (1999) apparently has this Bronze Age mortuary practice. or asimilar definition in mind when he talks of an 14 Including one example the excavators Erligang horizon but claims there was none for An- termed a“dragon kiln.” It was roughly 8mlong yang, his assertion is based on bronze vessels alone and 1mwide with five vents. It was associated with and an overly narrow interpretation of the term. As both earthenware and stoneware. Other types of Willey and Phillips go on to clarify, “it is recognized kilns at the site had stoneware and protoporcelain that horizons based on cultural criteria unsupport- sherds as well, however, so it does not seem to be ed by independent dating may have considerable the case that “dragon kilns” were necessary to fire temporal depth” (33). Indeed, horizons may last protoporcelain ceramics. centuries, as in the Middle Horizon of Peruvian ar- 15 The speculative nature of this narrative chaeology (Mosely 2001). can be seen in the fact that the increase of Pan- 21 While Huanbei’s wall only encloses an longcheng-related ceramic traits in the Wucheng area of 4.6 km2,compared to the 13 km2 of Zheng- assemblage during Wucheng II is said to be the zhou and the over 30 km2 of the Anyang site at its result of the abandonment of Panlongcheng and apogee, the actual size of the Huanbei site is still the collapse of Shang power in northern Hubei, unknown and quite possibly exceeded the walled whereas the spread of Erligang ceramic features in area (Jing Zichun, personal communication June the Erligang period is associated with the putative 2006). Being aburied site, it will take an extensive expansion of Shang power.Ifincreasing material program of coring before the extent of the occupa- cultural influence can be the result of either the ex- tion beyond the site walls becomes known. Huan- pansion or collapse of political entities, then read- bei also boasts the largest building known from ing such “influences” in terms of particular politi- the Central Plains Bronze Age (“Palace” 1) and the cal scenarios without strong supporting evidence longest wall of any Central Plains site up to that seems very arbitrary indeed. point. 16 ZSKY (2003) claims that the most of the 22 Indeed, history is littered with examples Shang-type bronzes from artifact pit 1date to the of strategic technologies (whether economic or Middle Shang, corresponding with the maximum military) spreading beyond their original centers expansion of the Panlongcheng variant. of invention despite the best efforts at secrecy and 17 Between 1955 and 1990, 14 locations in control by their inventors (Hittite iron, Chinese Yang and Chenggu Counties, Shaanxi, have pro- silk, and atomic weapons are just afew examples duced 26 groups of artifacts totaling 654 bronze that spring to mind). artifacts (ZSKY 2003). 23 If analogy to the Anyang period can be 18 Apparently, only “10 percent of the arti- made, it may be that new polities were set up by facts” fall into this category (ZSKY 2003:520). metropolitan elites in some of these areas or by 19 This category includes bronze zhang- elites in adjacent areas that shared common tra-

125 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

ditions with the metropolitan area. Or it may be that local elites adopted elements of Central Plains elite culture, even while trade, marriage, alliance, and conﬂict provided opportunities for individuals and artifacts to travel between regions. While it is obvious that there is intensive contact throughout an ever-wider area in this period, the political implications of this fact are much less clear.

126 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 4

127 READ ONLY/NO DOWNLOAD

Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5 The Anyang Period (ca. 1250–1050 BCE)

Anyang (Yinxu) intensively excavated site in China.1 The The best-known site of the Anyang, or site of Anyang was apparently the capi- Yinxu, period, is the type-site, Anyang. tal, or central place, of the last Shang Though long mined by tomb looters, kings.2 Anyang reached asize of over this site was ﬁrst excavated in 1928 by 30 km2 at its zenith (ZSKY 2003) with a archaeologists from the newly formed “palace-temple” precinct nearly 70 ha Academia Sinica. Archaeological work in area (Figure 5.2). The royal tombs at was stopped in 1937 for the Japanese Xibeigang, north of the palace-temple invasion of North China, but excava- complex across the Huan River, are or- tions resumed in 1950 and contin- ders of magnitude larger than any previ- ued sporadically until 1958, when the ously known, while the richest unlooted Anyang Work Team was set up by the tomb found at Anyang, that of King Wu Chinese Academy of Social Science’s Ding’s consort Fu Hao, surpasses, in Institute of Archaeology (ZSKY 2003). number and quality, the tomb furnish- Since then, excavations have continued ings of any other tomb found in China unabated to this day in the Anyang area, of this or any earlier period. Thus, de- making Anyang the longest and most spite the currently widespread notion READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.1. Anyang -period ceramic traditions (base map from Harvard Geospatial Library).

that Anyang and the polity of which it chitecture, burial practices, and sacri- formed the center were pale shadows ﬁce. Nevertheless, the Anyang period is of the Zhengzhou zenith (e.g., Bagley notable for two important new develop- 1999; Liu and Chen 2003, 2012; Yof- ments: writing and the introduction of fee2005),3 by nearly any measure An- the chariot. The ﬁrst of these, although yang was unsurpassed to that point in possibly having unpreserved anteced- East Asia and perhaps the world. ents (Keightley 2006; Bagley 2004 but see Smith 2008 for the argument that Much of the material culture and many the script could have developed rap- of practices in evidence at Anyang had idly), appeared in two forms in the An- direct antecedents in the Erligang- yang period: oracle-bone inscriptions and Xiaoshuangqiao-Huanbei–period and, starting in Anyang phase II, short metropolitan centers, including the inscriptions on bronze vessels usually bronze, ceramic, and stone industries, taking the form of ancestral dedica- large-scale rammed-earth courtyard ar- tions. By the end of the Anyang peri-

130 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

5.2. Anyang site map (redrawn from Campbell 2007). od, bronze inscriptions that recorded ing, form of elite social and symbolic events (usually the receipt of areward capital, forming afeature of high-elite from the king or other elite patrons) burial, and playing aprominent role in began to appear even as oracle-bone in- practices of both hunting and war.5 scriptions became pithier and narrower in subject matter (Keightley 1999). With the large corpus of oracle-bone inscriptions from Anyang, the chronol- Unlike writing, chariots were of nonin- ogy of the Anyang period proceeds from digenous origin4 and indicate intensive two sources: the internal periodiza- contacts with peoples to the north and tion of the oracle-bone inscriptions, west (Piggott 1974; Shaughnessy 1988; and ceramic seriation in conjunction Bagley 1999; see Anthony 2007 for the with stratigraphy and absolute dating Eurasian signiﬁcance of the chariot). methods (XSZDGZ 2000; ZSKY 2003). Moreover, whatever their origins, chari- Correlated with transmitted text-based ots soon became amajor, and endur- genealogies of the last Shang kings, the

131 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Phase Oracle-bone Period Kings Dates (BCE) II Early Wu Ding ca. 1250-1220 II Iand II Late Wu Ding, Zu Geng, Zu Jia ca. 1220-1160 III III and IV Lin Xin, Kang Ding, Wu Yi, WenDing ca. 1160-1102 IV VDiYi, Di Xin ca. 1101-1046

Table 5.1. Anyang periodization (based on ZSKY 2003:294; Keightley 1997:18; and Duandai 2000: 88).

Anyang period can be divided as seen in ponds such as have been found separat- Table 5.1.6 ing different residential districts at An- yang and was not acontinuous barrier The Development of Anyang around the palace-temple area (Tang and Jing 2009). Anyang’s other bronze Several large buildings in the palace- foundries commenced operation in temple area date from the ﬁrst phase this phase, as did bone workshops at (buildings yi -5, yi-7, and B-11), and Dasikongcun (ZSKY 2003) and Tiesanlu residential structures in Xiaotun lo- (Li et al. 2011). At the same time, the discus northeast and Xiaotun locus south tribution of the residential areas greatly were in use during this period. Out- expanded. The royal cemetery at Xibeig- side the palace-temple area, phase one ang was in use in this phase and, in gen- residential areas can be found in Hu- eral, residential areas expanded. The ayuanzhuang, Miaopu, Sipanmo, Hou- western “lineage cemeteries” (Zhong- gang, Dasikongcun, Wuguangcun, and guo Shehuikexueyuan Kaogu Yanjiu- beyond (Figure 5.2). The area of the suo, Anyang Gongzuodui [hereafter Yinxu site is estimated to have been ap- ZSKY,AG] 1979) began to see heavy use proximately 12 km2 in this period, and in this period (ZSKY 2003). bronze workshops in Miaopu North, Dasikongcun, and Xuejiazhuang were During phases III and IV,Anyang already in operation (see Figure 5.3). reached its zenith. The Miaopu foundry Burial areas were generally located near expanded to the west and south, dou- or interspersed with settlements and bling in size. The Xiaomintun and Xue- mostly concentrated in Miaopu north, jiazhuang foundries, as well as the bone Wuguancun, and Dasikongcun during workshops at Dasikongcun and Tie- this phase (ZSKY 2003). sanlu, all expanded in this period (ZSKY 2003). Anew bone workshop com- By the beginning of phase II, what was menced operation near Beixinzhuang once described as amoat (claimed to during phase III, and during phase IV, be 7–21 mwide and 3–10 mdeep) had anew jade and stone workshop began been dug around the palace-temple operation in Xiaotun northwest, in the area (ZSKY 1987). It is now believed palace-temple area. Residential sites in- that the “moat” was, in fact, aseries of creased in density and distribution, and

132 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

avarietyoffunctions.Thenorthernmost structures of the Bgroup are associated with rows of sacrificial pits representing separate sacrificial events and may have been an ancestral temple complex with acourtyard structure (Shih 1976b; ZSKY 2003; Du 2009). The southern structures of the Bgroup (B11, B12, B13) appear to have been three sides of another courtyard structure (the eastern side was eroded by the Huan River), with aseries 5.3. Duration of Anyang foundries (redrawn of two-story watchtowers to the south from Li 2003:311; fig. 7.9). (B14, B17, B21; Du 2009), which possibly served an administrative function. If the site reached its maximum size of this is so, then this arrangement seems over 30 km2 in this period (ZSKY 1994; to mirror that of foundations 1and 2at ZSKY 2003). Huanbei, where the northern complex had sacrificial pits and is thought to be The Palace-Temple Area aroyal lineage temple, while the southern courtyard structure lacked these Mostofthepalace-templeareawasexca- features (ZSKYAG, ZJHQKDK 2010). vated in the 1930s under difficult condi- tions by archaeologists of the fledgling The Cgroup of foundations to the Academia Sinica, who had little experi- southeast of the Bgroup were generally ence with the nuances of excavating soft small (around 10 m2)and show no signs features. Added to the limitations of of having ever had superstructures. the times was the interruption of work They were also associated with sacrifi- caused by the Japanese invasion and cial victims, and it has been suggested the loss of material during World War that they may have been sacrificial al- II, followed by the civil war (Shih 1933, tars (Shih 1961; Chang 1980). 1951, 1955, 1959, 1961, 1976; Creel 1937; Chang 1980; Thorp 2006). As are- To thenorth of the Bgroup were a sult, the relative dates of the large-scale number of rammed-earth structures (A rammed-earth platforms, their original group), which may have had aresiden- structures, and the overall layout of the tial function based on middens associ- palace-temple area are only crudely un- ated with them (Tang Jigen personal derstood despite several attempts at re- communication). At the southern end construction (Shih 1954, 1970, 1976a; of the palace temple area are anum- Du 2009; Figure 5.4). In terms of scale, ber of smaller structures hypothesized however, the Bgroup of buildings near to be houses of lower-ranking officials the center of the palace-temple area is or servants associated with the palace- the largest (5,000 m2). It contains nu- temple complex (Shih 1959; ZSKY,AG merous structures that probably served 1995).

133 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.4. Anyang palace-temple area (after Chang 1980:94, ﬁg.23).

134 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

Residential Areas roads with narrow-gauge cart tracks and large, irregular pits, hypothesized Despite over eighty years of work at An- to have been quarried for foundation yang and numerous excavations of do- earth, then used as ponds/cesspools. mestic structures, there has been no These ponds tend to be found around attempt, until very recently, to excavate the edge of residential clusters and large exposures or to investigate the may, in part, have served to demarcate internal structures of residential ar- them from other settlements (ZSKYAG eas (ZSKY 2003). What work has been 2009). The roads, by contrast, linked done is just being published, but apre- settlements together in alarge network liminary account of residential patterns centered on the palace-temple precinct can be made, based on excavations at (Tang and Jing 2009). Wells and large Baijiafen (ZSKY 2003), Liujiazhuang ritual deposits associated with residen- (ZSKYAG2009) and Xiaomintun (Yinxu tial clusters, on the other hand, sug- Xiaomintun Kaogudui [YXK] 2007). The gest mundane, as well as ritual, shared, structures at Baijiafen formed three intracommunity practices and spaces discrete clusters with two or three hun- (ZSKYAG2009; Tang and Jing 2009). dred meters between them. As noted in chapter 4, the only other major Central Large structures are also found outside Plains metropolitan site where resi- the palace-temple area, some over 50 dential structures have been systemati- m2 in area; they are frequently associ- cally (albeit preliminarily) analyzed is ated with foundation sacriﬁces. While Huanbei, which also shows aclustered most houses at Anyang are rectangu- settlement pattern. These residential lar surface dwellings (ZSKY 2003; Tang clusters are associated with burials as and Jing 2009), some semisubterra- well and are interpreted as lineage cem- nean and subterranean houses have eteries. The pattern of burial and resi- also been found (such as at Xiaomintun dential clusters is repeated all over the [YXK 2007]). Most houses are about 10 site of Anyang (Tang and Jing 2009). The m2 in size and generally have one or two composite impression of archaeology, rooms (ZSKY 2003), but some, like the inscriptions, and received texts, is that Beixujiaqiaocun cluster of courtyard these residential and mortuary clusters structures,7 are much larger (Figure are lineage settlements with associated 5.5). burials (Zhu 1991; Keightley 2000; ZSKY 2003; Tang 2004; Campbell 2007, 2009; Workshops Tang and Jing 2009). Workshops—their internal organiza- With areas of Anyang being studied tions and their relationship to the wider holistically for the ﬁrst time, surpris- urban center and polity—are currently ing new perspectives are emerging one of the foci of research interest at (ZSKYAG2009). Salvage excavations in Anyang. While much empirical work re- the Liujiazhuang area have discovered mains to be done, arecent article based not only residences and tombs, but also largely on inscriptional evidence argues

135 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.5. Beixujiaqiaocun courtyard structure cluster (after Meng 2003:68, ﬁg. 2).

foratwo-tieredproductionorganization specialization existed between found- model, with ahighly centralized royal ries, with some foundries producing production and then amoredecentral- afull range of artifacts, while others ized lineage production (He 2011). In seemed to specialize in just afew types some industries, such as bone working, (Table 5.2), none of the foundries has there is evidence for small-scale domes- been completely (or even systemati- tic production as well (ZSKY 1994; Li et cally) excavated, and overall, the bronze al. 2011). workshops show astriking redundancy.

Bronze Workshops Interestingly, despite the frequently There were as many as six foundries held belief that bronze vessels were in operation at Anyang at one time or strategic forms of symbolic capital another, and at least four in operation whose monopoly was closely protected at any given phase (Li 2003). Whereas by the king (Underhill 2002; Li 2003; acertain degree of division of labor or Liu and Chen 2003; Liu 2005; Li 2005,

136 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

Ritual Vessels Horse and Chariot Fittings Weapons Tools Dasikongcun x Xiaotun NE Xx xx Miaopu NX xx Xuejiazhuang Xx Xiaomintun SE X?? Xiaomintun Wx xXX

Table 5.2. Anyang foundries and their mold fragments (after Li 2003:304, table 7.2). etc.), there were at least four foundries court, and that weapon production was at Anyang producing bronze vessels too strategic not to supervise. Iwould at any one time, only one of which was argue, however, that the first assump- within the palace-temple area. While tion is not well-founded;9 the second, it is possible that all of the foundries largely untestable, given that the vast at Anyang were under strict royal su- majority of large Anyang tombs have pervision, their number and scattered been looted;10 and the third, unlikely, distribution across the site at amini- given the wide distribution of bronze mum would have worked against ease weapons in tombs both at Anyang and of supervision. Moreover, the redun- beyond (Campbell 2007). Moreover, dancy of production also suggests a while Anyang-period bronze inscrip- lack of centralized administration. Li tions themselves frequently commem- Yung-ti (2005) hypothesizes that this orate gifts from the king or other high pattern indicates that the king was not elites, bronze itself is never given, nor is directly involved in supervising casting, permission to cast the bronze ever men- but that the Shang court nonetheless tioned. In fact, it is the granting of cow- “controlled the procurement and dis- ries or other benefices that provides the tribution of raw materials for craft pro- occasion for having abronze vessel cast; duction,” as well as having “direct and the access to foundries is never present- perhaps total control over the finished ed as an issue. Thus, while both the ac- products” (10). While the first part of quisition of ores and finished bronzes Li’s hypothesis (that the king was not certainly depended on social networks directly involved in casting) is based on structured in gradients of power and re- the actual distribution of craft produc- stricted prestige (Campbell 2007, 2009), tion sites, the second, concerning con- the actual mechanisms by which metal trol of resources and finished products, was procured, casting patronized, and is entirely based on the assumptions its products distributed remain un- that large-scale resource use8 implies known. centralized supply, that the court could not have permitted large or finely cast bronze vessels to be used except by the

137 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.6. Xibeigang royal cemetery and sacriﬁcial pits (after Tang 2004, ﬁg. 7.5).

Bone Workshops from the enormous royal sacriﬁcial The only other relatively well-known rites that consumed thousands of head industry from Anyang is bone-working. per year.Production at the bone work- Major workshop areas are recognized shops was massive, and it is estimated from Beixinzhuang, Dasikongcun, and that Tiesanlu produced on the order Tiesanlu (near Miaopu).11 While Beix- of four million bone artifacts over the inzhuang and Dasikongcun have seen course of its operation (Campbell et al. numerous excavations since the 1950s, 2011). If one considers that the Tiesan- their assemblages have never been sys- lu worksite is estimated to be nearly 2 tematically analyzed, quantiﬁed, or ha in size, Dasikongcun 1ha, and Beix- studied by zooarchaeologists. Tiesanlu, inzhuang as large as 4ha, then the total on the other hand, excavated in 2002, bone tool production for the workshops 2006, and 2008, is currently the focus of might have been close to 15 million an international collaborative project artifacts (Li et al. 2011). The majority and provides amuchbetter basis for of these artifacts (80–90 percent) were understanding bone-working at Anyang “points” such as pins, awls, and arrow- (Li et al. 2010; Campbell et al. 2011; Li heads—quotidian artifacts that were et al. 2011). Anyang bone-working was widely distributed at Anyang and likely largely based on cattle limb bones, deer beyond (Campbell et al. 2011). Taken antler, and pig tusk (Campbell et al. together, these things suggest that, in 2011). The cattle likely derived in part addition to large workshops that large-

138 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5 ly catered to elite patrons, Anyang also Comparing residential and production had large industries that produced for a site data, it would seem that produc- wide range of consumers and likely sup- tion at Anyang was also segmented into plied trade networks beyond the center. multiple, discrete, and functionally redundant areas (i.e., producing the same While Dasikongcun and Tiesanlu were things). Combined with the transmitted in operation from at least phase II to textual evidence that Shang craft was or- phase IV,atBeixinzhuang bone-work- ganized on alineage basis (Chang 1980; ing seems to date from phases III to Haapanen 2005; He 2011), the most IV,based on the small amount exca- parsimonious explanation of multiple, vated (Li et al. 2011). It appears that, redundant industries might be that like bronze production (Li 2003), bone- lineage settlements were specialized working expanded in the second half of in occupation and that different work- Anyang’s existence. shops were under the control of different lineages. Specialization by lineage Other Workshops or settlement at Anyang would, at least Other workshops found at Anyang in- superﬁcially, mirror the phenomenon clude asmall jade and stone workshop of economically specialized villages inside the palace-temple compound. seen in Erlitou times at Huizui (chapter Until very recently almost nothing was 2), and in Anyang times at Guandimiao known about ceramic production at An- and Liwu (see below). If this is correct, yang (but see Haapanen 2005), and rel- then, Anyang as acluster of villages cen- atively few kilns had been reported. In tered on the palace-temple area (Tang 2008, what appears to be alarge pottery and Jing 2010), was also acluster of oc- production area with 10 kilns dating to cupationally specialized villages. the Anyang period was discovered to the south of Huayuanzhuang (Yue Zhanwei The Royal Cemetery and Clan personal communication). It is antici- Cemeteries pated that the study of this site will yield important new information about ce- The royal cemetery at Xibeigang is lo- ramic production at Anyang. cated 2.5 km northwest across the Huan River from the palace-temple complex Production Zones (Figure 5.6). Mostly excavated between The distribution of large workshop ar- 1934 and 1935, this 11 ha area contains eas at Anyang appears to form at least the largest tombs found in China of this three clusters outside of the palace- or any earlier period. The nine largest temple area. That bone and bronze in- tombs (eight with four ramps, one emp- dustries tend to be clustered together ty and presumably unﬁnished, without may partially have to do with their pol- ramps12)correspond to the number luting nature, but there may have been of kings that ruled at Anyang from the other factors (Li 2007), on which future time of Wuding to Di Xin (Table 5.1). In research will hopefully shed light. addition to the royal tombs, there are anumber of large double- and single-

139 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.7. Lineage cemeteries at Liujiazhuang (after Tang 2004, ﬁg. 4.10).

ramped tombs in the cemetery, includ- nese archaeologists. It is estimated ing 50WGM1, some “death attendant” that over 15,000 tombs of this type have burials, such as M259, and several been excavated at Anyang (Tang Jigen thousand human and animal sacrificial 2009 personal communication; see also pits (ZSKY 1994, 2003). On present evi- Tang 2004) in over 10 major burial are- dence it would seem that the royal cem- as. While the vast majority of the tombs etery was both the burial site of the last in the “lineage cemeteries” were small Shang kings, their nearer relatives, and rectangular pit burials, these cemeter- perhaps loyal servants, as well as avast ies also included larger tombs with one sacrificial ground dedicated to the cult or even two ramps, sometimes associ- of the royal ancestors, where the living ated with chariot burials (Figure 5.7). continued to sacrifice to the royal dead. The distribution of Anyang burials fits with the general tendency of workshops In addition to the royal cemetery, there and residences to occur in discrete clus- are numerous cemeteries scattered ters, suggesting the spatial marking of throughout the Anyang site, frequently horizontal rather than vertical social di- termed “lineage cemeteries” by Chi- visions.

140 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

As aﬁnal word of caution concerning the Anyang site, it should be noted that despite over 80 years of excavations, the urban layout is not well understood. Most of the work that has been under- taken at the site is salvage archaeology. Owing to the size of the site, the orientation of Chinese archaeology,13 and modern construction that sits atop more and more of it, the site has not been systematically surveyed and excavated. Never- theless, the Institute of Archaeology of the Chinese Academy of Social Sciences has sponsored amajor project under the current station director, Tang Jigen, to study Anyang’s site structure through asynthesis and mapping of all previous 5.8. Anyang-variant everyday ceramics seriation excavation and survey work done at An- (after ZSKY 2003:291, ﬁg. 6-3). yang. It is hoped that this project will shed new light on this crucial site. decreased to seven from the nine of the Xiaoshuangqiao-Huanbei period and Anyang Ceramic Traditions the eight of the Erligang period, mirror- ing both the reduction in geographic The Anyang (Yinxu) Variant distribution of Central Plains ceramic traditions and homogenization at the The Anyang variant of the Anyang ce- core. ramic tradition formed ametropolitan ceramic assemblage that attained un- The Anyang variant was centered on An- precedented homogeneity and distri- yang and distributed mainly in south- bution. Anyang-variant sites are more ern and central Hebei and northern and plentiful than those of any other An- central Henan. As such, it covered the yang-period variant, and the distribu- areas formerly occupied by the Xiaosh- tion of Anyang-variant sites is greater uangqiao-Huanbei period Baijiazhuang in geographical extent than any earlier and Caoyanzhuang variants, and so the Central Plains metropolitan variant. At metropolitan area of Erligang times re- the same time, however, the overall dis- mained part of the larger Anyang met- tribution of Central Plains ceramic tra- ropolitan ceramic tradition core (ZSKY dition sites (i.e., the variants other than 2003). The li-tripod remained the main that of Anyang) retracted during this pe- cooking vessel for this variant (as for riod, most noticeably in the south and all CPBA tradition variants; Figure 5.8). the west, expanding only in the east. The houses of Anyang-variant sites were The number of regional variants also mostly surface dwellings with rammed-

141 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Guandimiao Located near Zhengzhou and the contemporaneous site of Renmin Gongy- uan, Guandimiao is awell-preserved, Anyang-period, Anyang-variant village. As such, it presents arare opportunity for aholistic study of anonelite site. Ap- proximately 20 percent of this 10-ha site was excavated, revealing 22 houses, 20 kilns, hundreds of middens, and 228 small tombs dating from the Anyang period, as well as earlier and later occupations ranging from Yangshao to Qing times (HWKY 2008). Despite Guan- dimiao being an Anyang-variant site, the houses were all semisubterranean. The Anyang-period occupation was surrounded by aditch, and most of the burials were located outside of it. The residential and production remains were 5.9. Anyang mortuary ceramic seriation (after mostly located in the northern part of ZSKY 2003:293, fig. 6-4). the enclosed site, while sacrificial remains were found in the south (HWKY earth foundations, and there were few- 2008). The presence of ritual deposits er semisubterranean dwellings (ZSKY in asite with no evidence of elites, sug- 2003), showing aprogression toward gests, as do the sacrificial deposits in above-ground dwellings and the popu- the residential clusters at Anyang, that larization of building techniques that sacrifice was not the sole prerogative of were previously associated with elite metropolitan elites but rather that elite dwellings. Tombswith ramps began to and commoner ritual practices existed appear in this period, and waist-pits, in acontinuum (Campbell 2007, 2009). long adiagnostic feature of “Shang” The large number of kilns suggests to burials, reached unprecedented levels the excavators that Guandimiao was of popularity (around 50 percent; ZSKY specialized in ceramic production, an 2003). Evidence for human sacrifice and economic occupation that, as with Hui- death attendants also appears more fre- zui’s lithic production (chapter 2), had quently in this period. In addition, for possibly begun as early as the Long- most, but not all, Anyang-variant sites, shan period (HWKY 2008). Such sites gu and jue,whether ceramic or bronze, as Guandimiao, Huizui, and Liwu (see form the core of the mortuary assem- below) suggest that small settlements blage (Figure 5.9).14 played important roles in Central Plains regional and even interregional exchange.

