The Origins Of Political Order Part 9

Qing Qing .

Despite a nearly fifty-year interruption, however, the legitimacy of a centralized state had become so widely accepted by the end of the Tang that one of those military commanders, Zhao Kuangyin, was able to reunify the country in 960 as the emperor Taizu, founder of the great Song Dynasty. In many ways the Song was the most intellectually fertile of them all. While Buddhism and Daoism had made great inroads among both the Chinese people and elites during the Sui and Tang dynasties, Confucianism saw a huge revival at their expense during the Northern Song. Neo-Confucianism was a powerful intellectual movement that spread to the neighboring countries of Korea and j.a.pan and greatly influenced intellectual life throughout East Asia.3 At the same time, China began to experience a new series of invasions by tribal peoples that succeeded in conquering large parts of its territory, and finally the country as a whole.4 This began with the Khitan, a Turco-Mongol group from the Mongolian border region that set up the enormous Liao Empire to China's north and conquered sixteen key northern prefectures with ethnic Han Chinese populations. To the west of the Liao Empire, the Tanguts established the Xi Xia state, which included border regions that had been under Chinese control during earlier dynasties. Next to emerge were the Rurzhen (ancestors of the Manchus), a tribal people coming out of Manchuria who destroyed the Liao Empire and pushed the Khitan back into Central Asia. (They were pushed so far west that they eventually b.u.mped into the Russians, who thereafter referred to all Chinese as "Kitaiskiy.") In 1127, the Rurzhen sacked the Song capital of Kaifeng, took prisoner both the recently abdicated emperor and his son, and forced the entire Song court to move to southern China, inaugurating the Southern Song Dynasty. The Rurzhen state of Jin at its peak controlled roughly one-third of China, until it was in turn smashed in 1234 by another nomadic invader, the Mongols. This began with the Khitan, a Turco-Mongol group from the Mongolian border region that set up the enormous Liao Empire to China's north and conquered sixteen key northern prefectures with ethnic Han Chinese populations. To the west of the Liao Empire, the Tanguts established the Xi Xia state, which included border regions that had been under Chinese control during earlier dynasties. Next to emerge were the Rurzhen (ancestors of the Manchus), a tribal people coming out of Manchuria who destroyed the Liao Empire and pushed the Khitan back into Central Asia. (They were pushed so far west that they eventually b.u.mped into the Russians, who thereafter referred to all Chinese as "Kitaiskiy.") In 1127, the Rurzhen sacked the Song capital of Kaifeng, took prisoner both the recently abdicated emperor and his son, and forced the entire Song court to move to southern China, inaugurating the Southern Song Dynasty. The Rurzhen state of Jin at its peak controlled roughly one-third of China, until it was in turn smashed in 1234 by another nomadic invader, the Mongols.5 After taking northern China, the Mongols under Kublai Khan invaded from the southwest, and this time occupied the entire country. In 1279, the Mongols chased the Southern Song court down to Yaishan, an island in the far southeast, where thousands of courtiers committed suicide by jumping off of a cliff into the sea when finally surrounded by Mongol forces. After taking northern China, the Mongols under Kublai Khan invaded from the southwest, and this time occupied the entire country. In 1279, the Mongols chased the Southern Song court down to Yaishan, an island in the far southeast, where thousands of courtiers committed suicide by jumping off of a cliff into the sea when finally surrounded by Mongol forces.6 Kublai Khan became the first emperor of the new Yuan Dynasty until these foreign rulers were finally driven out in a nationalist uprising and replaced by a new indigenous Chinese dynasty, the Ming, in 1368. Kublai Khan became the first emperor of the new Yuan Dynasty until these foreign rulers were finally driven out in a nationalist uprising and replaced by a new indigenous Chinese dynasty, the Ming, in 1368.

While the prolonged period of military compet.i.tion during the Spring and Autumn and Warring States periods initiated an intense round of state building, foreign invasion during the Song Dynasty did not have remotely comparable effects on the Chinese political order. Despite the intellectual brilliance of the neo-Confucian school that arose during the Northern Song Dynasty, this was a rather dispiriting time when internal factional struggles within the Chinese court prevented the regime from preparing adequately to meet the clear and present danger arising on its borders. The reasons for this complacency lay in the fact that the source of military pressure was pastoral nomads at decidedly lower levels of social development than China itself. At this point in human history, political development did not necessarily confer on state-level societies decisive military advantages over tribal-level peoples organized as light cavalry. In the particular geography of China, the Middle East, and Europe, bordering as they did on the vast steppes of Central Asia, this led to the repeated cycle of decadence, barbarian conquest, and civilized renewal noted by the Arab philosopher Ibn Khaldun. The Khitans, Tanguts, Rurzhen, and Mongols all eventually adopted Chinese inst.i.tutions once they conquered Chinese territory; not one left behind a significant political legacy. It would take conquest by far more developed "barbarians" from Europe to stimulate the Chinese political system into more fundamental reform.

One of the broadest political developments to occur in China between the founding of the Sui in 581 and the later years of the Song Dynasty in the twelfth century was the reversal of patrimonial government and the restoration of centralized power operating through something that looked like the cla.s.sical bureaucracy of the Former Han Dynasty. By the end of this period, Chinese government was no longer dominated by a small circle of aristocratic families but was rather ruled by a gentry elite recruited from a much broader swath of society. The integrity of the bureaucracy as the guardian of Confucian values had been restored, laying the basis for the impressive governmental system of the Ming Dynasty in the fourteenth century. China's population had also increased enormously over this period, growing to fifty-nine million in 1000 and then to one hundred million by the year 1300.7 China's land area also expanded to something far closer to its present-day extent with the settlement of large frontier areas in the south. Commerce and communications across the whole of this huge region increased substantially through the building of ca.n.a.ls and roads. And yet, despite the size of the polity, China developed a centralized political structure that set rules and extracted taxes from across this complex society. No European state was to come close to governing so large a territory for more than another half millennium. China's land area also expanded to something far closer to its present-day extent with the settlement of large frontier areas in the south. Commerce and communications across the whole of this huge region increased substantially through the building of ca.n.a.ls and roads. And yet, despite the size of the polity, China developed a centralized political structure that set rules and extracted taxes from across this complex society. No European state was to come close to governing so large a territory for more than another half millennium.

