Chapter 1. The Economics of Huichang: Suppression and Selection¶

The preceding essay ended with China's reconstruction complete, awaiting a test imposed by the state. The test duly arrived in Huichang 5 (845). Essay 8, Chapter 3 established a comparison: China was a replication outside India of the group “struck by a blow, but possessing roots.” This chapter is the record of that replication. We first ask why the blow came, then what it destroyed, and finally what it failed to destroy. Once these three questions are answered, the judgment “roots explain death” will have evidence from the Chinese side.

Huichang was not the first. Emperor Taiwu of the Northern Wei destroyed Buddhism in 446, citing weapons and underground stores discovered in a monastery in Chang'an. When Emperor Wu of the Northern Zhou abolished Buddhism and Daoism in 574, his advisers laid their calculations on the table: seek soldiers among the monks, take land from beneath the monasteries and pagodas. Those returned to lay life numbered in the hundreds of thousands; apologetic writings inflated the figure to three million, to be read, as usual, at half face value.¹ The two cases share a feature: in the decrees, doctrine was only a pretext; what they actually targeted was resources—people, land, iron, copper. Each suppression of Buddhism in China began because it had accumulated too much, not because it had taught something wrong. This establishes the first contrast with its death in India. In India, no one signed a document ordering Buddhism's death. In China, every suppression came with an edict, an inventory, and an office in charge. The prohibition was perfectly clear, and was then never carried out to the end.

The Huichang case is the best documented. By the mid-ninth century, everything monasteries held appeared on the state's list of sensitive assets: tax-exempt estates, people exempt from service, commercial enterprises such as mills, lodgings and shops, and pawnshops, and, most conspicuous in an age of coin shortages, copper. A substantial share of the country's copper stood in halls as Buddha images, bells, and chimes, withdrawn from circulation. After the An Lushan rebellion, fiscal administration shifted to the two-tax system. To the revenue authorities, people and land that paid neither tax ceased to belong to a realm beyond worldly affairs and became things that had slipped through the net. Du Mu's account of the demolition calculated throughout how much of the people's wealth had drained away. Wuzong's proclamation was blunter: “If one man does not farm, someone suffers hunger; if one woman does not rear silkworms, someone suffers cold. Today the monks and nuns under heaven are beyond counting. All depend on farmers for food and on silkworm rearers for clothing.”² The proclamation denounced the monks' tax exemption, not the Buddha's doctrine. The second of the four questions is thereby answered: the main driving force was the treasury's deficit. Zhao Guizhen's and the Daoists' slanders, and Wuzong's personal likes and dislikes, merely supplied a fiscal decision with religious language. Essay 9, Chapter 3 noted that times of order meant quotas for Buddhism. Huichang was the quota in its extreme form: it reduced the quota to zero in one stroke.

In the eighth month of Huichang 5, Wuzong issued a proclamation summing up the results: more than 4,600 monasteries demolished, more than 40,000 zhaoti monasteries and lanruo hermitages destroyed, 260,500 monks and nuns returned to lay life and registered as households liable for the two taxes, 150,000 enslaved men and women confiscated, and tens of millions of qing of fertile land seized.² Essay 9, Chapter 3 already entered this line in its long table. Here we add one check on the arithmetic: the Tang empire's total cultivated land amounted to only a little over fourteen million qing. The “tens of millions of qing” of monastic land exceeded the farmland of the entire empire.³ This is no embellishment by someone retelling the story. The edict itself says precisely this: “Tens of millions of qing of rich, superior land were taken.” This volume reads official documents at half face value just as it does monastic biographies. Both sides exaggerated, only in opposite directions: one inflated miracles, the other official achievements. Even after discounting, the scale remains startling. The return of 260,000 people to lay life rests on monastic and household registers and represents almost the entire registered population of monks and nuns. This went beyond harassment. It was an institutional reset to zero.

A foreign eyewitness recorded the process. The Japanese monk Ennin, stranded in Chang'an, entered what he saw and heard in his diary month by month. The demolition was not a single night's violence but three years of administrative procedure. First came checks of ordination certificates, with those lacking certificates returned to lay life first. Next came successive reductions by conduct under the precepts and by age. Finally, foreign monks were sent home. Ennin himself was ordered to leave the country in this way.⁴ The means deserve attention: the Tang court dismantled Buddhism with registers, not weapons. Who had a certificate and who did not, who was over fifty and who under forty—all appeared in the Bureau of Sacrifices' records. One official order sent 260,000 people back into lay clothing. India's rulers never possessed such a register; to strike at Buddhism, they could only burn one monastery at a time. The Chinese state could work from its registers, because Buddhism had entered them two hundred years earlier. Registration had once been the price of Buddhism's place in China (Essay 9, Chapter 3). The same file now became the basis for demolition. The hand that could issue certificates could also revoke them.

