Chapter 1. Classification of the teachings: an operation to incorporate the entire inheritance¶

The previous essay ended with a question. A reunified empire looking back at a community of two million would demand to know its place within the order. The community answered with an intellectual ordering, called classification of the teachings. First consider the situation it had to resolve, one India had never encountered: making the successive simultaneous. A thousand years of Indian sedimentation, the Āgamas, Prajñāpāramitā, Yogācāra, tathāgatagarbha, and esoteric spells, were translated in China almost together within two or three hundred years. The translation workshops' customs office, described in Essay 9, Chapter 2, checked provenance, not doctrine. Scriptures legally admitted to the canon therefore contradicted one another. They taught both non-self and Buddha-nature, both emptiness and permanence, happiness, self, and purity, both three vehicles and one. Indians lived along a time axis, each generation cultivating its own field. The Chinese faced a warehouse of simultaneous contents, with the scriptures of every period on its shelves at once.

The method of classification was brilliant. It recast a historical problem as a curricular problem. Contradictions were not contradictions but stages. The scriptures seemed to fight one another, yet they were really the Buddha's teaching adapted to different capacities at different stages. Tiantai's five periods arranged this thought as the sequence of the Buddha's lifetime of teaching: the Avataṃsaka's first illumination, the Āgamas' inducement, the Vaipulya scriptures' rebuke, Prajñāpāramitā's sifting, and the Lotus and Nirvāṇa opening and bringing all together in one. Classifications in the Northern and Southern Dynasties were already known as the three southern and seven northern schemes. They culminated in Zhiyi's five periods and eight teachings and Fazang's five teachings and ten tenets, whose forms Volume 1 has described.¹

This volume has long been waiting for this step. Essay 4, Chapter 1 recorded two steps of the ratchet: new scriptures matched the existing corpus one level at a time, and apologists then turned a necessary condition into a sufficient one. Classification of the teachings was a third step. The aim was no longer to get past verification but to take it over, reweighting the corpus so that one's own texts became the standard. In India, classification was still piecemeal: definitive and nondefinitive meanings, the three periods. China made it into a complete system. Each school laid out a hierarchy, and at its highest rank stood, as it happened, that school's own central scripture. Tiantai judged the Lotus purely perfect; Huayan judged the Avataṃsaka the one vehicle of the distinct teaching. Each school assigned the highest rank to itself.

One point on which the schools actually agreed must be recorded: the first entry in the five periods. Tiantai placed the Avataṃsaka in the first three seven-day periods after awakening; Huayan placed its own scripture in the second seven-day period. The day counts differ slightly, but both put it first. The disagreement concerns the arrangement of the later periods (Volume 1, Part II, Essay 22, Chapter 1). Classification of the teachings was a shared institution. Their separate judgments took place within it.

Chinese classification had one property India lacked, and every subsequent story in this essay follows from it. It distinguished higher from lower, not authentic from spurious. Indian schools and Mahāyāna judged one another's scriptures false, as in the Jutuo (鞠多) case in Essay 4, Chapter 4, and the loser was excluded. Chinese classification judged its rival shallow, provisional, or an introductory course. Yet it remained spoken by the Buddha, remained in the canon, and remained available for study and practice. Turning an enemy into one's own preparatory class was a more complete incorporation than expulsion. The loser stayed in the game, merely demoted. No one therefore had to form a separate household, and no one could overturn anyone else. From then on, Chinese Buddhism was diversity within one system, not India's diversity of separate households. Essay 11 will consider how much this difference mattered in the centuries of persecution.

“Simultaneous arrival” summarizes the chapter’s account of a millennium of Indian textual layers becoming available together in China over two or three centuries. It does not mean that the texts arrived on the same day. The five shelves are this diagram’s summary of that corpus, not Tiantai’s five periods or Huayan’s five teachings. Placing mantra on the top shelf does not mean that either school ranked mantra as its highest teaching. The chapter’s distinctions between final, sudden, and perfect teaching remain in force.

The English image replaces the Chinese warehouse’s upward arrow with a double-headed one and moves its preparatory-class annex to the base. These changes do not alter the chapter’s claim: rival teachings are assigned lower ranks while remaining within the canon. The five-tier caste icon is an illustration, not a claim that India had five castes; “five distinct natures” refers to the classification of dispositions for awakening. The proposed connection between social classification and doctrinal classification remains explicitly the author’s conjecture, with the qualifications and tests stated in Chapter 1.