The idea that China had established (or reestablished) a far more modern political system not after its contact with the West in the seventeentheighteenth centuries but during the TangSong transition, was first put forward by the j.a.panese journalist-scholar Naito Torajiro after World War I.8 Naito argued that rule by aristocrats was swept away during the turbulent period after 750 when the Tang Dynasty experienced a number of internal rebellions and wars that empowered a series of military strongmen of nonn.o.ble background. After the Song Dynasty came to power in 960, the emperor's position was no longer threatened by n.o.ble families, and a much purer form of centralized despotism resulted. The examination system became a more open method of recruitment into the elite, and the position of commoners was improved by the ending of their serflike obligations to aristocratic landlords. A common mode of life was established throughout China, one less dependent on inherited privilege; the highly formal writing of the Tang period was replaced with a vernacular literature and easily accessible popular novels and histories. Naito drew explicit parallels to the early modern period in European history when feudal privileges were ended and equality of citizenship was introduced under the aegis of a strong absolutist state. Naito argued that rule by aristocrats was swept away during the turbulent period after 750 when the Tang Dynasty experienced a number of internal rebellions and wars that empowered a series of military strongmen of nonn.o.ble background. After the Song Dynasty came to power in 960, the emperor's position was no longer threatened by n.o.ble families, and a much purer form of centralized despotism resulted. The examination system became a more open method of recruitment into the elite, and the position of commoners was improved by the ending of their serflike obligations to aristocratic landlords. A common mode of life was established throughout China, one less dependent on inherited privilege; the highly formal writing of the Tang period was replaced with a vernacular literature and easily accessible popular novels and histories. Naito drew explicit parallels to the early modern period in European history when feudal privileges were ended and equality of citizenship was introduced under the aegis of a strong absolutist state.9 While much about the Naito hypothesis has been debated (particularly his effort to fit East Asian history into a Western periodization), many of his broad conclusions have been accepted by more recent scholars. While much about the Naito hypothesis has been debated (particularly his effort to fit East Asian history into a Western periodization), many of his broad conclusions have been accepted by more recent scholars.10 We can now turn to the four questions about China's political order posed at the beginning of the chapter, starting with the issue of despotism, and whether it was more severe in China than in other civilizations.

THE EVIL EMPRESS WU.

The story of Wu Zhao (624705), known to later Chinese memorialists as the "evil empress Wu," is compelling enough to deserve retelling quite apart from what it teaches us about the nature of Chinese politics. Empress Wu was the only woman to rule China in her own name and to establish her own dynasty. Her rise and fall is a chronicle of intrigue, brutality, terror, s.e.x, mysticism, and female empowerment. She was an extraordinarily gifted politician who gained power through sheer will and cunning, an achievement all the more striking given the resolutely antifemale nature of Confucian ideology.11 In the earlier discussion of the rule of law, I noted that it often applies initially to elites rather than to the broad ma.s.s of the population, who are not considered fully human beings subject to the law's protection. Where the rule of law does not exist, on the other hand, it is frequently the case that it is at times more dangerous to be a member of the elite than to be an ordinary person, given the high stakes and intense compet.i.tion for power at the top. This was the situation that played out under the reign of Empress Wu, who unleashed a wide-ranging terror against China's old aristocratic families.

A number of historians, particularly Marxist ones, have seen great social implications in the ascent of Empress Wu. Some have argued that she represented a rising bourgeois cla.s.s; others, that she was a champion of the ma.s.ses; still others, that she played an important role in pushing aside the patrimonial elites of the Sui and early Tang Dynasty period, replacing them with nonaristocratic officials. It is not clear that any of these theories is ultimately correct: she herself had an impeccable aristocratic lineage, being related to the royal Yang family of the Sui Dynasty. Far from promoting able commoners, she canceled the examinations for several years so she could pack the bureaucracy with her own favorites. To the extent that she contributed to the broader TangSong transition, it was because her purges of real and suspected aristocratic opponents decimated their numbers and weakened that cla.s.s as a whole, paving the way for a rebellion by An Lushan that marked the beginning of the end of the Tang Dynasty and set in train enormous social transformations of Chinese society.

Wu Zhao got her start, like many other women in the Chinese court, as a lowly concubine of the second Tang emperor, Taizong. Her father had been a supporter and later high official of the first Tang emperor, Gaozu, and her mother as noted was descended from the Sui royal family. She was rumored to have had an affair with Taizong's son Gaozong even before his father died. On her husband's death, she shaved her head and entered a Buddhist nunnery, but the new emperor Gaozong's senior consort, the empress w.a.n.g, wanted to distract him from another concubine and deliberately brought her to court as a rival.

This proved to be a deadly mistake. Emperor Gaozong was infatuated with Wu Zhao, and in the course of his long reign proved to be weak and easily manipulated by her. Wu Zhao had a daughter with the emperor, whom she arranged to have smothered after the childless Empress w.a.n.g had visited the child in the palace. The empress was accused of murdering Wu Zhao's daughter; w.a.n.g and a former favorite were demoted to commoner status and their families were exiled to a distant southern province. Wu Zhao then advanced to the position of senior consort. Upon becoming empress herself in 655, she had the former empress w.a.n.g and rival concubine chopped into pieces and stuffed in a wine vat. One by one, the court officials who had supported the former empress and opposed Wu Zhao's rise, including many who had loyally served former Tang emperors, found themselves exiled, or ended up dead.

While many Chinese women have exercised de facto power as regents or powers behind the throne for their sons or husbands, Empress Wu was determined to rule as a true coemperor and made increasingly public displays of her autonomous power. When the emperor accused her of witchcraft and sorcery as a means of getting out from under her domination, she confronted him and forced him to kill her accusers and purge all of their supporters from the court. She shocked the court by reviving a number of ancient ceremonials by which she honored herself as well as her husband, and moved the capital from Chang'an to Luoyang to escape ghosts of the many opponents she had murdered there. The empress had the heir apparent poisoned, then framed her own son who was next in line for the throne, on a charge of conspiracy to usurp his father, whereupon he was exiled and forced to commit suicide. When her husband finally died in 683, she had his successor (and her third son), Zhongzong, dragged from the throne and sequestered.

The empress's rise, not surprisingly, led to open rebellion in 684 on the part of a group of Tang aristocrats whose families she had degraded. The empress acted quickly to suppress the uprising and then unleashed a reign of terror against the entire n.o.ble cla.s.s by setting up a network of spies and informants who were lavishly rewarded for denouncing conspiracies. Her secret police engaged in what would now be called widespread "extrajudicial killings," and when the terror had run its course, she turned on her police officials and had them executed as well. This paved the way for her declaration of a new Zhou Dynasty in 690, ruling in her name alone and not that of any male relative.

