We seem to be getting closer to a fuller explanation for pristine state formation. We need the confluence of several factors. First, there needs to be a sufficient abundance of resources to permit the creation of surpluses above what is necessary for subsistence. This abundance can be natural: the Pacific Northwest was so full of game and fish that the hunter-gatherer-level societies there were able to generate chiefdoms, if not states. But more often abundance is made possible through technological advances like agriculture. Second, the absolute scale of the society has to be sufficiently large to permit the emergence of a rudimentary division of labor and a ruling elite. Third, that population needs to be physically constrained so that it increases in density when technological opportunities present themselves, and in order to make sure that subjects cannot run away when coerced. And finally, tribal groups have to be motivated to give up their freedom to the authority of a state. This can come about through the threat of physical extinction by other, increasingly well-organized groups. Or it can result from the charismatic authority of a religious leader. Taken together, these appear to be plausible factors leading to the emergence of a state in places like the Nile valley.14 Thomas Hobbes argued that the state or Leviathan came about as a result of a rational social contract among individuals who wanted to solve the problem of endemic violence and end the state of war. At the beginning of chapter 2 2 I suggested that there was a fundamental fallacy in this, and all liberal social contract theories, insofar as it presupposed a presocial state of nature in which human beings lived as isolated individuals. Such a state of primordial individualism never existed; human beings are social by nature and do not have to make a self-interested decision to organize themselves into groups. The particular form that social organization takes is frequently the result of rational deliberation at higher levels of development. But at lower ones, it evolves spontaneously out of the building blocks created by human biology. I suggested that there was a fundamental fallacy in this, and all liberal social contract theories, insofar as it presupposed a presocial state of nature in which human beings lived as isolated individuals. Such a state of primordial individualism never existed; human beings are social by nature and do not have to make a self-interested decision to organize themselves into groups. The particular form that social organization takes is frequently the result of rational deliberation at higher levels of development. But at lower ones, it evolves spontaneously out of the building blocks created by human biology.
But there is a flip side to the Hobbesean fallacy. Just as there was never a clean transition from an anomic state of nature to an orderly civil society, so there was never a complete solution to the problem of human violence. Human beings cooperate to compete, and they compete to cooperate. The birth of the Leviathan did not permanently solve the problem of violence; it simply moved it to a higher level. Instead of tribal segments fighting one another, it was now states that were the primary protagonists in increasingly large-scale wars. The first state to emerge could create a victor"s peace but over time faced rivals as new states borrowing the same political techniques rose to challenge its predominance.
WHY WEREN"T STATES UNIVERSAL?
We are now in a position to understand why states failed to emerge in certain parts of the world like Africa and Oceania, and why tribal societies persist in regions like Afghanistan, India, and the uplands of Southeast Asia. The political scientist Jeffrey Herbst has argued that the absence of indigenous states in many parts of Africa flows from the confluence of several familiar factors: "The fundamental problem facing state-builders in Africa-be they colonial kings, colonial governors, or presidents in the independent era-has been to project authority over inhospitable territories that contain relatively low densities of people."15 He points out that, contrary to popular imagination, only 8 percent of the continent"s land has a tropical climate, and that 50 percent receives inadequate rainfall to support regular agriculture. Though the human species got its start in Africa, human beings have thrived better in other parts of the world. Population densities had always been low throughout the continent until the arrival of modern agriculture and medicine; it was not until 1975 that Africa reached the population density that Europe enjoyed in the year 1500. Parts of Africa that are exceptions to this generalization, like the fertile Great Lakes region and the Great Rift Valley, have supported much higher population densities and indeed saw the early emergence of centralized states. He points out that, contrary to popular imagination, only 8 percent of the continent"s land has a tropical climate, and that 50 percent receives inadequate rainfall to support regular agriculture. Though the human species got its start in Africa, human beings have thrived better in other parts of the world. Population densities had always been low throughout the continent until the arrival of modern agriculture and medicine; it was not until 1975 that Africa reached the population density that Europe enjoyed in the year 1500. Parts of Africa that are exceptions to this generalization, like the fertile Great Lakes region and the Great Rift Valley, have supported much higher population densities and indeed saw the early emergence of centralized states.
The physical geography of Africa has also made the projection of power difficult. The continent has few rivers that are navigable over long stretches (again, exceptions to this rule like the lower Nile support this point, since it was home to one of the world"s first states). The great deserts of the Sahel are a huge barrier to both trade and conquest, in contrast to the less arid steppe lands of Eurasia. Those mounted Muslim warriors who did manage to cross this obstacle soon found their horses dying of encephalitis from the tsetse fly, which explains why the Muslim parts of West Africa are limited to the northern parts of Nigeria, Cote d"Ivoire, Ghana, and the like.16 In the parts of Africa that are covered by tropical forests, the difficulty of building and maintaining roads was an important obstacle to state building. The hard-surfaced roads the Romans built in Britain were still being used more than a millennium after the collapse of Roman power there; few roads can last more than a few seasons in the tropics. In the parts of Africa that are covered by tropical forests, the difficulty of building and maintaining roads was an important obstacle to state building. The hard-surfaced roads the Romans built in Britain were still being used more than a millennium after the collapse of Roman power there; few roads can last more than a few seasons in the tropics.
There are relatively few regions in Africa that are clearly circ.u.mscribed by physical geography. This has made it extraordinarily difficult for territorial rulers to push their administration into the hinterland and to control populations. Low population density has meant that new land was usually available; people could respond to the threat of conquest simply by retreating farther into the bush. State consolidation based on wars of conquest never took place in Africa to the extent it did in Europe simply because the motives and possibilities for conquest were much more limited.17 This meant, according to Herbst, that the transition from a tribal to a territorial conception of power with clearly conceived administrative boundaries of the sort that existed in Europe did not take place. This meant, according to Herbst, that the transition from a tribal to a territorial conception of power with clearly conceived administrative boundaries of the sort that existed in Europe did not take place.18 The emergence of states in parts of the continent that were circ.u.mscribed, like the Nile valley, is an exception fully consistent with the underlying rule. The emergence of states in parts of the continent that were circ.u.mscribed, like the Nile valley, is an exception fully consistent with the underlying rule.
