Soulstealers: The Chinese Sorcery Scare of 1768

by Philip A. Kuhn

  The Temple of Mercy in the lush silk country of Te-ch'ing sheltered a small community of very poor monks. The "incense fire was dim," it was said, meaning that the temple was little frequented by pilgrims or by devotees wanting masses celebrated for the dead, and so got few donations.43 Nearer the city, on Chien-yuan Mountain, was a prosperous temple: a Kuan-yin Hall, much favored by the local devout. Early in the spring of 1768, jealousy and privation had led the poor monks at the Temple of Mercy into uncharitable thoughts. A friend of the monks, an ingenious layman named Hsu, suggested that current popular fears about soulstealing might be turned to advantage. All the monks had to do was spread a rumor that masons were practicing sorcery in the vicinity of the prosperous temple, which would pollute the power of that establishment to bring blessings to worshipers.

  The story material was ready to hand. Down at the walled city, mason Wu and his men were already at work rebuilding the watergate. As it happened, Wu had won the contract over a crew of rival masons from another county. Might not the disappointed bidders try (as masons would) to harm their competitors by sorcery? Local folk believed that death pollution could be inflicted on an enemy by "burying death-magic" in his path.44 A slain rooster buried under a footpath would emit quite enough death pollution to do the job. All the monks had to do, suggested layman Hsu, was to spread the word that feuding masons had "buried death-magic" under paths leading to the Kuan-yin Hall. One of the monks who "knew a bit of writing" drew up the necessary posters: We have heard (he wrote) that last month a mason "buried death-magic" near Chien-yuan Mountain, and that persons passing over the spot on the way to the Kuan-yin Hall might be infected by it. Just now, when many people are going there to worship, we fear that they may come to harm. The Temple of Mercy is one of those long known as "pure gates of Buddha." It can "help one to approach good fortune and avoid calamity." Layman Hsu spread this intelligence about the county and received 500 copper cash for his trouble.

  With this story on their desks, the inquisitors asked mason Wu about the reported plot by rival masons to injure them by sorcery. This looked like a plausible origin for the spring panic in the Hangchow region: another frame-up, another attempt to injure rivals by an imputation of sorcery. But mason Wu, in his down-to-earth way, offered no helpful details.

  Mason Wu: Last year, masons from Hai-ning named Cheng Yuan-ch'en and Mao T'ien-ch'eng came to Te-ch'ing to contract for the bridge job. They couldn't agree on a price, so they left.

  Inquisitor: Was there any talk of "burying death-magic"?

  Mason Wu: This year we haven't laid eyes on Cheng or Mao and haven't heard that they bore us any grievance.'''

  Did this incident start the panic by fanning public fears of masonsorcerers? The grand councillors could not be sure. Yet the Temple of Mercy incident confirmed, for the skeptical among them, that "soulstealing" was a phantom conceived in ignorance and nourished in envy. Here was another case of cynical men manipulating popular fears for private ends. Mason Wu, at any rate, could not be held responsible. Along with Chu-ch'eng and the other monks, beggar Chi, and peasant Shen, he was to be escorted home and released. Here our original subjects, both victims and tormentors, drop gratefully from the historical record.

  CHAPTER 9

  Political Crime and

  Bureaucratic Monarchy

  We now have read several stories: about sorcery panic spreading among the common people; about a monarch becoming convinced that sorcery is a mask for sedition; about agnostic bureaucrats struggling to cope with demands from both sides but failing to satisfy either. These stories are layered one upon the another, several texts written on a single historical page. Beneath them lies another story, the hardest to read: how local events-including the sorcery scare-served as fuel for running the political system.'

  Sorcery played its part in the political system as the kind of event I shall call "political crime." Political crime included sedition in all its various guises, whether religious heterodoxy, literary innuendo, or outright revolt. Because it threatened the foundations of the system itself, political crime was considered distinct from the ever-present corruption, which merely reduced the system's efficiency. But if this were the case, why were not the bureaucrats as concerned about it as the monarch? It was, after all, their system too. The answer must lie at the core of bureaucratic monarchy itself, at least as we see it in the Chinese case. Documents from the sorcery crisis suggest why political crime was a monarch's issue and not a bureaucrat's issue. The heart of the problem was the relationship between routine and arbitrary power.