Empress Wu promoted a number of populist policies, reducing taxes and corvee labor, cutting back on lavish public expenditures, and distributing support to the aged and poor. She also promoted the writing of histories of Chinese women, raised the mourning duties owed mothers, and canonized her own mother as empress dowager. She did succeed in effecting a social revolution insofar as she killed off a large number of Tang aristocrats and Confucian scholars who had staffed the old administrative system. She replaced them, however, not with a cadre of talented commoners but with a series of favorites and sycophants, for whom she had to relax the examination and education requirements. The end of her reign was marked by mysticism, a series of lovers (often connected to her religious pa.s.sions), and openly venal patronage that she did not attempt to control. Nearly eighty years old, she was finally forced from power by a conspiracy that restored her son Zhongzong and the Tang Dynasty to power.

The empress Wu's behavior was hardly typical of all Chinese rulers, and subsequent Confucian moralists inveighed against her as a particularly bad ruler. But she was neither the first nor the last Chinese sovereign to behave despotically and to unleash a ma.s.sive reign of terror against the regime's own elites. Most European monarchs behaved in a more rule-bound way, even if their treatment of peasants and other commoners was often much crueler.

The empress Wu's rise also const.i.tuted a setback for the empowerment of Chinese women, since later writers took her to be an example of the bad things that happen when women get involved in politics. The Ming emperor had a metal plaque posted in his palace warning him and his successors against the intrigues of palace women. The latter had to return to the practice of manipulating their sons or husbands from behind the scenes.12 THE MANDATE OF HEAVEN.

The empress Wu's attempt to seize the throne and create her own dynasty raises the question of how Chinese monarchs acquired legitimacy in the first place. In Leviathan Leviathan, Thomas Hobbes argues that the sovereign derives his legitimacy from an unwritten social contract by which each individual gives up his natural liberty to do as he pleases in order to secure his own natural right to life, which would otherwise be threatened by the "warre of every man against every man." If we subst.i.tute "group" for "man," it is clear that many premodern societies operated on the basis of such a social contract, China's included. Human beings were willing to give up a huge amount of freedom and delegate a corresponding amount of discretion to an emperor who would rule them and guarantee social peace. They found this preferable to a state of war, which they had experienced repeatedly in their history, when powerful oligarchs fought each other and exploited their own people without restraint. This, then, was the meaning of the Mandate of Heaven: it was a conferral by Chinese society of legitimacy on a particular individual and his descendants to rule them with dictatorial authority.

What is perplexing about the Chinese system was not that the Mandate of Heaven existed in the first place, since a functional equivalent to it existed in all princely societies. The issue was rather procedural: How did a pretender to the throne know when he (or in the empress Wu's case, she) had the Mandate of Heaven? And once conferred, why did not other ambitious pretenders try to take it away at the first opportunity, given the enormous power and wealth that came along with being emperor?

The legitimacy of rulers in premodern societies can come from a number of sources. In hunter-gatherer and tribal societies, it usually is the result of some form of election, if not by the people as a whole, then by the leading lineages or tribal elders who hold a council and often vote on who will lead them. In feudal Europe, some form of elective procedure survived into early modern times, when bodies bearing names like the Estates-General or Cortes would be called upon to ratify the coming to power of a new dynasty. This occurred even in Russia, where a zemskiy sobor zemskiy sobor (a.s.sembly) of n.o.bles was called to legitimate the transfer of power to the Romanov dynasty in 1613. (a.s.sembly) of n.o.bles was called to legitimate the transfer of power to the Romanov dynasty in 1613.

The other major source of dynastic legitimacy was religion. In Christian Europe, the Middle East, and India, there were powerful religious establishments that could confer legitimacy on a ruler, or at times take it away (as in Gregory VII's conflict with the Holy Roman Emperor). Often these religious establishments were under the heel of the political authorities and had little choice but to confirm the ruling house. But in times of leadership struggle, these religious authorities could often tilt the balance in one direction or another through their ability to confer legitimacy on one of the contenders.

China was different from all of these other civilizations insofar as the Mandate of Heaven involved neither election nor religious legitimation. There was no Chinese inst.i.tutional equivalent of the Estates-General by which the elites of Chinese society could meet to formally ratify the selection of a new dynastic founder. Nor was there religious legitimation awarded by a religious hierarchy. There was no transcendental G.o.d in the Chinese system. The "heaven" in the Mandate of Heaven was not conceived of as a deity in the sense of the monotheistic religions Judaism, Christianity, and Islam, which laid down a clear set of written rules. Rather, it was more like Nature or the "grand order of things" that could be upset and required a return to equilibrium. Furthermore, there was no religious inst.i.tution that could award the mandate on behalf of heaven, the way that a Christian pope or Muslim caliph could legitimate a king or sultan.13 A change of dynasty always poses a major problem of legitimacy, since it is very frequently the case that the new dynasty has come to power through simple usurpation or violence. The concept of the Mandate of Heaven first appeared after the ShangZhou transition in the twelfth century B.C., since the Zhou kings clearly usurped the throne from its legitimate holder. China subsequently experienced a huge number of changes in dynasty during its more than four millennia of history. Not only were there major dynasties like the Qin, Han, Tang, Song, and Ming, but countless other lesser ones like the Three Dynasties following the fall of the Han, and the Five Dynasties succeeding the Tang. In addition, during the periods when China fell apart into separate regional states, each was ruled by its own dynasty.

There were no social prerequisites for becoming a dynastic founder. Some, like the founders of the Sui and Tang dynasties, were aristocrats and high officials in the previous regime. But others, like Liu Bang who founded the Han Dynasty, or Zhu Yuangzhang, who founded the Ming Dynasty, were commoners. Indeed, the first Ming emperor started out his life as an orphaned son of a peasant who barely survived famine and pestilence as a child, and went on to serve as a novice in a Buddhist monastery. He became a military commander in the Red Turban uprising, a religious movement of peasants, bandits, and adventurers who fought the injustices of local authorities. From there he went on to command ever larger armies of the growing anti-Mongol movement. Late Yuan China had fallen under the control of a series of local warlords, of whom Zhu Yuangzhang was one. Like many other dynastic founders, he was in some sense the warlord who proved to be the smartest and toughest and ended up on top.