The reason for the absence of states in aboriginal Australia may be similar to that which pertains to Africa. Australia is for the most part an extremely arid and undifferentiated continent; despite the length of time that human beings have lived there, population density has always been extremely low. The absence of agriculture and of naturally circ.u.mscribed regions may explain the failure of political structures above the level of tribe and lineage to emerge.
The situation in Melanesia is rather different. The region consists entirely of islands, so there is natural circ.u.mscription; in addition, agriculture there was invented long ago. Here the problem is one of scale and the difficulties of power projection, given the mountainous nature of most of the islands. The mountain valleys into which the islands are divided are small and capable of supporting only a limited population, and it is extremely difficult to project power over long distances. As noted earlier, the larger islands with more extensive fertile plains, such as Fiji and Hawaii, did see the emergence of chiefdoms and states.
Mountains also explain the persistence of tribal forms of organization in many of the world"s upland regions, including Afghanistan; the Kurdish regions of Turkey, Iraq, Iran, and Syria; the highlands of Laos and Vietnam; and Pakistan"s tribal agencies. Mountains simply make these regions very difficult for states and their armies to conquer and hold. Turks, Mongols, and Persians, followed by the British, Russians, and now the Americans and NATO forces have all tried to subdue and pacify Afghanistan"s tribes and to build a centralized state there, with very modest success.
Understanding the conditions under which pristine state formation occurred is interesting because it helps to define some of the material conditions under which states emerge. But in the end, there are too many interacting factors to be able to develop one strong, predictive theory of when and how states formed. Some of the explanations for their presence or absence begin to sound like Kipling Just So stories. For example, in parts of Melanesia the environmental conditions are quite similar to those of Fiji or Tonga-large islands with agriculture supporting potentially dense populations-where no state emerged. Perhaps the reason has to do with religion, or particular accidents of unrecoverable history.
It is not clear how important it is to develop such a theory, however, since the vast majority of states around the world were the products of compet.i.tive rather than pristine state formation. Many states were formed, moreover, in historical times for which we have a written record. Chinese state formation, in particular, began extremely early, somewhat after Egypt and Mesopotamia, and contemporaneously with the rise of states around the Mediterranean and in the New World. There are extensive written and archaeological records of early Chinese history, moreover, that give us a far more contextualized sense of Chinese politics. But most important, the state that emerged in China was far more modern in Max Weber"s sense than any of its counterparts elsewhere. The Chinese created a uniform, multilevel administrative bureaucracy, something that never happened in Greece or Rome. The Chinese developed an explicit antifamilistic political doctrine, and its early rulers sought to undermine the power of entrenched families and kinship groups in favor of impersonal administration. This state engaged in a nation-building project that created a powerful and uniform culture, a culture powerful enough to withstand two millennia of political breakdown and external invasion. The Chinese political and cultural s.p.a.ce extended over a far larger population than that of the Romans. The Romans ruled an empire, limiting citizenship initially to a relatively small number of people on the Italian peninsula. While that empire eventually stretched from Britain to North Africa to Germany to Syria, it consisted of a heterogeneous collection of peoples who were allowed a considerable degree of self-rule. By contrast, even though the Chinese monarch called himself an emperor rather than a king, he ruled over something that looked much more like a kingdom or even a state in its uniformity.
The Chinese state was centralized, bureaucratic, and enormously despotic. Marx and Wittfogel recognized this characteristic of Chinese politics by their use of terms like "the Asiatic mode of production" and "Oriental despotism." What I argue in succeeding chapters is that so-called Oriental despotism is nothing other than the precocious emergence of a politically modern state. In China, the state was consolidated before other social actors could inst.i.tutionalize themselves, actors like a hereditary, territorially based aristocracy, an organized peasantry, cities based on a merchant cla.s.s, churches, or other autonomous groups. Unlike in Rome, the Chinese military remained firmly under the state"s control and never posed an independent threat to its political authority. This initial skewing of the balance of power was then locked in for a long period, since the mighty state could act to prevent the emergence of alternative sources of power, both economic and political. No dynamic modern economy emerged until the twentieth century that could upset this distribution of power. Strong foreign enemies periodically conquered parts or the whole of the country, but these tended to be tribal peoples with less-developed cultures, who were quickly absorbed and Sinified by their own subjects. Not until the arrival of the Europeans in the nineteenth century did China really have to contend with foreign models that challenged its own state-centered path of development.
The Chinese pattern of political development differs from that of the West insofar as the development of a precociously modern state was not offset by other inst.i.tutionalized centers of power that could force on it something like a rule of law. But in this respect it also differed dramatically from India. One of Marx"s biggest mistakes was to lump China and India together under a single "Asiatic" paradigm. Unlike China but like Europe, India"s inst.i.tutionalization of countervailing social actors-an organized priestly cla.s.s and the metastacization of kinship structures into the caste system-acted as a brake on the acc.u.mulation of power by the state. The result was that over the past twenty-two hundred years, China"s default political mode was a unified empire punctuated by periods of civil war, invasion, and breakdown, whereas India"s default mode was a disunited system of petty political units, punctuated by brief periods of unity and empire.
The chief driver of Chinese state formation was not the need to create grand irrigation projects, nor the rise of a charismatic religious leader, but unrelenting warfare. It was war and the requirements of war that led to the consolidation of a system of ten thousand political units into a single state in the s.p.a.ce of eighteen hundred years, that motivated the creation of a cla.s.s of permanent trained bureaucrats and administrators, and that justified the move away from kinship as the basis for political organization. As Charles Tilly said of Europe in a later period, for China, "war made the state, and the state made war."
PART TWO.
State Building
6.
CHINESE TRIBALISM.
The origins of Chinese civilization; organization of tribal society in ancient China; characteristics of Chinese family and kinship; spread of feudalism under the Zhou and the nature of political authority
Tribalism has existed in China from the beginning of its recorded history. Segmentary lineages still remain in parts of southern China and Taiwan. When historians speak of Chinese "families," they are often referring not to nuclear units consisting of two parents and their children but to much broader groups of agnates that can number in the hundreds or even thousands. Since early Chinese history is relatively well doc.u.mented, we have a rare opportunity to observe the crystallization of states out of a tribal-level society.