The social setting of classification was the lecture culture of the Southern Dynasties. Doctrinal lectures were occasions shared by aristocratic families and emperors. Emperor Wu of Liang personally mounted the Dharma seat to expound scriptures. Four times he offered himself at Tongtai Monastery, and each time his ministers redeemed him with hundreds of millions in cash. The title "emperor-bodhisattva" carried the line of emperors taking bodhisattva precepts, recorded in Essay 4, Chapter 2, to its limit.² In such a setting, classification of the teachings was what gave a school its standing. To establish a school, one first had to establish a classification. A classification of its own marked a school's coming of age. Essay 3, Chapter 1 said that an Abhidharma of its own marked a school's coming of age. The continent and genre had changed; the structure remained intact.

The straightforward half of the account must be given its full weight, and setting it out is more interesting than criticism. Classification of the teachings was real hermeneutic labor. The contradictions were real, and grading them was an honest attempt. Moreover, the classifiers had remarkable discernment. The sequence of the five periods, Āgamas early, Prajñāpāramitā in the middle, and Lotus and Nirvāṇa later, broadly parallels the order of scriptural composition reconstructed by modern philology. They saw the strata but read geology as a curriculum. The growth of doctrines in succession was read as the Buddha's teaching in succession. In an age without philology, classification of the teachings was philology, only with the causal direction reversed. This book has no standing to mock that reversal, because it too classifies teachings. The Preface to Volume 2 acknowledged this long ago: it ranks five theories of liberation by the criterion of the cessation of suffering. Classification is unavoidable; only the criterion is open to choice. Zhiyi's criterion was perfection; this book's is cessation.³

The rules require an explicit marker. The following is this volume's conjecture. The motive for classifying the teachings was the honesty of defending the Dharma. Its proponents would not acknowledge contradictions between scriptures, because that would acknowledge a problem in "spoken by the Buddha" itself. Nor would they abandon any scripture, because the translation workshops had already produced it and believers had already embraced it. The only exit between these two refusals was grading. Classification is not the art of sophistry. In a system that does not permit the word "wrong" (错), it is the only possible way of writing that word: write "shallow" (浅) instead.

The conjecture has a deeper layer, which must be marked explicitly, with its weight stated. This is one of this volume's largest conjectures. The organization of knowledge reproduces the organization of society. This proposition has a history in the sociology of knowledge. Durkheim and Mauss's original formulation in their work on primitive classification was that "the classification of things reproduces the classification of people."⁴ This volume connects it to the classification of Buddhist teachings.

India was a politically divided world with extremely low class mobility. Caste was a hierarchy in which movement was exceedingly difficult. Its Buddhism showed the same structure. School boundaries behaved like caste boundaries. Refusing joint uposatha was like refusing shared meals; judging one another's scriptures false was like refusing intermarriage (Essays 2 and 4). The doctrine of five distinct natures, with one class lacking the nature and forever unable to become buddhas, was a caste system for souls. China was a unified dynasty with channels of mobility. A commoner could rise to ministerial rank, and in theory the ladder was open to everyone. Its Buddhism showed the same structure. Classification of the teachings posted examination results; it did not divide household registers. Bringing the three vehicles into one was the doctrinal version of "anyone can become a jinshi." There could be ranks, but no debased status.⁵

This conjecture has corroboration so good it seems almost a coincidence: the dispute over five distinct natures. Indian Yogācāra explicitly maintained that one class of beings would never become buddhas. This caste system for souls was brought to China by the most authoritative translator with the most complete texts, then rejected wholesale. Daosheng won the right to Buddhahood for icchantikas before the texts had all arrived (Essay 6, Chapter 2). Lingrun opened the attack in the early Tang, and Fabao and Huizhao fought for decades. The outcome was that universal Buddha-nature became an axiom of Chinese Buddhism. Zhanran extended Buddha-nature further, to plants, tiles, and rubble: insentient things possess the nature.