  Routine and Arbitrary Power in the Bureaucratic Monarchy

  Study of the Chinese political system under the late empires has produced two largely distinct literatures: on the structure, personnel, and values of the administrative bureaucracy;2 and on the development of the imperial institution, particularly the imperial communication system.' As a result, we now have a more sophisticated view of officialdom as a way of life; and a view of the ruler that makes him part of a political system, rather than a remote and all-powerful despot. I wonder, however, whether we have yet discovered how arbitrary power interacts with bureaucratic routine over a long period within a single system.`' We still tend to assume that the two are inversely related: the more of one, the less of the other; as one grows, the other shrinks. The tendency of social analysis since Max Weber is in fact to show that, in the long run, autocrats yield to bureaucrats. Yet I believe that arbitrary and routine authority may not have been incompatible in the Chinese system, and may indeed have found ways to live side by side.

  In his celebrated description of the Chinese polity, Max Weber actually avoids confronting the issue of how arbitrary and routine power interact. Instead, he characterizes the Chinese monarchy as incompletely centralized, and its operational norms as uncodified. The limitations of his data shielded him from a view of either arbitrary power or codified routine. The emperor himself is a shadowy figure in Weber's treatment of Chinese bureaucracy. Under the "average ruler," authority was not "centralized. 115 Weber presumably believed, however, that Chinese bureaucracy would be powerless when faced with a nonaverage ruler because it lacked specialization (only modern "bureaucratic experts" can compete effectively with the "absolute monarch," whom they can dismiss as a "dilettante").6 Although he uses the term "bureaucracy" in referring to the Chinese system, Weber actually includes that system not under "Bureaucracy," a subject heading he reserves for the specifically "modern" type, but rather under "Patriarchal and Patrimonial Domination."' Just as shadowy is Weber's notion of the codified routine through which the Chinese bureaucracy was disciplined and controlled. Though the "patriarchal" monarchy was able to achieve an "authoritarian and internalized bondage" of the officials by transferring them frequently and thereby keeping them from forming regional power-bases,8 the "patriarchal character of the political association ... was opposed to any development of formal law."" "Formal law," for Weber, must have included administrative codes by which the bureaucracy itself might be regulated. Although for these reasons Weber could not pose the problem sharply in the context of the Chinese state, his historical logic suggests that he saw arbitrary and routine power as incompatible. History tends to replace the former with the latter through routinization and rationalization.10

  In his classic treatise on the evolution of the Prussian state, Hans Rosenberg distinguishes between "dynastic absolutism" and "bureaucratic absolutism." By "absolutism," Rosenberg means power essentially unchecked by constitutional limits or by compromise with influential social strata. By "dynastic," he emphasizes the dominance of the monarch himself ("a royal bully," as he describes Frederick William I) over society at large, as well as over the corps of "royal servants" recruited to carry out his orders. This system Rosenberg also characterizes as an "experiment in royal monocracy." Although he does not describe in detail the interaction between the "monocrat" and his bureaucracy, the implication is that the "hideous spirit of fearful obedience to authority" that infused Prus
sian society at large was a projection of the bureaucracy's own state of mind." Nevertheless, Rosenberg asserts that even under the early Hohenzollerns, royal control relied upon minutely regularized procedures: the "public law" that governed the bureaucracy as well as the populace.12 We are left uncertain about how "monocracy" or dynastic absolutism preserved its freedom of action within a system of regulations that were designed to reduce the operation of government to a finely tuned routine.