Who benefited? The treasury came first: copper images, bells, and chimes went to the Salt and Iron Commissioner for minting; iron images became farm tools; fertile land entered the two-tax system; enslaved people were registered as taxpaying households. The Daoists came next, briefly. Zhao Guizhen's ascendancy lasted only a few years. In the very month Xuānzong took the throne, he had Zhao beaten to death. A victory won through an emperor's personal preferences ended with his death. Another class of beneficiaries is often overlooked: regional military governors who did not obey the edict. Ennin recorded that the circuits of Zhen, You, Wei, and Lu north of the Yellow River did nothing at all about the suppression order. Their military governors had calculations of their own: monasteries formed part of the established local order. Why dismantle one's own supports for Chang'an's sake?⁴ An irony lies here. Essay 10, Chapter 1 argued that a unified empire gave Chinese Buddhism its form of all-encompassing integration. In the Huichang year, what saved its life was precisely the empire's lack of unity. Only centralized power could demolish every monastery under heaven overnight; regional division kept the edict from getting through Hebei. Unity supplied the form; division left gaps. Chinese Buddhism benefited from both. Luck was at work here, scarcely strategy, and most instances of survival in institutional history are jointly written by luck and structure.

Then comes the most important measure in this test: the speed of recovery. Wuzong died in the third month of Huichang 6. In the first year of Xuānzong's Dazhong reign, an edict already allowed monks who could restore monasteries abolished in Huichang to reside in and manage them. Barely two years had passed.⁵ A little over twenty years later, in Yizong's Xiantong 14, the Buddha's bone relic was welcomed with a ceremonial escort more splendid than the precedent of the Yuanhe reign. The scene Han Yu had once risked his life to oppose returned at twice the scale. The emperor had only one thing to say: “If I live to see it, I shall have no regret even in death.”⁶ Fewer than thirty years separated institutional reduction to zero from the emperor's prostration. The Indian case in Essay 8 has no such curve. There, the curve points only downward. Huichang thus supplies a counterexample to the rule stated in Essay 8, that death is a process and its date merely the moment its end was recorded. The end seemed to have come, yet two years later monasteries were rebuilt on the same sites. In India, nothing was rebuilt.

But “recovery” must be taken apart. First look at the inventory. What returned were monasteries, images, monastic quotas, ordination platforms, and religious assemblies. Anything with an entry in the register could be restored by the state. What did not return were the teaching seats of the doctrinal schools; they never returned thereafter as a complete institutional system. Monks and nuns could come back, but doctrinal lineages could not. Four points explain this. First, the media were unequal. In the Dazhong restoration, the court rebuilt monasteries, ordained monks, and copied the canon. Yet the lifeblood of doctrinal learning lay outside the canon, not within it. Commentaries were not included in the canon, had very few copies, and clustered in the scripture libraries of official monasteries in the two capitals. One fire destroyed the last copies. Monastic registers were household records: a single edict could reactivate them. Commentaries were unique copies outside the canon: an edict could not write a lost book. Later, the Tiantai monk Yiji met Qian Chu through the recommendation of National Preceptor Deshao. He explained that the doctrinal works of Zhiyi, the Wise One, had suffered in the An Lushan wars and been burned in Huichang. Only fragments remained at home; complete copies existed only overseas. Wuyue therefore sent envoys to Goryeo, as Essay 10, Chapter 2 has described.⁷ Second, the human chain lost a generation. Training a master lecturer took twenty or thirty years, while the two capitals suffered two blows within thirty-five. Huichang's order of laicization left teachers dead or scattered among the townspeople. Before strength could recover, Huang Chao entered Chang'an in Guangming 1 (880), and the teaching seats of official monasteries went up in flames with the two capitals. The Tang then fell, followed by sixty years of continuous warfare in the north. Doctrinal learning had only one location, concentrated in the two capitals; both blows fell there. Third, no one needed doctrinal learning any longer. After the restoration, the court wanted blessings, ordination platforms, and religious assemblies, no longer doctrinal discussion. Translation had stopped after Xianzong's reign. Although Song Taizong reopened an Institute for the Translation of Scriptures, most works translated were esoteric texts. Once translated, they went straight into the canon, with no one lecturing on them. What had been restored was the ritual of translation, not the vocation of exposition.⁸ Fourth, surviving lineages were preserved by regional division. Chan and Tiantai remnants continued in Wuyue, where three generations of the Qian house supported their teachers as national preceptors and doctrinal texts flowed back from across the eastern sea. The lifeblood of doctrinal learning would later have to be broken and renewed once again. Yogācāra commentaries long lost at home had to wait a thousand years for Yang Wenhui to recover them from Japan: a second return flow.⁹ The four points yield one sentence: the state could restore what it had registered, but not what it had never registered. The capacity of the two kinds of lineage to survive disaster now diverged. A Chan lineage needed only two people and a sentence. A lineage of doctrinal learning needed, besides the two people, a building full of books, a teaching seat, and thirty peaceful years. After Huichang, two people were easy to assemble. The other three conditions never again came together at once.