142 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

5.10. Subutun-variant ceramics (after ZSKY 2003:308, ﬁg. 6-9).

Anyang Tradition, Subutun coring. The excavated tombs include Variant afour-ramped tomb (with 48 human “sacriﬁcial victims”), achariot pit, and Although this variant of the Anyang ce- asingle-ramped tomb. Both the burial ramic tradition covered the area pre- structures and assemblages are said to viously occupied by the Daxinzhuang closely resemble those of Anyang, even variant of the Erligang and Xiaoshuang- while some ceramics show “rich local qiao-Huanbeitraditions,itextendedfar- characteristics” (ZSKY 2003:314). Prob- ther east than the earlier Central Plains ably the most remarkable aspect of the tradition variants. Also, its center seems Subutun site is the four-ramped tomb, to have shifted east from the Jinan area the only one from the Anyang period to the Qingzhou (Subutun) area (ZSKY found outside of Anyang and generally 2003; Fang 2009; Figure 5.10). At the believed to be amarker of royal status. Subutun site itself, 10 tombs have been The interpretations of this tomb range excavated between 1965 and 1986, and from that of apolitical rival, to privi- many more tombs were found through leged favorite of the Anyang kings (Fang

143 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

2009:15).15 Given the strict sumptu- which can also be found at Anyang. Nev- ary rules that apparently governed the ertheless, and despite the evidence of number of ramps atomb could have elite contacts with the Anyang core, the at Anyang, together with the politically small tombs discovered at the Pingyin and militarily divisive Shang political Zhujiaqiao site do not have waist-pits or landscape depicted in the oracle-bone the same mortuary ceramics as Anyang, inscriptions,16 the rival scenario seems interring only ceramic guan-pots. Fang more likely. (2009) interprets this as indicating possible differences in the cultural or eth- Fang (2009) provided amorerecent nic affiliation between the Anyang or counterpoint to the ZSKY (2003) ac- Anyang-styled local elites and the com- count of northern Shandong in Anyang mon people. times. Instead of speaking of regional variants in blanket terms, he divided East of Jinan, and near the modern city up the Subutun-variant area into five of Zibo, is the third cluster, the largest settlement clusters. The first of these, site being Huantai Shijia. The site ap- Daxinzhuang, is centered around the proaches 30 ha in size and may have site of the same name, which was in ex- been moated (Fang 2009). Bronze ves- cess of 30 ha in area in Anyang times— sels have been found in the area since the largest thus far discovered in the the 1960s and oracle-bones and sacrifi- entire northern Shandong region (Fang cial pits have been discovered in recent 2009:8). Not only have inscribed oracle years. Oracle-bones and bronze vessels bones been discovered at Daxinzhuang have also been found at nearby sites, (Shandong Daxue Dongfang Kaogu suggesting this was an important area Yanjiu Zhongxin et al. 2003, 2004), but in Anyang times. The adjacent Chang- also Anyang-type bronze vessels have bai Mountains have Shandong’s larg- been found there with “clan insignia.”17 est copper deposits, suggesting mining Although there appears to be agreat and smelting in the area. Anyang-peri- deal of similarity between the material od ceramic tool molds have been found culture of Daxinzhuang and Anyang, at the nearby site of Zouping Langjun, the former differs from Anyang in that indicating some local casting. The ma- the ceramic gu and jue frequently seen terial culture of the sites in this cluster in Anyang burials do not appear, and a is said to be “classic Late Shang”(i.e., few bronze vessels show peculiarities Anyang variant), with afew local char- that suggest local casting (Fang 2009). acteristics (like unadorned li-tripods; Fang 2009:10). The second settlement cluster is the Xi- aotun cluster west of Jinan (Fang 2009). As for the Subutun cluster itself, Fang Anyang metropolitan-style bronze ves- (2009) notes that the ceramic assem- sels, chariot fittings, tools, and weap- blage of the large tomb M7 includes gu- ons were found in the area. Many of the and jue-vessels, rare in the Shandong bronzes had “clan insignia” on them. coastal area, apossible sign of close An- There were four different insignia, all of yang affiliation. Nevertheless, at anoth-

144 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5 er site in the area, Qingzhou Zhaopu, The ﬁfth of Fang’s clusters is the Lanjia an Anyang-period tomb was discovered cluster, north of the Yellow River on or showing pronounced local character- near what was then the coast. In addi- istics (body ﬂexed and on its side, de- tion to bronze vessels discovered in the ceased clasping aroebuck tooth18)and a area indicating the presence of Anyang- mixed assemblage of Central Plains tra- style elites, sites with large numbers of dition and non-Central Plains tradition thick-walled crucibles and salt-produc- ceramic vessels. Although neither sys- tion remains have been discovered near tematic nor necessarily representative, what was once the coastline, indicat- the presence of Shang elite and mixed ing alocal salt industry (Fang 2009).20 nonelite tombs is suggestive of possible Excavations at the village site of Liwu, elite-commoner cultural differences. moreover, have demonstrated that a In addition to tombs, salt production specialized but stable community of sites have been located among those salt producers lived there year-round sites of the Subutun cluster distributed (Shandong Sheng Wenwu Kaogusuo, along the ancient shoreline (Fang 2009; Beijing Daxue Zhongguo Kaoguxue Shandongsheng Wenwu Kaogu Yanjiu- Yanjiuzhongxin, Shandong Shifan suo, Beijing Daxue Zhongguo Kaogux- Daxue Qi Lu Wenhua Yanjiuzhongxin, ue Yanjiuzhongxin and Shouguangshi Binzhoushi Wenwu Guanlichu [SWK, Wenhuaju [SWKY,BDZKY,SW] 2010). BDZKY,SSDQLWY,BWG]2010), providing another example of specialized To the north of Subutun, but part of production tied into larger, likely inter- the same cluster of sites, is Shouguang regional, economic networks. Guchengcun, where over 60 bronze artifacts were discovered, and of those In summary, the two accounts provid- inscribed, most bore the inscription ed by ZSKY (2003) and Fang (2005) of 己 (Shouguangxian Museum, 1985). To northern Shandong in the Anyang pe- scholars who associate bronze insignia riod show some telling contrasts. The with polities, this suggests the presence ZSKY (2003) account, based as it is on of two polities in close proximity, that of homogenizing formal ceramic-tradi- Ya Chou 亞醜 and that of Qi 己 or Ji 纪. tion typologies and the assumption As Ihave argued elsewhere (Campbell these correspond to ethnic or even po- 2007), this one-to-one correspondence litical units, tends to describe areas in between bronze insignia and polity is blanket terms, combining what appear problematic. Nevertheless, if the “clan in Fang’s (2005) account to be overlap- insignia” do in fact reference some sort ping and intersecting material cultural of kindred as most scholars believe, tradition distributions into asingle then the presence of these two clusters Anyang variant. Fang’s approach on of insignia-bearing bronzes suggests the other hand, based on settlement two different elite lineages in the same clusters, bronze insignia, and resource area.19 distribution and interpreted through much later textual traditions, tries to identify ethnic groups and political ac-

145 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.11. Anqiu-variant ceramics (after ZSKY 2003:307, ﬁg. 6-9).

tors. Fang’s account allows us to see the presence of Shang “state” expan- how the formal typological account of sion. It could equally well indicate local ceramic traditions that we have been elite emulation or Anyang-independent summarizing tends to lump together political entities sharing the broader potentially diverse social-economic Central Plains metropolitan cultural ec- networks, downplaying diversity within umene. One need not go further afield its confines and exaggerating differ- than Eastern Zhou China to find com- ence at its boundaries. Though Fang’s peting polities with broadly similar elite account has its own limitations, it is a culture. reminder of the potentially very much more complicated maps that could be Anyang Tradition, Anqiu Variant drawn of economic networks, settlement patterns, and political groups if Distributed in western Shandong and only information was systematically eastern Henan, this variant is adevel- gathered on these aspects of the archae- opment of the Panmiao variant of the ological record in China. Combined, Xiaoshuangqiao-Huanbei period (ZSKY these accounts suggest an overall pic- 2003; Figure 5.11). The Anqiugudui site ture of tendrils of elite Anyang mate- itself had alonghistory of occupation rial culture penetration and ageneral with successive Longshan, Yueshi, Xia- dominance of Anyang ceramic types oshuangqiao-Huanbei, and Anyang lay- even while non-Central Plains tradition ers. The ceramics of this variant are not communities21 still apparently existed only similar in type but also in develop- in northern Shandong and especially mental history with corresponding ce- eastern Shandong. As for the political ramics at Anyang. In color, paste, and situation, it was likely even more com- thickness of vessels walls, however, An- plicated, but for current purposes, it is qiu ceramics are said to show marked enough to note that elite Anyang mate- differences, in addition to possessing rial culture does not necessarily equal afew ceramic styles that very rarely ap-

146 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

5.12. Chang Zi Koutomb: Human remains (after HWKY,ZW2000:11, ﬁg. 6). pear at Anyang (ZSKY 2003). This sug- of the burial—from awaist-pit with gests, at minimum, adifferent network human and dog attendants, to jade ar- of ceramic production/distribution tifacts and ceramics—suggest Anyang- than Anyang, as well as distinct but in- metropolitan, rather than Zhou cultural teracting22 potting communities. afﬁliation (HWKY,ZW2000). Interest- ingly, there are adozen protoporcelain An assemblage of bronze and jade ar- vessels indicating contact with the tifacts typical of an elite Anyang-type Yangzi region as well, while the unchar- burial and comparable to Anyang phase acteristically large ceramic assemblage Iartifacts was discovered in Huaiyang (209 vessels)24 does not include the li- County in 1981 (ZSKY 2003), indicating tripods commonly found in Shang and metropolitan-style elites in this area. In Zhou tombs. In summary, these discov- 1997 alarge two-ramped tomb of West- eries suggest that the material culture ern Zhou date was discovered at Luyi in this area (or at least some sites in this Taiqinggong (Henansheng Wenwu Ka- area), from common to elite, while with- ogu Yanjiusuo, Zhoukoushi Wenwuju in the Anyang orbit, nonetheless shows [HWKY,ZW] 2000;Figures 5.12, 5.13). some local variation. The ceramics in the tomb, however, are said to show Anqiu-variant character- Anyang Tradition, Qianzhangda istics (ZSKY 2003). Inscriptions on the Variant bronzes in the tomb identify the owner as Chang Zi Kou, or Kou, head of the Distributed in the southern half of Chang lineage,23 and various aspects Shandong Province, in part of the area

147 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.13. Chang Zi Koutomb: Grave goods (after HWYK, ZW 2000:16, ﬁg. 10).

of the Xiaoshuangqiao-Huanbei tradi- along with their remaining contents tion Panmiao variant, and centered on (they had been looted multiple times), the Qianzhangda site, the Qianzhangda the presence of Anyang, or Anyang- variant is distinguished by the pres- styled elites (Figure 5.15). Although the ence, in otherwise overwhelmingly form and contents of the burials are Central Plains bronze assemblages, of said to be similar to Anyang, “Yueshi- unadorned li and yan (generally associ- type” ceramics were also present, and ated with the Yueshi tradition), as well ceramic gu and jue were missing from as stamped hard-ware and protoporce- the assemblage. Taken together, these lain of southern origin (Figure 5.14; facts suggest that the adoption of Cen- ZSKY 2003; Fang 2009). Also unusual tral Plains metropolitan material cul- was the frequent use of lacquer artifacts ture was top-down, indicating ashared in tombs,25 perhaps suggesting south- elite culture overlying apossibly more ern contacts (Fang 2009). At the site of heterogeneous population. Bronze ves- Qianzhangda itself, of the 120 tombs sels have been discovered at many other excavated there, eleven were one- or sites in the area, some with insignia, two-ramp large tombs, ﬁve were chariot and all of Anyang style, in Anyang-type pits, and two were horse pits, signifying, assemblages (Figure 5.16). There is not,

148 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

5.14. Qianzhangda-variant ceramics (after ZSKY 2003:307, ﬁg. 6-9). however, much information concerning nonelite tombs or residential sites in general, so the account given here is necessarily partial.26 Nevertheless, looking at the ceramics recovered from midden deposits conﬁrms the general impression of the Qianzhangda variant as largely Central Plains metropolitan with local and nolocal, non-Central Plains elements.

149 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.15. Qianzhangda cemetery layout (after ZSKY 2005a:7, ﬁg. 3).

150 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

5.16. Bronze jue bearing insignia “shi” (after ZSKY 2005a:257, ﬁg. 182).

Anyang Tradition, Tianhu tradition that ZSKY (2003) is working Variant from, the anonymous authors note that, “in Late Shang times, Shang culture was This variant is known only from the rejected by local cultures and left the cemetery site of Tianhu. In the “ceramic banks of the Yangzi, retreating north of tradition equals cultural/ethnic group” the Tongbai Mountains” (318). Appar-

151 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.17. Tianhu-variant ceramics (after ZSKY 2003:309, ﬁg. 6-9).

ently as aresult, the Tianhu Anyang vari- The 20 or so Tianhu burials excavated ant was born. Whatever the social-polit- between 1980 and 1985 have been divid- ical story behind this change, Tianhu is ed into four grades or “classes,” ranging one of the southernmost Anyang-period from burials with over adozen bronze sites known with broadly Anyang-type vessels, to tombs with only ceramics material remains. While both ceramic vessels.27 The assemblages of Anyang- forms and styles are said to basically style bronze vessels (ding-cauldrons and resemble those found at Anyang, some pairs of gu and jue), the frequent use of Tianhu vessel styles are not found at waist-pits with dogs, and other aspects Anyang. The high percentage of coarse- of the burials suggest shared Central tempered brown and ﬁne-paste yellow Plains metropolitan burial practices ceramics also suggests local production and generally inclusion in awider Cen- techniques or clay sources somewhat tral Plains zone of interaction. At the different from those at Anyang (Figure same time, the frequency of lacquer 5.17; ZSKY 2003). vessels and the absence of ceramic gu or jue in tombs, suggest some local variation in mortuary practice as well.

152 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

5.18. Laoniupo-variant ceramics (after ZSKY 2003:308, ﬁg. 6-9).

Anyang Tradition, Laoniupo ity with those of Zhengzhou in the Erli- Variant gang and Xiaoshuangqiao-Huanbei periods had weakened or disappeared by By the Anyang period, Central Plains– Anyang times. However, given that we tradition sites were no longer found know nothing of the speciﬁc history of west of Xi’an, their western Shaanxi ceramic production or distribution in distribution having been taken over by this area, the underlying causes of these so called “Proto-Zhou” ceramic tradi- changes can only be speculated upon. tions, such as Zhengjiapo, Liujia, and Nianzipo. East of Xi’an, however, the The Laoniupo site (50 ha) itself had Er- Laoniupo variant of the Central Plains ligang- and Xiaoshuangqiao-Huanbei– metropolitan tradition survived un- period occupations, but most of the til the end of the Anyang period. The remains at the site date to the Anyang Laoniupo variant is said to be the most period (Liu 2001). Two badly preserved, dissimilar of all the Anyang-tradition large, rammed-earth foundations date variants (ZSKY 2003), and the Laoniupo to this period, the remains of one of site itself shows “obvious differences” them, almost 300 m2 in size (Liu 2001). between its Xiaoxhuangqiao-Huanbei Six kilns and the remains of abronze- and Anyang-period ceramics (Figure casting foundry have all been discov- 5.18; Liu 2001). All of this suggests that ered at the site. Ceramic molds for whatever mechanisms had brought bronze weapons, tools, masks, and ves- Central Plains ceramic traditions to sels place beyond adoubt the possibil- Shaanxi and maintained their similarity that the Anyang-type bronze vessels

153 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

were found, 7ofthe 38 were “medium”- sized, and over 80 percent of the tombs had waist-pits (Figures 5.19, 5.20). Gu and jue were the bronze vessel types most frequently interred in tombs and closely resemble vessels found at An- yang (Figure 5.21; ZSKY 2003). An unu- sually high percentage of the tombs had death attendants or human sacriﬁces (21 out of 38), and many of the tombs had multiple waist-pits and niches. The niches, as well as some features of the ceramic assemblage, show connections with non-Central Plains metropolitan traditions to the west (ZSKY 2003), while the presence of bovine and human bronze plaques shows connections with the Hanzhong area of southern Shaanxi (Figure 5.21) (although based on mold fragments found at the site, these were probably cast locally).

At the nearby site of Xi’an Yuanjiaya, an Anyang-period tomb was discovered with two bronze vessels (a gu and jue) and six ceramic vessels, three of which were “classic” Anyang-period metropolitan artifacts (ZSKY 2003) and the other three resembling Laoniupo ceramics. 5.19. Laoniupo Burial 86XLIII1M41 (after Liu Altogether, the Laoniupo variant sug- 2001:269, ﬁg. 234).2003:308, ﬁg. 6-9). gests abronze-casting tradition whose craftsmen and/or patrons were in contact with multiple nonlocal traditions discovered at the site could have been (including that of Anyang). Neverthe- cast locally, which further suggests that less, their ceramic tradition, though Anyang-type bronzes discovered at oth- apparently deriving from acommon er sites need not have been manufac- Erligang and Xiaoshuangqiao-Huanbei tured at Anyang. tradition, developed independently of Anyang. And yet, as the Yuanjiaya site Thirty-eight tombs, ahorse-pit, and a (and non-CPBA sites in the immediate chariot pit were excavated at Laoniupo area) show, ceramic and other mate- in 1986 (Liu 2001). Although no large rial cultural traditions are best seen as (by Anyang standards) or ramped tombs intersecting and overlapping networks

154 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5 of production and distribution rather than bounded homogenous spaces.

Beyond the Anyang Tradition Variants

The North

As noted above, the Anyang variant of the Anyang-period Central Plains metropolitan tradition is said to extend north into the southern part of Hebei Province. The situation farther north, however, is somewhat unclear.Onthe one hand, Anyang-type bronzes have been discovered at many sites excavated in and around Shijiazhuang and north—even to modern day Beijing. On the other hand, none of these excavations have proper excavation reports, and so it is difficult to say much 5.20. Laoniupo Chariot pit 86XLIII1M27 (after about the ceramic-tradition affiliations Liu 2001:273, fig. 237). in these areas. The discovery, in 1978, of Anyang phase Iceramics in amidden at Fangshan, southwest of Beijing, evidence for northern complex mate- suggests to the authors of ZSKY (2003) rial cultural elements in this area as far that, at least until Anyang I, people us- back as Erlitou times, and the fact that ing Central Plains metropolitan tradi- material culture does not equal ethnic tion ceramics were living in the area. or political group, at this point the best Complicating the picture however, are we can say is that the Central Plains nearby sites, such as Beifudi and Yan- metropolitan tradition northward tide shan, which had Anyang-period layers of material cultural “influence” seems with Weifang III–tradition affiliations to ebb and reverse in the Anyang period. (Jumahe Kaogudui 1988; see below). Moreover, whatever their political affili- The conclusion drawn by the authors ations or ethnic identities, there appear of ZSKY (2003) is that the Weifang III to have been communities with elites people, with their northern complex utilizing Central Plains metropolitan bronze culture, came down from the style bronze vessel assemblages at least YanMountains and pushed the Shang as far north as Shijiazhuang if not be- people south sometime after Anyang yond.28 I. Given the uncertainties concerning the material culture affiliations of cen- The Weifang III Tradition tral Hebei in the Anyang period, the The Weifang III tradition shows both

155 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.21. Laoniupo bronze burial goods: bronze mask from 86XLIII1M41 (after Liu 2001:296, fig. 256); gu- goblets from 86XLIII2M44 and 86XLIII1M33, respectively (after Liu 2001:284, fig. 245); and jue-vessels from 86XLIII2M44 and 86XLIII1M33, respectively (after Liu 2001:285, fig. 246).

Datuotou and Zhukaigou “inﬂuences” and eastward dramatically changed the and was distributed in more or less culture-historical (and by further impli- the same area as the Datuotou tradi- cation, political-strategic) picture along tion (Figure 5.22; ZSKY 2003; Jiang and the northern and western borders of the Wang 2010). The emerging culture-his- Central Plains metropolitan world. torical picture is of alate Xiaoshuang- qiao-Huanbei to early Anyang–period Unfortunately, there is little or no in- abandonment of the northern loop of formation on Weifang III inter- or in- the Yellow River by the Zhukaigou tra- trasettlement structure, production, or dition (chapter 2) with aconcomitant livelihood. Evidence of tool assemblag- appearance of Zhukaigou-type material es—including polished stone shovels, culture in an arc from northern Shaanxi axes, knives, and chisels—backs up the and Shanxi to northern Hebei and Li- hypothesis that the economy of people aoning (ZSKY 2003). The implication livingatWeifangIIIsiteswasatleastpar- is that, perhaps due to increasing arid- tially based on agriculture (ZSKY 2003). ity in the Ordos region, alarge-scale Houses were generally round or oval- migration of northerners southward shaped and semisubterranean, and ap-

156 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

5.22. Weifang III ceramics (after ZSKY 2003:610; fig. 8-49). parently no large sites or structures as- more common gu-and jue-vessels that sociated with this tradition have been form the core of Anyang bronze mor- discovered so far.Tombs take the form tuary assemblages) were found along of east-west oriented, rectangular buri- with circular or “trumpet-shaped” gold als with wooden coffins and bodies laid earrings (ZSKY 2003:610). The absence out face down. Anyang-style bronze ding of bronze gu and jue mirrors ageneral and gui vessels have also been found in lack of ceramic wine vessels in Weifang anumber of Weifang III burials, such as III–tradition assemblages, suggesting Pinggu Liujiahe M1 and four tombs at that the Anyang-style bronze vessels Jixian Zhangjiayuan. In these five exam- were being put to use in local practices ples, bronze ding and gui (but not the that were unrelated to Central Plains

157 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.23. Weiyingzi-tradition ceramics (after ZSKY 2003:616; ﬁg. 8-50).

metropolitan traditions. In addition to ZSKY 2003). Not many houseshave been Anyang-style bronze vessels, northern- excavatedatWeiyingzi sites,but those complex bronze tools and weapons also that have areeither round oroval-shaped appear in Weifang III sites along with semisubterranean or surface dwellings. gold earrings, gold bracelets, and hair- Relativelymoretombshave beenexca- pins (ZSKY 2003), broadly linking Wei- vated—oftenlocated besideresidenc- fang IIIwith metallurgical traditions es—alltaking theform of rectangular stretching from northern China and pits. Thelargesttombknown from this southern Siberia to Central Asia. tradition actuallydates to theWestern Zhouperiod.This isWeiyingzi tomb The Weiyingzi Tradition 7101 (LiaoningshengBowuguan Wenwu To the east of theWeifang III tradition, Gongzuodui 1977; ZSKY2003),which is the Weiyingzitraditionwas distributed almost9m2in area and hasinnerand in modern LiaoningProvince overmuch outer coffins, silks, pieces of bronzear- of what was formerlythe distribution of mor,chariot fittings,and gold bracelets. the LowerXiajiadian tradition. Perhaps The tomb wasoriented east-west,asare not surprisingly, the Weiyingzi tradition mostWeiyingzitombs. Bronzearmor, is said to show contactswiththe Weifang weapons, andevenCentral Plains–type III tradition,withinfluencesofLower bronze vesselshavealso been found in Xiajiadian,with Zhukaigou, as well as other Weiyingzi tombs.29 Bronzehoards with traditions farthereast(Figure 5.23; found inthe Weiyingzi areacontained

158 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

5.24. Zhenzhumen-tradition ceramics (after ZSKY 2003:462, fig. 8-4). assemblages of Central Plains bronzes, blanket replacement of Yueshi ceramic some with inscriptionsindicating their traditions with Central Plains mate- origin in the Western Zhou polity of Yan, rial culture is too simplistic and while based near modernBeijing. Sincethese there may still have been non-Central belong to theWesternZhouperiod,how- Plains ceramic production communi- ever, we will not discussthemhere.Wei- ties in subpeninsular Shandong, on the yingzisites have also turnedupweapons, Shandong Peninsula itself, the Yueshi- tools,and ornaments thatsuggest broad derived Zhenzhumen tradition definite- northern complex affiliations, suchas ly predominated. This tradition is said ring-pommeledknives,socketed axes, to develop out of the Zhaogezhuang andsocketedgedagger-axes, helmets, Yueshi regional variant and is located gold earrings, andgold bracelets. in roughly the same area. Since little excavation work has been done and even The East less published on sites of this tradition, information is limited (Figure 5.24). Ap- The Zhenzhumen Tradition parently few bronze artifacts have been As mentioned earlier, the east was the found at Zhenzhumen sites, and what one direction that saw acontinued ex- has been found all appears to be of Cen- pansion of the Anyang ceramic tradi- tral Plains origin (ZSKY 2003). In gen- tion. While the general picture of a eral, the picture that emerges from the

159 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

east is one of creeping Central Plains an ancient copper mine has been dis- metropolitan material cultural domi- covered at Tongling, Anhui, near the nance at the expense of local traditions, Yangzi. This mine may have been in awavethathad reached the coast but operation as early as the Anyang period not the Shandong Peninsula. This sim- (AWKY 1992; ZSKY 2003). In general, ple image, however, almost certainly however, it appears as though whatever belies amuchmorecomplicated web networks had brought and kept the ce- of social, economic, political, ethnic, ramics of Anhui south of the Huai River and military networks (Campbell 2007, within the Central Plains tradition orbit 2009). from the seventeenth through the four- teenth centuries BCE had weakened by The Southeast Anyang times. Whether this was due to aresurgence of local independence, the Aceramic tradition change occurred natural result of centrifugal Erligang during Anyang times in the general area and Xiaoshuangqiao-Huanbei expan- wherethe Xiaoshuangqiao-Huanbei sion, the movement of the metropoli- tradition Dachengdun variant was lo- tan center farther north, or changes in cated (central Anhui). While Central economic networks is not clear. Plains traditions dominate in northern Anhui, sites in the Lake Chao area In Jiangsu Province on either side of the show increasingly strong local charac- Yangzi River, the Hushu tradition was teristics (ZSKY 2003). Scattered finds still distributed more or less where it of tombs in the Yingshang County area was in Xiaoshuangqiao-Huanbei times, just north of the Huai River in the 1970s although there is areduction in the and 1980s produced numerous Anyang- number of sites in the Anyang period type bronze vessels in Anyang-type elite (ZSKY 2003). Hushu bronze-casting also assemblages, including bronze chariot appears to remain confined to small fittings, weapons, and tools. Neverthe- weapons and tools (ZSKY 2003). less, central Anhui is an archaeologically underdeveloped area, and more work The South needs to be done. Post-Panlongcheng Northern Hubei In the Lake Chao area, at the Dacheng- as mentioned earlier, by Anyang times dun site itself, the Anyang-period layer Panlongcheng had been abandoned (layer IV) is said to only contain early and the Middle Yangzi was beyond the Anyang-period ceramics, and even distribution of Central Plains ceramic these show marked local characteris- traditions. In the former area of the Pan- tics (Anhuisheng Wenwu Kaogu Yan- longcheng-variant distribution, afew jiusuo [AWKY]1987; ZSKY 2003). While sites with Anyang-period remains have apparently none of the Xiaoshuangqiao- been discovered, but their affiliation, Huanbei period Central Plains metro- other than being non-Central Plains, is politan tradition sites south of Lake unclear (ZSKY 2003:474). Nevertheless, Chao have Anyang–period remains, Anyang-type bronze vessels continue to

160 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

5.25. Zhouliangyuqiao-tradition ceramics (after ZSKY 2003:476; fig. 8-8). be found in the area, perhaps suggest- The Duimenshan–Feijiahe ing some level of continued contact Tradition with the Central Plains world. Farther south, to the east of Lake Dongting, the Erligang and Xiaosh- The Zhouliangyuqiao Tradition uangqiao-Huanbei outpost of Yueyang In roughly the area formerly occu- Tonggushan had been abandoned by pied by the Jingnansi tradition, the Anyang times, and the surrounding area Zhouliangyuqiao tradition emerged was apparently home to aflourishing during the Anyang period. While not bronze industry.Manyfamousbronze much is known about this tradition vessel finds have been discovered here, apart from its formal ceramic typologi- including the human-faced ding,the cal features, it is said to show “influenc- four-ram zun,and (possibly) the tiger- es” from northern Hubei (Figure 5.25; consuming-man-you.Large nao-bells ZSKY 2003:477). Bronze vessels have have also been discovered in this area, also been discovered at or near sites as- and Bagley (1999) notes that the general sociated with this tradition, such as the bronze repertoire is significantly dif- large zun discovered at Shashi Dongyue- ferent from that of the Central Plains cun and the two zun discovered at Jian- (bells, zun, lei,and animal-shaped ves- gling Cenhemiaoxingcun (ZSKY 2003). sels) as is their depositional context: pits and mountaintops, but generally

161 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.26. Duimenshan-Feijiahe–tradition ceramics (after ZSKY 2003:479; ﬁg. 8-9).

not burials. The discovery of the West- cauldrons and arelative abundance of ern Zhou site of Tanheli, however, puts stoneware (glazed and unglazed), which in doubt the Anyang-period date for the is nonetheless distinct from Jiangxi- bronzes found in the area. Instead, Xi- type stamped stoneware (Figure 5.26; ang (2006, 2008) cogently argues that ZSKY 2003). Unfortunately, other than the Tanheli site and some of the bronz- formal bronze and ceramic typologies, es (with Anyang clan insignia) found little or no information exists on the in the area indicate the presence of a settlements, production sites, or resi- group of post-Zhou-conquest Shang ref- dences of Duimenshan-Feijiahe ceram- ugees, possibly in part from one of the ic producers. southern polities of the Shang hegemo- ny and likely forming aminority of the The Baota Tradition local population. The resulting hybrid In northwestern Hunan, the Erligang- culture persisted through the Western and Xiaoshuangqiao-Huanbei–period Zhou period, but beyond the reach of Baota tradition underwent dramatic the Zhou hegemonic polity. change in the Anyang period, with local features becoming much more The Anyang-period ceramic tradition pronounced. The northern-derived associated with this area, Duimen- li-tripod all but disappeared from the shan–Feijiahe, is characterized by ding- ceramic assemblage, and vessel forms

162 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5 underwent changes that reﬂect the de- semblage31 included Erligang- to An- parture of Central Plains metropolitan yang-period bronze vessels, with the inﬂuence from the area (ZSKY 2003). latter being both more abundant and Nevertheless, these observations need showing more non-Central Plains char- not be seen mechanically in terms of acteristics (Figure 5.27) (Jiangxisheng amonolithic Shang state’s intrusion Wenwu Kaogu Yanjiusuo, Jiangxisheng and retreat, but rather the disintegra- Wenwu Bowuguan, Xinganxian Bowu- tion of the Xiaoshuangqiao-Huanbei– guan 1997). In general, the tomb shows period Panlongcheng-variant ceramic evidence both of widespread networks production and distribution networks of prestige goods (north to Henan, with their (as yet unclear and undoubt- west to Hubei, and perhaps beyond to edly dynamic) political, social, econom- southern Shaanxi32)and, at the same ic, and/or military connections to the time, evidence of local styles and of lo- northern metropolitan centers. cal uses of prestige goods. Moreover, Xingan is an indication of the wealth The Wucheng Tradition and sophistication of elites in northern During the Anyang period, Wucheng- Jiangxi, rivaling all but the highest ech- tradition ceramics, like all traditions elons of Anyang society.Nevertheless, in the Yangzi area, showed increasing- much more work needs to be done in Ji- ly local characteristics. The so called angxi Province and in the south in gen- yan-shaped vessel, probably derived eral before anything like an adequate from asimilar vessel that appeared picture emerges of social and political in neighboring Wannian tradition as- entities in the Wucheng-tradition area. semblages, became more common, How many large centers aside from even while northern-derived li-vessels Wucheng and Niutoucheng were there? became rarer, smaller, and stranger Were they simultaneously or serially oc- (from aCentral Plains perspective)—so cupied? What was the relationship be- small, in fact, that they likely no longer tween larger and smaller sites, and did functioned as cooking vessels (JBKY,ZB regional settlement patterns change 2005). Stonewares and protoporcelain over time? None of these questions can also increased at the expense of earth- be answered on present evidence, and, enwares. No bronze artifacts have been despite the passage of over 30 years discovered at the site of Wucheng from since the discovery of Wucheng, we still this period despite the discovery of have little beyond tantalizing clues. stone molds in phase III (Anyang-period) layers (JBKY,ZB2005). As in the Xiaoshuangqiao-Huanbei period, eastern Jiangxi was still apparently Also dating to the Anyang period is the occupied by the scattered, small sites famous Xin’gan tomb, located some (no larger than 10 ha) of the Wannian 5kmfrom alarge Wucheng-tradition tradition. This tradition was charac- walled site, Niutoucheng.30 This tomb terized by large quantities of stamped is the second richest burial yet discov- stoneware, glazed ceramics, and pro- ered from the Anyang period. Its as- toporcelains (ZSKY 2003; Peng 2004).