  If there was a purely "arbitrary" component to this system, it was unstable and short-lived. It fell victim to "an unremitting struggle for replacing arbitrary royal powers ... with general legal rules." Even under Frederick the Great, monocratic power was frustrated by officials who had "real power to obstruct and divert" by manipulating information and other acts of bureaucratic "sabotage." Under Frederick's weaker successors, the bureaucracy succeeded in securing itself against arbitrary sanctions by introducing life tenure and due process into the bureaucratic personnel system. The result, as Rosenberg describes post-Napoleonic Prussia, was it state ruled by career bureaucrats ("bureaucratic absolutism"); the monarch was simply the "top functionary."13 Here arbitrary and routine power were subject to a historical process that weakened the one to the advantage of the other-a process like Weber's "routinization" and "rationalization." In Rosenberg's Prussian case, arbitrary and routine power could not long coexist.

  "Bureaucratic monarchy" reads like an oxymoron. To the extent that it is "bureaucratic," what scope is left for the monarch? To the extent that it is monarchic, how can one man's autocratic power coexist with a system of universal rules? Both monarch and bureaucrat were caught in this dilemma; both were ambivalent toward formal administrative procedures. The monarch had to regulate his thousands of bureaucratic servants by written codes, to ensure that everyone stuck to the administrative procedures that underlay his own wealth and security. At the same time, he was naturally concerned to maintain his own distinctive position, his extra-bureaucratic power and autonomy. Consequently he had to struggle unceasingly to avoid becoming bureaucratized himself. Much of the normal business of government involved him in sanctioning decrees drafted for him by the Grand Secretariat, or in ratifying appointments of candidates presented to him by the Board of Civil Office. Faced by his document-drafters with a narrow range of choice, the busy monarch found himself "functioning" as a cog (albeit a bejeweled one) in a document-processing machine. How was he to break out of this trap and assert his position as master, not functionaryj14

  The bureaucratic official, for his part, was bedeviled by minute regulations on the form, timing, and routing of paperwork, fiscal and judicial deadlines, and the relations of superiors to subordinates. To break any of these regulations exposed him to impeachment, fines, transfer, or dismissal. Yet these onerous regulations at least drew certain boundaries around his responsibilities and offered him some protection from arbitrary demands by superiors and even by the monarch himself.''

  The Monarch's Control of Bureaucrats

  Rules yield predictability and standardization. They also limit the freedom of the one who applies them. In this sense they are a great leveler of status: those who apply and monitor the rules may become as entangled by them as those who are subject to them. The Ch'ing autocrats accordingly had to pick their way carefully between routine and arbitrary models of command. When rules were ineffective, the remedies included not only more rules but also procedures that rested upon arbitrary power. From early in his reign, Hungli was impatient with rules that did not work. His remedies included both tightening the screws of the routine bureaucratic machine and finding ways to inject his own arbitrary power into it. How he did this can be seen most readily in his efforts to evaluate his officials.

  Surveillance of Efficiency and Conduct

  At the heart of monarchic control lay the evaluation of officials: estimating their qualifications for appointment, surveying their conduct in office, and periodically evaluating their fitness for service. The history of Hungli's reign suggests how hard it is to force a bureaucracy to discipline itself. His despair at the system he inherited led him to seek alternative means of control.

  The essence of the official control system was the distinction between crime and administrative failure. Criminal penalties, for corruption or worse crimes, were handled by the Board of Punishments after the culprit had been impeached and removed from office. Administrative sanctions (ch'u fen) were handled by the Board of Civil Office. These penalties, which involved demotion in rank, transfer to a less desirable post, and monetary fines, covered a broad range of misdeeds, of which most were failures to meet deadlines or quotas (for solving criminal cases or collecting taxes), concealment of information, or other breaches of standard operating procedure. No official dossier was without its record of ch'u fen offenses. Here are some examples of typical offenses and their penalties, drawn from the 1749 edition of the Regulations of the Board of Civil Office, Administrative Sanctions:

  An official who fails to report the fact of a grain-transport boat's sinking: to he reduced one grade and transferred.

