From December 19, 1995, to January 20, 1996, we conducted the first systematic regional survey in the Yellow River valley with four colleagues from the Department of History (Archaeology Specialty) at Shandong University in Jinan city: Professors Cai Fengshu, Yu Haiguang, Luan Fengshi, and Fang Hui. Our colleagues are well known for their excavations at significant late Neolithic sites in Shandong province such as Dinggong and Yinjiacheng. We surveyed an area of approximately 35 km2 in southeastern Shandong near the city of Rizhao, including an important archaeological site called Liangchengzhen (Figure 1). Here we briefly describe the goals of our pilot project, methods, and results. Underhill and Bennett plan to continue doing fieldwork in the Liangchengzhen area with the archaeologists from Shandong University. Feinman and Nicholas generously agreed to offer their expertise in systematic regional survey and hope to continue to assist on this aspect of the work. Currently, all of us are preparing a report for publication in China and in the United States.
The major goal of our pilot project was to introduce the method of systematic regional survey to Shandong province. Regional survey has been a critical component of research programs on the development of complex societies in several areas of the world, yielding data on change in social, political, and economic relationships between settlements as well as environmental factors affecting site location. Research on the development of complex societies in Shandong requires data on topics such as change in regional settlement hierarchies, craft production, distribution and consumption, and change in regional demographic patterns. Southeastern Shandong has long been recognized as containing important late prehistoric sites. Liang-chengzhen was one of the first Longshan-period (ca. 2600-1900 B.C.) sites excavated in China (1936) as well as one of the richest, yielding diverse ceramic vessels (including extremely thin-walled, polished black pottery) and jade objects. Unfortunately, factors such as the Japanese invasion prevented further work at the site and extensive publication of the results. One of our major goals was to understand the regional context of this key site. Another goal was to understand the size of the site and the nature of the deposits. Previous fieldwork had indicated that the site may be ca. 90 ha in size. We also hoped that our survey would reveal evidence of earlier, poorly understood Neolithic cultures in southeastern Shandong, such as Dawenkou (ca. 4500-2600 B.C.) and Beixin (ca. 5500-4300 B.C.), as well as the post-Longshan Yueshi culture (ca. 1900-1100 B.C.?).
We believe that our collaborative regional survey was successful for several reasons: the extent of alluviation in the area was not severe, allowing sites to be recognized from surface artifacts on plowed or fallow fields; we achieved an initial understanding of the social and environmental context of the Liangchengzhen site; we found dozens of sites that had not been previously recognized; we examined the nature of the deposits at Liangchengzhen and met some people who had done surface collections there; team members shared their methodological expertise; and there was open debate about methodological issues. Feinman and Nicholas taught the team the full coverage survey methods they have used successfully in Oaxaca, Mexico. Our Chinese colleagues explained methods they have used for recognizing sites and for relative dating with pottery and stone tools. Three of us translated during discussions about method (Underhill, Bennett, Fang Hui). We planned the project for the winter, expecting that the limited vegetation would enhance survey conditions and that snow would be rare in this part of Shandong. Fortunately, it only snowed the day we left the survey area.
Our most common methodological debates were how to identify an archaeological site when surface artifacts are limited and how to recognize postdepositional processes affecting sites in the study area. Most of us agreed that postdepositional factors in the Liangchengzhen area should be investigated further, including natural factors such as river flooding and cultural factors such as removal of rich soil from archaeological sites by farmers. Also, the chronology for each site needs to be clarified, including the area of occupation for each phase. Some types of ceramics that we found are not common in other parts of Shandong, so relative dating from sherds will have to be refined in the future. However, we identified several sites of varying sizes that likely indicate differences in function (Liangchengzhen being the largest). Most sites are from the Longshan period, and other well-represented periods are the Zhou (ca. 1100-256 B.C.) and Han dynasties (ca. 206 B.C.-A.D. 220). More work needs to be done to identify sherds from the Beixin, Dawenkou, and Yueshi cultures in southeastern Shandong. Other goals for future research include determining whether Shang Dynasty (ca. 1700-1100 B.C.) settlements are present in the area and examining variability among sites in terms of lithics.
A grant from the High Risk Exploratory Research program of National Science Foundation, awarded to Underhill and Feinman, made this research possible. We are also grateful to several organizations in China that gave us permission to carry out our survey: the National Bureau of Cultural Relics in Beijing (Guojia Wenwu Ju), the Cultural Relics Bureau of Shandong Province (Shandong Sheng Wenwu Ju), the Rizhao City Bureau of Cultural Affairs (Rizhao Shi Wenhua Ju), and the Rizhao City Museum (Rizhao Shi Bowuguan).
Anne P. Underhill is in the Department of Anthropology at Yale University, New Haven, Conn. Gary Feinman is in the Department of Anthropology at University of Wisconsin, Madison. Linda Nicholas is an honorary fellow in the Department of Anthropology at the University of Wisconsin, Madison. Gwen Bennett is a doctoral graduate student in the Anthropology Department at the University of California, Los Angeles.