The genealogy of this step needs to be stated precisely. Zhanran's argument came from Huayan, and he marked its source. The claim of following conditions while remaining unchanged came from the great teaching; the claim that wood and stone lack mind came from a lesser school. His argument was a test of consistency. To allow following conditions while remaining unchanged and yet say that wood and stone lack mind is to contradict oneself. Fazang's discussion of universal and restricted scope in the Tanxuan ji had already stated that Buddha-nature and nature-origination extend to both the environment and its inhabitants, universally reaching the insentient. Chengguan narrowed the scope. His Yanyi chao states that this does not mean insentient things also possess an awakening nature and become buddhas like sentient beings. To allow their Buddhahood would be a false view. The conclusion of Volume 1, Part II, Essay 22, Chapter 5 is that, on inherent inclusion, Tiantai as a whole stands against Huayan; on scope, Zhanran and Fazang stand on the same side against Chengguan. Thus Zhanran did not go further than Huayan. He used the position of Fazang, an earlier Huayan patriarch, against Chengguan, its fourth patriarch.

A Tang commentary also transmits the story that Xuanzang, before translating, wanted to conceal from the eastern land the doctrine of a class lacking the nature, but Śīlabhadra rebuked him and stopped him. The story is discounted as hearsay, yet the anticipation it records is real. The translator himself foresaw that China would refuse it.⁶ The same doctrine, two societies: an axiom on one side, discarded on the other. Classification reproduces society; this case confirms it on both sides at once. The limits must also be stated. There are only two or three civilizational cases, the mechanism of reproduction is loose, and Durkheim's proposition is itself disputed in scholarship. Japan in Essay 13 will supply another example. For now, the matter rests here.

The echoes continue today. Emperor Xiaozong of Song's sentence, "Use Buddhism to cultivate the mind, Daoism to nourish life, and Confucianism to govern the world," was a functional classification of teachings on the scale of a civilization. The unity of the three teachings became a tide in the Ming. Modern scholars classify Chinese religiosity as diffused rather than institutional: no formal religious membership, weak boundaries, and use according to function.⁷ A practitioner today may do southern Buddhist insight meditation in the morning, dedicate merit toward Pure Land in the evening, and receive a Tibetan esoteric initiation on the weekend, finding everything reasonable and practicing a little of each. This is often treated as contemporary disorder. In this chapter's reading, it is precisely the continuing form of the classificatory habit: distinguish higher from lower, not authentic from spurious. The unified ranking is gone; the habit remains.

There were two costs. First, the ever-rising claim to the definitive. Once classification becomes a competitive genre, each later entrant must be more perfect. The contest for the highest rank has no end. The ordering of sudden and perfect teachings is not a suitable example of this contest. The Treatise on the Five Teachings orders them as small, initial, final, sudden, and perfect. Sudden is fourth and perfect fifth. The function of the sudden teaching is to reserve a place for those who do not depend on verbal teaching. On the Tiantai side, sudden belongs to the axis of methods of instruction. The Outline of the Four Teachings explicitly says that the doctrinal contents of the secret and indeterminate teachings are simply the tripiṭaka, shared, distinct, and perfect teachings. Sudden and gradual are ways of giving the Dharma; they carry no difference of doctrinal rank (Volume 1, Part II, Essay 22, Chapter 1, and Essay 21, Chapter 6).

A more concrete example is one of Fazang's judgments. He judged the Lotus using its own prized possession. The three carts at the gate are all the three vehicles; only the white ox cart given in the open ground beyond the boundary is the teaching of the one vehicle. He used the cart simile in Tiantai's central scripture to establish a higher rank above it. The opposition between classifications is sharpest at this point. The Chan of Chapter 3 of this essay is one product of this contest. The second cost is that the boundary of "spoken by the Buddha" dissolved completely. Everything could enter the canon, everything had a place, and questions of authenticity were permanently displaced by questions of depth. The ratchet described in Essay 4, Chapter 1 completed its last step here. The capacity to judge authenticity withered, and this would become visible only a thousand years later. When "Mahāyāna is not the Buddha's word" returned to China in the form of philology, it provoked thunderous anger, not argument, the main subject of Essay 14. The framework was in place; the systems were still missing. The next chapter examines how Tiantai and Huayan built complete systems on classification.

References¶