Human beings have lived in China for a very long time. Archaic humans like h.o.m.o erectus h.o.m.o erectus were present there as much as eight hundred thousand years ago, and were present there as much as eight hundred thousand years ago, and h.o.m.o sapiens h.o.m.o sapiens appeared first a few thousand years after their exit from Africa. Millet (in the north) and rice (in the south) were first cultivated at a very early point, and metallurgy and settled communities first appeared during the predynastic Yangshao period (50003000 B.C.). Walled cities and clear evidence of social stratification appeared during the Longshan period (30002000 B.C.). Before this point, religion was based on ancestor or spirit worship presided over by shamans who, as in most band-level societies, were not specialists but simply ordinary members of the community. But with the emergence of more stratified societies during the Longshan period, rulers began to monopolize control over shamanism and use it to bolster their own legitimacy. appeared first a few thousand years after their exit from Africa. Millet (in the north) and rice (in the south) were first cultivated at a very early point, and metallurgy and settled communities first appeared during the predynastic Yangshao period (50003000 B.C.). Walled cities and clear evidence of social stratification appeared during the Longshan period (30002000 B.C.). Before this point, religion was based on ancestor or spirit worship presided over by shamans who, as in most band-level societies, were not specialists but simply ordinary members of the community. But with the emergence of more stratified societies during the Longshan period, rulers began to monopolize control over shamanism and use it to bolster their own legitimacy.1 After the development of agriculture, perhaps the most critical technological development was the domestication of the horse. This may have happened first in Ukraine in the fourth millennium B.C., and spread to Western and Central Asia by the early second millennium. The transition to pastoral nomadism was completed by the beginning of the first millennium, which is when the first mounted tribal peoples started pushing their way into China.2 Much of subsequent Chinese history is dominated by this phenomenon. Much of subsequent Chinese history is dominated by this phenomenon.
The periodization of ancient China can be confusing (see Table 1 Table 1).3 Yangshao and Longshan are archaeological rather than dynastic categories, named after settlements on the middle and lower Yellow River in northern China. Dynastic China begins with the Three Dynasties, the Xia, Shang, and Zhou. The Zhou Dynasty in turn is divided into the Western and Eastern Zhou, a split that occurred in 770 B.C. when the Zhou moved their capital from Haojing in Shaanxi to Luoyang in modern western Henan province. The Eastern Zhou is then itself divided into two subperiods, the Spring and Autumn and Warring States periods. Yangshao and Longshan are archaeological rather than dynastic categories, named after settlements on the middle and lower Yellow River in northern China. Dynastic China begins with the Three Dynasties, the Xia, Shang, and Zhou. The Zhou Dynasty in turn is divided into the Western and Eastern Zhou, a split that occurred in 770 B.C. when the Zhou moved their capital from Haojing in Shaanxi to Luoyang in modern western Henan province. The Eastern Zhou is then itself divided into two subperiods, the Spring and Autumn and Warring States periods.
Table 1. Ancient China
Ancient China refers to the period from earliest prehistory up to the beginning of the Qin Dynasty, which marked the unification of China as a single empire. What we know about this period comes from extensive archaeological data, including large numbers of inscriptions on oracle bones (usually the shoulder bones of sheep), which were used for divination; inscribed bronze vessels; and bamboo strips on which court officials kept records of state affairs.4 Another source of information is the great cla.s.sics of Chinese literature composed in the last few centuries of the Eastern Zhou. Most important are the five canonical works whose study const.i.tuted the foundation of a Chinese Mandarin"s education in later centuries: the Another source of information is the great cla.s.sics of Chinese literature composed in the last few centuries of the Eastern Zhou. Most important are the five canonical works whose study const.i.tuted the foundation of a Chinese Mandarin"s education in later centuries: the Shi Jing Shi Jing, or Book of Odes Book of Odes; the Li Chi Li Chi, or Book of Rites Book of Rites; the Shu Jing Shu Jing, or Book of History Book of History; the I Jing I Jing, or Book of Changes Book of Changes; and the Chun Qiu Chun Qiu, or Spring and Autumn Annals Spring and Autumn Annals. The five cla.s.sics were said to have been compiled, edited, and transmitted by Confucius, and they and their voluminous interpretations were the basis of Confucian ideology, which shaped Chinese culture for millennia. The cla.s.sics were composed against the backdrop of growing civil war and political breakdown during the Eastern Zhou; the Spring and Autumn Annals Spring and Autumn Annals is an account of the reigns of twelve successive rulers of the state of Lu that to Confucius demonstrated the growing degeneracy of this period. The cla.s.sics, as well as works written by Confucius, Mencius, Mozi, Sun Tzu, and others in this time, contain a great deal of historical information, though the accuracy of these primarily literary works is unclear. is an account of the reigns of twelve successive rulers of the state of Lu that to Confucius demonstrated the growing degeneracy of this period. The cla.s.sics, as well as works written by Confucius, Mencius, Mozi, Sun Tzu, and others in this time, contain a great deal of historical information, though the accuracy of these primarily literary works is unclear.
There is clear evidence, however, that there was a tremendous reduction in the total number of political units in China, from approximately ten thousand at the beginning of the Xia Dynasty to twelve hundred at the onset of the Western Zhou, to seven at the time of the Warring States.5 Groundwork for the first truly modern state was laid in the western polity of Qin under Duke Xiao and his minister, Shang Yang. The process of state consolidation reached a conclusion when the king of Qin conquered all of his rivals and established a single empire, uniformly imposing inst.i.tutions first developed in Qin on much of northern China. Groundwork for the first truly modern state was laid in the western polity of Qin under Duke Xiao and his minister, Shang Yang. The process of state consolidation reached a conclusion when the king of Qin conquered all of his rivals and established a single empire, uniformly imposing inst.i.tutions first developed in Qin on much of northern China.
TRIBAL CHINA.
The transition from a tribal to a state-level society took place gradually in China, with state inst.i.tutions being layered on top of kinship-based social structures. What are sometimes referred to as "states" during the Xia and Shang dynasties are actually better characterized as chiefdoms or tribes with increasingly higher levels of stratification and centralized leadership. Up through the end of the Shang Dynasty, kinship remained the primary form of Chinese social organization. This began to change only under the Zhou Dynasty, when true states with standing armies and administrative structures began to emerge.