163 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.27. Wucheng-tradition bronze artifacts from the Xin’gan tomb (after ZSKY 2003:488, 489, ﬁgs. 8-12,13).

Aside from formal ceramic typology nuity with Sanxingdui. Houses, kilns, and the fact that they apparently did not tombs, and ceremonial areas have all have ametallurgical tradition that has been discovered, along with over 2,000 left any traces, little information exists artifacts of gold, bronze, jade, stone, concerning sites of this tradition. and ivory (SWKY 1999; ZSKY 2003). The famous artifact pits at Sanxingdui are Southwest also thought to date to early phases of the Anyang period. The Sanxingdui-JinshaTradition During the Anyang period the Sanx- Pits 1and 2atSanxingdui have been the ingdui tradition continued to ﬂourish topic of many articles in both Chinese in the Sichuan basin, although during and English and need no detailed intro- that same epoch, the political and cul- duction here. As noted above, they ap- tural center is thought to have shifted pear to have been exemplary examples from Sanxingdui to the Jinsha site of awider practice of burying artifacts (ZSKY 2003). The Jinsha site, discov- in pits in the Sichuan basin.33 In any ered in 2001, is 300 ha in size and shows case, the 1,700 plus gold, bronze, ivory, many signs of material cultural conti- jade, stone, bone, and ceramic objects

164 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5 collectively placed in these two pits, other resource that is hypothesized to with evidence of burning and breakage, have come to Anyang from Yunnan via clearly demonstrate both astrikingly Sichuan is lead. The argument, put for- unique assemblage of artifacts34 and ward in Jin et al. (1994, 1995, 1998), is radically different practices of terminal that many Anyang bronzes have ahigh use or “enclaving” (Kopytoff 1986). At percentage of arare lead isotope that is the same time, the inclusion of Central only found in Yunnan. Moreover, sites Plains-type bronze vessels35 in the pits in the Middle Yangzi (such as Xin’gan) both suggests apractice of collecting and the artifact pits of Sanxingdui have and raises the issue of the date of the even higher percentages of bronzes unique Sanxingdui-type bronze masks, with this isotope, an entirely logical figures, and other artifacts. distribution if Yunnan was, in fact, the source via Sichuan. Jin argues that lead Although social and political informa- ingots were being shipped from Yun- tion is largely lacking, in the Anyang pe- nan to Sichuan and beyond, perhaps riod, the Sichuan basin was still home along with copper, since there are na- to the Sanxingdui tradition, character- tive copper deposits in Yunnan, and, ized by large walled sites and elite craft the argument goes, it might have been industries that rivaled those of their more economical to ship easily mined known contemporaries. In terms of native copper from Yunnan than mine it contacts and networks, although Sanx- closer to home. Although this argument ingdui bronze masks, statues, and trees is controversial and obviously needs do not appear outside of Sichuan, large, more work toward demonstrating ac- “toothed” jade zhang are widely distrib- tual (rather than hypothesized) trade uted, and, given the quantity and size routes, the distribution of artifacts with of the those found in Sichuan, it may this isotope seems to support Jin’s hy- be that some of those found elsewhere pothesis (not to mention that the dis- originated here (see also So 2001). As tribution of artifacts with this isotope noted above, the Central Plains-style shrinks to Sichuan in the early Western bronze vessels found in the artifact pits Zhou and then only appears in Yunnan resemble Middle-Yangzi artifacts not before disappearing from the archaeo- only in design, but in type: predomi- logical record). Nevertheless, the exact nately zun-and lei-vessels. Interesting- source of this lead, in Yunnan or else- ly, many of the 4,700 plus cowrie shells where, has yet to be found. Other spe- discovered in the Sanxingdui artifact cialists have, moreover, questioned the pits were discovered inside bronze zun, results of Jin et al. (1994, 1995, 1998), lei,and masks, apractice also found in claiming instead that the source of the Yunnan Province in later times. This lead could be in the Qinling Mountains latter connection is doubly interesting of Shaanxi Province (Saito et al. 2002). If given that one of the major sources of these hypothesized networks in fact ex- the cowrie shells found in abundance isted, it would suggest that, despite no at Anyang36 is purported to be the In- evidence of direct contact, the Sanxing- dian Ocean, perhaps via Yunnan. An- dui polity(ies) may have played an im-

165 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.28. Lujiahe-tradition ceramics (after ZSKY 2003:515; ﬁg. 8-19).

portant, if indirect, role in some Anyang lished on this tradition (ZSKY 2003), economic networks. and much remains unclear.Asone might expect, the Lujiahe tradition is The Lujiahe Tradition said to show “influences” of both Sanx- ingdui tradition and Central Plains- Whereas the Chaotianzui variant of derived Panlongcheng tradition or its Sanxingdui tradition had flourished in Middle Yangzi inheritors (Figure 5.28). the Yangzi River gorges area during the Interestingly, “one of the most impor- “middle period of Sanxingdui” (ZSKY tant ritual vessels” of this tradition is a 2003:513), which perhaps included the ceramic lei that imitated Middle-Yangzi first half of the Anyang period, the Lu- bronze lei-vessels (ZSKY 2003:516). jiahe tradition is tentatively dated to the last half of the Anyang. Sites are said to be densely distributed but small in size (generally <100 m2). Despite anumber of excavations, very little has been pub-

166 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

5.29. Nianzipo-tradition ceramics and stone tools (ZSKY 2003:527, ﬁg. 8-23).

The Baoshan Tradition hypotheses that the Hanzhong bronzes were produced in amultitude of places During the Anyang period, the Baoshan with avariety of alloys from different tradition continued to ﬂourish, its ce- ores, including some local artifact types ramics showing contacts with the Lujia- (scepter and sickle-shaped objects), he, Sanxingdui, and Shaanxi traditions, which have distinctive alloys as well as even while its bronze repertoire links stylistic features (Mei et al. 2009; Chen this area to ageneral sphere of Cen- et al. 2009). tral Plains-derived bronze industry, to Middle Shaanxi (including Laoniupo), The West Sichuan, and even Jiangxi37 (XBX 2002; ZSKY 2003; Cao 2006; Luo 2011). Alocal As mentioned above, by Anyang phase tradition is also suggested by objects, II, Central Plains metropolitan tradi- such as bovine and human plaque/ tion sites had disappeared from their masks, triangular ge or kui,and sickle- Xiaoshuangqiao-Huanbei–period dis- shaped ge38 (Xibeidaxue Wenbo Xueyu- tribution in western Shaanxi, to be an [XWX] 2002; ZSKY 2003; but see Luo found only east of Xian from that point 201139). Recent archaeometric studies until their total disappearance at the of casting technique, alloy composi- end of Anyang period. During this time, tion, and ore sources seem to back the anumber of new traditions emerged in

167 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

western Shaanxi, some occupying sites tripods,” this ceramic tradition is said that had previously seen Central Plains to date from Anyang Iand perhaps even metropolitan tradition (Beixin variant) earlier and lasts until the beginning occupation in the Xiaoshuangqiao- of Anyang phase IV (Figure 5.29; ZSKY Huanbei period. 2003:525–528). Some archaeologists have argued that this tradition should West-Central Shaanxi be considered Proto-Zhou (Zhongguo Shehuikexueyuan Kaogu Yanjiusuo By Anyang phase II, Central Plains ce- Jing-Wei Gongzuodui 1989), while oth- ramic traditions had disappeared from ers have argued it is avariant of the Liu- the area west of Xian, replaced by jia tradition (Zhou 1988). The ceramics “lo-cal” traditions (or types),40 such as of this tradition also show similarities Nian-zipo, Liujia, and Zhengjiapo. In with the Siwa tradition to the west and Anyang times, Guanzhong, west of the Lijiaya tradition to the northeast. Xi’an, was apatchwork of interpenetrating tradi-tions, many of The Liujia Type which were probably produced by Distributed south to the northern ethnic and political groups that would slopes of the Qinling Mountains and participate in the coalition that north to Pingliang, east to Fufeng, and overthrew the Shang king at An-yang west to Tianshui in Gansu, this tradition and ushered in the Zhou dynasty.41 overlaps with Nianzipo to some degree, although generally its distribution is The Nianzipo Type farther south. Fewsites of this tradition The southern limit of this type of site is have been excavated (ZSKY 2003), and Qishan, while it extends west to Pingli- what is known about it comes mostly ang, Gansu and east as far as the Jing from burials (Figure 5.30). These tend River: essentially the valleys and up- to be small, generally “cave” or “niche” lands north of the western end of the burials, which are dug into the sides of WeiRiver Valley in Shaanxi. The houses pits with aramp. Burial goods mostly of this tradition take the form of subter- comprise ceramic vessels, frequently a ranean, semisubterranean, and surface single li-tripod, but sometimes avaria- constructions, with subterranean hous- blenumberofothervessels(ZSKY2003). es built into embankments being the Bronze tubes, bells, and beads are also most common. The burials discovered sometimes found in Liujia tombs. The associated with this tradition are all cemetery at Liujia itself (in the Zhouy- small and generally rectangular pit bur- uan) can be divided into three phases, ials, although “cave” burials dug into the first two of which are of the Liujia embankments also appear.Although tradition and the third of which is of the most of the burials have wooden coffins, Zhengjiapo tradition (ZSKY 2003). The some have stone coffins. Most burials Liujia phases are dated to Anyang phase have only one or two ceramic vessels, II and III respectively.The ceramics of generally li-ordou-vessels. Character- the Liujia tradition show many similari- ized by “divided crotch, bag-footed li- ties with those of Nianzipo (including

168 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

5.30. Liujia-type tomb, bronze bell and tubes, and ceramics (after ZSKY 2003: 529; ﬁg. 8-24). divided crotch, bag-footed li-tripods), Zhengjiapo tradition was distributed prompting some to argue that they are mostly between Xian and the Zhouy- in fact the same tradition. uan, in the area formerly occupied by Xiaoshuangqiao-Huanbei Beicun–vari- The Zhengjiapo Type ant sites. By Anyang phase IV,however, By approximately Anyang II, the Zhengjiapo-tradition sites could be

169 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.31. Zhengjiapo-type artifacts (after ZSKY 2003:532; ﬁg. 8-25).

found as far north and west as Gansu the most common (like Nianzipo-type Qingyang (ZSKY 2003). The houses of sites). Burials tend to be placed in dis- this tradition include subterranean, crete cemeteries, and the majority of semisubterranean, and surface dwell- them are small rectangular pits. Most ings, with subterranean houses being burials have either asingle ceramic

170 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5 li-tripod or a li-tripod and guan-pot. A the other neighboring traditions. The few tombs also have bronze ding-or pre-Zhou remains are said to date to the gui-vessels, while others have bronze ge end of the Anyang period, when, based dagger-axes and/or bronze beads. The on transmitted traditions, King Wenof joined crotch li-tripods and other as- the Zhou is supposed to have founded pects of the ceramic assemblage distin- his capital, Feng, at this site. Interest- guish this tradition from neighboring ingly though, the putative predynastic traditions, while nonetheless showing Zhou tombs are basically Central Plains Liujia, Nianzipo, Central Plains ceram- metropolitan-style tombs, although the ic, and Central Plains and Northern bronze vessel assemblage (ding-and Complex bronze tradition “inﬂuences” gui-vessels) lack the gu and jue typical (Figure 5.31). The Zhengjiapo tradi- of Anyang bronze mortuary assemblag- tion has been implicated in the search es, and the ceramics show non-Central for the predynastic Zhou people; two Plains tradition origins (Figure 5.32). prominent scholars claim that only Zhengjiapo strata containing divided If there is aconsensus concerning crotch, bag-footed li-tripods should be Shaanxi west of Xian during the 150 or considered predynastic Zhou, while lay- so years during the rise of the Zhou and ers without this diagnostic ceramic type their subsequent conquest of the Shang belonging to the Western Zhou (Zhang world, it is that the material cultural and Liang 1989). The fact that there picture is complicated. But then, given is debate concerning whether certain the textual tradition that the Zhou arose Zhengjiapo deposits are pre- or postdy- among apanoply of non-Shang peo- nastic Zhou suggests astrong similarity ples and led alarge alliance against the between Zhengjiapo and Western Zhou Shang, this confusing material cultural ceramic traditions, whatever the ethnic picture is, in fact, in line with Zhou rep- or political afﬁliation of the potters who resentations of their own origins. made them, people who distributed them, or consumers who used them. Farther west, the Xindian and Siwa traditions were still distributed in east- The Fenghao Site ern Gansu, Qinghai, and even western Both tombs and living areas with what Shaanxi in the case of Siwa. From the are believed to be predynastic Zhou re- perspective of material culture and mains have been discovered distributed burial practices, Xindian and Siwa sites along arise between Keshengzhuang appear to share at least some similari- and Zhangjiapo. The four pre-Zhou ties with Nianzipo and Liujia type sites. tombs discovered in the area contained What this might mean in political, waist-pits and ledges with death attend- economic, or cultural terms, however, ants, as well as bronze vessels, weap- awaits future research. ons, and chariot equipment. The ceramics excavated from the burials and living sites include types that resemble Zhengjiapo ceramics, as well as those of

171 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.32. Fenghao site pre-Zhou ceramics (after ZSKY 2003:534; ﬁg. 8-26).

The Northwest insignia bing “丙”(Shanxisheng kaogu yanjiusuo 2006). Interestingly, while the West of the Taiheng Mountains, very burials were basically of Anyang-style, few Anyang-period sites have been dis- the ceramics of this and other Anyang- covered in southern and southeastern period sites in Shanxi show non-Cen- Shanxi. In the former distribution area tral Plains characteristics, including of the Erligang Dongxiafeng and Xia- “Shaanxi inﬂuences” (ZSKY 2003:322), oshuangqiao-Huanbei Xiaoshen vari- even while some of the bronze artifacts, ants, only afew scattered ﬁnds of mid- including an animal-headed knife and dens and burials attest to an Anyang ascepter, also suggest Northern Com- period Central Plains tradition pres- plex connections (Figure 5.34; see also ence in the area. Thorp 2006).

At Lingshi Jingjiecun, three large un- In middle Shanxi, the Jinzhong tradi- ramped (7–9 m2)Anyang-period tombs tion shows increasing Central Plains were discovered (Figure 5.33). Two of metropolitan influence during the lat- the tombs had waist-pits (containing ter part of the Xiaoshuangqiao-Huanbei dogs) and death attendants; all three and beginning of the Anyang periods, had inner and outer coffins and assem- followed by alarge influx of Northern blages of Anyang-style bronze vessels (in Complex features via the Lijiaya tra- addition to weapons, tools, and chariot dition (ZSKY 2003). In terms of sites, equipment), many of which carried the Taigu Baiyan, Xinzhou Liansigou, and

172 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

5.33. Lingshi M1 (after Shanxisheng kaogu yanjiusuo 2006:16, 86, figs. 8, 93).

Fenyang Xinghuacun all have Anyang- remains) and later Jinzhong-tradition period Jinzhong-tradition remains. remains dating to the latter part of the Interestingly, Xinhuacun has some Anyang period (Jinzhong kaogudui early Anyang-period burials (and more 1989; ZSKY 2003). Jinzhong ceramic li- extensive Xiaoshuangqiao-Huanbei tripods are said to come in three styles:

173 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.34. Bronze artifacts showing Northern Complex influence or affiliation from the Lingshi tombs (after Shanxisheng kaogu yanjiusuo 2006:134, 178, 84, 78 figs. 150, 203, 91, 85).

Central Plains metropolitan-style, a local characteristics. These ﬁnds can local “intermediate” style, and abag- be divided into Central Plains metro- footed, Shaanxi-type li.Unfortunately, politan-type bronze vessels and weap- the provenance of vessels in each style ons (Figure 5.35); Northern Complex is unknown, and so, whether the variety weapons, tools, and ornaments; and is due to trade, aheterogeneous popu- vessels of “mixed” type (Figure 5.36; lation, or experimentation by local pot- ZSKY 2003:585). Although assemblages ters, also is not known. Tenbronze ves- vary in terms of ratios of Central Plains sels discovered at Xinzhou, however, versus Northern zone bronze artifacts, were all Anyang-style and most likely their distribution interpenetrates even imports (ZSKY 2003:573). while the Northern zone-type artifacts show common characteristics, sug- The Lijiaya Tradition gesting that the Jin-Shaan Plateau was Since the 1950s, Anyang-period bronze azoneofcontact between the Central artifacts have been discovered in over Plains and Northern Complex bronze twenty places along the course of the traditions. Yellow River as it runs south between Shanxi and Shaanxi provinces.42 These While none of the bronze artifact ﬁnds artifacts are generally thought to date of the Jin-Shaan Plateau area were ar- to the second half of the Anyang period chaeologically excavated and, thus, can- (phases III–IV), and many show striking not be directly related to particular sites

174 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

5.35. CPBA-style bronzes from the Jin-Shaan Plateau area (after ZSKY:2003: 587, ﬁg. 8-40). or cultural strata, survey and excavation although afew tombs contained bronze in the region has turned up anumber axes (ZSKY 2003). of sites to which have been assigned the name Lijiaya “culture” (ZSKY 2003). At another nearby site, Suide Bijiaqu, The Qingjian Lijiaya site itself, on the subterranean houses, storage pits, and Shaanxi side of the Yellow River, is a medium-sized tombs were discovered. roughly 6.7 ha fortress, with 100 mcliffs Unfortunately the tombs were com- to the north and south, cliffs and walls pletely looted. Nevertheless, bronze to the west, and awallofpiled stone and artifacts are known to have been dis- earth blocking the eastern approach.43 covered at the site and at two other lo- Kilns, tombs, and houses were all found cations nearby.Scapulimancy with pre- within the site, including arelatively pared and chiseled sheep bones was large rammed-earth building (approx. also practiced at Bijiaqu (ZSKY 2003). 50 m2),within arammed-earth walled courtyard (approx. 1,000 m2)with gate- The ceramics of these sites show strong houses at the southern entrance. The similarities with each other, suggesting burials were almost all rectangular pits acommon tradition (Figure 5.37; ZSKY with cofﬁns and few or no grave goods, 2003). The Lijiaya tradition, moreover,

175 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

5.36. Northern Complex bronze and gold artifacts of the Late Shang Jin-Shaan Plateau (after ZSKY 2003:588–-589; ﬁg. 8-41).

appears to derive from the Zhukaigou and II oracle-bone inscriptions, which tradition, which disappeared from the show an overwhelming military and po- Ordos region about the same time that litical concern with the north and west. the Lijiaya tradition appeared afew Indeed, if Xia’s (2005) reconstruction hundred kilometers downriver. of the Gongfang wars are correct, it was at the end of Anyang period II that the In general, the northern and western Anyang court and its local allies lost the frontiers of the Central Plains material war in Shanxi, permanently removing cultural world show dramatic changes Anyang political influence in the area, and the influx of non-Central Plains tra- roughly the same period that Lijiaya ditions into areas where Central Plains sites, and mixed bronze assemblages variants were formerly distributed. appear on the Jin-Shaan Plateau. The Anyang period is also the period in which the chariot appears, an im- Conclusion port from the steppe, and these intru- sive ceramic traditions all show some The Anyang period, rather than aperi- degree of affiliation or contact with od of political devolution or of arecon- Northern Complex bronze traditions. stituted but weak “Shang state” over- Moreover, this material-cultural situ- shadowed by the glory of its Zhengzhou ation is paralleled in Anyang period I predecessor, was aperiod that saw the

176 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

5.37. Lijiaya-tradition ceramics (after ZSKY 2003:590; ﬁg. 8-42). rise of “the Great Settlement Shang” at than any other second-millennium BCE Yinxu. Yinxu, Anyang was asite signiﬁ- site in China, it is also true that Anyang cantly larger than Zhengzhou and was metropolitan style bronze vessels vastly at the center of ametropolitan tradition outnumber Erligang or Xiaoshuang- variant of ceramic production, bronze- qiao-Huanbei bronzes excavated or in casting, ritual practices, and architec- collections around the world, suggest- ture of unmatched homogeneity and ing that the bronze industry at Anyang distribution (Figure 5.1). The site of reached an entirely new scale. Anyang Yinxu, Anyang, was not only of unprece- elite crafts also show marked progress dented size, but also contained the rich- in technical virtuosity, suggesting an est burials, largest foundry sites, and uninterrupted development under the palace-temple area of any known con- continued patronage of powerful elites temporaneous or earlier site in East Asia from Erlitou to Anyang. Whatever the (Table 5.3). While it is true that there is political narrative may have been over amuch fuller range of data from Yinxu the course of the Central Plains Bronze

177 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Center Size Circumscribed Palace- Largest Palace- Bronze Casting Temple Area Temple ERLITOU 300 ha 11 ha 9000 m2 1location Yanshi 200 ha 4.5 ha 9000 m2 2(?) locations ZHENGZHOU 290 ha (inner 37 ha 2000 m2 2locations walls);1300 ha (outer walls); 2500 ha estimated total site Panlongcheng 75 ha none (?) 6000 m2 ? Xiaoshuangqiao 150 ha none 500 m2 1location HUANBEI 470 ha 20+ ha 16,000 m2 ? ANYANG 3000 ha 70 ha 5000m2(?) 6locations

Table 5.3. Comparison of major sites, 1800–1050 BCE.

Age, it apparently did not involve acol- horizon gave way to asouthern inter- lapse. mediate period with continued interaction and exchange. In the West, Central Nevertheless, while both Anyang and its Plains ceramic tradition inﬂuences con- metropolitan ceramic variant distribu- tinued to ebb even while some contact tion were of unparalleled size, the An- was apparently maintained between yang period saw the retreat of Central the elites of Anyang and at least some of Plains metropolitan ceramic tradition those of the WeiRiver Valley.Thisagain variants on all fronts except the east. suggests that political relationships Given the sudden appearance of the and elite material culture circulated in chariot, the encroachment of northern networks different from those respon- complex inﬂuences in the northwest sible for producing ceramic traditions. and north, and the early period oracle- Moreover, an Anyang-period shrinkage bone inscriptions that indicate trou- in the distribution of Central Plains bled northern and western borders, the tradition ceramics should be no more reduced distribution of Central Plains considered an unambiguous sign of a tradition ceramics in the north and weakened state and decentralized polit- west may have had less to do with the ical situation than the Xiaoshuangqiao- internal weakness of the Shang polity Huanbei period “maximum” distribu- centered at Anyang than with external tion the sign of astrong, centralized factors. The situation in the south af- and expansionist polity.The Anyang-pe- ter the demise of the widespread Pan- riod ebb of Central Plains metropolitan longcheng variant by the end of the Xi- ceramic tradition distribution in the aoshuangqiao-Huanbei period is one of south, west, and north may be partially multiple interacting regional traditions related to the movement of the metro- with no one center of cultural gravity. politan area (and its attendant popu- One might say that the Panlongcheng lation nucleation) north and east, fol-

178 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5 lowed on by military pressure from the at least Anyang phase II, while tin may north and west. have been procured from Yunnan or Ji- angxi (Jin 1990). Some of the jade found In contrast to ceramic traditions, An- in Anyang tombs has been claimed yang metropolitan-style bronze ves- by some scholars to be nephrite from sels continued to circulate widely even Hetian in Xinjiang (Zhang 1982; Shen while local casting traditions ﬂourished 1991) and, if true, would constitute an on the peripheries of the Central Plains exchange network thousands of kilom- metropolitanworld.IfBagley(1999:208) eters long. Even longer exchange routes can claim, in reference to bronze-cast- have been suggested for cowrie shells, ing, that there was “no Late Shang ho- which, based on their modern distribu- rizon,” it nonetheless remains true that tion, may have originated in the Indian the distribution of metropolitan-style Ocean (Li 2003; Peng and Zhu 1995). Central Plains bronze artifacts was no Protoporcelain and stamped and glazed less extensive in the Anyang period than stoneware are also generally thought it was in the Erligang period. The differ- to have been imported from the south, ence between the Erligang and Anyang with Wucheng apossible site of produc- periods rather resides in the greater dis- tion (Chen et al. 1999), as well as sites persion of bronze-casting both within in Zhejiang (ZWKY et al. 2011). Lacquer and beyond the metropolitan ecumene and shell inlay may also have been ob- by Anyang times. This trend toward the jects of exchange (Fang 2009), not to proliferation and spread of bronze-cast- mention the undoubtedly numerous ing, moreover, can already be seen in things that left no trace in the archaeo- Erligang times with casting occurring at logical record. Nevertheless, while it is multiple sites in Zhengzhou, at Yanshi, apparent that Anyang was aperiod of and probably at Panlongcheng. long-range regional exchange, the nature of the networks, their participants, There is, moreover, evidence of contin- and facilitators, or even the routes re- ued contact and long distance exchange main all but unknown. networks linking Yinxu with its non- Central Plains metropolitan tradition neighbors. Firstly, if Yinxu was the site of an unprecedented bronze industry, not only casting more, but larger, and thicker-walled vessels than at any time previously, where did it obtain its metals? The Yangzi area and Shandong were both possible sources for copper, while lead isotope studies have suggested that Anyang shared alead source with the bronze industries of Zhengzhou, Wucheng, Panlongcheng, and Sanx- ingdui (Jin et al. 1994, 1995, 1998) until

179 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Endnotes sonal communication]). Nevertheless, they were 1See Li (1977) for an understanding of used in hunting, as oracle-bone examples attest, so Anyang based on prewar excavations (and abibli- presumably they were at least effective as mobile ography of the voluminous Academia Sinica pub- archery platforms. lications on the early excavations and analyses 6Despite astrong later textual tradition based on them). Chang (1980) builds on this with that identifies the Shang king PanGeng with the later research, but much work has been done in establishment of acapital at Anyang and the long- the intervening quarter-century since Chang wrote held belief that the Anyang period begins with his Shang Civilization,rendering some of its conclu- reign, there is little or no evidence, either in oracle- sions out of date. Bagley (1999), Keightley (1999), bones, the royal cemetery, or the palace-temple Thorp (2006), and Liu and Chen (2012) supple- complex, for kings prior to Wuding at Anyang. ment Chang’s (1980) account but collectively still Some scholars have suggested instead that Huan- lag behind the current Chinese understanding of bei, just across the river to the north, might be the the site. My brief description of Anyang here will capital that Panlong was supposed to have estab- thus follow more recent developments. lished in the area (Yang and Tang 1999). 2While there is some lingering debate 7Infact, given their size, they are unlikely among oracle-bone scholars about this issue, the to have been ordinary residences and may well consensus among archaeologists is that Anyang have served some function other than dwelling. was acapital site. From an archaeological perspec- 8The idea,moreover, that there were tive, the palace-temples, large-scale industry, mon- large flows of metal coming into Anyang and ear- umental tombs, writing, not to mention the sheer lier centers for bronze-casting is probably based size of the site, make it difficult to see Anyang as on modern industrial assumptions. It is certainly anything but acentral place and an unprecedent- not founded on any estimate of contemporaneous edly large and prosperous one at that. metal consumption. My own preliminary order of 3Yoffee(2005) includes atable comparing magnitude estimate, based on extrapolating from site sizes of early urban spaces and lists Anyang as bronzes buried in tombs at Anyang, is around 19 km2,apparently based on personal communica- 1,000 tons over 200 years (Campbell 2013), or tion with Liu Li and Yates (1997). More recent esti- about 1,000,000 kgs ÷200 years =5,000kg/year, or mation of the size of the Anyang site put it at over approximately 14 kg per day.Inother words, asin- 30 km2,surpassing any earlier or contemporane- gle person-load of metal per day could likely have ous site, including Zhengzhou. supplied the Anyang foundries. Considering that 4Although cart tracks (for avehicle nar- Anyang far outstripped Erligang or Erlitou bronze- rower than achariot) have been discovered at Er- casting, the idea that some centrally administered litou (Erlitou Team 2005a), there is no evidence for “state” apparatus had to have controlled and ad- chariots or horses in the Central Plains before they ministered the flows of metal into the center in suddenly began to appear across northern China these cases seems extremely dubious. in the Anyang period (Yuan and Flad 2003). 9The idea that large-scale production re- 5There is debate about the nature of quires asingle, centralized source of resources as chariot use, however, especially in the case of war. opposed to multiple, perhaps independent sourc- Many authors are not convinced that Anyang chari- es seems to also underlie Liu and Chen’s (2003) ots would be very useful in fighting as opposed to depiction of early complex polities in China as cen- simply marking status and/or providing mobile trally controlled resource extraction mechanisms command platforms (Bagley 1999; Tang Jigen [per- (what might be termed “the Exxon model”).