At the other end, who suffered least? Chan was in the mountains, farming for its own food. Its property was not in official monasteries, its income did not depend on grants of estates, and what the state's hand could not reach it could not confiscate. Pure Land was in the family hall: one name to recite occupied no official slot, required no ordination certificate, and appeared on no demolition list. The two structures described in Essay 10, Chapters 2 and 3—three levels of dependence and Chan's resilience to disaster—were both borne out at Huichang. The more a school depended on, the worse its injuries; the less it depended on, the more easily it escaped. This chapter therefore accepts only half of what “selection” implies. Huichang did not choose the winners. Before Huichang, the winners had already been determined by their structures of dependence. Fire merely revealed the result. Tang Yongtong summarized the aftermath as the decline of doctrinal learning and the flourishing of Chan and Pure Land. This volume adopts his conclusion and pushes the cause one step back: fire did not select Chan and Pure Land; their forms could already withstand fire.¹⁰

The recovery inventory omits another layer, with good reason: the popular layer needed no recovery because it had never been interrupted. Three kinds of unearthed objects bear witness. First, scripture pillars. The Uṣṇīṣavijayā-dhāraṇī Sūtra promised that if dust from a pillar touched someone's body, or its shadow passed over them, one could avoid even the consequences of grave offenses warranting a bad rebirth. Erecting pillars for deceased relatives thus became a common meritorious act. Nearly 150 Tang stone inscriptions of this scripture survive. The pace of pillar erection rose rather than fell between the reigns of Wenzong and Yizong, spanning Huichang itself. Building continued through the Five Dynasties; Qian Liu, king of Wuyue, erected three in a single year.¹¹ A single stone pillar could deliver the dead without any monastic community to tend it. Huichang could demolish a monastery, but not the stones beside a road. Second, the Ten Kings. In the late Tang, the monk Zangchuan of the Dashengci Monastery in Chengdu composed the Scripture Spoken by the Buddha on the Ten Kings. It combined the Buddhist intermediate state lasting seven sevens of days with the Confucian mourning observances at one hundred days, one year, and three years, making ten ritual offerings. The underworld was organized into ten courts, each with a king, each king with judges, each judge with documents and case files. Kṣitigarbha presided over them all. The dead passed from court to court; their families made offerings at each observance. From the Five Dynasties and Song onward, the scripture circulated throughout north and south and became the base text of Chinese funeral rites. By the fourteenth century, no one doubted the Ten Kings any longer.¹²¹³ The conjecture in Essay 10, Chapter 1 gains another piece of evidence here, produced in China itself; as usual, no separate conjecture is filed. India's afterlife ran automatically according to karma. China's afterlife was equipped with offices, case files, and procedures for appearing in court. The departed souls of a bureaucratic dynasty passed through a bureaucratic underworld. Third, community associations. Dunhuang manuscripts preserve hundreds of association rules and circulating notices. Help with burials and mutual contributions for funerals were their main activities, alongside lighting lamps, constructing caves, holding ritual feasts, and copying scriptures. Breaches of agreement drew penalties, and admission made no distinction between monks and laypeople.¹⁴ This layer held no ordination certificates, entered no monastic registers, and appeared on no demolition list. However precise the power of a register, it could not reach beyond that register. Essay 10, Chapter 4 argued that Pure Land made illiteracy and lack of time premises of its design. This popular layer went further: it kept working without a monastic community at all. When Chan and Pure Land later put down deep roots, this was the soil. Daily recitation entered the family hall; Ten Kings rites entered funerals. Buddhism's place in funerary and memorial rites, discussed in Chapter 3, was built on this layer.