In this early phase of Chinese history, society was organized as lineages, agnatic groups claiming descent from a common ancestor. The basic military unit consisted of males from approximately one hundred households making up a lineage, grouped under a flag or banner and led by the lineage chief. Lineages could flexibly combine into clans or higher-order lineages, and the king was the supreme head of all lineages in a particular area.6 During the period of the Three Dynasties, ritual behavior within lineages was codified in a series of laws. The rites revolved around worship of the lineages" common ancestor and took place at the ancestral temple that held the tablets inscribed with the ancestor"s name. There were several sections of these temples, corresponding to the level of lineage or sublineage organization. Lineage leaders reinforced their authority through their control over the rites; failure to correctly observe either the rites or military orders led to severe punishment by the king or higher lineage heads. Correspondingly, if an enemy was to be truly vanquished, it was important to break up its ancestral temple, loot its symbolic treasures, and then kill off the enemy"s male progeny to break the "rope of descent."7 As in other tribal societies, China in this period was subject to increasing and decreasing levels of social organization. On the one hand, lineages, based in settled villages, combined for purposes of war, self-defense, or commerce. Sometimes alliances were voluntary and based on common economic interest; sometimes they were due to the ritual respect of a particular leader; quite often they were due to coercion. Warfare became increasingly common, as evidenced by the spread of rammed-earth-walled towns that began to proliferate during the Longshan period.8 On the other hand, lineage society was subject to constant fission, as younger descendants sought new land and established their own branches of the kindred. At this time, China was spa.r.s.ely populated, and families could escape the authority of an established lineage simply by moving to a new place.9 Thus, as the theories of state formation predict, low population density and lack of circ.u.mscription worked against the formation of states and hierarchy. Thus, as the theories of state formation predict, low population density and lack of circ.u.mscription worked against the formation of states and hierarchy.
Nonetheless, in the older parts of the Yellow River valley, population densities rose, along with agricultural productivity. Increasing levels of hierarchy during the Shang Dynasty are seen in the severe punishments that could be imposed by leaders on their followers, and by the spread of slavery and human sacrifice. Oracle bone inscriptions mention five types of punishment: branding the forehead, cutting off the nose, cutting off the feet, castration, and death.10 Many burial sites from this period contain eight to ten prostrate, headless skeletons, probably of slaves or prisoners of war. Higher-level leaders were buried with as many as five hundred sacrificial victims; as many as ten thousand sacrificial victims have been uncovered in burial pits in Yinxu, along with large numbers of horses, chariots, tripods, and other valuable artifacts. Appeasing dead ancestors thus deprived the living of huge amounts of resources, human, animal, and material. Many burial sites from this period contain eight to ten prostrate, headless skeletons, probably of slaves or prisoners of war. Higher-level leaders were buried with as many as five hundred sacrificial victims; as many as ten thousand sacrificial victims have been uncovered in burial pits in Yinxu, along with large numbers of horses, chariots, tripods, and other valuable artifacts. Appeasing dead ancestors thus deprived the living of huge amounts of resources, human, animal, and material.11 Clearly, a shift was beginning to take place from a tribal to a more hierarchical form of polity. Clearly, a shift was beginning to take place from a tribal to a more hierarchical form of polity.
CHINESE FAMILY AND KINSHIP.
One of the great constants in Chinese history is the importance of family and kinship to social organization. The rulers of Qin tried to suppress kin ties in favor of a more impersonal form of administration, both in their own kingdom and for China more broadly once they had established a unified empire. When the Chinese Communist Party came to power in 1949, it too tried to use its dictatorial power to eliminate traditional Chinese familism and bind individuals to the state. Neither of these political projects worked as well as their authors hoped; the Chinese family proved very resilient, and agnatic descent groups still exist in parts of China.12 After the brief Qin Dynasty, impersonal administration was finally established during the Former Han Dynasty (206 B.C.A.D. 9). But kinship made a big comeback toward the end of the Later Han, Sui, and Tang dynasties. Impersonal state administration was restored only during the Song and Ming dynasties beginning in the second millennium A.D. Particularly in southern China, lineages and clans remained strong up to the twentieth century. On a local level, they played a quasi-political function and partially displaced the state itself as a source of authority over many matters. After the brief Qin Dynasty, impersonal administration was finally established during the Former Han Dynasty (206 B.C.A.D. 9). But kinship made a big comeback toward the end of the Later Han, Sui, and Tang dynasties. Impersonal state administration was restored only during the Song and Ming dynasties beginning in the second millennium A.D. Particularly in southern China, lineages and clans remained strong up to the twentieth century. On a local level, they played a quasi-political function and partially displaced the state itself as a source of authority over many matters.
There is a huge literature on Chinese kinship, much of it written by anthropologists who have studied contemporary communities in Taiwan and southern China, and have made use of kinship records going back to the nineteenth century for these areas.13 There are also studies of family relationships for earlier periods of Chinese history based on the extraordinarily detailed records that individual kin groups have left behind. We have much less information on kinship in ancient China, and there is some danger in projecting modern trends that far back in time. Some scholars argue that contemporary lineages were the product of deliberate policies engineered by neo-Confucians during the TangSong transition and that kinship was different prior to the second millennium A.D. There are also studies of family relationships for earlier periods of Chinese history based on the extraordinarily detailed records that individual kin groups have left behind. We have much less information on kinship in ancient China, and there is some danger in projecting modern trends that far back in time. Some scholars argue that contemporary lineages were the product of deliberate policies engineered by neo-Confucians during the TangSong transition and that kinship was different prior to the second millennium A.D.14 Nonetheless, certain features of kinship organization have remained constant over the centuries of Chinese history. Nonetheless, certain features of kinship organization have remained constant over the centuries of Chinese history.
Kinship in Chinese society is strictly patrilineal or agnatic. The lineage has been defined by one anthropologist as "a corporate group corporate group which celebrates which celebrates ritual unity ritual unity and is based on and is based on demonstrated descent demonstrated descent from a common ancestor." from a common ancestor."15 While some modern lineages trace ancestry to an ancestor twenty generations removed, historical ones usually went no deeper than five generations. By contrast, the clan was a much broader grouping of kin encompa.s.sing several lineages and often based on fictive kinship. They and related surname a.s.sociations often existed only to define exogamy. While some modern lineages trace ancestry to an ancestor twenty generations removed, historical ones usually went no deeper than five generations. By contrast, the clan was a much broader grouping of kin encompa.s.sing several lineages and often based on fictive kinship. They and related surname a.s.sociations often existed only to define exogamy.16 As in other agnatic societies, succession and inheritance pa.s.s only through males. A woman is not considered a permanent part of her own lineage but is rather a resource to be used by the family in arranging alliances with other important families. When she marries, she breaks her ties with her birth family, and in many periods of Chinese history could return to visit them only on carefully prescribed days. The wife no longer worships at her birth family"s temple but at that of her husband. Because the "rope of descent" pa.s.ses only through males, she has no status in her new family until she herself bears a male offspring. Indeed, her soul is not secure until she has sons who will pray for her spirit along with her husband"s once she is dead. In more practical terms, her sons are her source of social security in old age.