180 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5

10 That is to say, although the largest vessels Shang-Zhou transition or early Western Zhou. and richest unlooted tombs have all been found 17 These are 辛又 and 役 respectively, the in the royal cemetery and the palace-temple com- former being discovered not at Daxingzhuang but plex, and it does generally seem to be true that at the nearby Liujiazhuang (Fang 2009). larger tombs are better furnished, no one has done 18 Interring roebuck canines is an ancient astudy of the correlation between bronze vessel burial practice in the Shandong region going back quality and size and other markers of rank in buri- to Dawenkou times (Li 2008). als, so the claim that elites outside the “Shang 19 My argument, in brief, is that while I court” could not have possessed vessels of acertain believe that the bronze insignia refer to kinship kind is untested. Moreover, given the fragmentary groups, these group names, even if they corre- nature of the sample, it is difficult to say what the spond to the ruling lineage, correspond only rarely upper end of the sumptuary rules spectrum looked to polity names as seen in the oracle-bone inscrip- like when so few unlooted high elite tombs have tions. Iargue that polity names are generally de- been discovered. rived from place names and that the practice of 11 While it had been asserted that alarge associating bronze insignia directly with polity midden containing cattle bone excavated at Huay- names is mistaken. Thus there may indeed have uanzhuang was also the remains of bone-working, been two separate Anyang-period political entities this conclusion now seems unlikely based on area- in the Subutun area, but they were probably not nalysis of the report (Li et al. 2011). called Chou or Ji by contemporaries (or however 12 That M1567, which appears to be unfin- those graphs were pronounced in Anyang times), ished and yielded no artifacts, was Di Xin’s tomb just as the Zhou dynasts were not called the “Ji dy- accords with traditional accounts of his suicide nasty” or the “Ji polity.” after defeat at the hands of the Zhou. If this is cor- 20 Both Liu and Chen (2003) and Fang (2005) rect then the building of royal tombs began in the make reference to asalt official (lu xiaochen 卤小 lifetime of the ruler. 臣)recorded in the oracle-bones, but neither ref- 13 Under the current directorship of Tang erences the inscription number.Myown study of Jigen, these orientations have been rapidly chang- Xiaochen officials in the Anyang period based on ing, at least at Anyang, but much of that recent both the CHANT database and Leizhuan turned work has yet to be completed or published. up only one example that could be read that way 14 In some sites in Hebei, such as Xiaqiyuan, (5596). This inscription is unique, provides no con- and Zhaoyao, li-tripods continued to form the core text concerning the individual, and the graph in of ceramic mortuary assemblages. question can be read in other ways than lu 卤 (salt). 15 Several bronzes from this and other Moreover, even if the graph in question should tombs in the cemetery bore the inscription Ya be read as lu it might refer to aplacename (as it Chou 亚醜,whichmany scholars have identified appears to in heji 36756), which would translate with Xiaochen Chou 小臣醜,who supposedlyap- as “a minor servitor of Lu”, or something else en- pears in the oracle-bone inscriptions. In fact, the tirely.This underlines the dangers inherent in at- inscription in question, which provides no context tempting to interpret unique or rarely appearing besides the name, actually reads Xiaochen Gui 小 oracle-bone graphs or phrases, and demonstrates 臣鬼,making the identification even more tenu- the need for amore rigorous approach than is fre- ous. quently current in Chinese paleographic circles. 16 All of this is assuming that this tomb did, The pernicious practice of then basing interpreta- in fact, date to the Anyang period rather than the tions of wide-ranging socio-political implication

181 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

upon such tenuous evidence is obviously even gu and jue; II. Two bronze ding-cauldrons, two more problematic. pairs of bronze gu and jue; III. One bronze ding- 21 It is important to note that whereas “non- cauldron, one pair of gu and jue;IV. Ceramic ves- Central Plains tradition communities” only refers sels only. to their material culture and certain cultural prac- (ZSKY 2003:319). This grading of tombs apparently tices, it may very well be that people producing or bears the anachronistic interpretive influence of using this material culture were members of politi- Middle Western Zhou and later more stringent cal communities affiliated with Central Plains elite sumptuary regulations where the number of cer- culture users or even the Anyang polity itself. tain types of bronze vessels are supposed to rigidly 22 Of course, the million-dollar question is reflect the status of the deceased. This, however, “what sort or sorts of interaction?” but given the was not the case for the Anyang period (Campbell orientation of Chinese archaeology toward formal 2007). typology rather than social or technical process, 28 Note, however, as the Laoniupo example the evidence that might resolve this question does shows, the presence of Anyang-style bronzes does not presently exist. not necessarily mean they were produced at An- 23 It has also been argued that this name yang. should be read “Wei Zi Qi” and associated with 29 Such as Kazuo Heshanggou Tomb 1 the grandson of the last Shang king as recorded in which had bronze hu-and you-vessels, the latter such early chronicles as the Shiji and the Zhushu- filled with cowrie shells (ZSKY 2003). jinian (Wang 2002; Matsumaru 2004). 30 This site is currently undergoing excava- 24 In fact, it is odd for large Anyang-period tion and has not been published. Central Plains metropolitan tombs to have large 31 There were in excess of 1,900 artifacts quantities of pottery (Bagley 1999). Ceramic ves- in the tomb, including more than 480 bronze ar- sels tend to be replaced with bronze vessels in tifacts, 1,072 stone and jade artifacts, and 480 large tombs so that there is an overall negative cor- ceramic vessels (about 20 percent of which were relation between tomb size and number of ceramic glazed stoneware or protoporcelain). vessels (Campbell 2007). In this sense Taiqinggong 32 Abronze head found in the tomb closely is similar to the Anyang-period elite tomb at Xin- resembles masks found at Laoniupo and even gan and perhaps suggests aHuai-Yangzi burial tra- more abundantly in the Hanzhong area of south- dition differing from Central Plains norms. ern Shaanxi. 25 Although lacquer objects do appear in the 33 Table 8-1 (ZSKY 2003) lists ten artifact royal tombs at Anyang, they are not very common pits that were discovered. The contents of the other in Anyang tombs of asimilar size and richness to pits, most of which are only partially reported, ap- those found at Qianzhangda. pear to be mainly jade or stone artifacts, although 26 Even elite remains mostly come in the at least two of the other pits had bronze plaques. formoflootedtombsorpoorlyprovenancedbronze 34 Representing, moreover, amonumental finds by farmers or workers in the 1970s and 1980s, expenditure of skilled and unskilled labor by the when central government control meant that farm- standards of Bronze Age China. ers were more likely to turn over finds to authori- 35 Nevertheless these vessels likely origi- ties than to try to sell them on the black market. nated in the middle Yangzi area and, thus, repre- 27 More specifically the burials were graded sented an indirect Central Plains influence. as follows: 36 The tomb of Fuhao alone has more than I. Three bronze ding-cauldrons, five pairs of bronze 7,000 cowries in it, and cowrie shells form almost

182 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 5 the only form of reward recorded on Anyang-period bronze vessel gifting inscriptions. 37 Not only was aBaoshan-type bronze- mask artifact found at Xin’gan, but also asickle-ge. 38 In addition to sickle-ge,“sickle-shaped” weapons have also been found in the Hanzhong area (65 altogether, in three separate ﬁnds), which are, apparently, unique to the Hanzhong area (ZSKY 2003:518). Sickle-ge have also been found in burials at Anyang. 39 Luo Tai2010, argues that the human or “primate” and bovine masks were probably made at Laoniupo and traded to the Hanzhong region. 40 ZSKY (2003) prefers to call Nianzipo, Liu- jia, and Zhengjiapo “types” because of the intense debate that surrounds the issue of the predynastic Zhou and the implications (in Chinese archaeological circles) of distinct ethnic groups that would arise by calling them separate archaeological “cultures.” 41 This of course, does not justify attempt-ing to match specific ethnic or political groups recorded in later texts with ceramic types as some authors in China have attempted to do. Named political groups may well have included heterogeneous populations or homogeneous populations incorporating heterogeneous potting traditions. 42 See ZSKY (2003:584, 585) for bibliographic references. 43 This wall had inner and outer layers and was reinforced during the occupation of the site (ZSKY 2003).

183 READ ONLY/NO DOWNLOAD

Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 6 Conclusion: The Central Plains Bronze Age from Erlitou to Anyang

Although certain Anyang material cultural aquantum leap in scale, it is not between traditions had earlier or later origins, Er- Erlitou and its Longshan predecessors, litou stands at the headwaters for much such as Taosi or Zhoujiazhuang, but rather of the complex of elite material cultural between Erlitou and Zhengzhou. Another practices that deﬁne Central Plains Bronze marked difference between the Erlitou and Age civilization. Moreover, whatever the Erligang periods (other than the re-align- political relationship between the Zheng- ment of major architecture) is the appear- zhou and Erlitou centers, the Erlitou ex- ance of anumber of walled sites in the early pansion saw no retraction but rather was stages of the Erligang period. Moreover, the incorporated into the expansion of Shang disparity in size between Zhengzhou and material cultural distribution. Erlitou, too, these sites, which otherwise share orienta- seems to serve as amodel for acapital site, tion, building techniques,and styles of ma- with its contemporaneously unparalleled jor architecture and burials, suggests their size, circumscribed palace-temple district, subsidiary status within acommon network enclosed courtyard architecture built to of elite material cultural practices. While monumental scale, and bronze-casting this phenomenon might suggest expan- workshops. At the same time, if there is sion, conquest, and, perhaps, colonization READ ONLY/NO DOWNLOAD

from the Zhengzhou core, the political re- tions cannot simply be read as areflection lationships between elites and the mecha- of conquest, colonization, and population nisms that produced and maintained them replacement. Indeed, where more detailed remain obscure beyond the apparently information exists, the picture is often one dominant status of Zhengzhou. At the same of mixed practices and selective adoption time that the urban center at Zhengzhou rather than outright replacement. Unfor- exerted expanding cultural and political in- tunately, finer-grained ceramic research fluence, it was also the site of acentripetal aimed at uncovering the practices of pro- pull that brought together diverse groups duction and networks of exchange (social, of people whose material culture homog- political, cultural, technological, and eco- enized over the roughly 200 years of Zheng- nomic) behind ceramic tradition changes zhou’s occupation. Zhengzhou,then, was is only just beginning. This research will not only the origin of metropolitan tradi- no doubt rewrite the story of Bronze Age tions, but also the site of their genesis from China, but for now we must make do with a diverse origins. much hazier picture.

Fromthepointofviewofceramictraditions, Looked at from the perspective of Central metropolitan traditions and their variants Plains metropolitan bronze industry and continued to expand in distribution from its products, there was acontinuing expan- Erlitou through to the Xiaoshuangqiao- sion in distribution throughout the second Huanbei period, after which the Anyang half of the second millennium BCE. In Er- period saw both acontraction of the met- litou times, bronze vessels and compound ropolitan ceramic tradition distribution mold casting seem to have had averyre- and an expansion of the core variant. As stricted distribution. In the Erligang pe- mentioned earlier, this phenomenon may riod, metropolitan style bronze vessels see have been related to the relocation of the agreat expansion in distribution, while metropolitan core to Anyang, followed by casting took place in multiple locations. political/military pressure from the north In the Xiaoshuangqiao-Huanbei period, and west. Thus, what were the distributions metropolitan style bronzes were even more of three separate ceramic traditions in Er- widely distributed, and foundries on the litou times (Erlitou, Luwangfen-Songyao, peripheries of the Central Plains material Xiaqiyuan), three separate Erligang vari- cultural world started casting their own ves- ants in Erligang times (Erligang, Liulige, sels in styles showing increasing heteroge- Taixi ), and two variants in Xiaoshuangqiao- neity.InAnyang times, metropolitan styles Huanbei times (Baijiazhuang, Caoyan- were still widely distributed, but local inzhuang) coalesced into asingle metropoli- dustries beyond the distribution of Shang tan ceramic tradition in the Anyang period ceramic traditions were in full swing. Given (Yinxu). At the same time, what were once the multiple centers for bronze-casting dur- the southern and western peripheries of the ing Erligang times, the expanding scale and Xiaoshuangqiao-Huanbei ceramic horizon the continued technical development of saw the development of non-Central Plains bronze-casting, along with the existence of metropolitan traditions. Nevertheless, the huge metropolitan centers throughout the expansion or contraction of ceramic tradi- Xiaoshuangqiao-Huanbei and Anyang peri-

186 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Chapter 6 ods, the long-term centrifugal progression of favor by the Anyang period. Whether this of bronze-casting technology is more likely is due to changes in patterns of warfare or the result of hundreds of years of competi- other reasons is unknown (and, of course, tive emulation of astrategic form of sym- preservation and discovery may also be fac- bolic capital on the part of local elites and tors). It was not due to alack of warfare in technologically advanced societies on the the Anyang period, nor, as evidenced by the periphery than evidence for the collapse increased scale of everything else at Yinxu, of strong Erlitou and Zhengzhou-centered due to alack of labor or resources. polities and their bronze monopolies as some authors have argued (e.g., Bagley Contra Wheatley (1971), Chang (1983, 1985, 1999; Liu and Chen 2003, 2012). 1986), and others (e.g., Shen 1994, 2003),1 Central Plains Bronze Age cities were no Combining distributions of elite and more “king’s cities” or “cult centers” than nonelite material culture across this ex- later Imperial capitals were “emperor’s cit- panse of time, at least from the Erligang ies” or ritual centers. That is not to say that period, although perhaps beginning in Er- they did not serve an important symbolic litou times, the expansion of ceramic and purpose or that nothing changed between bronze traditions was accompanied by the Shang and Qin-Han times, but that Central building of settlements sharing metropoli- Plains Bronze Age urban spaces were also tan cultural features. Distributions of Cen- centers of production and exchange even tral Plains metropolitan ceramic traditions if the particulars remain sketchy (Camp- and various forms of metropolitan elite ma- bell et al. 2011). As per Yoffee (2005), they terial culture, however, do not completely brought populations together in unprece- coincide. dented densities, undoubtedly forging new forms of social networks and identities. Politically, the archaeological picture of Part of this story is the processes of homog- North China in the second millennium enization of material culture seen at Zheng- BCE suggests the intrusion of metropoli- zhou and Anyang and the role of the metro- tan elites or the incorporation of local elites pole as cultural, political, and sacred center into metropolitan networks of elite prac- of the world. At the same time, however, at tices. This may have involved colonization, Anyang, there is evidence of kinship-based conquest, alliance, peaceful incorporation, clusters of residences and burials, suggest- or measures of each. Comparing the Erli- ing that the basic divisions of society and so- gang with the Xiaoshuangqiao and Anyang cial identity were predicated on kin groups, periods, an obvious difference can be seen whatever their speciﬁc constitution. in the building of walled Shang sites on the peripheries in the Erligang period. Howev- Endnote er, whether this phenomenon was related to 1Shen(2003), moreover, makes the erro- establishing dominion in newly conquered neous claim that early Chinese cities before the areas or to the defensive concerns of sub- eighth century BCE were “occupied mainly by royal sidiary metropolitan elites is not known. families and the ruling class” (290). Forsomereason, building walls around settlements in North China appears to fall out

187 READ ONLY/NO DOWNLOAD

ABBREVIATIONS /ACRONYMS

AT,IA, CASS | Anyang Team, Institute of Archaeology,Chinese Academy of Social Sciences AWKY | Anhuisheng wenwu kaogu yanjiusuo EFT,IA, CASS | Erlitou Fieldwork Team, Institute of Archaeology,Chinese Academy of Social Sciences HKY | Henansheng kaogu yanjiusuo HWKY | Henansheng wenwu kaogu yanjiusuo HWKY, ZW | Henansheng wenwu kaogu yanjiusuo, Zhoukoushi wenwuju HbWKY | Hubeisheng wenwu kaogu yanjiusuo HbWY | Hebeisheng wenwu yanjiusuo JBKY, ZB | Jiangxisheng bowuguan kaogu yanjiusuo, zhangshushi bowuguan LWF-SY | Luwangfen–Songyao tradition NZWKY, EB | Neimenggu Zizhiqu Wenwu Kaogu Yanjiusuo, Erduosi Bowuguan SWKY | Sichuansheng Wenwu Kaogu Yanjiusuo SWY,BDZKY,SSDQLWY,BWG | Shandongsheng Wenwu Kaogu Yanjiusuo, Beijing Daxue Zhongguo Kaoguxue Yanjiuzhongxin, Shandong Shifan Daxue Qi Lu Wenhua Yanjiuzhongxin, Binzhoushi Wenwu Guanlichu SWY,BDZKYZSW | Shandongsheng wenwukaogu yanjiusuo, Beijing daxue Zhongguo kaoguxue yanjiu zhongxin Shouguangshi wenhuaju XBX | Xibeidaxue Wenbo Xueyuan XSZDGZ | XiaShang Zhou Duandai Gongcheng Zhuanjiazu XWX | Xibeidaxue wenbo xueyuan YXK | Yinxu Xiaomintun Kaogudui ZDW, KWG | Zhengzhou Daxue Wenboyuan, Kaifengshi Wenwu Gongzuodui ZSKY | Zhongguo Shehuikexueyuan Kaogu Yanjiusuo ZSKY, AG | Zhongguo Shehuikexueyuan Kaogu Yanjiusuo, Anyang Gongzuodui. ZSKYAG, ZJHQKDK | Zhongguo shehuikexueyuan kaogu yanjiusuo Anyang gongzuodui, Zhong Jia Huanheliuyu quyu kaogu diaocha ketizu ZSKYEG | Zhongguo Shehuikexueyuan Kaogu Yanjiusuo, Erlitou Gongzuodui ZSKYHDG | Zhongguo shehuikexueyuan kaogu yanjiusuo Henan diyi gongzuodui ZSKY,HEG | Zhongguo shehuikexueyuan kagu yanjiusuo, Henan dier gongzuodui ZSKYNG | Zhongguo Shehuikexueyuan Kaogu Yanjiusuo Neimenggu Gongzuodui ZSKYSD | Zhongguo Shehuikexueyuan Kaogu Yanjiusuo Shanxi Dui ZSKY,SG|Zhongguo Shehuikexueyuan Kaogu Yanjiusuo, Shandong Gongzuodui ZSKY,ZLB,SKY | ZhongguoShehuikexueyuan Kaogu Yanjiusuo, Zhongguo Lishi Bowu- guan, Shanxisheng Kaogu Yanjiusuo ZWKY | Zhejiangsheng Wenwu Kaogu Yanjiusuo ZW,ZB, HW | Ziboshi Wenwuju, Zhiboshi Bowuguan, Huantaixian Wenwuguanlisuo

188 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Bibliography

Allan, Sarah. 2007. “Erlitou and the Formation of Chinese Civilization: Toward aNew Paradigm.” Journal of Asian Studies 66:461–496.

An Jinhuai 安金槐.1993. “Guanyu Zhengzhou Shangdai qingtongqi jiaocangkeng xingzhi de tantao” 关于郑州青铜器窖藏坑性质的探讨 (Discussion Concerning the Nature of the Zhengzhou Shang-Period Bronze Artifact Storage Pits). In Zhengzhou Shangcheng kaogu xinfaxian yu yanjiu:1985–1992 郑州商城考古发现与研究：1985－1992 (New Archaeologi- cal Discoveries and Research at the Shang Walled Site at Zhengzhou: 1985–1993), edited by Henan Sheng Wenwu Yanjiusuo 河南省文物研究所,pp. 60–63. Zhengzhou: Zhongzhou Guji Chubanshe.

———. 1997. “Zailun Zhengzhou Shangdai qingtongqi jiaocangkeng de xingzhi yu niandai” 再论郑州商代青铜器窖藏坑的性质与年代 (Further Discussion of the Date and Character of the Zhengzhou Shang Period Bronze Artifact Storage Pits). Huaxia kaogu 1997.1:73-82.

Anhuisheng Wenwu KaoguYanjiusuo 安徽省文物考古研究所.1987. “Anhui sheng Hanshan Dachengdun yizhi fajue baogao” 安徽省含山大城墩遗址发掘报告 (Excava- tion Report on the Dachengdun site, Hanshan, Anhui Province). Kaoguxue jikan 6:83–99.

———. 1992. “关于铜陵古铜矿初步考察研究的报告 (Preliminary Investigative Research Concerning the Ancient Copper Mine at Tongling).” In Zhongguo gu tongdu Tongling 中国古铜都铜陵 (China’s Ancient Copper Capital, Tongling), edited by Tongling Shizheng Xiewenshi Weiyuanhui 铜陵市政协文史委员会.Tongling: Tongling Shizheng Xiewenshi Weiyuanhui, Tonglingshi.

Anthony, David. 2007. The Horse, the Wheel, and Language: How Bronze-Age Riders from the Eurasian Steppes Shaped the Modern World.Princeton: Princeton University Press.

Anyang Team, Institute of Archaeology, Chinese Academy of Social Sciences. 2004a. “Survey and Test Excavation of the Huanbei Shang City in Anyang.” Translated by Zhicun Jing. Chinese Archaeology 4:1–20. (Originally published in Kaogu 2003.5:3–16).

———. 2004b. “A Brief Report on the Excavation of Palatial Compound F1 at the Huanbei Shang City in Anyang, Henan.” Chinese Archaeology4:21–28. (Originally published in Kaogu 2003.5:17–23).

Appadurai, Arjun. 1986. The Social Life of Things: Commodities in Cultural Perspective.Cambridge, UK: Cambridge University Press.

189 READ ONLY/NO DOWNLOAD

Bagley, Robert. 1999. “Shang Archaeology.” In The Cambridge History of Ancient China,edited by Michael Loewe and Edward L. Shaughnessy, pp. 124–231. Cambridge: Cambridge Univer- sity Press.

———. 2004. “Anyang Writing and the Origin of the Chinese Writing System.” In The First Writing: Script Invention as History and Process,edited by Stephen Houston, pp. 190– 249. Cambridge: Cambridge University Press.

Baines, John, and Norman Yoffee. 1998. “Order, Legitimacy, Wealth in Ancient Egypt and Mesopotamia.” In Archaic States,edited by Gary Feinman and Joyce Marcus, pp. 199– 260. Santa Fe:SchoolofAmerican Research Press.

Beijing Daxue KaoguXiShang Zhou Zu北京大学考古系商周组,and Shaanxisheng KaoguYanjiusuo陕西省考古研究所.1994. “Shaanxi Yaoxian Beicun yizhi 1984 nian fajue baogao” 陕西耀县北村遗址 1984 年发掘报告 (Report on the 1984 Excavations at Beicun, Yaoxian, Shaanxi). Kaoguxue yanjiu 2.

Blanton, Richard. 1998. “Beyond Centralization: Steps Toward aTheory of Egalitarian Behavior in Archaic States.” In Archaic States,edited by Gary Feinman and Joyce Marcus, pp. 135–172. Santa Fe:SchoolofAmericanResearch Press.

Campbell, Roderick. 2007. “Blood, Flesh, and Bones: Kinship and Violence in the Social Economy of the Late Shang.” Unpublished PhD dissertation, Departments of Anthro- pology and East Asian Languages and Civilizations, Harvard University.

———. 2009. “Toward aNetworksand Boundaries Approach to Early Complex Polities:The Late Shang Case.” Current Anthropology 50 (6):821–848.

———. nd. “A Preliminary Estimation of Anyang Bronze Production.” Unpublished manuscript.

Campbell, Roderick, Li Zhipeng, He Yuling, and Yuan Jing. 2011. “Consumption, Exchange, and Production at the Great Settlement Shang: Boneworking at Tiesanlu, An- yang.” Antiquity 85:1279–1297.

Cao Dazhi 曹大志.2004. “Erlitou leixing muzang de zai renshi” 二里头类型墓葬的再认识 (Another Look at Erlitou-Variant Tombs). Unpublished manuscript.

Cao Nan 曹楠.2008. “Sandai shiqi chutu bingxing yuqi yanjiu” 三代时期出土柄型玉器研究 (Research on Three Dynasties–Period Handle-Shaped Jade Artifacts). Kaoguxuebao 2008.2:141–174.

Cao Wei 曹玮 (ed). 2010. Hanzhong chutu Shang dai qingtongqi 汉中出土商代青铜

190 Roderick B. Campbell READ ONLY/NO DOWNLOAD

器(ShangBronzes Unearthed Hanzhong). Chengdu: Ba Shu Shushe.

Chang, K. C. 1980. Shang Civilization. New Haven: Yale University Press.

———. 1983. Art, Myth, and Ritual: The Path to Political Authority in Ancient China. Cambridge, MA: Harvard University Press.

———. 1985. “Guanyu Zhongguo chuqi ‘chengshi’ zhege gainian” 关于中国初期‘城市’这个概念 (On the Concept of ‘City’ in Ancient China). Wenwu 1985.2:61–67.

———. 1986. The Archaeology of Ancient China.4th edition. New Haven and London: Yale University Press.

Chen Kunlong, Thilo Rehren,Jianjun Mei, and Congcang Zao. 2009. “Special Alloys from Remote Frontiers of the Shang Kingdom: Scientiﬁc Study of the Hanzhong Bronzes from Southwest Shaanxi, China.” Journal of Archaeological Science 2009.36:2108–2118.

Chen, Tiemei, George Rapp, Jing Zhichun, He Nu. 1999. “Provenance Studies of the Earliest Proto-Porcelain Using Instrumental Neuron Activation Analysis.” Journal of Archaeo- logical Science 26:1003–1015.

Chen Xingcan. 2005. “Lithic Production of Early States in China: An Examination of the Development of Craft Specialization.” Paper presented at Workshop on Early Chinese Civiliza- tion,March 10–12, 2005, University of British Columbia, Vancouver.

Chen, Xingcan, Li Liu, and Chunyan Zhao. 2010. “Salt from Southern Shanxi and the Development of Early States in China.” In Salt Production in China in aComparative Perspec- tiv: Salt Archaeology in China,vol. 2, edited by Li Shuicheng and Lothar von Falkenhausen, pp. 42–65. Beijing: Kexue Chubanshe.

Childe, V. Gordon. 1963 [1930]. The Bronze Age.New York: Biblio and Tannen.

______. 1950. “The Urban Revolution.” Town Planning Review 21: 3–17.

Cohen, David. 2001. “The Yueshi Culture, the Dong Yi, and the Archaeology of Ethnicity in Early Bronze Age China.” Unpublished PhD dissertation, Harvard University.

Creel, Herlee G. 1937. The Birth of China: AStudy of the Formative Period of Chinese Civilization.New York: Reynal and Hitchcock.

Du Jinpeng 杜金鹏.2005. “A Preliminary Investigation of Palace Foundation No. 1atthe Huanbei Shang City.” Translated by RodCampbell. In Chinese Archaeology

191 READ ONLY/NO DOWNLOAD

5:192–199. (Originally published in Chinese in Wenwu 2004.5:50–64).