Did might then make right for the Chinese? Was the Mandate of Heaven simply an after-the-fact ratification of a power struggle between warlords? To a large extent, it was. Characteristically, there is a large Chinese literature on the subject, such as the essay of Ban Biao from the first century A.D. that explains why certain rulers deserved the Mandate and others did not. But it is very hard to extract from these writings a clear set of principles or procedures for awarding the Mandate that could not be applied afterward to any particular holder of the office who succeeded in coming to power.14 The awarding of the t.i.tle "dynasty" to the rule of a particular leader was often conferred by historians long afterward, legitimating a regime that had been regarded as highly dubious at the time. The historian Frederick Mote points out that there was very little to distinguish the usurpations of Guo Wei, founder of the little-noted Later Zhou Dynasty, and Zhao Kuangyin, who a decade later founded the mighty Song Dynasty. Both came to power as a result of betrayal and deceit; Guo Wei's dynasty folded early only because his son Guo Rong died unexpectedly at the age of thirty-eight. Had the latter lived, Zhao Kuangyin might have gone down in history as an able commander who tried to stage a treasonable putsch. The awarding of the t.i.tle "dynasty" to the rule of a particular leader was often conferred by historians long afterward, legitimating a regime that had been regarded as highly dubious at the time. The historian Frederick Mote points out that there was very little to distinguish the usurpations of Guo Wei, founder of the little-noted Later Zhou Dynasty, and Zhao Kuangyin, who a decade later founded the mighty Song Dynasty. Both came to power as a result of betrayal and deceit; Guo Wei's dynasty folded early only because his son Guo Rong died unexpectedly at the age of thirty-eight. Had the latter lived, Zhao Kuangyin might have gone down in history as an able commander who tried to stage a treasonable putsch.15 But the moral distance between an emperor and a powerful warlord is still enormous. The former is a legitimate ruler whose authority is willingly obeyed; the latter is a violent usurper. The Chinese elites themselves had a sense of which leaders were qualified to hold the Mandate of Heaven and which were not, even if this could not be articulated in a precise set of procedural rules. The Confucian idea of the Rectification of Names meant that emperors had to live up to ideal types of predecessors. They had to possess something like Machiavelli's quality of virtu virtu that characterized the successful prince. A would-be emperor obviously had to be a born leader, someone who could inspire others to follow his authority, and could take risks to achieve his goals. Leadership was most often exercised in the domain of military affairs, which is why so many dynastic founders got their start as military officers. But China prized military prowess to a much lesser degree than did other civilizations. The Confucians very much had in mind an ideal of an educated scholar-bureaucrat and not an uncouth warlord. A pretender who did not exhibit both deference toward Confucian values and a certain subtlety born of education would not attract the support of the various factions around the court. Mote contrasts the Ming founder Zhu Yuangzhang to another warlord-pretender against whom he successfully competed, Zhang Shicheng: that characterized the successful prince. A would-be emperor obviously had to be a born leader, someone who could inspire others to follow his authority, and could take risks to achieve his goals. Leadership was most often exercised in the domain of military affairs, which is why so many dynastic founders got their start as military officers. But China prized military prowess to a much lesser degree than did other civilizations. The Confucians very much had in mind an ideal of an educated scholar-bureaucrat and not an uncouth warlord. A pretender who did not exhibit both deference toward Confucian values and a certain subtlety born of education would not attract the support of the various factions around the court. Mote contrasts the Ming founder Zhu Yuangzhang to another warlord-pretender against whom he successfully competed, Zhang Shicheng:

Zhang Shicheng's liability in the eyes of potential elite advisors and political a.s.sociates was that he was a smuggler and a bandit, a ruffian whose career had given little evidence that he could become more than that ... Zhu Yuangzhang took great pleasure in a literati joke played on Zhang Shicheng by some of his early scholar-advisors. In devising elegant-sounding formal names for Zhang and his brothers, they had given Zhang the name Shicheng, not telling him that in the book of Mencius Mencius there is a well-known line where those two words appear in sequence. With a slight adjustment in the punctuation the line in there is a well-known line where those two words appear in sequence. With a slight adjustment in the punctuation the line in Mencius Mencius can be made to read: "Shicheng is a cad." This ingenious display of contempt toward Zhang Shicheng made Zhu laugh, until he grew suspicious that his literati advisers in all likelihood had similarly ingenious ways of denigrating him. can be made to read: "Shicheng is a cad." This ingenious display of contempt toward Zhang Shicheng made Zhu laugh, until he grew suspicious that his literati advisers in all likelihood had similarly ingenious ways of denigrating him.16

While elites in Chinese society did not vote to ratify a new dynasty, they exercised considerable behind-the-scenes influence in the power struggles between potential rulers. The Mandate of Heaven was not simply something always awarded to the most ruthless and brutal warlord, though such people did periodically come to power in China.

Many would-be dynastic founders, like the empress Wu, went through the rituals required to invest themselves with imperial authority-choosing a temple name for themselves as well as the name of the age that their dynasty was to initiate-but were then quickly deposed. The Chinese system was, however, capable of extraordinary inst.i.tutionalization. Once there was a general consensus within the society that a particular individual held the Mandate of Heaven, the emperor's legitimacy was not generally challenged except under extraordinary circ.u.mstances. In this respect the Chinese political system was far more developed than that of the tribal societies surrounding it.

When an emperor received the Mandate of Heaven, his power was virtually unlimited. And yet, Chinese emperors seldom used their powers to the fullest extent possible. Tyranny was always a possibility, but often not a reality. Why that is so is the subject of the following chapter.

21.

STATIONARY BANDITS.