There is a high degree of tension between the young wife and her mother-in-law, doc.u.mented in countless Chinese novels and plays over the centuries, since the latter is allowed to tyrannize the former until a son is born. But after bearing a son, a woman can achieve very high status as the mother of the heir in an important lineage. Many of the court intrigues in Imperial China revolved around the efforts of powerful dowagers to improve their sons" political standing. In the Former Han Dynasty, empress dowagers were able to choose the heir to the throne on at least six occasions.17 One of the sad truths about premodern societies was the difficulty of producing a male offspring who survived into adulthood. In an era before modern medicine, high status and wealth made very little difference in this quest. The history of monarchies around the world attests to the state of perpetual political crisis that attended the failure of queens or other royal consorts to produce male offspring. Many contemporary j.a.panese anxiously followed the travails of Crown Prince Naruhito"s wife, Masako, in trying to conceive a son after their marriage in 1993. This paled in comparison to a string of earlier emperors, however: only three of the fifteen children of Emperor Ninko (18001846) survived past the age of three, and only five of Emperor Meiji"s (18521912) fifteen offspring reached adulthood.18 In China, as in other societies, this problem was traditionally addressed through concubinage, by which high-status men could effectively acquire second, third, and even more wives. China developed a complex, formalized system for determining succession in such situations. For example, the son of a primary wife had superior inheritance rights to the son of a concubine, even if he were younger, though some emperors violated this rule. Despite the system of rules, uncertainties about succession fueled a great deal of court politics. In 71 B.C., Huo Xian, wife of a prominent official, had the empress Xu murdered while pregnant and had her own daughter subst.i.tuted. In A.D. 115, the emperor Andi"s childless empress Yan had a secondary consort put to death for having given birth to a son.19 As in the case of the Greeks and Romans described by Fustel de Coulanges, the Chinese kinship system was intimately connected with the system of private property. Initially during the Zhou Dynasty, all land was declared to be the property of the state, but Zhou kings were too weak to enforce this, and property increasingly became private and subject to sale or alienation.20 The lineage as a whole owned the property housing the ancestral temple or hall. In addition, the richer lineages could invest in common property like dams, bridges, wells, and irrigation systems. Individual families owned their own plots, but their ability to alienate them was severely entailed by the ritual obligations of the lineage. The lineage as a whole owned the property housing the ancestral temple or hall. In addition, the richer lineages could invest in common property like dams, bridges, wells, and irrigation systems. Individual families owned their own plots, but their ability to alienate them was severely entailed by the ritual obligations of the lineage.21 Growth of the lineage always posed problems with regard to inheritance of property. A system of primogeniture existed during the early Zhou Dynasty, but it was replaced by a rule of splitting inheritances equally among male children that prevailed for most of the rest of Chinese history down to the twentieth century.22 Under this system, family land often got divided into smaller and smaller parcels, leading to economically unviable holdings. The Chinese developed the ideal of the joint family, in which multiple generations of male descendants lived under the same roof. As the sons grew older, they started their own residences on subdivided family land, or else sought to acquire new land nearby. Descendants, however, still had a stake in the lineage"s common property and owed worship duties to common ancestors, which could prevent them from moving too far away or freely selling their land. Under this system, family land often got divided into smaller and smaller parcels, leading to economically unviable holdings. The Chinese developed the ideal of the joint family, in which multiple generations of male descendants lived under the same roof. As the sons grew older, they started their own residences on subdivided family land, or else sought to acquire new land nearby. Descendants, however, still had a stake in the lineage"s common property and owed worship duties to common ancestors, which could prevent them from moving too far away or freely selling their land.23 Strong regional differences later emerged with regard to property and coresidence. In northern China, the power of lineages declined over time; members of lineages moved to different, widely separated villages and lost their common sense of ident.i.ty. In the south, however, lineage and clan members continued to live and work near each other, sometimes with entire villages bearing the same clan surname. There has been a great deal of speculation on the reasons for these different outcomes, including the fact that the south was an unsettled frontier for many centuries, which facilitated lineages remaining together even as they expanded, and the constant wars and displacement that occurred in the north, which tended to break up coresident kinship groups.
It is important to remember that lineage organization was in many respects a privilege of the well-to-do. Only they could afford large estates capable of subdivision, communal property, and the multiple wives and concubines sometimes required to produce heirs. In fact, when the rules of the lineage system were first codified during the Zhou Dynasty, they applied only to certain elite families. Poor families could afford fewer children, and in some cases compensated for the absence of a male heir by adopting a son who would give up his lineage name in favor of his wife"s-a practice that became common in j.a.pan but was frowned upon in China.24 CHINA"S "FEUDAL" PERIOD The Shang people were conquered by the Zhou tribes who had settled to their west along the Wei River (in contemporary Shaanxi province), a process that began early in the eleventh century B.C. The conquest took several years to complete, the Shang forces having to simultaneously battle horse-mounted nomads to their east in Shandong. The Zhou king killed the Shang heir and murdered his own brothers to seize power, and in the end established a new dynasty.25 This conquest set the stage for what many scholars have labeled China"s feudal period, in which political power was held on a highly decentralized basis by a series of hierarchically ranked clans and lineages. Throughout the Western and early Eastern Zhou dynasties, kinship remained the primary principle of social organization. But states began to coalesce all over China as a result of the incessant wars that were fought between these kinship groupings during the Spring and Autumn and Warring States periods. We can follow in great detail the factors driving Chinese state formation, based increasingly not on archaeological reconstructions but on historical evidence.