  If an official supervising the collection of the land tax falls short [of the quota] by an amount less than one-tenth, he is to be blocked from promotion and fined a year's [nominal] salary. If he is short a tenth or more, he is to be reduced in rank by one grade ... and if he is short five-tenths or more he is to be dismissed from office.

  If 'a local official, fearing to be disciplined for laxity in arresting criminals, under some pretext intimidates a plaintiff and forces him to avoid using the word "robbery" and not report it as such, . . . he is to be removed from office.'s

  Although Chinese government has long included special organs to investigate and impeach officials for incompetence and wrongdoing, their history since medieval times has been one of decline. The branch of government generally called "the Censorate" (under the Ch'ing, to-ch'a-yuan) historically had duties of both remonstrating with the emperor about his conduct and keeping an eye on the bureaucracy. At least as early as the seventh century A.D., "remonstrance" upward was secondary to surveillance downward. But over time even the independent surveillance function was eroded. The Manchu conquerors inherited from their Ming predecessors a Censorate that had largely lost its ability to supervise field administration. "Surveillance offices" (an-ch'a-ssu) in the provinces had, by the late sixteenth century, already assumed the regular judicial work of provincial government. The Manchus completed their incorporation into the provincial bureaucracy, and we now refer to these officials as "provincial judges."17 Although there were censorial offices in the capital to check on the work of metropolitan officials, they were largely engaged in combing documents for irregularities. And although there were "provincial censors" charged with overseeing provincial administration, these men were actually stationed in Peking, which meant that the "eyes and ears" of the sovereign were considerably dimmed outside the capital. Accordingly, the job of surveillance in both capital and provinces mainly fell to line bureaucrats, each of whom was responsible for watching the conduct of his subordinates. To symbolize how administration and surveillance were melded, a provincial governor bore the brevet title of vice-president of the Censorate, to indicate his special responsibility to scrutinize the conduct of his subordinates. In effect, the bureaucracy was really watching itself. 18

  This kind of in-house bureaucratic surveillance followed two modes: ad hoc impeachment (for both incompetence and criminality), and periodic evaluation leading to triennial fitness reports for all officials, reports that also served as the basis for impeaching substandard officials. In both these modes, the process relied largely on the work of line bureaucrats and rather little upon the Censorate. Of 5,151 impeachment cases in the Ch'ien-lung reign, less than 8 percent were initiated by the Censorate, with the rest by line officials in Peking or the provinces.19 Though Hungli believed that both modes worked badly, he identified the problem most clearly in the triennial fitness reports.

  The Triennial Evalua
tions

  Periodic evaluation of officials has a history as long as that of Chinese government.20 The Manchus inherited the system from the Ming and had installed it even before the conquest.21 By the mid-eighteenth century the basic elements of the evaluation for civil officials22 were the Capital Investigation (ching-ch'a) which included all Peking officials except those of the three highest ranks, and the Grand Accounting (ta-chi), which included provincial officials except for governors-general, governors, and provincial treasurers and judges.

  For both the capital and provincial systems, the cumbersome procedure was that every year an official would be rated (k'ao-ch'eng) by his superior officer. These ratings served as raw material for the triennial evaluations. In the capital, the triennial registers would be aggregated by the heads of the Six Boards, and in the provinces by the governors. The registers (bound traditionally in imperial yellow) were then forwarded to a review commission consisting of officials from the Board of Civil Office and the Censorate, along with one Han and one Manchu grand secretary. The commission would then review the "yellow registers" and decide who should be promoted, demoted, or retained in office. The cases of men due for promotion or demotion would then be the subjects of separate memorials to the Throne from the Board of Civil Office. Men whom the Throne approved for promotion as "outstanding" (cho-i) still had to be recommended in separate memorials by their superiors. Strict accountability applied in these cases of promotion for merit. In the case of lower-level officials, recommendations had to note whether there were any outstanding treasury shortages or unresolved court cases that might block promotion. If any were subsequently found after promotion, the recommender himself would be punished by demotion and transfer.