in the North China Plain and Yangzi lowlands and to the early agricultural sites. Only very much later did urbanization come to the even more isolated Manchurian valleys. The great Amur River has yet to achieve centrality.
The chiefdoms, or local complex societies, in the most favored localities did well in terms of wealth and population and could both get rich through trade and conquer the less favored ones. If circumscription was sufficient to make it impossible for the less favored to escape or to unite in a large oppositional force, the central social unit would become larger, wealthier, and more populous. It would then be forced, at some point, to develop a ruling elite, law code, and other trappings of a state; informal rule and simple kinship would simply not provide enough structure. Military organization and financing the military, in particular, would require central organization and some sort of ruling group.
All this leads to a necessity for the government to show off its wealth and power by having huge buildings, if only as defensive structures—but usually they are much more than that: they show off administrative power. Moreover, the government moves to control religion, ceremony, spectacle, great holidays, and other solidarity-building institutions. As the Marxists point out, a happy harmony prevails among the elite when the rulers, the army, and the priests are all in agreement. Even if there are dissident factions, they can unite around the goal of keeping the people docile and taxpaying. They can also insure that the elites will get most of the rapidly increasing wealth that urban civilization and trade bring to the city gates. One sees why a natural fence is needed to keep the people in.
The only thing that can disarrange this neat picture is a situation in which a marginal area produces better and better-organized fighters—such that the core’s superior numbers are neutralized. China was to learn all about this, to its enormous cost, when the steppes became organized into chiefdoms and then into true states. (The rise of independent city-states dominating trade could also disrupt this situation but this never happened in China, though it was a continual experience in the Mediterranean and Southeast Asia.)
On the other hand, such governments do often succeed in delivering relative peace and order, so the people may not always want to escape. Complex, densely populated nonurban societies (“chiefdoms”)—like many societies in Native America and Oceania 500 years ago—are particularly conflict-ridden. States slowly but surely reduced the incidence of violent death (Pinker 2011). And of course there is much to be said for the spectacle, excitement, and variety of urban life. The Chinese speak of the “heat and noise” (renao) of cities as a positive and desirable thing, however much romantic and rural Westerners may view urban heat and noise as mere pollutants. The Germans say, “city air makes people free.” (Early German elites lived in the countryside on their estates, however, whereas Chinese elites generally preferred to live in the biggest cities they could find, making Chinese city air rather less free than the German form.)
Civilization, like agriculture in the Old World, appears first in the Near East and then about a thousand years later in China. With agriculture, we can safely assume independent invention, but this is not true of cities. It seems eminently possible that the idea of cities and civilization diffused across Central Asia to China (see Mair 2005—though this source deals mainly with later centuries). We know that bronze technology, horses and chariots, and funeral rites spread from west to east in early civilized times. This being the case, it seems likely that the whole idea of urbanization spread similarly. By 2000 BCE, when the first signs of civilization appear in China, cities and urban life were well established all the way from the entire Nile Valley to northern and central India and the western edges of Central Asia. Large, sophisticated towns flourished throughout much of Central Asia, though they were to vanish in the dark ages that afflicted much of the Western world in the few centuries just before and after 1000. (Climate was one reason—it turned colder and drier—but there were other poorly known factors.)
On the other hand, it seems fairly clear that the Chinese independently invented writing (on early Chinese writing, see Li Feng and Branner 2011). The Chinese would not have invented such a difficult and problematic system if they had had access to the cuneiform or alphabetic options that arose in the Near East. (I realize that this claim could earn me a charge of bias, but I cannot see how it could have been otherwise. The value of the Chinese writing system today is that it distinguishes the countless homonyms of Mandarin. Reconstructions show that the ancient Chinese did not have anything like the current problem in that regard.) Nor would their writing have shown such clear evidence of slow and organic development in place.
Chinese civilization arose in a core area in the western parts of the North China Plain and the adjacent Wei Valley. Until recently, it seemed to be a civilization that began in one area and spread in discrete rings outward, like the ripples from a stone cast in a pond. This neat scenario was early questioned by Wolfram Eberhard (one of my teachers). Today, we know Eberhard was right.
The people of the Yangzi Valley were as advanced as those on the North China Plain, if not more so, from earliest times onward. By 2000 BCE they had large towns and sophisticated art, similar to and culturally related to the proto-civilization of the North China Plain (Underhill and Habu 2006. Sichuan is also providing dramatic new finds that show a related but distinctive early civilization there (Bagley 2001). Urban-size sites extended from the far north to the Yangzi and inland to Chengdu by 2500–2000 BCE. Many had huge walls and large public buildings. Differentiated occupations, complex religion, and other features of civilization are attested all over north and central China. Shao Wangping (2002) believes that neither the view of a West Asian origin for Chinese civilization nor the view of Chinese culture as spreading from a point source on the Yellow River can be sustained any longer. However, Western inspiration for urbanization is not ruled out. The spread of writing (at least) from the central Yellow River area is clear.
Moreover, stunning recent finds in north and northeast China reveal utterly unexpected cultures there. The mysterious and controversial Hongshan culture (4500–3000 BCE) had intensive agriculture, as well as pig burials (Nelson 1994, 1995). It produced many large towns long before China had dynasties. “A huge ritual complex, about 8 by 10 km2, was discovered at the late Hongshan period (ca. 3500–3000 BCE) site of Niuheliang in western Liaoning province…. It contains stone platforms interpreted as altars, stone foundations that could have been temples,” sculptures, images, jewels, shamanic figures, “pig-dragons,” and much more (Underhill and Habu 2006: 131). Perhaps more striking is a statuette of a woman with inset eyes of pale blue jade (Morris 2010: 126). She was presumably blue-eyed. (This does not necessarily mean the people of the town were blue-eyed. In Chinese folk belief, spirit beings were often white-eyed or blue-eyed.) She has been regarded as a “goddess,” and the whole complex called a goddess temple. Hongshan declined (Liu and Chen 2012) and the great sites were no more, but the Liao valleys continued to be important cultural foci.
Although their monumental architecture is huge, the communities were small, perhaps a thousand people. The subsequent Xiajiadian culture created huge stone walls, evidently for defense (Shelach et al. 2011). Magnificent photographs of these finds are now available and show a site a great deal like a city (Zhang Zhongpei 2002: 79–80). If writing were present, no one would hesitate to call it one. However, no writing or anything comparable is associated with these sites. The Hongshan culture remains totally mysterious. Its people may have spoken ancestral Chinese, ancestral Korean, or some lost language. Could they be among the Rong? We will probably never know.
Similar monumental settlements are now being found in northwest China and Inner Mongolia. These finds appear to be greeted with enthusiasm by local people. In one recent case, a road was being built between the Inner Mongolian towns of Chifeng and Chaoyang. Construction turned up a large town 4,000 years old, with a huge wall and several major structures. The choice was made, all the way up from the local archaeologist to Beijing, to delay the road and save the buildings. The mayor of Chifeng gave his opinion in a line that should be circulated to all archaeology projects: “We, people of Chifeng, would rather travel to Chaoyang by donkey than destroy this site.” The site was saved by building an underpass below it (Carver 2011: 714).
Moving back to the focal area on the central Yellow River: Taosi in Shanxi reached 3 square km—the size of a middling Classic Mayan city or small early Near Eastern one—by 2300 BCE or so. It was nearly abandoned by 1900, possibly