———. 2009. Yinxu yi ershiyi jizhi deng jianzhuyicun yanjiu 殷墟乙二十一基址等建筑遗存研究 (Research on Yinxu Foundation B-21 and Related Structures). Kaogu 2009.11:59–69.

Du Jinpeng 杜金鹏,Wang Xuerong 王学容,and Zhang Liangren 张良仁.1999. “Shilun Yanshi Shangcheng xiaocheng de jige wenti” 试论偃师商城小城的几个问题 (A Preliminary Discussion of Several Questions Concerning the Inner Wall at the Yanshi Shangcheng Site). Kaogu 1999.2:35–40.

Erlitou Fieldwork Team, Institute of Archaeology, Chinese Academy of Social Sciences. 2005a. “Prospecting and Excavation of the Palace Complex and the Outside Roads ofthe Erlitou Site in Yanshi, Henan.” Translated by Zhang Liangren. Chinese Archaeology 5:1–9. (Originally published in Kaogu 2004.11:3–13).

———. 2005b. “A Large Turquoise Dragon-form Artifact Discovered at the Erlitou Site.” Translated by Zhang Liangren. Chinese Archaeology 5:10–12. (Originally published in Kaogu2005.7:15–20).

Falkenhausen, Lothar von. 1993. “On the Historiographical Orientation of Chinese Archaeology.”Antiquity 67(257):839–849.

———. 2006. “The External Connections of Sanxingdui.” Journal of East Asian Archaeology 5:191–245

Fang Hui. 2009. “Shang wangchao jinglue dongfang de kaoguxue guancha” 商王朝经略东方的考古学观察.(Archaeological Observations Concerning the Eastern Strategy of the Shang Dynasty). In Duowei shiyu –Shang wangchao yu Zhongguo zhaoqi wenming yanjiu ,edited by Jing Zhichun 荆志淳,Tang Jigen唐际根,and Ken-ichi Takashima高嶋謙一,pp. 70–84. Beijing: Kexue Chubanshe.

Fitzgerald-Huber, Louisa G. 1995. “Qijia and Erlitou: The Question of Contacts with Distant Cultures.” Early China 20:17–68.

———. 2003. “The Qijia Culture: Paths East and West.” Bulletin of the Museum of Far Eastern Antiquities 75:55–78.

Flad, Rowan. 2002. “Ritual or Structure? Analysis of Burial Elaboration at Dadianzi, Inner Mongolia.” Journal of East Asian Archaeology 3:23–51.

———. 2008. “Divination and Power:A Multiregional View of the Development of Oracle Bone Divination in Early China.” Current Anthropology 49(3):403–438.

192 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Flad, Rowan, and Chen Pochan. 2013. Ancient Central China: Centers and Peripheries Along the Yangzi River.Cambridge: Cambridge University Press.

Ford, Anne. 2004. “Ground Stone Tool Production at Huizui, China: An Analysis of a Manufacturing Site in the Yiluo River Basin.”.Indo-Paciﬁc Prehistory Association Bulletin 24:71–78.

Ge, Yan, and Katheryn Linduff.1990. “Sanxingdui: ANew BronzeAge Site in Southwest China.” Antiquity 64:505–513.

Haapanen, Mina H. 2005. “From aCommunity to Communities of Practice: The Late Shang Dynasty Site of Miaopu Locus North at Anyang, Henan Province, China.” Unpublished PhD dissertation, Cotsen Institute of Archaeology, University of California, Berkeley.

He Yuling 何毓灵.2011. “Yinxu shougongye shengchan guanli moshi tanxi” 殷墟手工业生产管理模式探析 (A Discussion of Yinxu Craft Production Organization). Sandai kaogu 4:279–289.

Hebeisheng Wenwu Yanjiusuo 河北省文物研究所.1985. Gaocheng Taixi Shang dai yizhi 藁城台西商代遗址 (The Shang-Period Site of Gaocheng Taixi). Beijing: Wenwu Chu- banshe.

Henansheng Wenwu KaoguYanjiusuo 南省文物考古研究所.1993. “Zhengzhou Xiaoshuangqiao yizhi de diaocha yu shijue” 郑州小双桥遗址的调查与试掘 (Survey and test excavation at the Xiaoshuangqiao site, Zhengzhou). In Zheng- zhou Shangcheng kaogu xin faxian yu yanjiu:1985–1992 郑州商城考古新发现与研究.Edited by Henan Sheng Wenwu Yanjiusuo, pp. 242–271. Zhengzhou: Zhongzhou Guji Chubanshe.

———. 1996. “1995 nian Zhengzhou Xiaoshuangqiao yizhi de fajue” 1995年郑州小双桥遗址的发掘 (The 1995 Excavations at the Xiaoshuangqiao Site, Zhengzhou). Huaxia kaogu 1996.3:1–23.

———.2001. Zhengzhou Shangcheng: 1953–1985 nian kaogu fajue baogao 郑州商城：1953 年–1985年考古发掘报告 (Zhengzhou Shangcheng: 1953–1985 Excavation Report). Beijing: Wenwu Chubanshe.

———. 2008. “Henan Xingyang shi Guandimiao yizhi Shang dai wanqi yicun fajue jianbao” 河南荥阳市关帝庙遗址商代晚期遗存发掘简报 (Preliminary Excavation Report on the Late Shang Remains at the Guandimiao Site, Xingyang City, Henan). Kaogu 2008.7:32–46.

Henansheng Wenwu KaoguYanjiusuo, Zhengzhou Shi Wenwu KaoguYanjiusuo 河南省文物考古研究所, 郑州市文物考古研究所.1999. Zhengzhou Shang dai tongqi jiaocangkeng

193 READ ONLY/NO DOWNLOAD

郑州商代铜器窖藏坑 (Storage Pits with Shang-Dynasty Bronze Artifacts in the Zhengzhou Area). Beijing: KexueChubanshe.

Henansheng Wenwu KaoguYanjiusuo 河南省文物考古研究所,and Zhoukoushi Wenwuju 周口市文物局.2000. Luyi Taiqinggong Changzikou mu 鹿邑太清宫长子口墓 (The- Tomb of Chang Zi KouatTaiqinggong, Luyi). Zhengzhou: Zhongzhou guji Chubanshe.

Hu Qianying 胡谦盈.1979. “Lun Siwa wenhua” 论寺洼文化 (A Discussion of the Siwa Cul- ture). Wenwu jikan 2:139–153.

Huang Zhanyue 黄展岳.2004. Gudai renxing renxun tonglun. 古代人牲人殉通论 (A Discussion of Ancient Human Sacriﬁce and Attendants-in-Death). 2nd edition. Beijing: Wenwu.

Hubeisheng Wenwu KaoguYanjiusuo 湖北省文物考古研究所.2001. Panlongcheng: 1963– 1994 nian kaogu fajue baogao 盘龙城:1963–1994年考古发掘报告 (Panlongcheng: 1963–1994 Excavation Report). Beijing: KexueChubanshe.

Jiang Gang蒋刚王.2008. 夏商西周文化对其西方和北方地区文化渗透的方向性与层级性 Xia Shang Zhou wenhua dui qi xifang he beifang diqu wenhua shentou de fangxiangx- ing yu cenjixing (The Directional and Hierarchical Nature of the Absorption of Northern and Western Area Cultures by Xia, Shang, and Zhou Cultures). Kaogu 12: 53–65.

Jiang Gang蒋刚王,and Wang Zhigang 志刚.2010. “Guanyu Weifang sanqi wenhua he Zhangjiayuan shangceng wenhua de zai renshi” 关于围坊三期文化和张家园上层文化的再认识 (A New Understanding Concerning the Weifangsanqi Culture and the Upper Zhangji- ayuan Culture). Kaogu 2010.5:67–78.

Jiangxisheng Wenwu KaoguYanjiusuo 江西省文物考古研究所,Jiangxi Sheng Bowuguan 江西省博物馆,and Xingan Xian Bowuguan 新干县博物馆.1997. Xingan Shangdai damu 新干商代大墓 (The Large Shang-Period Tomb at Xin’gan). Beijing: Wenwu Chubanshe.

Jiangxisheng Bowuguan KaoguYanjiusuo, Zhangshu Shi Bowuguan 江西省博物馆考古研究所, 樟树市博物馆.2005. Wucheng: 1973~2002 nian kaogu fajue baogao 吴城 1973– 2002年考古发掘报告 (Wucheng: The 1973–2002 Excavation Report). Beijing: Kexue Chu- banshe.

Jiang Yude 江雨德 (Roderick B. Campbell). 2011. “Guo zhi da shi: Shangdai wanqi zhong de lizhi gailiang” 国之大事：商代晚期中的礼制改良 (The Great Affairs of the State: The Mid–Late Shang Ritual Reforms).InYinxu yu Shang wenhua –Yinxu kexue fajue 80 zhou nian jinnian wenji 殷墟与商文化－殷墟科学发觉80周年纪念文集 (Yinxu and Shang Culture: Collection of Papers Commemorating the Eightieth Anniversary of the Scientiﬁc Excavation of Yinxu), edited by Zhongguo Shehuikexueyuan KaoguYanjiusuo 中国社会科学院考古研究

194 Roderick B. Campbell READ ONLY/NO DOWNLOAD

所,pp. 267–276. Beijing: Kexuechubanshe.

Jianjun Mei, Kunlong Chen, and WeiCao. 2009. “Scientiﬁc Examination of Shang- Dynasty Bronzes from Hanzhong, Shaanzi Province, China.” Journal of Archaeological Science 2009.36:1881–1891.

Jin Zhengyao 金正耀.1990. “Wan Shang Zhongyuan qingtongqi de kuangliao laiyuan” 晚商中原青铜的矿料来源 (Mineralogical Sources of Late Shang Bronzes from the Central Plains). Disanjie guoji Zhongguo kexueshi taolunhui lunwenji 第三届国际中国科学史讨论会论文集 (Collected Papers of the Third International Conference on the History of Chinese Science). Beijing: KexueChubanshe.

Jin Zhengyao 金正耀,W.T.Chase,Hirao Yoshimitsu 平尾良光,Mabuchi Hisao 马渊久夫, Yang Xizhang 杨锡璋,and Miwa Karoku 三轮嘉六.1998. “Zhongguo lianghe liuyu qingtong wenming zhijian de guanxi –yichutuShang qingtongqi de qian tongweisu bizhi yanjiu jieguo wei kaocha zhongxin 中国两河流域青铜文明之间的联系 – 以出土商青铜器的铅同位素比值研究结果为考察中心 (Relationships between Bronze Civilizations in the Two River Val- leys of China: Focusing on the Results of the Study of Lead Isotopes Ratios from Unearthed Shang Bronzes). Reprinted in Shang wenhua lunwenji 商文化论文集 (Collected Studies of the Shang Culture). 2volumes. 2003. Edited by Li Boqian, pp. 594–602. Beijing: Wenwu Chu- banshe.

Jin Zhengyao 金正耀,W.T.Chase, Hirao Yoshimitsu 平尾良光,Peng Shifan 彭适凡, Mabuchi Hisao马渊久夫,Miwa Karoku 三轮嘉六,and Zhan Kaixun 詹开逊.1994. “Xingan Shang dai damu qingtongqi de qiantongweisu bizhi yanjiu” 新干商代大墓青铜器的铅同位素比值研究 (A Study of Lead Isotope Ratios in Bronzes from the Shang Tomb at Xin’gan). Kaogu 1996.8:744–747.

Jin Zhengyao 金正耀,Mabuchi Hisao 马渊久夫,W.T.Chase, Chen De’an 陈德安, Miwa Karoku 三轮嘉六,Hirao Yoshimitsu 平尾良光,and Zhao Dianzeng 赵殿增.1995. “Guanghan Sanxingdui yiwukeng qingtongqi de qian tongweisu 广汉三星堆遗物坑青铜器的铅同位素比值研究 (A Study of the Lead Isotope Ratios in the Bronzes from the Sanxingdui Artifact Pits). Wenwu 1995.2:80–85.

Jing Zhicun 荆志淳,XuGuangde 徐广德,HeYuling 何毓灵,and Tang Jigen 唐际根. 2007. “M54 chutu yuqi de dizhi kaoguxue yanjiu” M54 出土玉器的地质考古学研究 (Geoar- chaeological Research on the Jade Artifacts from M54). In Anyang Yinxu Huayuanzhuang dongdi Shang dai muzang 安阳殷墟花园庄东地商代墓葬 (Shang-Dynasty Tombs at Huayuan- zhuang Locus East, Yinxu, Anyang), edited by Zhongguo Shehuikexueyuan KaoguYanjiusuo 中国社会科学院考古研究所,pp. 345–387. Beijing: Kexue chubanshe.

Jing Zhicun 荆志淳,Tang Jigen 唐际根,HeYuling 何毓灵 and Xu Guangde 徐广德. 2011. “Shangdai yong yu de wuzhi xing” 商代用玉的物质性 (The Materiality of Jade Use in

195 READ ONLY/NO DOWNLOAD

the Shang Dynasty). In Yinxu yu Shang wenhua –Yinxu kexue fajue 80 zhou nian jinnian wenji 殷墟与商文化－殷墟科学发觉80周年纪念文集 (Yinxu and Shang Culture: Collection of Papers Commemorating the 80th Anniversary of the Scientiﬁc Excavation of Yinxu), edited by Zhongguo Shehuikexueyuan KaoguYanjiusuo 中国社会科学院考古研究所,pp. 86–117. Beijing: Kexuechubanshe.

Jinzhong Kaogudui 晋中考古队.1989. “Shanxi Fenyang, Xiaoyi liangxian kaogu diaocha he Xinghuacun yizhi de fajue” 山西汾阳,孝义两县考古调查和杏花村遗址的发掘 (Ar- chaeological Survey of Fenyang and Xiaoyi Counties, Shanxi, and Excavation of the Xinghua- cun Site). Wenwu 1989.4:22–30.

Jumahe Kaogudui 拒马河考古队.1988. “Hebei Yixian Laishui guyizhi shijue baogao” 河北易县涞水古遗址试掘报告 (Report of Trial Excavations at the Laishui Ancient Site in Yi County, Hebei). Kaoguxuebao 1988.4:421–454.

Keightley, David. 1973. “Religion and the Rise of Urbanism.” Journal of the American Oriental Society 4: 527–538.

———. 1999. “The Shang: China’s First Historical Dynasty.” In The Cambridge History of Ancient China,edited by Michael Loewe and Edward L. Shaughnessy. Cambridge, UK: Cambridge University Press.

———. 2000. The Ancestral Landscape: Time, Space, and Community in Late Shang China.Berkeley, California: The Institute of East Asian Studies, University of Califor- nia and the Society for the Study of Early China.

———. 2006. “Marks and Labels: Early Writing in Neolithic and Shang China.” In Archaeology of Asia,edited by Miriam Stark, pp. 177–201. Malden: Blackwell.

Kopytoff, Igor.1986. “The Cultural Biography of Things: Commoditization as Process.” In The Social Life of Things: Commodities in Cultural Perspective,edited by Arjun Ap- padurai, pp. 64–94. Cambridge: Cambridge University Press.

Lee, Gyoung-ah, Gary Crawford, Li Liu, and Xingcan Chen. 2007. “Plants and People from the Early Neolithic to Shang Periods in North China.” Proceedings of the National Acad- emy of Sciences 104(3):1087–1092.

Li Chi. 1977. Anyang.Seattle: University of Washington Press.

Li Zhongli 李仲立.1981. “Shilun Xian-Zhou wenhua de yuanliu” 试论先周文化的渊流 (A Hypothesis Concerning the Origins of the Predynastic Zhou). Shehuikexue 1981.1:49– 59.

196 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Li Min. 2008. “Conquest, Concord, and Consumption: Becoming Shang in Eastern China.” Unpublished PhD dissertation, Department of Anthropology, University of Michi- gan, Ann Arbor.

Li Shuicheng 李水城.2005. “Xibei yu Zhongyuan zaoqi yeytongye de quyu tezheng ji jiaohu zuoyong” 西北与中原早期冶铜业的区域特征及交互作用 (Regional Characteristics and Interaction of Early Copper Metallurgy in the Northwestern Region and the Central Plains). Kaogu Xuebao 2005.3:239–278.

Li Yung-ti. 2003. “The Anyang Bronze Foundries: Archaeological Remains, Casting Technology, and Production Organization.” Unpublished PhD dissertation, Department of Anthropology, Harvard University, Cambridge, MA.

———. 2005. “The Social-political Implications of Multiple Craft-production Areas at Yinxu.” Paper presented at Workshop on Early Chinese Civilization, March 10–12, 2005, University of British Columbia, Vancouver.

———. 2007 “Co-Craft and Multicraft: Section-Mold Casting and the Organization of Craft Production at the Shang Capital of Anyang.” In Craft Production in Complex Societies: Multicraft and Producer Perspectives. Edited by Izumi Shimada, pp. 184–223. University of Utah Press, Salt Lake City.

Li Zhipeng 李志鹏,Jiang Yude 江雨德 (Rod Campbell), Yuan Jing 袁靖,HeYuling 何毓灵.2010. Yinxu Tiesanlu zhiguzuofang yizhi chutu zhigu yicun de chubu fenxi yu renshi 殷墟铁三路制骨作坊遗址出土制骨遗存的初步分析与认识 (A Preliminary Analysis of the Bone- working Remains from the Bone Workshop Site of Tiesanlu, Yinxu). Zhongguo Wenwubao, Sept. 17, p. 7.

Li Zhipeng 李志鹏,HeYuling 何毓灵,and Jiang Yude 江雨德 (Roderick Campbell). 2011. Yinxu WanShang zhigu zuofang yu zhigu shougongye de yanjiu huigu yu zai tantao 殷墟晚商制骨作坊与制骨手工业的研究回顾与再探讨 (Reﬂections on Research Concerning Late Shang Bone Workshops and Bone Working). Sandai Kaogu 4:471–486.

Liaoningsheng Bowuguan Wenwu Gongzuodui 辽宁省博物馆文物工作队.1977. “Liaoning Chaoyang Weiyingzi Xizhou mu he guyizhi” 辽宁潮阳魏营子西周墓和古遗址 (Western Zhou Tombs and the Ancient Site at Weiyingzi, Chaoyang, Liaoning). Kaogu 5:306– 312.

Liu Li. 2004. The Chinese Neolithic:Trajectories to Early States. Cambridge, UK: Cambridge University Press.

———. 2009. “Zhongguo zaoqi guojia zhengzhi geju de bianhua” 中国早期国家格局的变化 (The Changing Pattern of Early Chinese States). In Duowei shiyu –Shang wangchao yu

197 READ ONLY/NO DOWNLOAD

Zhongguo zhaoqi wenming yanjiu ,edited by Jing Zhichun 荆志淳,Tang Jigen唐际根 and Ken- ichi Takashima高嶋謙一,pp. 152–170. Beijing: KexueChubanshe.

Liu Li and Chen Xingcan. 2001. “Cities and Towns: The Control of Natural Resources in Early States, China.” Bulletin of the Museum of FarEastern Antiquities 73:5–47.

———. 2003. State Formation in Early China.London: Duckworth.

Liu Li, Chen Xingcan, and Li Baoping. 2007. “Non-state Crafts in the Early Chinese State: An Archaeological View from the Chinese Hinterland.” Indo-Paciﬁc Prehistory Associa- tion Bulletin 27:93–102.

Liu Li and Xu Hong. 2007. “Rethinking Erlitou: Legend, History, and Chinese Archaeology.” Antiquity 81:886–900.

Liu Shi’e 刘士莪 (Editor). 2001. Laoniupo: Xibei Daxue kaogu zhuanye tianye fajue baogao 老牛坡: 西北大学考古专业田野发掘报告 (Field Report on the 1985–1988 Excavations at Laoniupo by the Northwest University Archaeology Program). Xi’an: Shaanxi Renmin Chu- banshe.

Liu Senmiao 刘森淼.2002. “Panlongcheng waiyuan daizhuang hangtu yiji de chubu renshi” 盘龙城外缘带状夯土遗址的初步认识 (A Preliminary Understanding of the Belt of Rammed-Earth Foundations Outside the Walls of Panlongcheng). Wuhan Wenbo 1:12–15.

Liu Xu 刘绪.2001. “Youguan Xiadai niandai he Xia wenhua cenian de jidian kanfa”. 有关夏代年代和夏文化测年的几点看法 (A FewThoughts Concerning the Chronology of the Xia Dynasty and the Dating of the Xia Culture). Zhongyuan wenwu 2001.2:32–33.

Luan Fengshi 栾丰实.1997a. “Yueshi wenhua de fenqi yu leixing” 岳石文化的分期与类型 (The Chronology and Typology of the Yueshi Culture). In Haidai diqu kaogu yanjiu. 海岱地区考古研究 (Archaeological Research in the Haidai Area), pp. 364–374. Jinan: Shandong Daxue Chubanshe.

———1997b. “Lun Yueshi wenhua de laiyuan.” 论岳石文化的来源 (On the Origins of the Yueshi Culture). In Haidai diqu kaogu yanjiu 海岱地区考古研究 (Archaeological Research in the Haidai Area), pp. 318–347. Jinan: Shandong Daxue Chubanshe.

Luo Tai 罗泰 (Lothar von Falkenhausen). 2010. “Hanjiang shangliu pendi de qingtong shidai de ruogan guancha” 漢江上流盆地的青銅時代的若干觀察 (The Bronze Age in the Upper Han River Basin: Some Observations). In Hanzhong chutu Shang dai qingtongqi 汉中出土商代青铜器 (Shang Bronzes from Hanzhong), 4vols., edited by Cao Wei 曹玮,pp. 517–569. Chengdu: Ba Shu Shushe.

198 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Mei, Jianjun, Kunlong Chen, WeiCao. 2009. “Scientiﬁc Examination of Shang-Dynasty Bronzes from Hanzhong, Shaanxi Province, China.” Journal of Archaeological Science 36:1881–1891.

Moseley, Michael. 2001. The Incas and Their Ancestors: The Archaeology of Peru. London:Thamesand Hudson.

Nan Yuquan 南玉泉.1989. “Xindian wenhua xulie ji qi yu Kayue, Siwa de guanxi” 辛店文化序列及其与卡约、寺洼文化的关系 (The Xindian Culture Sequence and Its Relation- ship to the Kayueand Siwa Cultures). In Kaoguxue leixingxue de lilun yu shijian考古学类型学的论理与实践 (The Theory and Practice of Archaeological Classiﬁcation), edited by Yu Weichao 俞伟超,pp. 73–109. Beijing: Wenwu Chubanshe.

Neimenggu Zizhiqu Wenwu KaoguYanjiusuo, Eerduosi Bowuguan 内蒙古自治区文物考古研究所，鄂尔多斯博物馆.2000. Zhukaigou: Qingtongshidai zaoqi yizhi fajue baogao 朱开沟: 青铜时代早期遗址发掘报告 (Zhukaigou: Excavation Report on an Early Bronze Age Site). Beijing: Wenwu Chubanshe.

Peng Ke and Zhu Yanshi. 1995. “New Research on the Origins of Cowries Used in Ancient China.” Sino-Platonic Papers,no. 68. Philadelphia: Department of Asian and Middle Eastern Studies, University of Pennsylvania.

Peng, Minghan 彭明瀚.2004. “Ganjiang Poyanghuqu Shangdai wenhua de quxi leixing yanjiu” 赣江鄱阳湖区商代文化的区系类型研究 (A Regional-Typological Study of Shang-Period Cultures in the Ganjiang River and Poyang LakeArea). Kaogu 2004.3:68–79.

Piggott, Stuart. 1974. “Chariots in the Caucasus and in China.” Antiquity 48:16–24.

Qiu Shihua 仇士华,Cai Lianzhen 蔡莲珍,Xian Ziqiang 冼自强,BoGuancheng 薄管成.1983. “Youguan suowei Xia wenhua de tanshisi niandai ceding de chubu baogao”有关所谓夏文化的碳十四年代测定的初步报告 (A Preliminary Report on the Carbon-14 Dates of the So-Called ‘Xia Culture’). Kaogu 1983.10:923–926.

Qiu Shihua 仇士华,Cai Lianzhen 蔡莲珍,and Zhang Xuelian 张雪莲.2006. “Guanyu Erlitou wenhua de niandai wenti” 关于二里头文化的年代问题 (Concerning the Problem of the Dates of the Erlitou Culture). In Erlitou yizhi yu Erlitou wenhua yanjiu 二里头遗址与二里头文化研究 (Research on the Erlitou Site and Erlitou Culture), edited by Du Jinpeng and Xu Hong, pp. 321–333. Beijing: Kexue Chubanshe.

Rawson, Jessica (ed.). 1996. Mysteries of Ancient China: New Discoveries from the Early Dynas- ties.New York: George Braziller.

RenXianghong 仁相宏.1996. “Yueshi wenhua de faxian yu yanjiu” 岳石文化的发现与研究

199 READ ONLY/NO DOWNLOAD

(The Discovery and Study of the Yueshi Culture). Zhongguo kaoguxuehui dibaci nianhui lunwenji. 中国考古学会第八次年会论文集 (Collected Papers From the Eighth Annual Meeting of the Chinese Archaeological Society). Beijing: Wenwu Chubanshe.

Saito Tsutomu, Rubin Han, Shuyun Sun, Congqiang Liu. 2002. “Preliminary Consideration of the Source of Lead Used for Bronze Objects in Chinese Shang Dynasty–Was It Really from the Boundary Along Sichuan, Yunnan, and Guizhou Provinces?” In BUMA-V: Proceedings of the Fifth International Conference on the Beginnings of the Use of Metals and Alloys. Messages from the History of Metals to the Future Metal Age: April 21–24, 2002, Gyeo- ngju, Korea.Edited by Gyu-Ho Kim, Kyung-Woo Yi, and Hyung-Tai Kang, pp.xx. Seoul: Korean Institute of Metals and Materials.

Shangdong Daxue Dongfang KaoguYanjiu Zhongxin, Shandongsheng Wenwu Kaogu Yanjiusuo, Jinanshi KaoguYanjiusuo 山东大学东方考古研究中心, 山东省文物考古研究所, 济南市考古研究所.2003. “Jinanshi Daxinzhuang yizhi chutu Shangdai jiaguwen” 济南市大辛庄遗址出土商代甲骨文 (Shang-Period Oracle-Bones Discovered at the Daxinzhuang Site, Jinan). Kaogu 2003.6:3–6.

———. 2004. “Jinan Shi Daxinzhuang Shangdai Juzhi yu Muzang” 济南市大幸庄商代居址与墓葬 (Settlement Site and Tombs of the Daxinzhuang Site, Jinan). Kaogu 2004.7:25–33.

Shandong Sheng Wenwu KaoguYanjiusuo山东省文物考古研究所,Beijing Daxue Zhongguo Kaoguxue Yanjiu Zhongxin北京大学中国考古学研究中心 and Shouguangshi Wen- huaju 寿光市文化局.2010. “Shandong Shouguangshi Shuangwangcheng yanye yizhi 2008 nian de fajue” 山东寿光市双王城盐业遗址2008 年的发掘 (The 2008 Excavations at the Salt- Production Site of Shuangwangcheng, Shouguang City, Shandong). Kaogu 2010.3:18–36.

Shandong Sheng Wenwu Kaogusuo山东省文物考古研究所,Beijing Daxue Zhongguo Kaoguxue Yanjiu Zhongxin北京大学中国考古学研究中心,Shandong Shifan Daxue Qi Lu Wenhua Yanjiu Zhongxin山东师范大学齐鲁文化研究中心,Binzhoushi Wenwu Guanlichu 滨州市文物管理处.2010. “Shandong Yangxinxian Liwu yizhi Shangdai yicun fajue jianbao” 山东阳信县李屋遗址商代遗存发掘简报 (Preliminary Excavation Report of the Shang-Period Remains from the Site of Liwu, YangxinCounty, Shandong). Kaogu 2010.3:3–17.

Shanxisheng kaogu yanjiusuo 山西省考古研究所.2006. “Lingshi Jingjie Shangmu” 灵石旌介商墓 (The Shang Tombs at Jingjie, Lingshi). Beijing: Kexue Chubanshe.

Shaughnessy, Edward. 1988. “Historical Perspectives on the Introduction of the Chariot into China.” Harvard Journal of Asiatic Studies 48.1:189–237.

Shelach, Gideon. 1994. “Social Complexity in North China During the Early Bronze Age: AComparative Study of the Erlitou and Lower Xiajiadian Cultures.” Asian Perspectives

200 Roderick B. Campbell READ ONLY/NO DOWNLOAD

33(2):261–291.