Whether all states are predatory, and whether the Chinese state in Ming times deserved to be called that; examples of arbitrary rule drawn from later periods in Chinese history; whether good government can be maintained in a state without checks on executive authority

In an influential article, the economist Mancur Olson posited a simple model of political development.1 The world was initially ruled by "roving bandits," like the various warlords of early twentieth-century China, or the ones operating in Afghanistan and Somalia at the beginning of the twenty-first century. These bandits were purely predatory and sought to extract as many resources from the population as possible, often with very short time horizons so they could quickly move on to other victims. At a certain point one bandit would emerge stronger than all the others and come to dominate the society: "These violent entrepreneurs naturally do not call themselves bandits but, on the contrary, give themselves and their descendants exalted t.i.tles. They sometimes even claim to rule by divine right." In other words, the king, who claimed a legitimate t.i.tle to rule, was simply a "stationary bandit" with motives no different from those of the roving bandits he displaced. The stationary bandit realizes, however, that he can become even richer if, instead of going for short-term plunder, he provides stability, order, and other public goods to his society, thereby making it richer and liable to higher taxes in the long run. From the standpoint of the ruled, this represents an advance on the roving bandits. But "exactly the same rational self-interest that makes a roving bandit settle down and provide government for his subjects also makes him extract the maximum possible amount from the society for himself. He will use his monopoly of coercive power to obtain the maximum take in taxes and other exactions." The world was initially ruled by "roving bandits," like the various warlords of early twentieth-century China, or the ones operating in Afghanistan and Somalia at the beginning of the twenty-first century. These bandits were purely predatory and sought to extract as many resources from the population as possible, often with very short time horizons so they could quickly move on to other victims. At a certain point one bandit would emerge stronger than all the others and come to dominate the society: "These violent entrepreneurs naturally do not call themselves bandits but, on the contrary, give themselves and their descendants exalted t.i.tles. They sometimes even claim to rule by divine right." In other words, the king, who claimed a legitimate t.i.tle to rule, was simply a "stationary bandit" with motives no different from those of the roving bandits he displaced. The stationary bandit realizes, however, that he can become even richer if, instead of going for short-term plunder, he provides stability, order, and other public goods to his society, thereby making it richer and liable to higher taxes in the long run. From the standpoint of the ruled, this represents an advance on the roving bandits. But "exactly the same rational self-interest that makes a roving bandit settle down and provide government for his subjects also makes him extract the maximum possible amount from the society for himself. He will use his monopoly of coercive power to obtain the maximum take in taxes and other exactions."

Olson goes on to posit that there is a rate of tax extraction at which the stationary bandit can maximize his revenues, comparable to the monopolist's price in microeconomics. If rates are raised beyond this limit, they undermine incentives to produce, thereby causing total tax revenues to fall. Olson argues that autocratic rulers inevitably set taxes at that maximum rate, but that democratic regimes, because they have to appeal to a "median voter" who bears the brunt of taxation, tax at a lower rate than their autocratic counterparts.

Olson's view of rulers as stationary bandits who extract as much as they can from society in taxes unless somehow politically prevented from doing so is a pleasingly cynical concept of the way that government works. It very much fits with the efforts of economists to extend their model of rational, utility-maximizing behavior into the political realm and to see politics as nothing more than an extension of economics. It also accords nicely with the antistatist traditions of American political culture, which have always regarded both government and taxation with great suspicion. And it provides an elegant predictive model of both political economy and political development, one that has been greatly expanded by other social scientists in recent years.2 The only problem with Olson's theory is that it isn't correct. The rulers of traditional agrarian societies often failed to tax their subjects at anything close to Olson's posited maximizing rate. It is of course extremely difficult to do a retroactive estimate of what a maximal tax rate would have been for incompletely monetized societies with poor historical data on incomes and tax revenues. But we do know that premodern rulers often raised their tax rates substantially to meet specific spending needs like financing wars and lowered them again once the emergency had pa.s.sed. Only at certain points did rulers push their societies toward a counterproductive breaking point, and this usually occurred in response to desperate conditions at the end of a dynasty. During normal times they must have been taxing their societies at levels well below the maximum.

There is no better ill.u.s.tration of the inadequacies of Olson's model than China during the Ming Dynasty, where there is a broad scholarly consensus that tax rates were set far below their theoretical maximum, and indeed far below a level that was necessary to provide the minimal public goods, particularly defense, that were needed to keep the society viable. What was true for Ming China was true for other agrarian societies as well, like the Ottoman Empire and the various monarchies in Europe, and provides the components of an alternative theory as to why these traditional regimes seldom taxed subjects at maximal rates.3 It was not only on matters of taxation that emperors did not use their powers to the degree theoretically possible. Despotism of Empress Wu's sort was a periodic but not a continual phenomenon. Many Chinese rulers exhibited what might charitably be labeled leniency or forbearance toward their subjects, or what a Confucian would call "benevolence." China has had a long history of tax protests, and a strong Confucian tradition maintained that high taxes represented a moral failing of the state. The Shi Jing Shi Jing, or Book of Odes Book of Odes, contains the following poem:

Big rat, big rat, don't eat my millet!

Three years I've served you but you won't care for me.

I'm going to leave you and go to that happy land, happy land, happy land where I'll find my place.4