The process of Chinese state formation is particularly interesting in a comparative perspective, since it sets precedents in many ways for the process Europe went through nearly one thousand years later. Just as the Zhou tribes conquered a long-settled territory and established a feudal aristocracy, so too did the Germanic barbarian tribes overrun the decaying Roman Empire and create a comparably decentralized political system. In both China and Europe, state formation was driven primarily by the need to wage war, which led to the progressive consolidation of feudal lands into territorial states, the centralization of political power, and the growth of modern impersonal administration.26 There were a number of important differences between China and Europe, however, which are masked by the use of terms like "feudal," "family," "king," "duke," and "n.o.bility" to label parallel Chinese inst.i.tutions in English-language histories of dynastic China. We therefore need to define these terms carefully and indicate both where there were important parallels and where the civilizations diverged.
Among the most confusing and misused terms are "feudal" and "feudalism," which have been rendered largely meaningless as a result of promiscuous use by both scholars and polemicists.27 In a tradition that starts with Karl Marx, "feudalism" is often taken to refer to an exploitative economic relationship between lord and peasant that existed in medieval Europe, centering around the manor. The rigidity of a lot of Marxist historiography has made scholars in that tradition look for a feudal stage of development as an inevitable precursor to the rise of modern capitalism in a host of societies where the concept is not relevant. In a tradition that starts with Karl Marx, "feudalism" is often taken to refer to an exploitative economic relationship between lord and peasant that existed in medieval Europe, centering around the manor. The rigidity of a lot of Marxist historiography has made scholars in that tradition look for a feudal stage of development as an inevitable precursor to the rise of modern capitalism in a host of societies where the concept is not relevant.28 A more historically accurate definition of feudalism was laid out by the historian Marc Bloch, focusing on the inst.i.tutions of the fief and va.s.salage as they existed in medieval Europe. The fief was a contractual agreement between lord and va.s.sal by which the latter was given protection and a plot of land in return for serving the lord in a military capacity. The contract was solemnized in a ceremony in which the lord placed the va.s.sal"s hands within his own and sealed the relationship with a kiss. The relationship of dependency entailed clear obligations on both sides and needed to be renewed annually.29 The va.s.sal could then create subfiefs out of his lands and enter into relationships with his own va.s.sals. The system generated its own complex set of ethical norms concerning honor, loyalty, and courtly love. The va.s.sal could then create subfiefs out of his lands and enter into relationships with his own va.s.sals. The system generated its own complex set of ethical norms concerning honor, loyalty, and courtly love.
From the standpoint of political development, the critical aspect of European feudalism was not the economic relationship between lord and va.s.sal but the decentralization of power it implied. In the words of the historian Joseph Strayer, "Western European feudalism is essentially political-it is a form of government ... in which political authority is monopolized by a small group of military leaders, but is rather evenly distributed among members of the group."30 This definition, also a.s.sociated with Max Weber, is the one I will use throughout this volume. The core of the inst.i.tution was the grant of the fiefdom, benefice, or appanage, a delineated territory over which the va.s.sal exerted some degree of political control. Despite the theoretical revocability of feudal contracts, European va.s.sals over time turned their fiefdoms into patrimony, that is, property that they could hand down to their descendants. They acquired political rights over these territories to raise armies, tax residents, and administer justice free from interference of the nominal lord. They were thus in no way the lord"s agent but rather lords in their own right. Marc Bloch points out that the patrimonial character of later feudalism actually represented a degeneration of the inst.i.tution. This definition, also a.s.sociated with Max Weber, is the one I will use throughout this volume. The core of the inst.i.tution was the grant of the fiefdom, benefice, or appanage, a delineated territory over which the va.s.sal exerted some degree of political control. Despite the theoretical revocability of feudal contracts, European va.s.sals over time turned their fiefdoms into patrimony, that is, property that they could hand down to their descendants. They acquired political rights over these territories to raise armies, tax residents, and administer justice free from interference of the nominal lord. They were thus in no way the lord"s agent but rather lords in their own right. Marc Bloch points out that the patrimonial character of later feudalism actually represented a degeneration of the inst.i.tution.31 But it is precisely this distributed character of political power within a feudal system that makes it unique. But it is precisely this distributed character of political power within a feudal system that makes it unique.
In this sense, the Zhou Dynasty China was a feudal society.32 It bore no resemblance to a centralized state. Like many conquest dynasties before and after, the Zhou king found that he did not have the forces or resources under his personal control to rule the territories he had acquired. This was particularly true in the west, where the Zhou were under pressure from steppe nomads, and in the frontier areas to the south that would later become the state of Chu. So he distributed fiefdoms or appanages to his retainers and cowarriors who, given the tribal nature of Zhou society, were his kinsmen. The Zhou king set up seventy-one fiefdoms, of which his kinsmen ruled fifty-three. The others were distributed to defeated Shang lords who pledged loyalty to the new dynasty, or to other Zhou administrators or military commanders. The va.s.sals to whom these lands were granted thereby obtained substantial autonomy to rule them as they pleased. It bore no resemblance to a centralized state. Like many conquest dynasties before and after, the Zhou king found that he did not have the forces or resources under his personal control to rule the territories he had acquired. This was particularly true in the west, where the Zhou were under pressure from steppe nomads, and in the frontier areas to the south that would later become the state of Chu. So he distributed fiefdoms or appanages to his retainers and cowarriors who, given the tribal nature of Zhou society, were his kinsmen. The Zhou king set up seventy-one fiefdoms, of which his kinsmen ruled fifty-three. The others were distributed to defeated Shang lords who pledged loyalty to the new dynasty, or to other Zhou administrators or military commanders. The va.s.sals to whom these lands were granted thereby obtained substantial autonomy to rule them as they pleased.33 There were a number of important differences between Chinese feudalism under the Zhou and its European variant. In Europe, segmentary, tribal inst.i.tutions were destroyed at the beginning of Europe"s feudal period, usually within a couple of generations after a barbarian tribe"s conversion to Christianity. European feudalism was a mechanism for binding unrelated lords to unrelated va.s.sals, facilitating social cooperation in a society where complex kinship no longer existed. In China, by contrast, the primary political actors were not individual lords but lords and their kinship groups and their kinship groups. Within a European lord"s domain, impersonal administration had already begun to take root, in the form of the feudal contract between lord and peasant. Authority was vested in the lord himself and not in the lord"s clan. The fief was a possession of his family but not of a larger corporate descent group.