Huichang also supplied a parallel experiment, cleaner even than the Indian control group. Buddhism was not the only religion banned: four were suppressed in succession. Manichaeism died first, and is the most revealing case. It depended on the Uyghurs. In Huichang 3, the Kirghiz defeated the Uyghurs. As soon as its patron fell, the estates, houses, money, and goods of Manichaean monasteries in Chang'an were confiscated, and their monks were referred to the Secretariat-Chancellery for a detailed disposition. Only two years later, with the proclamation of the eighth month of Huichang 5, did the other three face action: more than three thousand people designated Da Qin, Muhu, and Xian were ordered back to lay life, so as not to adulterate Chinese customs. There were three outcomes. The Church of the East never recovered in the Central Plains and never returned. Zoroastrianism fell silent on entering the Song. Manichaeism moved into popular society, changed its name to the Religion of Light, and became an enemy of the court.¹⁵ One reign, one empire, four banned religions, three outcomes: two years later Buddhism was back under state administration, Manichaeism went underground, and the Church of the East and Zoroastrianism disappeared. The difference was not the superiority of their doctrines but where each had rooted itself before Huichang. The lifeblood of the three foreign religions was official permission and communities of foreign merchants. Zoroastrian shrines barred outsiders; donors to the Church of the East rose and fell with trade routes. Neither religion had a place beside Chinese graves, a daily practice in the family hall, or a seat in community associations. The comparison with the Church of the East in Essay 9, Chapter 1 reaches its result here. Both entered under borrowed cover; Buddhism put down roots in China, while the Church of the East remained dependent on foreign merchant communities. When the prohibition came in 845, its congregations were emptied overnight. Essay 8's judgment—the blow explains the date of death; the roots explain the death—finds its clearest verification in Huichang's own edicts.

The shared heading compresses two prohibitions. Chapter 1 and its note distinguish the action against Manichaeism in Huichang year 3 (843) from the edict of year 5 (845), which did not name Manichaeism. The four outcomes belong to the chapter’s comparison across that reign, not to four religions named in one edict. “Back under state administration” means Buddhism’s return to the state’s registered establishment.

The Chinese image places China and India in separate panels; the English image stacks the curves and places 845 at the Chinese trough, whereas the Chinese image marks it on the descent. These are schematic contrasts, not statistical series. The Indian curve has no dated scale and does not share the Chinese dates 845, 847, and 873. The unbroken popular layer belongs to the Chinese case: dhāraṇī pillars, Ten Kings rites, and community mutual-aid associations, including assistance with funerals. The chapter’s distinction between the destructive blow and the survival of roots remains its explanation of the contrast.

After Huichang, there was only half of one further event of this kind. In Xiande 2 of the Later Zhou (955), Shizong abolished more than thirty thousand monasteries without imperially granted plaques and took copper images for minting. He neither burned scriptures nor killed monks, and even offered a dignified justification for destroying images: “The Buddha transforms people through the good way. If one sets one's heart on goodness, one serves the Buddha. How could those copper images be what is meant by the Buddha?”¹⁶ The Buddha resides in goodness, not copper, so the copper may be mine. This was no longer the suppression of Buddhism but a state's seizure of property from a permanent institution. Even this half-suppression failed to cross the boundaries of regional division. Wuyue obeyed the edict and, after obeying it, destroyed nothing. Qian Hongchu inspected Hangzhou's monasteries as ordered and found 480 with imperial plaques—more than 17 percent of the monasteries left standing nationwide, with not one liable for abolition. Huichang's edict could not get across the Yellow River; Xiande's could not cross the Qiantang. Regional division left room for Buddhism a second time.⁷ For the next thousand years, China saw no further nationwide suppression of Buddhism. Why did it stop with the Three Wus and One Zong? Marked explicitly, as required, the following is this volume's conjecture: both parties had finished their lessons. The state learned that demolishing monasteries yielded a one-time return, while administering them yielded revenue every year. With ordination certificates and monastic quotas in hand, why set another fire? Buddhism learned not to grow so large that it became a fiscal problem. Buddhism henceforth deliberately kept the scale of monastic property and the number of monks within the bounds set by the state. Huichang and Xiande together settled an exchange: Buddhism's right to permanent residence in China, obtained in return for economic restraint and complete administrative submission. The next chapter offers two pieces of evidence against which this conjecture can be tested: the price list for ordination certificates and the tax rules for monastic land.

Finally, set out the results of this experiment. Of the three comparison groups in Essay 8, Chapter 3, the Chinese group yields these results: the blow was no less severe than any in India, and more administratively precise; death did not occur, because something else lay beneath the point of impact. Indian Buddhism was the basket of eggs described in Essay 8, Chapter 2: destroy the estates and everything was destroyed. Before Huichang, Chinese Buddhism was already spread across several places: the mountains, in Chan farming; the family hall, in daily recitation; the graveside, in funeral rites. Song people would add another place, on paper, through woodblock printing, the subject of the next chapter. Huichang took Buddhism's real estate but could not take its habits. The state understood this too. Restoration was therefore an acknowledgment, not repentance: better to rebuild the monasteries and return habits to a place where they could be registered than to leave 260,000 people with those habits scattered among registered households. The blow explains the date of death; the roots explain the death. That judgment has now been verified on two continents. The next chapter examines what Chinese Buddhism relied on in peaceful times: new paper, new donors.

References¶