———. 1996. “The Emergence of Complex Society in Northeast China from the Fourth to the FirstMillennia B.C.: APerspective from the Chifeng Area of Inner Mongolia.” PhD dissertation, University of Pittsburgh, PA.

———. 1999. Leadership Strategies, Economic Activity, and Interregional Interaction: Social Complexity in Northeast China.New York:Kluwer/Plenum.

Shen Bin 申斌.1991. “‘Fuhaomu’ yuqi cailiao tanyuan”‘妇好墓’ 玉器材料探源 (Investigation into the Source of Jade Artifacts in the Fu Hao Tomb). Zhongyuan wenwu 1991.1:73–77.

Shen Chen. 1994. “Early Urbanization in the Eastern Zhou in China (770–221 B.C.): An Archaeological View.” Antiquity 68:724–744.

———. 2003. “Compromises and Conﬂicts: Production and Commerce in the Royal Cities of Eastern Zhou, China.” In The Social Construction of Ancient Cities,edited by Monica Smith, pp. 290–310. Washington: Smithsonian Books.

Shih Chang-ju 石璋如.1933. “Diqici Yinxu fajue: E-qu gongzuo baogao” 第七此殷墟發掘：E區工作報告 (The Seventh Excavation at Yinxu: Report on Work in Area E). Anyang fajue baogao 4:709–728.

———. 1951. “Xiaotun C-qu de muzangqun” 小屯C-區的墓葬羣 (The Burial Group at Area C, Xiaotun). Bulletin of the Institute of History and Philology 23:447–487.

——. 1954. “Yindai di shang jianzhu fuyuan zhi yi li” 殷代地上建築復原之一例 (Reconstructing Yin-Dynasty Surface Architecture: One Example). Annals of the Academia Sinica 1954.1:267–280.

———. 1955. “Xiaotun Yindai de jianzhu yiji” 小屯殷代的建築遺蹟 (Architectural Remains at Yin-Dynasty Xiaotun). Bulletin of the Institute of History and Philol- ogy26:131–188.

———. 1959. Yinxu jianzhu yicun 殷墟建築遺存 (Architectural Remains from Yinxu). Taipei: Institute of History and Philology, Academia Sinica.

———. 1961. “Xiaotun Yindai bingzu muzhi jiqi you guan xianxiang” 小屯殷代丙組墓址及其有關現象 (The Yin-Dynasty Bing Group Tombs at Xiaotun and Some RelatedPhe- nomenon). In Qingzhu Dong Zuobin xiansheng liushiwu sui lunwenji 慶祝董作賓先生六十五歲論文集(Papers Presented to Mr.Tung Tso-pin on His Sixty-ﬁfth Birthday), pp. 781–802. Taipei: Institute of History and Philology, Academia Sinica.

201 READ ONLY/NO DOWNLOAD

———. 1970. “Yindai di shang jianzhu fuyuan de dier li” 殷代地上建築復原的第二例 (Reconstructing Yin-Dynasty Surface Architecture: ASecondExample). Bulletin of the Insti- tute of Ethnology,Academia Sinica 29:321–341.

———. 1976a. Yindai di shang jianzhu fuyuan de disan li 殷代地上建築復原的第三例 (Reconstructing Yin-Dynasty Surface Architecture: AThird Example). Bulletin of the De- partment of Archaeology and Anthropology,National Taiwan University 39/40:140–157.

———. 1976b. Yiqu jizhi shang xia de muzang 乙區基址上下的墓葬(The Tombs Above and Below the Foundations in Area Yi). Taipei: Institute of History and Philol- ogy, Academia Sinica.

Sichuan Sheng Wenwu KaoguYanjiusu 四川省文物考古研究所 (ed.).1999. Sanxingdui jisikeng 三星堆祭祀坑 (The Sanxingdui Sacriﬁcial Pits). Beijing: Wenwu Chu- banshe.

Smith, Adam. 2008. “Writing at Anyang: The Role of the Divination Record in the Emergence of Chinese Literacy.” Unpublished Ph.D. dissertation, Cotsen Institute of Archae- ology, University of California, Los Angeles.

Smith, Monica. 2003. “Introduction: The Social Construction of Ancient Cities.” In The Social Construction of Ancient Cities,edited by Monica Smith, pp. 1–36. Washington, D.C.: Smithsonian Books.

So, Jenny.2001. “Jade and Stone at Sanxingdui.” In Ancient Sichuan: Treasures from a Lost Civilization,edited by Robert Bagley, pp. 153–176. Seattle and Princeton: Seattle Art Mu- seum and Princeton University Press.

Sun Hua 孙华.2000. Sichuan pendi de qingtong shidai 四川盆地的青铜时代 (The Bronze Age of the Sichuan Basin). Beijing: Kexue Chubanshe.

______. 2009. Shang dai qianqi de guojia zhengti --- cong Erligang wenhua chengzhi he gongshi jianzhu jizhi de jiaodu 商代前期的国家政体 ——从二里冈文化城址和宫室建筑基址的角度 (The Early Shang-Dynasty State Regime:Fromthe Perspective of Erligang- Culture Wall and Palace Architectural Remains). In Duowei shiyu –Shang wangchao yu Zhong- guo zhaoqi wenming yanjiu ,edited by Jing Zhichun 荆志淳,Tang Jigen 唐际根,and Ken-ichi Takashima 高嶋謙一,pp. 171–197. Beijing: Kexue Chubanshe.

Tang Jigen. 2004. “The Social Organization of Late Shang China: AMortuary Perspective.” Unpublished Ph.D. dissertation, Archaeology, University College London.

Tang Jigen 唐际根 and Jing Zhicun 荆志淳.2009. “Anyang de ‘Shang yi’ yu ‘Da Yi

202 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Shang’ ” 安阳的“商邑”与“大邑商” (Anyang’s “Shang Settlements” and “Great Settle- ment Shang”). Kaogu 2009.9:70–80.

Tang Jigen 唐际根,Jing Zhicun 荆志淳,HeYuling 何毓灵.2010. “Huanbei Shangcheng gongdianqu yi, erhao hangtu jizhi jianzu fuyuan yanjiu” 洹北商城宫殿区一、二号夯土基址建筑复原研究 (Architectural Reconstruction Research on Rammed-Earth Foun- dations 1and 2within the Palace Area of Huanbeishangcheng). Kaogu 2010.1:23–35.

Thorp, Robert. 1988. “The Archaeology of Style at Anyang: Tomb 5inContext.” Archives of Asian Art 41:47–69.

———. 1991. “Erlitou and the Search for the Xia.” Early China 16:1–38.

———. 2006. China in the Early Bronze Age: Shang Civilization.Philadelphia: University of Pennsylvania Press.

Tong Weihua 佟伟华.2003.“Shangdai qianqi Yuanqu pendi tongzhi zhongxin – Yuanqu Shangcheng” 商代前期垣曲盆地统治中心 – 垣曲商城 (The Early Shang Period Center of PoliticalControl in the Yuanqu Basin: The Shang Walled Site at Yuanqu). In Shang wenhua lunwenji 商文化论文集 (Collected Studies of the Shang Culture). 2volumes. Edited by Li Boqian 李伯谦,pp. 455–469. Bejing: Wenwu Chubanshe.

Underhill, Anne P. 2002. Craft Production and Social Change in Northern China.New York: Kluwer Academic Publishers.

———. 2006. “Warfare and the Development of States in China.” In The Archaeology of Warfare: Prehistories of Raiding and Conquest,edited by Elizabeth Arkush and Mark Allen, pp. 253–285 .Gainesville: University of Florida Press.

Wang Jinxian 王进先.1982. “Shanxi Zhangzhishi jianxuan zhengji de Shangdai qingtongqi” 山西长治市拣选征集的商代青铜器 (Selections of Shang-Period Bronze Vessels Collected in Zhangzhi City, Shanxi). Wenwu 1982.9:49–52.

Wang Lixin 王立新.1998. Zao Shang wenhua yanjiu 早商文化研究 (Research on Early Shang Culture). Beijing: Gaodeng Jiaoyu Chubanshe.

———. 2009. “Yetan wenhua xingcheng de xiehouxing –yizhaoShang yu Erlitou de zhihouxing” 也谈文化形成的滞后性－以早商文化与二里头文化为例.(Another Discussion of the Time Lag in Cultural Formation: Using the Early Shang and Erlitou as Ex- amples). Kaogu 2009.12:47–55.

Wang Ming-ke. 1992. “The Ch’iang of Ancient China through the Han Dynasty: Ecological Frontiers and Ethnic Boundaries.” Unpublished PhD dissertation, East Asian

203 READ ONLY/NO DOWNLOAD

Languages and Civilizations, Harvard University.

Wang Wei 王巍.2005. “Yanshi Shangcheng kaogu xin jinzhan ji qi xiangguan wenti” 偃师商城考古新进展及其相关问题 (New Archaeological Advances at Yanshi Shangcheng and Some RelatedIssues). In Yanshi Shangcheng yizhi yanjiu 偃师商城遗址研究 (Research on the Yanshi Shangcheng Site), edited by Du Jinpeng 杜金鹏,Wang Xuerong 王学容,pp. 249–262. Beijing: KexueChubanshe.

Wang Weilin王炜林,Sun Zhouyong孙周勇.2011. “Shimao yuqi de niandai ji xiangguan wenti” 石峁玉器的年代及相关问题 (The Date of the Shimao Jades and Related Questions). Kaogu yu wenwu 2011.4:40–49.

Wang Xuerong 王学容.1999. “Yanshi Shangcheng buju de tansuo he sikao” 偃师商城布局的探索和思考 (Investigation and Thoughts Concerning the Layout of Yanshi Shangcheng). Kaogu 1999.2:24–34.

———. 2000. “Henan Yanshi Shangcheng di er hao jianzhuqun yizhi yanjiu” 河南偃师商城第II号建筑群遗址研究 (Research on the Site of Building Group II at Yanshi Shangcheng in Henan Province). Huaxia kaogu 2000.1:41–60.

Wang Xun 王迅.1994. Dongyi wenhua yu Huaiyi wenhua yanjiu.东夷文化与淮夷文化研究 (Re- search on the Eastern Yi and Huai Yi Cultures). Beijing: Beijing Daxue Chubanshe.

Webb, John, Anne Ford, and Justin Gorton. 2007. “Inﬂuences on Selection of Lithic RawMaterial Sources at Huizui, aNeolithic/Early Bronze Age Site in Northern China.” Indo- Paciﬁc Prehistory Association Bulletin 27:76–86.

Wheatley, Paul. 1971. Pivot of the Four Quarters: APreliminary Inquiry into the Origin and Character of the Ancient Chinese City.Chicago: Aldine Publishing Company.

Willey, G. R. ,and P. Phillips. 1958. Method and Theory in American Archaeology.Chicago and London: University of Chicago Press.

Xibeidaxue Wenbo Xueyuan 西北大學文博學院.2002. Chenggu Baoshan—1998 nian fajue baogao 城固宝山—1998 年发掘报告 (Chenggu Baoshan: 1998 Excavation Report). Bei- jing: Wenwu Chubanshe.

Xia Hanyi 夏含夷 (Shaughnessy, Edward L.). 2005. Gushi yiguan 古史異觀 (Alternative Views on Ancient History). Shanghai: Shanghai Guji Chubanshe.

Xia Nai 夏鼐.1949. “Lintao Siwashan fajueji” 临洮寺洼山发掘记 (Memoirs of the Excavation of Siwashan, Lintao). Kaoguxuebao 1949.4:71–138.

204 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Xia Shang Zhou Duandai Gongcheng Zhuanjiazu 夏商周断代工程专家组.2000. Xia Shang Zhou duandai gongcheng: 1996–2000 nian jieduan chengguo baogao 夏商周断代工程：1996–2000年阶段成果报告 (简本)(The Xia, Shang, Zhou Chronology Project: 1996–2000 Report (abridged version)). Beijing: Shijie tushu Chubanshe.

XiangTaochu向桃初.2006. Xiangjiang liuyu Shang Zhou qingtong wenming yanjiu de zhongyao tupo湘江流域商周青铜文明研究的重要突破.(An Important Breakthrough in the Study of the Shang-Zhou Bronze Civilization of the Xiang River). Nanfang wenwu 2006.2:68– 80.

———. 2008. Xiangjiang liuyu Shang Zhou qingtong wenming yanjiu 湘江流域商周青铜文化研究 (A Study of the Shang-Zhou Bronze Civilization of the Xiang River). 2vols. Bei- jing: Xianzhuang shuju.

Xinhua, October 30, 2012. “Chinese Scientists Identify Largest Neolithic City.” Xinhua 2012, http://www.china.org.cn/china/Off_the_Wire/2012-10/30/content_26940312. html.

Xu Hong 许宏,Chen Guoliang 陈国梁,Zhao Haitao 赵海涛.2004. “Erlitou yizhi juluo xingtai de chubu kaocha” 二里头遗址聚落形态的初步考察 (A PreliminaryInvestigation of the Settlement Pattern of the Erlitou Site). Kaogu 2004.11: 23–31.

———. 2005. “A Preliminary Investigation of the Settlement Pattern of the Erlitou Site.” Translated by Zhang Liangren. Chinese Archaeology 5:13–18. (Originally published in Kaogu2004.11:23–31).

Xu,Jay.2001a. “Introduction Part I: Sichuan Before the Warring States.” In Ancient Sichuan: Treasures from aLost Civilization,edited by Robert Bagley, pp. 21–38. Seattle and Princeton: Seattle Art Museum and Princeton University Press.

———. 2001b. “Bronze at Sanxingdui.” In Ancient Sichuan: Treasures from aLost Civilization,edited by Robert Bagley, pp. 59–152. Seattle and Princeton: Seattle Art Museum and Princeton University Press.

———. 2006. “Deﬁning Archaeological Cultures at the Sanxingdui Site.” Journal of East Asian Archaeology 5:149–190.

———. 2008. “The Sanxingdui Site: Art and Archaeology.” Unpublished PhD dissertation, Department of Art and Archaeology, Princeton University.

Xu Xusheng 徐旭生.1959. “1959 nian xia Yuxi diaocha ‘Xiaxu’ de chubu baogao” 1959 年夏豫西调查“夏墟”的初步报告 (Preliminary Report of the Surveys in the Ruins of Xia in 1959). Kaogu 1959.11:592–600.

205 READ ONLY/NO DOWNLOAD

YanWenming 严文明.1986. “Jiaodong yuanshi wenhua chutan” 胶东原始文化初探 (A Preliminary Discussion of Jiaodong Primitive Culture). In Shandong shiqian wenhua lunwenji 山东史前文化论文集 (Collected Papers on Shandong Prehistory), edited by Shandong Sheng “Qi Lu KaoguCongkan” Bianjibu 山东省《齐鲁考古丛刊》编辑部,pp. 63–95. Jinan: Qi Lu Shushe.

Yang Xizhang 杨锡璋 and Tang Jigen 唐际根.1999. “Yubei Jinan diqu de zhong Shang yicun yu Pangeng yi qian de Shangdu qiantu 豫北冀南地区的中商遗存与盘庚以前的商都迁徙 (Middle Shang Remains in Northern Henan, Southern Hebei, and the Movements of the Shang Capital Before Pangeng).” In Sandai wenming yanjiu 1:248–256. Beijing: Kexue Chubanshe.

Yates, Robin. 1997. “The City-State in Ancient China.” In The Archaeology of City- States: Cross Cultural Approaches,edited by Deborah Nichols and Thomas Charlton, pp. 71–90. Washington: Smithsonian Institute Press.

Yinxu Xiaomintun Kaogudui. 2007. “Henan Anyangshi Xiaomintun Shangdai fangzhi 2003–2004 nian fajue jianbao” 河南安阳市小民屯商代房址2003－2004年发掘简报 (Prelimi- nary Report on the 2003–2004 House-Features at Xiaomintun, Anyang City, Henan). Kaogu 2007.1:3–13.

Yin Wei-chang. 1986. “A ReexaminationofErh-li-t’ou Culture.” In Studies of Shang Archaeology.Edited by K. C. Chang. New Haven: Yale University Press.

Yoffee, Norman. 2005. Myths of the Archaic State.Cambridge, UK: Cambridge University Press.

YouRende 尤仁德.2002. Gudai yuqi tonglun 古代玉器通论 (A General Survey of Ancient Jade Artifacts). Beijing: Zijinchen Chubanshe.

Yuan Guangkuo袁广阔 and Qin Xiaoli 秦小丽.2000. “Henan Jiaozuo Fucheng yizhi fajue baogao” 河南焦作府城遗址发掘报告 (Excavation Report for the Fucheng Site, Jiaozuo, Henan). Kaogu xuebao 2000.4:501–536.

Yuan Guangkuo 袁广阔 and Zeng Xiaomin 曾晓敏.2004. “Lun Zhengzhou Shangcheng neicheng he waiguocheng guanxi” 论郑州商城内城和外郭城关系 (Some Problems on the Relationship between the Inner and Outer Walls of the Shang City at Zhengzhou). Kaogu 3:59–67.

Yuan Guankuo and Zeng Xiaomin. 2005. “Relationship between the Inner and Outer Walls at the Zhengzhou Shang City.” Translated by RodCampbell. Chinese Archaeology 5:188–191. (Originally published in Kaogu 2004.3:59–67).

206 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Yuan Jing and RowanFlad. 2003. “Two Issues Concerning Ancient Domesticated Horses in China.” Bulletin of the Museum of FarEastern Antiquities 73:111–127.

———. 2005. “New Zooarchaeological Evidence for Changes in Shang-Dynasty Animal Sacriﬁce.” Journal of Anthropological Archaeology 24(3):252–70.

Yuan Jing 袁靖,Huang Yunping 黃蕴平,Yang Mengfei 杨梦菲,LüPeng 吕鹏,Tao Yang 陶洋,and Yang Jie 杨杰.2007. “Gongyuanqian 2500 nian ~gongyuanqian 1500 nian Zhongyuan diqu dongwu kaogu yanjiu: yi Taosi, Wangchenggang, Xinzhai he Erlitou yizhi wei li” 公元前 2500 ~ 公元前 1500 年中原地区动物考古研究 ——以陶寺，王城岗，新砦和二里头遗址为例 (Research into the Zooarchaeology of the Central Plains, 2500–1500 BC: The Cases of Taosi, Wangchenggang, Xinzhai, and Erlitou). Keji Kaogu 2:12–34.

Yuan Jing and RoderickCampbell. 2009. “Recent Archaeometric Research on the ‘Origins of Chinese Civilization.’ ” Antiquity 83:96–109.

Zhang Changshou 张长寿 and Liang Xingpeng 梁星彭.1989. “Guanzhong Xianzhou qingtong wenhua de leixing yu Zhou wenhua de laiyuan” 关中先周青铜文化的类型与周文化的渊源 (The Typology of Guanzhong Proto-Zhou Bronze Culture and the Origins of Zhou Culture). Kaoguxuebao 1989.1:1–24.

Zhang Guoshuo 张国硕.1989. “Yueshi wenhua laiyuan chutan” 岳石文化来源初探(Prelimi- nary Discussion of Origins of the Yueshi Culture). Zhengzhou daxue xuebao 1:1–5.

Zhang Li 张莉.2012. “Cong Longshan dao Erlitou—yi Songshan nan bei wei zhongxin” 从龙山到二里头——以嵩山南北为中心 (Social Transformation from the Longshan Period to the Erlitou Period: Songshan and Beyond). Unpublished PhD dissertation, School of Archaeol- ogy and Museology, Peking University.

Zhang Xuezheng 张学正,Shui Tao, 水涛,Han Chongfei 韩翀飞.1993. “Xindian wenhua yanjiu” 辛店文化研究 (Research into Xindian Culture). In Kaoguxue wenhua lunwenji 考古学文化论集 (Collected Papers on Archaeological Cultures), edited by Su Bingqi 苏秉琦,pp. 122–152. Beijing: Wenwu Chubanshe.

Zhao Huacheng 赵化成.1989. “Gansu dongbu Qin he Qiang Rong wenhua de kaoguxue tansuo” 甘肃东部秦和羌戎文化的考古学探索 (Archaeological Exploration of Qin and Qiang- Rong Culture in Eastern Gansu). In Kaoguxue leixingxue de lilun yu shijian 考古学类型学的论理与实践 (The Theory and Practice of Archaeological Classiﬁcation), edited by Yu Weichao 俞伟超,pp.145–176. Beijing: Wenwu Chubanshe.

Zhao Zhijun 赵志军.2007. “Gongyuanqian 2500 nian ~gongyuanqian 1500 nian Zhongyuan diqu nongye jingji yanjiu” 公元 2500 年～公元前 1500 年中原农业经济研究

207 READ ONLY/NO DOWNLOAD

(Researchinto the Agricultural Economy of the Central Plains, 2500–1500 BC). Keji Kaogu 2:1–11.

Zhao Zhiquan 赵芝荃.1986. “Guanyu Erlitou wenhua leixing yu fenqi de wenti.” 关于二里头文化类型与分期的问题.(Concerning the Issues of Erlitou Culture Variants and Chronology). In Zhongguo kaoguxue yanjiu— XiaNai Xiansheng kaogu wushinian jinianhui wenji (er). 中国考古学研究——夏鼐先生考古五十年纪念文集 (二)(Research in Chinese Archaeology: Collected Papers in Commemoration of Xia Nai’s Fifty Years of Archaeology [2]), edited by Zhongguo Kaoguxue Yanjiu Bianweihui 中国考古学研究编委会pp. xx. Beijing: Kexue Chubanshe.

Zhejiang Sheng Wenwu KaoguYanjiusuo浙江省文物考古研究所,Huzhou Shi Bowuguan 湖州市博物馆,and Deqingxian Bowuguan 德清县博物馆.2011. “Zhejiang dong Tiaoxi zhongliu Shangdai yuanshici yaozhicun” 浙江东苕溪中流商代原始瓷窑址群 (The Shang-Dynasty-Period Protoporcelain Kiln Site Cluster in the Middle Reaches of the Tiaoxi, Eastern Zhejiang. Kaogu 2011.7:3–23.

Zheng Guang 郑光.2005. “Yanshi Erlitou Yizhi” 偃师二里头遗址 (The Yanshi Erlitou Site). Reprinted in Yanshi Erlitou yizhi yanjiu 偃师二里头遗址研究 (Research on the Yanshi Erlitou site), edited by Du Jinpeng 杜金鹏 and Xu Hong 许宏,p.795. Beijing: KexueChu- banshe.

Zheng Jiexiang 郑杰祥.1988. Xiashi chutan. 夏史初探 (Preliminary Investigations into Xia History). Zhengzhou: Zhongzhou Chubanshe.

Zhengzhou Daxue Wenboyuan郑州大学文博院andKaifengshi Wenwu Gongzuodui 开封市文物工作队.2000. Yudong Qixian fajue baogao 豫东杞县发掘报告 (Excavation Report from Qi County, Eastern Henan). Beijing: KexueChubanshe.

Zhongguo Shehuikexueyuan KaoguYanjiusuo 中國社會科學考古研究所.1956. Huixian fajue baogao 辉县发掘报告 (Excavation Report from Hui County). Beijing: Kexue Chubanshe.

———. 1980. Yinxu Fuhao mu 殷墟妇好墓 (The Tomb of Fu Hao of Yinxu). Beijing: Wenwu Chubanshe.

———. 1987a. Yinxu fajue baogao (1958–1961) 殷墟发掘报告 (1958–1961) (Yinxu Excavation Report [1958–1961]). Beijing: Wenwu Chubanshe.

———. 1987b .“Yinxu 259, 260 hao mu fajue baogao” 殷墟 259, 260 号墓发掘报告 (Yinxu 259, 260 hao mu fajuebaogao). Kaoguxuebao 1987.1:99–117.

208 Roderick B. Campbell READ ONLY/NO DOWNLOAD

———. 1994. Yinxu de faxian yu yanjiu 殷墟的发现与研究 (The Discovery and Research of Yinxu). Beijing: KexueChubanshe.

———. 1996. Dadianzi ----- Xiajiadianxiaceng wenhua yizhi yu mudi fajue baogao 大甸子——夏家店下层文化遗址与墓地发掘报告 (Dadianzi: Excavation of aLower Xiajiadian Culture Site and Cemetery). Beijing: KexueChubanshe.

———. 1998. Anyang Yinxu Guojiazhuang Shangdai muzang: 1982 nian~1992 nian kaogu fajue baogao 安阳殷墟郭家庄商代墓葬：1982年 ~1992年考古发掘报告 (Shang-Dynas- ty TombsatGuojiazhuang, Yinxu, Anyang: 1982–1992 Excavation Report). Beijing: Zhongguo Da Baike Quanshu Chubanshe.

———. 1999. Yanshi Erlitou: 1959~1978 nian kaogu fajue baogao 偃师二里头： 1959年 ~1978年考古发掘报告 (Yanshi Erlitou: 1959–1978 Excavation Report). Beijing: Zhongguo Da Baike Quanshu Chubanshe.

———. 2002a. Anyang Xiaotun 安阳小屯 (Xiaotun, Anyang). Beijing: Shijie Tushu Chuban Gongshe.

———.2002b. “Henan Yanshi Shangcheng Shangdai zhaoqi wangshi jisi yizhi” 河南偃师商城商代早期王室祭祀遗址 (The Early Shang RoyalSacriﬁcial Site at Yanshi Shangcheng, Henan). Kaogu 2002.7:6–8.

———. 2003. Zhongguo kaoguxue: XiaShang Juan 中国考古学夏商卷 (Chinese Archaeology: Xia and Shang). Zhongguo shehuikexue Chubanshe, Beijing.

———. 2005a. Anyang Yinxu chutu yuqi 安阳殷墟出土玉器 (Jades From Yinxu). Science Press, Beijing.

———. 2005b. Tengzhou Qianzhangda damudi 滕州前掌大墓地 (The Qianzhangda Cemetery in Tengzhou). 2vols. Wenwu Chubanshe, Beijing.

Zhongguo Shehuikexueyuan KaoguYanjiusuo, Anyang Gongzuodui. 中国社会科学院考古研究所, 安阳工作队.1979. “Yinxu xiqu mudi” 殷墟西区墓地 (Cemeteries at Western Yinxu).”Kaogu xuebao 1979.1:27–146.

———. 1995. “1973 nian Xiaotun nandi fajue baogao” 1973 年小屯南地發掘報告 (1973 Xiaotun South Excavation Report). Kaoguxue jikan 9:45–137.

———. 2009. “Henan Anyang shi Yinxu Liujiazhuang beidi 2008 nian fajue jianbao” 河南安阳市殷墟刘家庄北地 2008 年发掘简报 (Preliminary Excavation Report for Liuji- azhuang North Locus, Yinxu, Anyang City, Henan). Kaogu 2009.7:20–38.

209 READ ONLY/NO DOWNLOAD

Zhongguo Shehuikexueyuan KaoguYanjiusuo Anyang Gongzuodui中国社会科学院考古研究所安阳工作队 and Zhong Jia Huanheliuyu Quyu KaoguDiaocha Ketizu 中加洹河流域区域考古调查课题组.2010. “Henan Anyangshi Huanbei shangcheng yizhi 2005–2007 nian kancha jianbao” 河南安阳市洹北商城遗址2005~2007 年勘察简报 (Preliminary Report of the 2005–2007 Survey of the Huanbei Shang Walled Site, Anyang, Henan). Kaogu 2010.1:3–8.

Zhongguo Shehuikexueyuan KaoguYanjiusuo Erlitou Gongzuodui 中国社会科学古研究所二里头工作队.1974. “Henan Yanshi Erlitou zao Shang gongdian yizhi fajue jianbao” 河南偃师二里头早商宫殿遗址发掘简报 (Brief Report of Excavations of the Early Shang Pal- aces at Erlitou, Yanshi, Henan). Kaogu 1974.4:234–248.

———. 1992. “1987 nian Yanshi Erlitou yizhi muzang fajue jianbao” 1987 年偃师二里头遗址墓葬发掘简报 (Preliminary Report of the 1987 Tomb Excavations at Yanshi Erlitou). Kaogu 1992.4:294–303.

———. 2005. “Henan Yanshi Erlitou yizhi zhongxinqu de kaogu faxian”河南偃师市二里头遗址中心区的考古新发现 New Archaeological Discoveries in the Central Area of the Erlitou Site, Yanshi, Henan). Kaogu 2005.7:15–20.

Zhongguo Shehuikexueyuan KaoguYanjiusuo Henan Dier Gongzuodui中国社会科学院考古研究所河南第二工作队.1998. “Henan Yanshi Shangcheng dongbei’ou fajue jianbao” 河南偃师商城东北隅发掘简报 (Preliminary Report of Excavations at the Northeast Corner of Yanshi Shangcheng, Henan). Kaogu 1998.6:1–8.