In China, on the other hand, fiefdoms were granted to kinship groups, who could then subinfeudate their lands to sublineages or collateral branches of the tribe. The authority of an individual Chinese n.o.ble was therefore less hierarchical and weaker than that of a European lord, because he himself was embedded in a larger kinship framework that limited his discretion. I noted earlier that in tribal societies, leadership is often achieved rather than ascribed-it has to be earned by the leader rather than being given by birth. In Zhou China, leadership was evolving in a more hierarchical direction, but it still remained constrained by kin networks and thus appeared more "tribal" than its European variant. According to one observer, during the Spring and Autumn period, "The state resembled an enlarged household; the ruler reigned but did not rule. Ministers were important not because they held their offices; they were important and received offices because they were kin to the ruler or because they were heads of prominent families."34 The king was more first among equals than a true sovereign: "Various stories tell of n.o.bles who upbraided the ruler in public and spat at him without being reprimanded or punished, who rejected requests for precious objects, who played board games with the ruler in the midst of his harem, who helped themselves uninvited from the ruler"s table, or who called on the ruler to share dinner, only to find him out back shooting birds." The king was more first among equals than a true sovereign: "Various stories tell of n.o.bles who upbraided the ruler in public and spat at him without being reprimanded or punished, who rejected requests for precious objects, who played board games with the ruler in the midst of his harem, who helped themselves uninvited from the ruler"s table, or who called on the ruler to share dinner, only to find him out back shooting birds."35 In the clan-based organization of Zhou society, armies were themselves segmented, with no centralized command and control. Each lineage raised its own forces and combined (like Nuer segments) into larger units. "Accounts of campaigns reveal that in the field these levies remained under their own commanders, that major decisions were generally made by group consultation among the leaders, and that the detachments were only loosely bound together so that a commander might lead off his own men without regard for the rest of the army."36 There were numerous cases in which a subordinate overrode the orders of a nominal lord because no strict command-and-control hierarchy existed. According to the anthropological categories developed in chapter There were numerous cases in which a subordinate overrode the orders of a nominal lord because no strict command-and-control hierarchy existed. According to the anthropological categories developed in chapter 5 5, early Zhou polities were tribes or at best chiefdoms and not states.
Chinese feudal society in the Zhou Dynasty was similar to its European counterpart insofar as it developed sharp cla.s.s divisions and an aristocracy set off by a moral code built around honor and the risk of life in violent struggle. Early tribal societies start off relatively egalitarian, with various leveling mechanisms to prevent the emergence of sharp status differences. Then certain individuals begin to distinguish themselves in the hunt. There is a continuity, stretching as we have seen all the way back to the human species" primate ancestors, between the hunt and warfare. In hunting and conquest, hierarchy a.s.serts itself because some individuals and groups are simply better hunters and warriors than others. Those who excel in the hunt tend to excel in war; the cooperative skills needed for hunting evolve into military tactics and strategy. Through victory in battle, some lineages acquire higher status than others, and within each lineage, warriors who distinguish themselves emerge as leaders.
So too in China. The continuity between hunting and warfare was preserved in a series of rituals that served to legitimate the social status of the warrior aristocracy. Mark Lewis argues that during the Spring and Autumn period, "The actions that set the rulers apart from the ma.s.ses were the "great services" of those altars, and these services were ritually directed violence in the form of sacrifices, warfare, and hunting."37 Hunting brought animals to be sacrificed to the ancestors, while warfare brought human sacrifices, a Shang practice that was continued under the Zhou until the fourth century B.C. Military campaigns began at temples with sacrifices and prayers to ensure the success of the campaign. Meat was ritually shared during the rites, prisoners" blood was shed to consecrate war drums, and particularly hated enemies were turned into a meat sauce to be consumed by members of the court or army. Hunting brought animals to be sacrificed to the ancestors, while warfare brought human sacrifices, a Shang practice that was continued under the Zhou until the fourth century B.C. Military campaigns began at temples with sacrifices and prayers to ensure the success of the campaign. Meat was ritually shared during the rites, prisoners" blood was shed to consecrate war drums, and particularly hated enemies were turned into a meat sauce to be consumed by members of the court or army.38 Aristocratic warfare in early Zhou China became highly ritualized. Wars were fought for the purpose of making another clan recognize one"s dominance, or to avenge slights to one"s honor. Armies went forth to defend the "inherited achievements of ancestors"; failure to do so would deny a leader the proper rites when he became a dead ancestor. They could achieve this through ceremonial tests of strength and honor rather than an all-out fight to the death. Battles were often prearranged between groups of aristocrats, who observed a complex set of rules. The appearance of the enemy on the battlefield required an army to engage or suffer dishonor, and it was sometimes regarded as dishonorable not to attack the enemy"s strongest point. Conversely, armies would withdraw from the field of battle when the opposing lord died, so as not to increase the other side"s mourning duties. In the early Spring and Autumn period, aristocrats did most of the fighting from chariots that were expensive and required a high degree of skill to operate and maintain.39 Obviously, the military strategist Sun Tzu"s advice to use the "indirect method," in which surprise and deceit were important components, came from a later period in Chinese history. Obviously, the military strategist Sun Tzu"s advice to use the "indirect method," in which surprise and deceit were important components, came from a later period in Chinese history.
China during the early Zhou Dynasty had evolved into something between a tribal and a chiefdom-level society. None of the units commonly referred to as "states" in the histories were true states. Zhou China was a textbook example of a patrimonial society. That is, the entire country was "owned" by a series of local lords and their kin groups. Within the constraints of China"s agnatic kinship rules, the land and the people living on it were patrimony or heritable property that was pa.s.sed down to descendants. There was no distinction in this society between public and private; each ruling lineage raised armies, imposed taxes, and dispensed justice as it saw fit. All this, however, soon changed.
7.
WAR AND THE RISE OF THE CHINESE STATE.
How the Chinese state arose out of military compet.i.tion; Shang Yang"s modernizing reforms; the doctrine of Legalism and its critique of Confucian familism; why political development was not accompanied by economic or social development
During the Eastern Zhou Dynasty (770256 B.C.), genuine states began to coalesce in China. They established standing armies that were capable of enforcing rules throughout a defined territory; they created bureaucracies to collect taxes and administer laws; they mandated uniform weights and measures; and they created public infrastructure in the form of roads, ca.n.a.ls, and irrigation systems. One state in particular, the kingdom of Qin, embarked on a remarkable modernizing project whose direct target was the kinship-based, patrimonial social order of the early Zhou. It democratized the army by bypa.s.sing the warrior aristocrats and directly conscripting ma.s.ses of peasants, it engaged in large-scale land reform by dispossessing patrimonial landowners and giving land directly to peasant families, and it promoted social mobility by undermining the power and prestige of the hereditary n.o.bility. As "democratic" as these reforms sound, their only purpose was to increase the power of the Qin state and thus create a remorseless dictatorship. The strength of these modern political inst.i.tutions allowed Qin to defeat all of the other contending states and unify China.