Whatever the constraints on the power of the Chinese emperor in Ming times, they were not not based on law. As we have seen in the case of the empress Wu, Chinese rulers, unlike their European counterparts, did not have to seek permission from sovereign courts or parliaments in order to raise taxes. Not only could they arbitrarily set tax rates through simple executive order, they could also confiscate property at will. Unlike the "absolutist" monarchs of early modern France and Spain who had to proceed very gingerly when confronting powerful elites (see chapters based on law. As we have seen in the case of the empress Wu, Chinese rulers, unlike their European counterparts, did not have to seek permission from sovereign courts or parliaments in order to raise taxes. Not only could they arbitrarily set tax rates through simple executive order, they could also confiscate property at will. Unlike the "absolutist" monarchs of early modern France and Spain who had to proceed very gingerly when confronting powerful elites (see chapters 23 23 and and 24 24), the first Ming emperor, Taizu, simply confiscated the lands of the largest landowners in the realm. He was said to have liquidated "countless" affluent households, particularly in the Yangtze delta, where he believed he faced particularly strong opposition.5 The real constraints on Chinese power were different, and were of three basic sorts. The first was a simple lack of incentives to create the administrative capacity to carry out orders and in particular to extract a high level of taxes. China was already a huge country at the beginning of the Ming Dynasty, with a population of more than 60 million in 1368 that grew to 138 million by the seventeenth century.6 The challenges of collecting taxes over so vast a territory were daunting. In the fourteenth century there was very little money in circulation, so the basic agricultural tax that was supposedly leveled on every inhabitant of China was collected in kind. The challenges of collecting taxes over so vast a territory were daunting. In the fourteenth century there was very little money in circulation, so the basic agricultural tax that was supposedly leveled on every inhabitant of China was collected in kind.7 In-kind payments were usually made in grain, but they could take the form of silk, cotton, timber, or other goods. There was no consolidated monetary system for recording these payments or converting them into a common unit of measure. Many payments were consumed (that is, "budgeted") locally; others had to be physically shipped to granaries at successively higher levels of administration and ultimately to the capital (first at Nanjing, and later in Beijing). Taxpayers were charged the costs of shipping their taxes to the government, a surcharge that often exceeded the value of the underlying goods. There were no clear distinctions between local and central revenues and budgeting. One scholar has compared the system to an old-fashioned telephone switchboard, in which wires would come out of different holes and into others in a system of confusing spaghetti-bowl-like complexity. In-kind payments were usually made in grain, but they could take the form of silk, cotton, timber, or other goods. There was no consolidated monetary system for recording these payments or converting them into a common unit of measure. Many payments were consumed (that is, "budgeted") locally; others had to be physically shipped to granaries at successively higher levels of administration and ultimately to the capital (first at Nanjing, and later in Beijing). Taxpayers were charged the costs of shipping their taxes to the government, a surcharge that often exceeded the value of the underlying goods. There were no clear distinctions between local and central revenues and budgeting. One scholar has compared the system to an old-fashioned telephone switchboard, in which wires would come out of different holes and into others in a system of confusing spaghetti-bowl-like complexity.8 The ministry of revenue was so understaffed that it was unable to control or even understand this system. The cadastral surveys that were supposed to be the basis of the land tax were incompletely performed early on in the dynasty, and not updated, so that with subsequent population growth, changes in ownership, or even physical geography (flooding or the reclaiming of land), the basic population registers soon became hopelessly out of date. The Chinese, like other peoples, were extremely good at hiding a.s.sets from the tax collector and engaging in schemes to in effect launder income. The ministry of revenue was so understaffed that it was unable to control or even understand this system. The cadastral surveys that were supposed to be the basis of the land tax were incompletely performed early on in the dynasty, and not updated, so that with subsequent population growth, changes in ownership, or even physical geography (flooding or the reclaiming of land), the basic population registers soon became hopelessly out of date. The Chinese, like other peoples, were extremely good at hiding a.s.sets from the tax collector and engaging in schemes to in effect launder income.9 The draconian powers of taxation and confiscation held by the emperor also tended to be a wasting commodity. It could be used early on in a dynasty when the emperor was consolidating power and settling scores with former opponents. But as time went on, the palace found it often needed the cooperation of those same elites and dramatically reduced tax rates in the areas it had earlier confiscated property.

Lack of administrative capacity limited tax revenues not only on the supply side; there were also limitations to the amounts of revenues demanded by different emperors. Olson's a.s.sumption that any ruler would want to maximize revenues reflects the common a.s.sumption of modern economics that maximization is a universal characteristic of human behavior. But this is an anachronistic projection of modern values backward onto a society that didn't necessarily share them. The first Ming emperor, Taizu, was an austere autocrat who cut the size of the central government and avoided foreign wars; his granaries actually ran surpluses. This was not true of his successor, Chengzu (13601424), who launched an ambitious program of ca.n.a.l construction and palace building. Chengzu was also the emperor who funded the voyages of the eunuch naval commander Zheng He (13711435), who sailed a fleet of giant ships as far as Africa and possibly beyond. Expenditures ran at two to three times the level of the first Ming emperor. Surtaxes and labor requisitions were raised accordingly, which led to tax revolts and discontent throughout the empire. As a result, the third emperor and his successors lowered tax rates to a level closer to those of the first emperor and made other political concessions to an offended gentry cla.s.s. 10 10 For much of the dynasty, the land tax was set at a low 5 percent of total yield, a figure significantly lower than those of other agrarian societies. For much of the dynasty, the land tax was set at a low 5 percent of total yield, a figure significantly lower than those of other agrarian societies.11 Chinese monarchs, no less than rulers of other premodern societies, often exhibited what the economist Herbert Simon has labeled "satisficing" rather than maximizing behavior.12 That is, in the absence of an urgent need for revenue, such as a war, they were often content to let sleeping dogs lie and collect only the amount of revenues required for their regular needs. That is, in the absence of an urgent need for revenue, such as a war, they were often content to let sleeping dogs lie and collect only the amount of revenues required for their regular needs.13 A truly determined emperor could decide to behave like a maximizer, and some, like Chengzu, did, but the idea that all autocratic political leaders automatically maximize is manifestly not true. A truly determined emperor could decide to behave like a maximizer, and some, like Chengzu, did, but the idea that all autocratic political leaders automatically maximize is manifestly not true.

A third limitation on the power of Chinese emperors operated in domains well beyond taxation and fiscal policy, which was the need for delegation. All large organizations, whether governments or private corporations, have to delegate authority, and when they do, the "leader" sitting at the top of the administrative hierarchy loses an important degree of control over the organization. The delegation can be to functional specialists like budgeting officers or military logisticians, or it can be regional, to a cadre of provincial, prefectural, munic.i.p.al, and local authorities. These delegations are necessary because no ruler can ever have enough time or knowledge to make all of the important decisions in his realm.

But with the delegation of authority goes power. The agents to whom power has been delegated have authority over the delegator in the form of knowledge. This can be either the technical knowledge that goes with the running of a specialized ministry or agency or the local knowledge of particular conditions existing in a certain region. It is for this reason that organizational specialists like Herbert Simon have argued that authority in any large bureaucracy does not flow only from the top to the bottom, but oftentimes in a reverse direction as well.14 Chinese emperors experienced this problem much as modern presidents and prime ministers do, in the form of unresponsive and sometimes outright rebellious bureaucracy. Ministers objected to policies proposed by their boss, or quietly failed to implement them. Of course, Chinese rulers had certain tools that modern executives don't: they could administer vicious floggings on the bare b.u.t.tocks of even their most senior ministers, or casually imprison or execute them.15 But this kind of coercive solution to the princ.i.p.al-agent problem didn't solve the underlying issue of information. Bureaucrats often didn't carry out the wishes of their leader because they had better knowledge of the real conditions of the empire-and could hide their activity from him. But this kind of coercive solution to the princ.i.p.al-agent problem didn't solve the underlying issue of information. Bureaucrats often didn't carry out the wishes of their leader because they had better knowledge of the real conditions of the empire-and could hide their activity from him.