Zhongguo Shehuikexueyuan KaoguYanjiusuo Henan Diyi Gongzuodui 中国社会科学院考古研究所河南第一工作队.2010a. “Henan YanshishiHuizui yizhi xizhi 2004 nian fajue jianbao” 河南偃师市灰嘴遗址西址2004 年发掘简报 (Preliminary Report of the 2004 Excava- tions at the Site of Huizui, Yanshi City,Henan). Kaogu 2010.2:36–46.

Zhongguo Shehuikexueyuan KaoguYanjiusuo Henan Diyi Gongzuodui 中国社会科学院考古研究所河南第一工作队.2010b. “2002–2003 nian Henan Yanshi Huizui yizhi de fajue” 2002－2003年河南偃师灰嘴遗址的发掘 (The 2002–2003 Excavation of the Huizui Site, Yan- shi, Henan). Kaoguxuebao 2010.3:393–422.

Zhongguo Shehuikexueyuan KaoguYanjiusuo Henan Xinzhaidui, Zhengzhoushi Wenwu KaoguYanjiuyuan 中国社会科学院考古研究所河南新砦队，郑州市文物考古研究院. 2009. “Henan Xinmishi Xinzhai yizhi fajue jianbao” 河南新密市新砦遗址发掘简报 (Prelimi- nary Excavation Report for the Site of Xinzhai, Xinmi City, Henan). Kaogu 2009.2:3–15.

Zhongguo Shehuikexueyuan KaoguYanjiusuo Jing-Wei Gongzuodui 中国社会科学院考古研究所泾渭工作队.1989. “Shaanxi Changwu Nianzipo Xianzhou wenhua yizhi fajue jilu” 陕西长武碾子坡先周文化遗址发掘纪略 (Record of Excavations at the Proto-Zhou Site of Nian- zipo, Changwu, Shaanxi).” Kaoguxue jikan 6:123–142.

210 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Zhongguo Shehuikexueyuan KaoguYanjiusuo Neimenggu Gongzuodui

中国社会科学院考古研究所内蒙古工作队.1961. “Neimenggu Chifeng Yaowangmiao, Xiajia- dian yizhi shijue jianbao” 内蒙古赤峰药王庙，夏家店遗址试掘简报(Preliminary Report of Trial Excavations at the Yaowangmiao and Xiajiadian sites, Chifeng, Inner Mongolia). Kaogu 1961.2:77–81.

———. 1974. “Chifeng Yaowangmiao, Xiajiadian yizhi shijue baogao” 赤峰药王庙，夏家店遗址试掘报告 (Trial Excavation Report for the Yaowangmiao and Xiajiadian Sites, Chifeng). Kaoguxuebao 1974.1:111–144.

Zhongguo Shehuikexueyuan KaoguYanjiusuo Shandong Gongzuodui

中国社会科学院考古研究所山东工作队.1985. “Shandongsheng Changdaoxian Tuojidao Da- kou yizhi”山东省长岛县砣矶岛大口遗址 (The Dakou Site on Tuoji Island, Changdao County, Shandong Province). Kaogu 1985.12:1068–1084.

Zhongguo Shehuikexueyuan KaoguYanjiusuo, Shanxi Dui

中国社会科学院考古研究所山西队.1980. “Shanxi Xiangfenxian Taosiyizhi fajue jianbao” 山西襄汾县陶寺遗址发掘简报 (Preliminary Report of Excavations at the TaosiSite, Xiangfen county, Shanxi).” Kaogu 1980.1:18–31.

———. 1983. “1978–1980 nian Shanxi Xiangfen Taosimudi fajue jianbao” 1978– 1980 年山西襄汾陶寺墓地发掘简报 (Preliminary Report of the 1979–1980 Cemetery Excava- tions at Taosi, Xiangfen, Shanxi).” Kaogu 1983.1:30–42

———. 1989. “Jinnan kaogu diaocha baogao” 晋南考古调查报告 (Southern Shanxi Archaeological Survey Report). Kaoguxue Jikan 6:1–51.

Zhongguo Shehuikexueyuan KaoguYanjiusuo, Henan Xinzhai Dui, Zhengzhoushi Wenwu KaoguYanjiuyuan 中国社会科学院考古研究所河南新砦队, 郑州市文物考古研究院. 2009. Henan Xinmishi Xinzhai yizhi 2002 nian fajue jianbao 河南新密市新砦遗址2002 年发掘简报 (Preliminary Excavation Report for the Xinzhai site, Ximishi, Henan). Kaogu 2:3–15.

Zhongguo Shehuikexueyuan KaoguYanjiusuo, Zhongguo Lishi Bowuguan, Shanxisheng KaoguYanjiusuo 中國社會科學考古研究所, 中国历史博物馆, 山西省考古研究所 1988. Xiaxian Dongxiafeng 夏縣东下冯 (Dongxiafeng Site in Xia County). Beijing :Wenwu Chubanshe.

Zhu Fenghan 朱凤瀚.1991. Shang Zhou jiazu xingtai yanjiu 商周家族形态研究 (A Study of the Family Patterns of Shang and Western Zhou China). 2nd ed., 2004. Tianjin: Tianjin Guji Chubanshe.

211 READ ONLY/NO DOWNLOAD

Ziboshi Wenwuju 淄博市文物局,Ziboshi bowuguan 淄博市博物馆,Huantaixian Wenwu Guanlisuo桓台县文物管理所.1997. “Shandong Huantaixian Shijia yizhi Yueshi wenhua mu- goujia jisi qiwukeng de fajue” 山东桓台县史家遗址岳石文化木构架祭祀器物坑的发掘 (Exca- vation of aWood-Framed Sacriﬁcial Artifact Pit at the Yueshi Site of Shijia, Huantai County, Shandong). Kaogu 1997.11:1–18.

Zou Heng 邹衡.1980. “Shilun Xia wenhua”试论夏文化 (A Preliminary Discussion of Xia Culture). In XiaShang Zhou kaoguxue lunwenji 夏商周考古学论文集 (Collected Papers on Xia, Shang, and Zhou Archaeology). Beijing: Wenwu Chubanshe.

———. 1988. “Zailun Xianzhou wenhua” 再论先周文化 (A ReexaminationofProto- Zhou Culture). In Zhou Qin Han Tang kaogu yu wenhua guoji xueshu huiyi lunwenji 周秦汉唐考古与文化国际学术会议论文集 (Collected Papers from the International Conference on Zhou, Qin, Han, and Tang Archaeology and Culture). Xibeidaxue xuebao.

212 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Index

Index Erlitou, 26, 27,31 Hushu, 91 Note: Page numbers in bold italics indicate Panlongcheng, Erligang, 86 illustrations or tables. Yanshi Shangcheng, 77 Zhanghe, 45 adzes ash-pits Erlitou, 27 at Dongxiafeng, 36 Haojiazhuang, 50 at Erlitou, 28 Maqiao, 53 awls Wanbei, 51 Anyang, 138 agriculture Erlitou, 27,31 in Erlitou Period, 19, 32–33, 40 Yinjiacheng, 51 Qijia, 58 axes. See also dagger-axes Weifang III, 156 Chaotianzui, 98 Yangzhuang, 40 Erlitou, 26,31 ancestral dedications, 130 Lijiaya, 175 Anhui, 120 Maqiao, 53 animal mask plaques, at Erlitou, 27 Panlongcheng, Erligang, 85 Anqiugucheng, 114–115 Wanbei, 51 Anqiugudui site, 146 Weifang III, 156 Anqiugudui variant, 47, 50,51 Yinjiacheng, 51 Anqiu variant, 130, 146,146–147 Yueshi, 109 antler, 138 Anyang Period Baijiafen, 135 ceramic traditions in, 130, 141,141–155, 142, Baijiazhuang site, 108 143, 146, 149, 151, 152, 153, 154, 155, 156 Baijiazhuang variant, 108, 112, 113, 186 chariots in, 131 Baimiaozhu, 53 chronology of, 131–132, 132 Baiyan culture, 41. See also Jinzhong tradition innovations in, 130–131 Baodun tradition, 54, 55 residential areas in, 135, 136 Baoshan tradition workshops in, 135–139 Anyang, 130, 166 writing in, 130–131 Xiaoshuangqiao-Huanbei, 108, 119, 121, Anyang Sanjiazhuang, 107 121–122 Anyang site, 129–130, 130, 131,180n2 Baota tradition Anyang variant, 130, 141,141–142 Anyang, 130, 162–163 Aohan banner, 46 Erligang, 92,92–93, 116 architectural component, at Xiaoshuangqiao, 109 beads architecture. See houses; palace; rammed-earth Dongxiafeng, 64n14 structures Erlitou, 27 architecture orientation, 27, 100, 104n28, 111 Liujia, 168 arrowheads Zhengjiapo, 171 Anyang, 138 Beicun variant bone, 138 Erligang, 83, 85

213 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Xiaoshuangqiao-Huanbei, 108,116, 117,118, bronze vessels. See also speciﬁc vessels 122 Anyang, 142, 142,144 Beixin variant, 168 Beicun, 116, 118 Beixinzhuang, 138 Erlitou, 25, 26,31 bells Subutun, 145 Duimenshan-Feijiahe, 161 Xiaoshen, 117,118 Erlitou, 27,31 Xiaoshuangqiao-Huanbei, 108–109, 125n9 Liujia, 168, 169 burials. See mortuary Wucheng, 116 bi-ladles, Zhanghe, 45 Caoyanzhuang variant, 108, 112,113–114, 186 bird-headed ladles, Sanxingdui, 56 cattle, 33, 58 blades. See also dagger-axes; knives cattle bones, 138 Baoshan, 125n19 cattle scapula, 58–59, 79 Erlitou, 26,31 cattle skulls, in burials, 40 Sanxingdui, 56,56–57 cauldrons Sanxingdui-Jinsha, 165 Dongxiafeng, 39 bone tools Doujitai, 51 at Erlitou, 31 Duimenshan-Feijiahe, 161, 162 at Yanshi Shangcheng, 77 Erlitou, 27 bone workshops, 132, 138–139, 181n11 at Erlitou, 27 bovine skulls, in burials, 40 Fenghao, 171 bowls, Lutaigang, 45 Lower Dianjiangtai, 53 boxes, lacquer, at Erlitou, 29 Maqiao, 53 bracelets Tianhu, 152 Erlitou, 26,31 Weifang III, 157 Weifang III, 158 Zhengjiapo, 171 Weiyingzi, 158 cave houses. See also houses Bronze Age, as term, 13–14 Dongxiafeng, 36, 38 bronze foundry Jinzhong, 88 at Anyang, 132, 133,136–137, 137 Zhanghe, 45 at Erlitou, 29–30 ceramic production. See also kiln; workshops at Nanguanwai, 70 Anyang, 139 at Yanshi Shangcheng, 74 Erligang, 72, 77 at Zijingshan North, 71 Erlitou, 31 bronze hoards Guandimiao, 142 Baoshan, 121 Qijia,58 Chengdonglu, 124n4 Yanshi Shangcheng, 77, 103n19 Panlongcheng, 115 ceramic traditions Weiyingzi, 158–159 Anqiugudui, 47, 50,51 Xiaoshuangqiao-Huanbei, 109 Anqiuvariant, 130, 146,146–147 Zhengzhou, 71–72 Anyang, 141,141–155, 142, 143, 146, 149, 151, bronze knives, Luwangfen-Songyao, 42–43 152, 153, 154, 155, 156 bronze metallurgy, Qijia, 59 Anyang variant, 130, 141,141–142

214 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Index

Baijiazhuang variant, 108,112, 113, 186 Erligang, 88,91, 95 Baimiaozhu, 53 Erlitou, 52, 53, 65n32 Baodun, 54 Xiaoshuangqiao-Huanbei, 120 Baoshan, 166 Jingnansi Anyang, 130 Erligang, 90,91–92, 116 Xiaoshuangqiao-Huanbei Period, 108,119, Erlitou, 20,53 121,121–122 Jinzhong, 119 Baota Anyang, 130 Anyang, 130 Erligang, 87–88 Erligang, 92,92–93, 116 Erlitou, 20, 38, 41, 41,45, 65n20 Beicun variant Laoniupo variant, 130, 153,153–155, 156 Erligang, 83, 85 Liangzhu, 53 Xiaoshuangqiao-Huanbei, 108,116, 117, Lijiaya, 130,174–176, 175, 175 118,122 Liujia, 130,153, 168–169,169, 183n42 Caoyanzhuang variant, 108, 112,113–114, 186 Liuligevariant, 80–81, 113, 186 Chaotianzui variant, 91 Liuwan variant, 65n41 Dachengdun variant Longshan, 33, 41, 42, 53, 60 Anyang, 160 Lower Dianjiangtai, 20,51, 52–53 Erlitou, 78,87, 90 Lower Xiajiadian, 20, 41, 46,46–47, 47,65n28, Xiaoshuangqiao-Huanbei, 108,115, 115, 88,89 120 Lujiahe, 130,166, 166 Datuotou Lutaigang variant, 43, 45, 65n25 Erligang, 89 Luwangfen-Songyao, 42–43, 43,45, 62, 65n22, Erlitou, 20,47, 48 80, 81, 186 Xiaoshuangqiao-Huanbei, 119, 125n10 Maqiao Daxinzhuang variant Erligang, 91 Anyang, 144 Erlitou, 20, 52,53 Erligang, 78,87 Metropolitan, 20, 33, 79–80 Xiaoshuangqiao, 108,114, 114 Nianzipo, 130,153, 167, 168,183n42 Dongxiafeng variant Niujiaogang variant, 20,33, 35,40 Erlitou, 20,33, 34,34–40, 37, 39,63n8 Panlongcheng Xiaoshuangqiao-Huanbei, 118–119 Erligang, 78, 83, 84,85–86,91, 104n23, 116 Doujitai, 20,51, 87,89–90 Erlitou, 20, 53, 79 Duimenshan-Feijiahe, 130,161–162, 162 post-, Anyang, 130,160–161 Erligang, 70–71, 77–87, 78 Xiaoshuangqiao-Huanbei, 108, 115–116, Erlitou, 20, 33–37,33–59, 39, 41–44, 46–50, 52, 116, 118,120, 121 54–58 Panmiao variant, 108, 114–115, 115,146 Fenghao, 130,171, 172 Qianzhangda variant, 130,147–149, 149, 151 Haojiazhuang,47, 48, 49,50 Qijia, 20, 57, 57,57–59, 60, 62, 99 Hougang II s, 43 Qilidun variant, 57, 58 Huangniangniangta variant, 65n41 Qinweijia, 65n41 Hushu Sanxingdui Anyang, 160 Erligang, 91–92, 95,95–98, 96

215 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Erlitou, 20,53–57, 54, 55, 56,60 Zhanghe variant, 43, 45 Xiaoshuangqiao-Huanbei, 121, 122 Zhaogezhuang variant, 20,47, 48, 49 Sanxingdui-Jinsha, 130, 164–166 Zhengjiapo, 130, 153, 169–171, 170,183n42 Shijiahe, 57, 86 Zhenzhumen tradition, 130, 159, 159–160 Shizhaocun variant, 58 Zhouliangyuqiao, 130,161, 161 Siwa Zhukaigou Anyang, 130 Erligang, 88, 89 Erligang, 98–99 Erlitou, 20, 41–42, 42,45, 59,62 Xiaoshuangqiao-Huanbei, 122 Xiaoshuangqiao-Huanbei, 119 Subutun variant, 130, 143, 143–146 Changbai Mountains, 144 Taixi variant, 186 Chang Zi Koutomb, 147, 148 Erligang, 81, 113 Chaohu, 87 Xiaoshuangqiao-Huanbei, 108,114, 114 Chaotianzui variant, 91, 166 as term, 16 chariots, in Anyang period, 131, 140, 140, 155,176, Tianhu variant, 130,151–152, 152 180nn4–5 Wanbei variant, 20, 47–48, 50, 51 Chengdonglu, 124n4 Wangwan III, 33, 63n6 Chengdu plain, 54, 57, 121 Wannian Chenggu Baoshan, 121 Anyang, 130, 163–164 Chengziya site, 65n31 Erligang, 94, 95, 104n26 Chifeng Xiajiadian, 46 Weifang III, 119, 130, 155–158, 157 Chinese Academy of Social Sciences, 14 Weiyingzi, 130,157–159, 158 Chinese Bronze Age, as term, 13–14 Wucheng, 163–164, 164 chisels Anyang, 130 Erlitou, 27 Erligang, 72, 93,93–94,104nn23–25 Maqiao, 53 Xiaoshuangqiao-Huanbei, 116, 120–121 Weifang III, 156 Xiaoshen, 108, 117, 118 Yinjiacheng, 51 Xiaoshuangqiao-Huanbei, 108, 112, 113–119, chronology 114, 115, 116, 117 of Erligang, 73,74, 102n3 Xiaqiyuan, 40, 43–44, 44, 62, 65n23, 80, 81, of Erlitou, 63n1 186 of Sanxingdui, 54, 55 Xiawanggang variant, 20, 33, 36, 40–41 cinnabar, Dongxiafeng, 64n14 Xindian “clan insignias,” 144, 145, 162, 181n19 Anyang, 130 cofﬁns. See also mortuary Erligang, 98, 99, 99 Baoshan, 122 Xiaoshuangqiao-Huanbei, 122 Erlitou, 29 Yangzhuang variant, 20,33, 35, 40 Lijiaya, 175 Yinjiacheng variant, 20, 47, 50, 51 Lingshi Jingjiecun, 172 Yuegezhuang variant, 43 Nianzipo, 168 Yueshi, 40, 42,45, 47–51, 49, 50,53, 62, 65n29, Qijia,59 65n32, 67n50, 79, 80, 87 Siwa, 98 Anyang, 148 Weifang III, 157 Xiaoshuangqiao-Huanbei, 120 Weiyingzi, 158

216 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Index colonization, 185–186 Zhengjiapo, 171 compass orientation, of architecture, 27, 100, Zhengzhou, 102n4 104n28, 111 Dakou site, 48 cooking utensils, at Yanshi Shangcheng, 77 Dasikongcun site, 131,132, 138, 139 cooking vessels. See also li-tripods; yan-steamers Datuotou tradition/variant Anyang, 141 Erligang, 89 Dachengdun, 87 Erlitou, 20,45, 47, 48 Daxinzhuang, 87 Xiaoshuangqiao-Huanbei, 119, 125n10 Dongxiafeng, 34, 38–39, 39 Dawenkou culture, 29 Doujitai, 51 Daxinzhuang variant Erlitou, 33,38–39, 39, 64n13 Anyang, 144 Lower Dianjiangtai,53 Erligang, 78, 87 Lutaigang, 45 Xiaoshuangqiao-Huanbei, 108, 114, 114 Luwangfen-Songyao, 42, 43 death attendants, 59, 98, 142, 154 Wannian, 95 deer antler, 138 Wucheng, 163 deer scapula, 40, 58–59 copper, 86, 100, 144, 160, 165 ding-cauldrons courtyards Dongxiafeng, 39 at Erlitou, 27–28 Doujitai, 51 at Yanshi Shangcheng, 76, 76 Duimenshan-Feijiahe, 161, 162 cowrie shells. See also shells Erlitou, 27 Dongxiafeng, 64n14 Fenghao, 171 Lower Xiajiadian, 47 Lower Dianjiangtai, 53 Sanxingdui-Jinsha, 165 Maqiao, 53 Weiyingzi, 182n29 Tianhu, 152 craft production. See also bronze foundry; ceramic Weifang III, 157 production Zhengjiapo, 171 at Dongxiafeng, 36–37, 38 Dinggong site, 65n31 at Erlitou, 29–32 ding-steamers, Lower Dianjiangtai, 53 at Panlongcheng, 86 dishes, Sanxingdui, 56 at Yanshi Shangcheng, 77 divination, bone, 79, 130–131. See also scapulamancy Dachengdun variant Di Xin, 139, 181n12 Anyang, 160 dog, 58, 103n15, 147 Erlitou, 78, 87, 90 Dongtaibao culture, 41. See also Jinzhong tradi- Xiaoshuangqiao-Huanbei, 108,115, 115,120 tion Dadianzi, 46, 47, 47 Dongxiafeng site, 35–38, 37,64n10, 83–84 dagger-axes Dongxiafeng variant Baoshan, 122, 166 Erligang, 78, 82–84 at Erlitou, 26, 27,31 Erlitou, 20, 33, 34,34–40, 37, 39,63n8 Maqiao, 53 Xiaoshuangqiao-Huanbei, 118–119 Panlongcheng, Erligang, 85 Dong Yi, 67n50 Weiyingzi, 159 donkey, 58

217 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

double-eared, bag-footed tripods, 98 Erlitou Period double-eared guan-pots, 58 bias towards identification of, 64n8 double large-eared guan-pots, 58 ceramic traditions in, 20, 33–37,33–59, 39, Doujitai tradition, 20,51, 87, 89–90 41–44,46–50, 52, 54–58 dou-vessel Xia in, 19 Lutaigang, 45 Erlitou variant, 20, 33 Nianzipo, 168 exchange networks, 61, 179 Sanxingdui, 56 dragon, turquoise, at Erlitou, 28, 30–31 Fangshan, 155 “dragon” kiln, 91, 125n14 Fenghao tradition, 130,171, 172 drinking vessels. See also gu-drinking beakers fish hooks, at Erlitou, 27 in burials, 29 flat-bottomed pots, Sanxingdui, 56 Dongxiafeng, 64n14 “forbidden cities,” 25 Erlitou, 29, 31, 56 foundry Lower Xiajiadian, 47, 47 at Anyang, 132, 133,136–137, 137 Sanxingdui, 56 at Erlitou, 29–30 Tianhu, 152 at Nanguanwai, 70 drums, at Erlitou, 29 at Yanshi Shangcheng, 74 Duimenshan-Feijiahe tradition, 130,161–162, 162 at Zijingshan North, 71 Fucheng, 80, 81 Early Shang. See Erligang Period Fufeng, 168 earrings Fufeng Baijiayao, 116 Datuotou, 47 Fufeng-Qishan, 116 Qijia, 59 Fufeng Yijiabu, 116 Weifang III, 157, 158 Fu Hao, 129 egg-shaped urns, 45, 83, 88 Fujian, 95 elite artifacts, at Erlitou, 26,31 fu-pots, Maqiao, 53 elites, burial of, 29, 38, 131, 143–144, 145, 148, 182n26 gang-urns, Panlongcheng, 86 Erligang Period Gansu, 60, 89, 98, 101, 122, 168, 171 architecture orientation in, 27 Gaocheng Taixi, 107 bias towards identification of, 64n8 ge-dagger-axes ceramic traditions, 70–71, 77–87, 78 Baoshan, 122, 166 chronology, 73 at Erlitou, 26, 27, 31 as horizon, 126n20 Maqiao, 53 Erligang site, 69. See also Zhengzhou Panlongcheng, Erligang, 85 Erlitou Weiyingzi, 159 as ceremonial center, 25 Zhengjiapo, 171 expansion, 19 Zhengzhou, 102n4 as horizon, 59 gold earrings, 157 palace/temple, 22, 22, 23,24, 24,25–28, 63n4 Guandimiao site, 142 phases at, 24–25 Guangshe culture, 41. See also Jinzhong tradition site, 20, 21, 22,64n9 guan-jars, Siwa, 98

218 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Index guan-pots Sanxingdui, 56 Beicun, 85 hoards Dongxiafeng, 33, 34,39 Baoshan, 121 double large-eared, 58 Chengdonglu, 124n4 Doujitai, 51 Panlongcheng, 115 Erligang, 85 Weiyingzi, 158–159 ﬂat-bottomed, 56 Xiaoshuangqiao-Huanbei, 109 Haojiazhuang, 50 Zhengzhou, 71–72 Lutaigang, 45 hoes Luwangfen-Songyao, 42, 43 Haojiazhuang, 50 Qijia, 58 Yueshi, 48 round-bellied, 33,39, 85 Hongshan culture, 46 saddle-mouthed, 99 hooks, at Erlitou, 27 Sanxingdui, 56 horizon single-eared, 34,39 deﬁned, 126n20 Xindian, 99 Erligang as, 126n20 Zhanghe, 45 Erlitou as, 59 Zhaogezhuang, 48 horse, 58 Zhengjiapo, 171 horse-pit, 148, 154 gu-drinking beakers Hougang II tradition, 43 Anyang, 142, 142 Hougang site, 131,132 Erlitou, 31 houses. See also cave houses Laoniupo, 154, 156 Anyang, 133, 135, 136,141–142 Sanxingdui, 56 Daxinzhuang, 114 Tianhu, 152 Dongxiafeng, 36, 64n10 gui-vessel Erligang, 73 Fenghao, 171 Haojiazhuang,48 Lower Xiajiadian, 47, 47 Huanbei, 135 Weifang III, 157 Jinzhong, 88 Zhengjiapo, 171 Lijiaya, 175 Niujiaogang, 40 hairpins, Weifang III, 158 Panlongcheng, Erligang, 86 hand coiling, of ceramics, 38 Qijia, 58 handle-shaped objects, at Erlitou, 26,31, 100 Sanxingdui, 55 Hanshan Dachengshou, 87 semisubterranean, 41, 58, 77, 88, 135, 142 Hanzhong, 57, 121, 122 subterranean, 45, 88, 135 Haojiazhuang variant, 47, 48, 49,50 Weifang III, 156–157 Hebei, 43, 88 Weiyingzi, 158 Henan, 43, 45, 78 Wucheng, 120 he-vessels Xiawanggang, 40 Erlitou, 27 Yanshi Shangcheng, 77, 103n14 Lower Xiajiadian, 47, 47 Zhanghe, 44–45 Qijia, 59 Zhaogezhuang, 48

219 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Zhengjiapo, 170 Jiangling Cenhemiaoxingcun, 161 Zhukaigou, 41 Jiangxi, 122 Huanbei Shangcheng, 110–113, 111 Jiangxi Ruichang, 91 Huanbei site, 107, 122, 126n21 Jinan Daxinzhuang, 107 Huangniangniangtai variant, 65n41 Jingnansi tradition Huantai Shijia site, 144 Erligang, 90,91–92, 116 Huizui, tool production at, 32 Erlitou, 20,53 human remains. See mortuary Jinzhong tradition Hunan Yueyang, 91 Anyang, 130, 172–174 husbandry.See agriculture; individual animals Erligang, 87–88 Hushu tradition Erlitou, 20,38, 41, 41,45, 65n20 Anyang, 160 Xiaoshuangqiao-Huanbei, 119 Erligang, 88,91, 95 Jixian Zhangjiayuan, 157 Erlitou, 52, 53, 65n32 jue-hoes, Yueshi, 48 Xiaoshuangqiao-Huanbei, 120 jue-vessels Anyang, 142, 142 inscriptions, in Anyang Period, 130–131, 144, 147, Erlitou, 27 181n15, 181n20 Laoniupo, 154, 156 ivory Lower Xiajiadian, 47, 47 Erlitou, 24 Qianzhangda, 151 Sanxingdui-Jinsha, 56, 164 Subutun, 144 Tianhu, 152 jade Anqiu, 147 kaogu wenhua, translation of phrase, 16 Anyang, 139, 179, 182n31, 182n34 Kazuo Heshanggou Tomb 1, 182n29 Daxinzhuang, 114 kiln(s). See also ceramic production development of industry in, 60 at Dongxiafeng, 38, 39, 67n46 Dongxiafeng, 64n14 “dragon,” 91, 125n14 Erligang, 96, 100 at Erlitou, 30, 31 Erlitou, 24, 25, 26,28, 30–31, 60, 64n14, 67n44 at Guandimiao, 142 Lijiazui, 86 Maqiao, 91 Lower Xiajiadian, 47 protoporcelain and, 125n14 Qijia, 58 Sanxingdui-Jinsha, 164 Sanxingdui, 56, 56,57 at Wucheng, 120, 121 Sanxingdui-Jinsha, 164, 165 at Zhengzhou, 72, 103n19 Zhengzhou, 77 knives. See also blades; dagger-axes jars, Siwa, 98 Beicun, 85 jewelry bronze, 42 Datuotou, 47 Datuotou, 47 Erlitou, 26,31 Daxinzhuang, 87 Qijia, 59 Erlitou, 27 Weifang III, 157, 158 Haojiazhuang, 50 Weiyingzi, 158 Hushu, 91