WAR AND STATE BUILDING.
The political scientist Charles Tilly has famously argued that European state building was driven by the need of European monarchs to wage war.1 The correlation between war and state building is not a universal one; this process has not, by and large, played out in Latin America. The correlation between war and state building is not a universal one; this process has not, by and large, played out in Latin America.2 But war was without question the single most important driver of state formation during China"s Eastern Zhou Dynasty. Between the beginning of the Eastern Zhou in 770 B.C. and the consolidation of the Qin Dynasty in 221 B.C., China experienced an unremitting series of wars that increased in scale, costliness, and lost human lives. China"s transition from a decentralized feudal state to a unified empire was accomplished entirely through conquest. And virtually every modern state inst.i.tution established in this period can be linked directly or indirectly to the need to wage war. But war was without question the single most important driver of state formation during China"s Eastern Zhou Dynasty. Between the beginning of the Eastern Zhou in 770 B.C. and the consolidation of the Qin Dynasty in 221 B.C., China experienced an unremitting series of wars that increased in scale, costliness, and lost human lives. China"s transition from a decentralized feudal state to a unified empire was accomplished entirely through conquest. And virtually every modern state inst.i.tution established in this period can be linked directly or indirectly to the need to wage war.
When compared to other warlike societies, China"s b.l.o.o.d.y record during the Eastern Zhou stands out. One scholar has calculated that in the 294-year duration of the Spring and Autumn period, more than 1,211 wars were fought between and among Chinese "states." Throughout this entire period, there were only 38 years of peace. More than 110 political units were extinguished during this time. During the 254 years of the subsequent Warring States period, 468 wars took place, with only 89 peaceful years. The total number of wars declined only because the number of states had fallen dramatically through conquest and incorporation. During the Warring States period, sixteen states were extinguished by the seven that survived during this period. But the wars that did occur increased dramatically in scale and duration. During the Spring and Autumn period, some wars consisted of a single battle and were concluded in a day. Toward the end of the Warring States period, sieges could last for months and wars for years, and involve armies as large as five hundred thousand troops.3 Compared to other militaristic societies, China under the Zhou was remarkably violent. By one estimate, the state of Qin succeeded in mobilizing 8 to 20 percent of its total population, compared to only 1 percent for the Roman Republic and 5.2 percent for the Greek Delian League. Rates of mobilization were even lower in early modern Europe.4 Casualties were also of unprecedented scale. Livy reports that the Roman Republic lost approximately 50,000 soldiers in their defeats at Lake Trasimene and Cannae; a Chinese memorialist claims that 240,000 soldiers died in one battle in 293 B.C. and 450,000 in another in 260. In all, the state of Qin was said to have killed more than 1.5 million soldiers of other states between 356 and 236 B.C. All of these figures are regarded by historians as wildly inflated and unverifiable, but it is still remarkable that the Chinese ones are a full order of magnitude higher than their Western counterparts. Casualties were also of unprecedented scale. Livy reports that the Roman Republic lost approximately 50,000 soldiers in their defeats at Lake Trasimene and Cannae; a Chinese memorialist claims that 240,000 soldiers died in one battle in 293 B.C. and 450,000 in another in 260. In all, the state of Qin was said to have killed more than 1.5 million soldiers of other states between 356 and 236 B.C. All of these figures are regarded by historians as wildly inflated and unverifiable, but it is still remarkable that the Chinese ones are a full order of magnitude higher than their Western counterparts.5 INSt.i.tUTIONAL INNOVATIONS BROUGHT ON BY CONSTANT WARFARE.
Intensive warfare created incentives powerful enough to lead to the destruction of old inst.i.tutions and the creation of new ones to take their place. These occurred with regard to military organization, taxation, bureaucracy, civilian technological innovation, and ideas.
Military Organization The first consequence of this high level of warfare was, unsurprisingly, an evolution in the military organizations of the warring states.
As noted earlier, wars in the early Spring and Autumn period were fought by aristocrats riding chariots. Each chariot required a driver and at least two warriors, and was accompanied by an extensive logistics train of up to seventy soldiers. Driving a chariot and firing from it were difficult skills requiring substantial training and thus suitable as aristocratic occupations. 6 6 Infantry in this period served only as auxiliaries. Infantry in this period served only as auxiliaries.
The transition from chariot to infantry/cavalry warfare took place gradually at the end of the Spring and Autumn period. Chariots were of limited use in the southern states of Wu and Yue, which had many lakes and swamps, and they were not effective in mountainous areas. Cavalry made its first appearance at the beginning of the Warring States period, evidently based on experience with the mounted western steppe barbarians. Infantry became more useful with the proliferation of iron weapons, crossbows, and lamellar (plated) armor. The western state of Qin was one of the first to reorganize its army and eliminate chariots in favor of a mixture of cavalry and foot soldiers, due partly to terrain and partly to constant pressure from barbarians. The state of Chu was the first to conscript the people of another state when it defeated Chen and forced its farmers into military service. These troops were not organized by kinship group but by administrative units arranged in clear hierarchies with fixed numbers of subordinate units.7 The first all-infantry army was deployed in the mid-sixth century B.C. and infantry completely displaced chariot armies over the next two centuries. Ma.s.s conscription of peasants became common practice by the beginning of the Warring States period. The first all-infantry army was deployed in the mid-sixth century B.C. and infantry completely displaced chariot armies over the next two centuries. Ma.s.s conscription of peasants became common practice by the beginning of the Warring States period.8 The shift from chariots to infantry as the core of a Chinese army"s striking power has clear parallels with the shift from the heavily armored mounted knights to infantry armies made up of bowmen and pikemen in Europe. Neither of these developments enhanced the social position of the aristocracy, who const.i.tuted the charioteers and knights. In both civilizations, it was only the aristocratic elite that could afford to equip themselves for the older style of warfare and who had the requisite specialist training fo