A large country like China had to be governed by delegation to local authorities, but then these local authorities would commit abuses, become corrupt, or even conspire against the central government. The normal administrative hierarchy was not adequate to deal with this problem, because while orders flowed downward, information did not necessarily flow back up. The most dictatorial emperor would not be able to discipline a wayward official if he didn't know an abuse was occurring.

This limitation of princely power was discussed in premodern China under the heading of the relative merits of the "feudal" versus the "prefectural" forms of administration. Feudal ( fengjian fengjian) in this sense carries none of the complex connotations of European feudalism; it simply means that authority was decentralized, compared to the prefectural system, where local officials were agents of the center. According to the Ming scholar Gu Yanwu (16131682),

The fault of feudalism was its concentration of power on the local level, while the fault of the prefectural system is its concentration of power at the top. The sage rulers of antiquity were impartial and public-minded in their treatment of all men, parceling out land to them and dividing up their domains. But now the ruler considers all the territory within the four seas to be his own prefecture, and is still unsatisfied. He suspects every person, he handles every affair that comes up, so that each day the directives and official doc.u.ments pile higher than the day before. On top of this, he sets up supervisors, provincial governors and governors-general, supposing that in this way he can keep the local officials from tyrannizing over and harming the people. He is unaware that these officials in charge are concerned only in moving with utmost caution so as to stay out of trouble until they have the good fortune to be relieved of their posts, and are quite unwilling to undertake anything of profit to the people.16

The typical solution that Chinese rulers devised to get around the problem of unresponsive administrative hierarchies was to superimpose on them a parallel network of spies and informants who were completely outside the formal governmental system. This explains the great role played by eunuchs. Unlike normal bureaucrats, eunuchs had direct access to the imperial household and often came to be trusted to a far greater degree than the regular administrators. The palace therefore sent them out on missions to spy on and discipline the regular hierarchy. By the end of the Ming Dynasty, there were an estimated one hundred thousand eunuchs a.s.sociated with the palace.17 From 1420 on they were organized into an Orwellian secret police organization known as the Eastern Depot, under the direction of the eunuch director of ceremonial, which became "an organ of totalitarian terrorism" in the later years of the dynasty. From 1420 on they were organized into an Orwellian secret police organization known as the Eastern Depot, under the direction of the eunuch director of ceremonial, which became "an organ of totalitarian terrorism" in the later years of the dynasty.18 But the emperor found he could not control the eunuchs themselves, who made their own policy, staged coups, and conspired against him despite the existence of a "Eunuch Rectification Office." But the emperor found he could not control the eunuchs themselves, who made their own policy, staged coups, and conspired against him despite the existence of a "Eunuch Rectification Office."19 The political system didn't have any downward mechanisms of political accountability-that is, there were no local elections or independent media to keep officials honest. As a consequence, the emperor had to pile one centralized system of top-down control on top of another. Even so, he was not able to achieve a strong degree of control over his realm. The political system didn't have any downward mechanisms of political accountability-that is, there were no local elections or independent media to keep officials honest. As a consequence, the emperor had to pile one centralized system of top-down control on top of another. Even so, he was not able to achieve a strong degree of control over his realm.

The Ming Dynasty's unwillingness and inability to extract the taxes it needed led ultimately to its collapse. Whereas China had been largely free of foreign threats for the first two centuries of Ming rule, the security situation began to deteriorate sharply toward the end of the sixteenth century. j.a.panese pirates began to raid the wealthy southeastern seaboard, and the shogun Toyotomi Hideyoshi invaded Korea in 1592. The same year a war started in Inner Mongolia, and there were uprisings by aboriginal peoples in the south. The most serious development of all was that the Manchu to the north were becoming stronger and better organized, and making incursions along the northeastern frontier.

The government's response to the crisis was completely f.e.c.kless. In the face of rising expenses, it depleted its silver reserves but refused to raise taxes on the gentry cla.s.s until too late. Acc.u.mulated tax delinquencies continued to rise through the first decades of the seventeenth century as the military threat became more intense. The emperor even declared a number of tax amnesties, apparently in recognition of the fact that the state had no chance of collecting back taxes. The soldiers on the frontier, who had formerly been organized into self-sufficient military colonies, could no longer support themselves but became dependent on payments from the central government that had to be delivered over long supply lines. The regime failed to organize an adequate logistics system and thus failed to pay its soldiers on time. The dynasty stumbled on until 1644, when the government in Beijing was weakened by the Han Chinese rebel Li Zicheng, and then finally fell to a Manchu army from the north working together with disgruntled remnants of the Ming army.

GOOD GOVERNMENT, BAD GOVERNMENT.

The Ming Dynasty was the last fully indigenous regime to rule China until the twentieth century, in which the traditional Chinese political system developed to its greatest extent. It was characterized by inst.i.tutions that in retrospect were amazingly modern and effective, and others that were unbelievably backward and dysfunctional.

In the first category was the recruitment system into the imperial bureaucracy. The examination system's roots went all the way back to the Han Dynasty, but throughout the Sui, Tang, and early Song dynasties entry into the bureaucracy tended to be controlled by a small circle of elite families. It was only during the Ming Dynasty that the examination system became the main avenue for entry into government and acquired a level of prestige and autonomy that made it a model for all subsequent exam systems.