220 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Index

Luwangfen-Songyao, 42–43 Qianzhangda, 148 Qijia, 59 Siwa, 98 stone, 42, 48, 50, 85, 87 Zhanghe, 45 Yueshi, 48 Zhengjiapo, 171 Zhanghe, 45 Liujia tradition, 130,153, 168–169, 169,183n42 kui, Baoshan, 166 Liujiazhuang site, 131,132, 135 Liulige variant, 80–81, 113, 186 lacquer Liuwan variant, 65n41 Anyang, 182n25 Liwu, 145 Dongxiafeng, 64n14 Longshan traditions, 33, 41, 42, 43, 53, 60 Erlitou, 24 Lower Dianjiangtai tradition, 20, 51, 52–53 ladles Lower Xiajiadian tradition, 20,41, 46, 46–47, 47, Erlitou, 29 65n28, 88, 89, 158 Sanxingdui, 56 Lujiahe tradition, 130,166, 166 Zhanghe, 45 Luodamiao period, 102n3 Lake Chao, 160 Luoyang, 78 Lake Dongting, 92 Lutaigang variant, 43, 45, 65n25. See also Xiaqi- Lake Tai, 20, 53 yuan tradition languages, 13. See also inscriptions; writing Luwangfen-Songyao tradition, 42–43, 43,45, 62, Lanjia cluster, 145 65n22, 80, 81, 102n2, 186 Laoniupo, 122 Luyi Taiqinggong, 147, 182n24 Laoniupo variant, 130, 153,153–155, 156 lei Maqiao tradition Lujiahe, 166 Erligang, 91 Sanxingdui-Jinsha, 166 Erlitou, 20, 52,53 Liangchengzhen, 63n2 masks Liangzhu tradition,53 Baoshan, 122, 167 Liaoning, 60, 89, 101 Erlitou, 27 Lijiaya tradition, 130,174–176, 175,176 Hanzhong, 122 Lijiazui cemetery, 85 Laoniupo, 153, 156 “lineage cemeteries,” 140, 140. See also mortuary Sanxingdui, 55 Lingshi Jingjiecun, 172, 173, 174 Sanxingdui-Jinsha, 165 li-tripods Meiyuanzhuang site, 131,132 Anyang, 141 Mengzhuang, 80 Baota, 163 metropolitan tradition Dachengdun, 87 in Anyang Period, 130 Daxinzhuang, 87 in Erligang Period, 79–80 Erligang, 70 in Erlitou Period, 20, 33 Jinzhong, 173–174 in Xiaoshuangqiao-Huanbei Period, 107–108 Liujia, 168, 169 Miaopu site, 131,132 Lutaigang, 45 mica, 48 Luwangfen-Songyao, 42 middens Nianzipo, 168 Anyang, 133, 172

221 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Daxinzhuang, 114 Panlongcheng Lijiazui, 85 Erlitou, 25, 26, 28, 31 Qianzhangda, 148, 150 Guandimiao, 142 Qijia, 59 Huayanzhuang, 181n11 Sanxingdui-Jinsha, 164–165 Wucheng, 120 Siwa, 98 Zhengzhou, 72, 73 Subutun, 143–145 Middle Shang, 107. See also Xiaoshuangqiao- Taixi, 114 Huanbei Period Tianhu, 152 millet, 19, 32 Weifang III, 157 mirrors Weiyingzi, 158 Erlitou, 61 Wucheng, 120, 125n13, 163, 164 Qijia, 59 Xiawanggang, 40, 64nn16–17 Mojiaoshan Liangzhu, 63n2 Yanshi Shangcheng, 77 molds Zhaogezhuang, 48 Dongxiafeng, 36, 84 Zhengjiapo, 170–171 Erligang, 84 Erlitou, 30 Nanguanwai, 70, 102n2, 108 Laoniupo, 153 Nanjing, 120 Wucheng, 121, 163 Nanshunchengjie, 124n6 Yanshi, 77 nao-bells, Duimenshan-Feijiahe, 161 Zouping Langjun, 144 nephrite, 179 mortuary. See also waist pits Nianzipo tradition, 130,153, 167,168, 183n42 Anqiu, 147, 147, 148 Niujiaogang variant, 20,33, 35,40 Anyang, 129, 132, 135, 138,139–141, 141,142, 181n15, 182nn24–37 oracle-bone divination, 79, 130–131. See also Baoshan, 122 scapulamancy chariots in, 140, 155 oracle-bone inscriptions, 130–131, 144 death attendants in, 59, 142 Ordos region, 41, 156, 176 distinction, 29 Dongxiafeng, 36, 38, 82 palace(s) elite, 29, 38, 131, 143–144, 145, 148,182n26 at Anyang, 129, 132, 133, 134, 178 Erligang, 71, 73, 82, 85 at Erligang, 69 Erlitou, 25, 27–28, 28–29, 63n3, 64n13 at Erlitou, 22, 22,23, 24, 24,25–28,63n4, 178 Fenghao, 171 at Huanbei Shangcheng, 111 jade in, 31 at Panlongcheng, 178 lacquer in, 31 at Xiaoshuangqiao, 178 Laoniupo, 154, 154, 155 at Yanshi Shangcheng, 74,75–76, 102n3, Lijiaya, 175 102n7, 103nn10–13, 178 Lingshi Jingjiecun, 172, 173, 174 at Zhengzhou, 178 Liujia, 168 pan-dishes, Sanxingdui, 56 Lower Dianjiangtai, 53 PanGeng, 180n6 Lower Xiajiadian, 47 Panlongcheng site, 20, 82, 85,86, 103n18,120 Nianzipo, 168 Panlongcheng tradition/variant

222 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Index

Erligang, 78, 83, 84,85–86, 91, 104n23, 116 Wucheng, 93, 163 Erlitou, 53, 79 post-, Anyang, 130,160–161 Qi, 145 Xiaoshuangqiao-Huanbei, 108, 115–116, 116, Qianzhangda site, 148 118,120,121 Qianzhangda variant, 130,147–149, 149, 151 Panmiao variant, 108,114–115, 115,146 Qijia tradition, 20,42, 57, 57,57–59,60, 62, 99 pendants, 51 Qijiazhuang site, 131 pig, 32–33, 58 Qilidun variant, 57,58 pig scapula, 40, 79 Qinghai, 99, 171 pig tusk, 138 Qingjian Lijiaya site, 175 Pinggu Liujiahe M1, 157 Qingzhou Zhaopu site, 145 Pingliang, 168 Qinling Mountains, 165, 168 ping-vases, Sanxingdui, 56 Qinweijia variant, 65n41 pins, Anyang, 138 Qishan Wangjiazui, 118 plaques 59, 130–131 radiocarbon dating. See also chronology Baoshan, 167 of Erligang, 74, 102n3 Erlitou, 27 of Erlitou, 63n1 Laoniupo, 154 of Sanxingdui, 54 Sanxingdui, 56 rammed-earth structures Sanxingdui-Jinsha, 182n34 Anyang, 133, 141–142 plastrons, turtle, 79 Erligang, 73, 74, 75, 77, 79, 102n3 political organization, 187 Erlitou, 22, 24, 25, 26 Anyang, 137, 145–146 Haojiazhuang, 48 bronze production and, 137 Laoniupo, 153 Erligang, 100 Lijiaya, 175 Erlitou, 42, 61–62, 65n21 Liulige, 81 Xiaoshuangqiao-Huanbei, 123–124 Panlongcheng, 86 ponds, 75, 132, 135 Xiaoshuangqiao, 109, 113, 124n5 postholes Yinjiacheng, 51 at Erlitou, 30 Yinxu, 111 at Wucheng, 125n12 Yueshi, 48 post-Panlongcheng tradition, 130,160–161 remains. See mortuary pots. See also fu-pots; guan-pots residential architecture. See houses production. See agriculture; ceramic production; rice, 19, 32, 40 craft production; workshops rings protoporcelain Datuotou, 47 Erligang, 80, 91 Qijia, 59 Erlitou, 24 ritual area, Wucheng, 120, 125n12 kilns and, 125n14 ritual remains, Sanxingdui, 55–56 Lower Dianjiangtai, 53 roads, 135 Maqiao, 91 roebuck tooth, 145, 181n18 Wannian, 95 round-bellied guan-pots, 33,39, 85

223 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Lower Xiajiadian, 47 sacriﬁce Sanxingdui-Jinsha, 165 animals in, 33, 138 Weiyingzi, 182n29 at Anyang, 133, 138,138, 140 Yinjiacheng, 51 at Erlitou, 27–28 Shenmu, 57 at Huanbei Shangcheng, 111 Shijiahe tradition, 57, 86 Siwa, 98 Shijia site, 65n31 saddle-mouthed jars, 98 Shijiazhuang, 155 saddle-mouthed pots, 99 Shimao, 57, 60, 61, 63n2 salt, 100 Shizhaocun variant, 58 salt production, 145 Shouguang Guchengcun site, 145 salt storage, at Dongxiafeng, 84 shovels Sanliqiao, 63n8 Anqiugudui, 51 Sanxingdui-Jinsha tradition, 130,164–166 Erlitou, 26,31 Sanxingdui site, 55, 55,61, 65n36 Weifang III, 156 Sanxingduitradition sickles Erligang, 91–92, 95, 95–98, 96 Baoshan, 122 Erlitou, 20,53–57, 54, 55, 56,60 Beicun, 85 Xiaoshuangqiao-Huanbei, 121, 122 Luwangfen-Songyao, 42 saws, at Erlitou, 27 single-eared guan-pots, 34,39 scapulamancy, 40, 58, 79. See also oracle bone Sipanmo site, 131,132 divination Siwatradition scepters Anyang, 130 Baoshan, 122, 167 Erligang, 98–99 Lingshi Jingjiecun, 172 Xiaoshuangqiao-Huanbei, 122 script, Wucheng, 93. See also inscriptions; writing skulls, cattle, in burials, 40 semisubterranean houses soybeans, 19, 32 Anyang, 135, 142 spade Jinzhong, 88 Panlongcheng Erligang, 86 Qijia, 58 Yinjiacheng, 51 Yanshi Shangcheng, 77 spatulas, at Yanshi Shangcheng, 77 Zhukaigou, 41 spears Shandong, 179 Maqiao, 53 in Anyang Period, 144 Panlongcheng, 85–86 burials at, 29 spindle whorl, at Erlitou, 27 in Xiaoshuangqiao-Huanbei Period, 123 spoons, at Erlitou, 31 Shang people, 15, 45, 63n1, 69, 101, 129, 155, 162 steamers Shanxi, 40, 44, 78, 88, 89, 101, 119, 123 Dachengdun, 87 sheep, 33, 58 Daxinzhuang, 87 sheep scapula, 40, 58, 79 Dongxiafeng, 38–39 shells Haojiazhuang, 50 Dongxiafeng, 64n14 Lower Dianjiangtai, 53 Erlitou, 24, 28 Maqiao, 53

224 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Index

Qianzhangda, 148 Nianzipo, 168 Sanxingdui, 56 Qianzhangda, 148 Wannian, 95 Siwa, 98 Zhaogezhuang, 48 Zhanghe, 45 storage area, at Yanshi Shangcheng, 75, 76 Zhengjiapo, 171 subterranean houses. See also houses turquoise Anyang, 135 Dongxiafeng, 64n14 Dongxiafeng, 36, 38 (See also houses) Erlitou, 24, 28, 30–31, 61, 64n13, 64n14 Jinzhong, 88 Qijia, 59 Zhanghe, 45 Sanxingdui, 56 Subutun variant, 130, 143,143–146 turquoise dragon, at Erlitou, 28, 30–31 Suide Bijiaqu, 175 turtle plastrons, 79 tusk, 138 Taigu Baiyan site, 172 Taihang Mountains, 81, 172 urn burials, at Xiawanggang, 40, 64n16. See also Taixi variant, 186 mortuary Erligang, 81, 113 urns Xiaoshuangqiao-Huanbei, 108, 113, 114, 114 Dongxiafeng, 83 Taiyi Mountains, 120 egg-shaped, 45, 83, 88 talc, 48, 50 Erligang, 83 Tanheli site, 162 gang, 86 Taosi, 29, 32, 63n2, 63n3, 63n8 Jinzhong, 88 temple. See palace(s) Panlongcheng, 86 textiles, at Erlitou, 31 Zhanghe, 45 Tianhu variant, 130,151–152, 152 Uruk expansion, 19 Tianshui, 168 Tiesanlu, 132,138, 139 vases, Sanxingdui, 56 tin, 29, 61, 100, 179 tombs. See mortuary waist pits. See also mortuary Tongbai Mountains, 85 Anqiu, 147 Tonggushan, 116 Anyang, 142 Tongling, 94, 104n23, 160 Daxinzhuang, 114 tooth, roebuck, 145 Dongxiafeng, 82 tripods Laoniupo, 154, 154, 155 Anyang, 141 Lingshi Jingjiecun, 172 Baota, 163 overview of, 103n15 Dachengdun, 87 Taixi, 114 Daxinzhuang, 87 Tianhu, 152 Erligang, 70 Wucheng, 120–121 Jinzhong, 173–174 Zhengzhou, 77 Liujia, 168, 169 Wanbei variant, 20,47–48, 50,51 Lutaigang, 45 Wangwan III tradition, 33, 63n6 Luwangfen-Songyao, 42 Wannian tradition

225 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Anyang, 130,163–164 ceramic traditions in, 108, 112,113–119, 114, Erligang, 94,95, 104n26 115,116, 117 wealth, mortuary and, 29. See also elites, burial of metropolitan traditions, 107–108 weapons. See also arrowheads; dagger-axes; knives phase I, 108–109 Erlitou, 30 phases II and III, 110–113, 111 Panlongcheng, Erligang, 85 Xiaotun, 107, 144 Qijia, 59 Xia people, 15, 19 Weifang III tradition, 119, 125n10, 130,155–158, Xiaqiyuan tradition, 40, 43–46, 44,62, 65n23, 80, 157 81, 102n2, 186 Weiyingzi tradition, 130,157–159, 158 Xiawanggang site, 40, 64n16 weng-urns, Zhanghe, 45 Xiawanggang variant, 20,33, 36, 40–41 Western Zhou period, 158, 159 Xibeigang cemetery, 129, 138,139–140 wheat, 19, 32 Xindian tradition wheel, in ceramic production, 31, 38, 56, 58, 63n5 Anyang, 130 workshops. See also ceramic production; foundry Erligang, 98, 99, 99 at Anyang, 132, 135–139 Xiaoshuangqiao-Huanbei, 122 bone, 132, 138–139 Xin’gan, 116, 122 at Erligang, 72 Xin’gan tomb, 163, 164 at Erlitou, 29–32 Xinghuacun site, 173 at Zhengzhou, 71 Xingtai Caoyanzhuang, 107 writing. See also inscriptions Xinzhai, 32, 33, 63n6 in Anyang Period, 130–131, 144 Xinzhai tradition, 33 Wucheng, 93 Xinzhou Liansigou, 172 Wucheng II, 125n15 Xuejiazhuang site, 132 Wucheng site, 125n11, 179 Wucheng tradition Ya Chou, 145 Anyang, 130,163–164, 164 Yangshao, 74 Erligang, 72, 93, 93–94, 104nn23–25 Yangzhuang variant, 20,33, 35,40 Xiaoshuangqiao-Huanbei, 116, 120–121 Yanshi, 60, 61, 109 Wu culture, 91 Yanshi Erlitou, 20, 21 Wu Ding, 129, 139, 180n6 Yanshi Shangcheng, 74,74–77, 76, 102n3, 109 Wuguangcun site, 131,132 yan-steamers Daxinzhuang, 87 Xiadynasty, 20, 65n22, 67n50 Dongxiafeng, 38–39 Xi’an Laoniupo, 85, 122 Haojiazhuang, 50 Xi'an Yuanjiaya site, 154 Lower Dianjiangtai, 53 Xiaoheyan culture, 46 Maqiao, 53 Xiaojiashan-Jiaoshan tradition. See Wannian Qianzhangda, 148 tradition Sanxingdui, 56 Xiaomintun site, 131,132, 135 Wannian, 95 Xiaoshen variant, 108, 117, 118 Zhaogezhuang, 48 Xiaoshuangqiao, 55, 71, 107, 109–110 Yaowangcheng, 63n2 Xiaoshuangqiao-Huanbei Period Yaoxian Beicun, 85

226 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Index

Yinjiacheng variant, 20,47, 50,51 Xiaoshuangqiao-Huanbei, 119, 158 Yinxu, 110, 177, 179, 186. See also Anyang Zhumadian Yangzhuang, 40 you, Duimenshan-Feijiahe, 161 Zijingshan, 108 Yuanqu IV,119 Zijingshan North, 71 yue-axes. See also axes Zouping Langjun site, 144 Erlitou, 27 zun-vessel Panlongcheng, Erligang, 85 Duimenshan-Feijiahe, 161 Yuegezhuang variant, 43. See also Xiaqiyuan tradi- Lutaigang, 45 tion Sanxingdui-Jinsha, 165 Yueshi tradition Zhouliangyuqiao, 161 Anyang, 148 Erligang, 89, 102n2 Erlitou, 40, 42, 45, 47–51, 49, 50,53, 62, 65n29, 65n32, 67n50, 79, 80, 87 Xiaoshuangqiao-Huanbei, 120 Yunnan, 165 zhang-blades Baoshan, 125n19 Sanxingdui, 56–57 Sanxingdui-Jinsha, 165 Zhanghe variant, 43, 44–45. See also Xiaqiyuan tradition Zhangjiashan, 91 Zhangzhai nan jie, 124n4 Zhaogezhuang variant, 20,47, 48, 49 Zhejiang, 63n2 Zhengjiapo tradition, 130,153, 169–171, 170, 183n42 Zhengzhou as center, 185–186 in Erligang Period, 60, 61, 62, 69, 71,101 in Xiaoshuangqiao-HuanbeiPeriod, 108–109, 122, 123 Zhengzhou Shangcheng, 74, 102n3 Zhenzhumen tradition, 130, 159,159–160 Zhoujiazhuang, 63n2 Zhouliangyuqiao tradition, 130, 161, 161 Zhou people, 15, 171 Zhukaigou artifacts, 42, 42 Zhukaigou tradition Erligang, 88, 89 Erlitou, 20,41–42, 42,45, 59, 62

227 READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

COTSEN INSTITUTE OF ARCHAEOLOGY PRESS MONOGRAPHS

Contributions in Field Research andCurrent Issues in Archaeological Method and Theory

Monograph 79 | Archaeology of the Chinese Bronze Age: From Erlitou to Anyang, Roderick B. Campbell Monograph 78 | Visions of Tiwanaku,AlexeiVranich and Charles Stanish (eds.) Monograph 77 | Advances in Titicaca Basin Archaeology–2,AlexeiVranich and Abigail R. Levine (eds.) Monograph 76 | The Dead Tell Tales,María Cecilia Lozada and Barra O’Donnabhain (eds.) Monograph 75 | The Stones of Tiahuanaco,Jean-Pierre Protzen and Stella Nair Monograph 74 | Rock Art at Little Lake: An Ancient Crossroads in the California Desert, Jo Anne VanTilburg, Gordon E. Hull, and John C. Bretney Monograph 73 | The History of the Peoples of the Eastern Desert,HansBarnard and Kim Duistermaat (eds.) Monograph 71 | Crucible of Pueblos: The Early Pueblo Period in the Northern South- west,Richard H. Wilshusen, Gregson Schachner, and James R. Allison (eds.) Monograph 70 | Chotuna and Chornancap: Excavating an Ancient Peruvian Legend, Christopher B. Donnan Monograph 69 | An Investigation into Early Desert Pastoralism: Excavations at the Camel Site,Negev, Steven A. Rosen Monograph 68 | The Chanka: Archaeological Research in Andahuaylas (Apurimac), Peru,Brian S. Bauer, Lucas C. Kellett, and Miriam Aráoz Silva Monograph 67 | Inca Rituals and Sacred Mountains: AStudy of the World’s Highest Archaeological Sites,Johan Reinhard and Maria Costanza Ceruti Monograph 66 | Gallinazo: An Early Cultural Tradition on the Peruvian North Coast, Jean-François Millaire with Magali Morlion Monograph 65 | Settlement and Subsistence in Early Formative Soconusco,Richard G. Lesure (ed.) Monograph 64 | The South American Camelids,Duccio Bonavia Monograph 63 | Andean Civilization: ATribute to Michael E. Moseley,Joyce Marcus and Patrick Ryan Williams (eds.) Monograph 62 | Excavations at Cerro Azul, Peru: The Architecture and Pottery, Joyce Marcus Monograph 61 | Chavín: Art, Architecture, and Culture, William JConklin and Jeffrey Quilter (eds.) Monograph 60 | Rethinking Mycenaean Palaces II: Revised and Expanded Second Edi- tion, Michael L. Galaty and William A. Parkinson (eds.) Monograph 59 | Moche Tombs at Dos Cabezas, Christopher B. Donnan Monograph 58 | Moche Fineline Painting from San José de Moro, Donna McClelland,

228 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Donald McClelland, and Christopher B. Donnan Monograph 57 | Kasapata and the Archaic Period of the Cuzco Valley,Brian S. Bauer (ed.) Monograph 56 | Berenike 1999/2000,StevenE.Sidebotham and Willeke Wendrich (eds.) Monograph 55 | Roman Footprints at Berenike: Archaeobotanical Evidence of Subsist- ence and Trade in the Eastern Desert of Egypt,René T. J. Cappers Monograph 54 | Advances in Titicaca Basin Archaeology 1,Charles Stanish, Amanda B. Cohen, and Mark S. Aldenderfer Monograph 53 | Us and Them: Archaeology and Ethnicity in the Andes,Richard Martin Reycraft Monograph 52 | Archaeological Research on the Islands of the Sun and Moon, Lake Titi- caca, Bolivia: Final Results from the Proyecto Tiksi Kjarka,Charles Stanish and Brian S. Bauer (eds.) Monograph 51 | Maya Zooarchaeology: New Directions in Theory and Method,Kitty F. Emery (ed.) Monograph 50 | Settlement Archaeology and Political Economy at Tres Zapotes, Veracruz, Mexico, Christopher A. Pool (ed.) Monograph 49 | Perspectives on Ancient Maya Rural Complexity,Gyles Iannone and Samuel V. Connell (eds.) Monograph 48 | Yeki bud, yeki nabud: Essays on the Archaeology of Iran in Honor of William M. Sumner,Naomi F. Miller and Kamyar Abdi (eds.) Monograph 47 | Archaeology in the Borderlands: Investigation in Caucasia and Be- yond, Adam T. Smith and Karen S. Rubinson (eds.) Monograph 46 | Domestic Ritual in Ancient Mesoamerica,Patricia Plunket (ed.) Monograph 45 | Pathways to Prismatic Blades,Kenneth Hirth and Bradford Andrews (eds.) Monograph 44 | Ceramic Production and Circulation in the Greater Southwest, Donna M. Glowacki and Hector Neff (eds.) Monograph 43 | Pottery of Postclassic Cholula, Mexico, Geoffrey McCafferty Monograph 42 | Pompeian Households: An Analysis of the Material Culture, Penelope M. Allison Monograph 41 | Rethinking Mycenaean Palaces: New Interpretations of an Old Idea, Michael L. Galaty and William A. Parkinson (eds.) Monograph 40 | Prehistory of Agriculture: New Experimental and Ethnographic Ap- proaches, Patricia C. Anderson (ed.) Monograph 39 | Recent Advances in the Archaeology of the Northern Andes: In Memory of Gerardo Reichel-Dolmatoff,Augusto Oyuela-Caycedo and J. Scott Raymond (eds.) Monograph 38 | Approaches to the Historical Archaeology of Mexico, Central and South America, Janine Gasco, Greg Charles Smith, and Patricia Fournier-Garcia Monograph 37 | Hawaiian Adze Production and Distribution: Implications for the De- velopment of Chiefdoms,Barbara Lass Monograph 36 | New Light on Old Art: Recent Advances in Hunter-Gatherer Rock Art READ ONLY/NO DOWNLOAD

Archaeology of the Chinese Bronze Age

Research,D.W.Whitley and L. L. Loendorf (eds.) Monograph 35 | Pottery of Prehistoric Honduras: Regional Classiﬁcation and Analysis, J. S. Henderson and M. Beaudry-Corbett Monograph 34 | Settlement Archaeology of Cerro de las Mesas, Veracruz, Mexico, Barbara Stark (ed.) Monograph 33 | Girikihaciyan: AHalafian Site in Southeastern Turkey, P. J. Watson and S. LeBlanc Monograph 32 | Western Pomo Prehistory: Excavations at Albion Head, Nightbirds’ Retreat and Three Chop Village, Mendocino County, California, Thomas N. Layton Monograph 31 | Investigaciones Arqueológicos de la Costa Sur de Guatemala,David S. Whitley and Marilyn P. Beaudry (eds.) Monograph 30 | Archaeology of the Three Springs Valley, California: AStudy in Functional Cultural History, Brian D. Dillon and Matthew A. Boxt Monograph 29 | Obsidian Dates IV: ACompendium of Obsidian Hydration Readings from the UCLA Obsidian Hydration Laboratory, Clement W. Meighan and Janet L. Scalise (eds.) Monograph 28 | Archaeological Field Research in the Upper Mantaro, Peru, 1982–1983: Investigations of Inka Expansion and Exchange, Timothy Earle et al. (eds.) Monograph 27 | Andean Archaeology: Papers in Memory of Clifford Evans,Ramiro Matos M., Solveig Turpin, and Herbert Eling, Jr. (eds.) Monograph 26 | Excavations at Mission San Antonio 1976–1978, Robert L. Hoover and Julio J. Costello (eds.) Monograph 25 | Prehistoric Production and Exchange in the Aegean and Eastern Medi- terranean, A. Bernard Knapp and Tamara Stech (eds.) Monograph 24 | Pots and Potters: Current Approaches in Ceramic Archaeology,Pru- dence Rice Monograph 23 | Pictographs and Petroglyphs of the Oregon Country, Part 2,J.Malcolm Loring and Louise Loring Monograph 22 | The Archaeology of Two Northern California Sites, Delmer E. Sanburg, F. K. Mulligan, Joseph Chartkoff, and Kerry Chartkoff Monograph 21 | Pictographs and Petroglyphs of the Oregon Country, Part 1,J.Malcolm Loring and Louise Loring Monograph 20 | Messages from the Past: Studies in California Rock Art, ClementW. Meighan (ed.) Monograph 19 | Prehistoric Indian Rock Art: Issues and Concerns,JoAnne Van Tilburg and Clement W. Meighan (eds.) Monograph 18 | Studies in Cypriote Archaeology, Jane C. Biers and David Soren Monograph 17 | Excavations in Northern Belize, Central America,Raymond Sidrys Monograph 16 | Obsidian Dates III: ACompendium of Obsidian Hydration Determina- tions Made at the UCLA Obsidian Hydration Laboratory,Clement Meighan and Glenn Russell Monograph 15 | Inland Chumash Archaeological Investigations,David S. Whitley, E. L. McCann, and C. W. Clewlow, Jr. (eds.)

230 Roderick B. Campbell READ ONLY/NO DOWNLOAD

Monograph 14 | Papers in Cycladic Prehistory, Jack L. Davis and John F. Cherry (eds.) Monograph 13 | Archaeological Investigations at the Ring Brothers Site Complex, Thou- sand Oaks, California,C.W.Clewlow, Jr., David S. Whitley, and Ellen L. McCann (eds.) Monograph 12 | The Running Springs Ranch Site: Archaeological Investigations at VEN-65 and VEN-261, Jack Prichett and Allen McIntyre Monograph 11 | The Archaeology of Oak Park, Ventura County, California, C. William Clewlow, Jr., and David S. Whitley (eds.) Monograph 10 | Rock Art of East Mexico and Central America: An Annotated Bibliogra- phy,Matthias Strecker Monograph 9 | The Late Minoan IDestruction of Crete: Metal Groups and Stratigraphic Considerations,HaraGeorgiou Monograph 8 | Papers on the Economy and Architecture of the Ancient Maya,Raymond Sidrys (ed.) Monograph 7 | History and Prehistory at Grass Valley, Nevada,C.W.Clewlow, Jr., Helen F. Wells, and Richard Ambro (eds.) Monograph 6 | Obsidian Dates II: ACompendium of Obsidian Hydration Determina- tions Made at the UCLA Obsidian Hydration Laboratory,C.W.Meighan and P. I. Vanderhoeven (eds.) Monograph 5 | The Archaeology of Oak Park, Ventura County, California,C.W. Clewlow, Jr., Allen Pastron, and Helen F. Wells (eds.) -

231 READ ONLY/NO DOWNLOAD

232 READ ONLY/NO DOWNLOAD Ar chaeolo gy fthe of

On the front cover: Zun in the shape of a water buffalo, Huadong Tomb 54 ( image courtesy of the Chinese Academy of Social Sciences, Institute for Archaeology).

MONOGRAPH 79 COTSEN INSTITUTE OF ARCHAEOLOGY PRESS Roderick B. Campbell