The examination system was linked to a much broader educational establishment. There was a network of Confucian schools all over the country, to which ambitious parents could send their children. The best students were recommended by their teachers to go on to the national universities in Beijing and Nanjing, where they would prepare to take the civil service exams. (Teachers who recommended students who failed to perform well were punished, something that modern universities might consider as a means of combating grade inflation.) It was still possible for elite families to get their children placed within the system through a category known as "students by purchase." But these Chinese precursors of contemporary legacy admittees to Harvard or Yale (that is, children of wealthy alumni) seldom made it to the highest reaches of the bureaucracy, which remained heavily meritocratic.20 The highest possible honor was to place first in the three successive levels of exams: provincial, metropolitan, and palace. Only one individual, Shang Lu, managed to achieve this in the entire history of the dynasty; he went on to reach the very top of the hierarchy as senior grand secretary in the late fifteenth century. The highest possible honor was to place first in the three successive levels of exams: provincial, metropolitan, and palace. Only one individual, Shang Lu, managed to achieve this in the entire history of the dynasty; he went on to reach the very top of the hierarchy as senior grand secretary in the late fifteenth century.21 The Chinese bureaucracy established a model that would eventually be replicated by virtually all modern bureaucracies. There was a centralized system of appointment and promotion, based on ranks from 1 at the top to 9 at the bottom (much like the General Service schedule in the American bureaucracy). Each of these ranks was divided into an upper and lower section, so one would expect a promotion from, say, rank 6a to 5b. Officials making it through the examination system were appointed to low-ranking offices in various parts of the country, always in a region different from the one in which he grew up. If relatives happened to be a.s.signed to the same office, the junior one usually had to withdraw. After three years, a bureaucrat was rated by the head of his agency, who pa.s.sed the evaluation on to the central personnel office. Lateral entry into the bureaucracy was discouraged. The officials who survived this system and were promoted to the top of the hierarchy tended to be extraordinarily well qualified.22 These highly qualified and well-organized bureaucrats served, however, at the whim of an autocrat who was not himself rule bound in any way and could with the stroke of a pen undermine carefully formulated policies. They were subject to capricious punishments and purges by the sovereign, and only a minority of senior bureaucrats succeeded in finishing their terms without being humiliated in one way or another. Some of the worst decisions were those made by the first Ming emperor, Taizu, who, growing suspicious of his own grand counselor, not only abolished the office but also forbade any of his successors to reestablish the office on pain of death. This meant that no succeeding Ming emperor was allowed to have the equivalent of a prime minister but instead had to deal directly with the tens of ministries and agencies that did the actual work of government. This system was barely workable for an extremely energetic and detail-oriented emperor like Taizu, and a disaster for subsequent rulers of lesser capabilities. In one ten-day period, Taizu had to respond to 1,660 different official doc.u.ments dealing with 3,391 separate matters.23 One can imagine what his successors thought about the work load he imposed on them. One can imagine what his successors thought about the work load he imposed on them.

Many later emperors were not up to snuff. By tradition, one of the worst was the emperor Shenzong (otherwise known as the Wanli emperor), whose long rule between 1572 and 1620 corresponded to the period of the dynasty's decline.24 In the second half of his rule, he refused to see ministers or preside over the court. He allowed thousands of reports and memoranda to pile up in his office, unread and unanswered. Indeed, he failed to come out of his palace at all for years at a time, during which important governmental decisions simply failed to be made. He was also extremely greedy, raiding the state treasury to meet personal expenses like building a magnificent tomb. At the time of the early seventeenth-century military crisis, when state reserves were reduced to some 270,000 taels of silver, the emperor had acc.u.mulated more than two million taels in his personal account. Despite repeated requests from the minister of revenue, he refused to release more than nominal amounts of funds to the government for purposes like paying the troops. In the second half of his rule, he refused to see ministers or preside over the court. He allowed thousands of reports and memoranda to pile up in his office, unread and unanswered. Indeed, he failed to come out of his palace at all for years at a time, during which important governmental decisions simply failed to be made. He was also extremely greedy, raiding the state treasury to meet personal expenses like building a magnificent tomb. At the time of the early seventeenth-century military crisis, when state reserves were reduced to some 270,000 taels of silver, the emperor had acc.u.mulated more than two million taels in his personal account. Despite repeated requests from the minister of revenue, he refused to release more than nominal amounts of funds to the government for purposes like paying the troops.25 His actions led directly to the growth of Manchu power, which would ultimately destroy the dynasty. His actions led directly to the growth of Manchu power, which would ultimately destroy the dynasty.

THE "BAD EMPEROR" PROBLEM Of the three components of political development that we have been following-state building, rule of law, and accountability-the Chinese got the first right at a very early point in their history. In a sense, they invented good government. They were the first to design an administrative system that was rational, functionally organized, and based on impersonal criteria for recruitment and promotion. Perhaps because Chinese society is so familistic, Chinese state builders saw their particular task as freeing the government from patrimonial or nepotistic influences that were the source of enormous corruption.

Creating such a system in the cauldron of the Warring States period was one thing; keeping it going over the next two millennia was another. The modernity of the bureaucracy, achieved early on, fell victim to decay and repatrimonialization as the state fell apart and was appropriated by wealthy aristocratic families. Decay of the state took place over many centuries, and restoration of the bureaucracy to something like the design originally intended by its Qin and Han creators also took centuries to accomplish. By the time of the Ming Dynasty, the cla.s.sical system had been perfected in many ways. It was more meritocratic and exercised control over a society that was far larger and more complex than the one that had existed in Han times.

In other respects, however, the Chinese political system was underdeveloped. It never generated a rule of law or mechanisms of political accountability. Society outside the state continued as before to be far less organized for political action than its counterparts in Europe or India. There was no landed, independent aristocracy, and no independent cities. The dispersed gentry and peasantry could pa.s.sively resist the government's orders, and periodically broke out into violent uprisings that were suppressed with great savagery. But it was never able to inst.i.tutionalize itself as a corporate group to demand rights from the state, as the peasantry in Scandinavia was to do. Independent religious orders had sprung up during the Sui and Tang dynasties, with the spread of Buddhism and Daoism. At different times in Chinese history, these religious orders acted to oppose the state, from the Red Turbans to the Taiping rebels. But religion continued to be a sectarian phenomenon that was viewed with suspicion by the orthodox Confucian authorities and never represented a powerful social consensus that could limit the state's power through its custodianship of law.

One of dynastic China's great legacies, then, is high-quality authoritarian government. It is no accident that virtually all of the world's successful authoritarian modernizers, including South Korea, Taiwan, Singapore, and modern China itself, are East Asian countries sharing a common Chinese cultural heritage. It is very hard to find authoritarian rulers with qualities like those of Lee Kuan Yew of Singapore or Park Chung Hee of South Korea in Africa, Latin America, or the Middle East.

But the experience of the Ming Dynasty, as well as other periods of Chinese history, raises troubling questions about the durability of good governance under conditions where there is no rule of law or accountability. Under the leadership of a strong and capable emperor, the system could be incredibly efficient and decisive. But under capricious or incompetent sovereigns, the enormous powers granted them often undermined the effectiveness of the administrative system. The empress Wu purged the bureaucracy and packed it with her own unqualified supporters; the emperor Taizu abolished the prime ministership and locked his successors into this awkward system; the emperor Shenzong ignored the bureaucracy altogether and government collapsed. The Chinese recognize this as the problem of the "bad emperor."

There was a form of accountability in the Chinese system. Emperors we