Chapter 3 · The Root Schism: The Ten Practices, Mahādeva, and Two Visions of Buddhism¶

At some point around a century after the Buddha's passing, the saṅgha split into two branches, the Sthaviras and the Mahāsāṃghikas: the event known as the root schism. The later eighteen schools and more all branched off from these two. But this juncture must first be separated into its parts in accordance with Volume I's examination of the evidence, because the sources supply three archives that do not agree. The Southern Vinaya records the dispute over the ten practices at Vaiśālī, but the Vinaya's own account ends with all ten ruled unlawful and the seven hundred reciting together; it says not a word about schism. The schism was attached to the ten practices by later religious histories in the Dīpavaṃsa line.¹ The Northern Sarvāstivāda line attributes the schism to Mahādeva's five points, but this archive more likely concerns a later dispute within the Sarvāstivāda that was moved back to the schism's starting point. The Mahāsāṃghikas' own Śāriputraparipṛcchā offers another account: an elder wished to expand the Vinaya-piṭaka, most monks refused, and that caused the separation. Volume I examined the disagreements among these three archives one by one. What could be established there was only one external anchor and an order of magnitude: Aśoka's schism edicts show that by the middle of the third century BCE, division was already an accomplished fact requiring royal orders to suppress it. The schism itself more likely arose from disputes over the extent of the disciplinary rules and the organization of the saṅgha than from a clear doctrinal opposition. Its date can be placed only on the scale of one or two centuries after the Buddha's passing.² This volume does not repeat that examination. It takes its conclusion and uses these mismatched archives in two ways: first the ten practices, a conflict that did not produce a split; then the retrospective narratives of the schism. The cause remains a mystery, but every account gives away something about its author.

Look first at the ten practices. The Vajjian monks of Vaiśālī followed ten practices. Yaśas, an elder arriving from the west, encountered the most conspicuous of them: on the day of uposatha, they placed a copper bowl in the assembly and asked laypeople to put money into it, saying it was needed by the saṅgha. Yaśas publicly declared this unlawful. The local saṅgha instead sentenced him to apologize to the laypeople and then expelled him. He traveled west to seek help, enlisted Revata and other elders, and seven hundred assembled at Vaiśālī. East and west each nominated four adjudicators to deliberate point by point, and all ten practices were ruled unlawful. The list deserves to be read in full: storing salt in a horn for later use; eating when the shadow had passed noon by two fingerbreadths; eating again in a neighboring village after finishing a meal; holding separate uposathas within the same boundary; performing a formal communal act before everyone had assembled and obtaining consent afterward; following a preceptor's precedent; drinking unchurned milk; drinking palm liquor not yet fully fermented; relaxing the requirements for sitting cloths; and, tenth, accepting and keeping gold and silver. What is most striking after reading it through is the nature of the list, not the severity or leniency of any one item. Not one concerns doctrine; every one concerns subsistence. How to store salt, whether mealtime can be extended by two fingerbreadths, whether money may be accepted: this is a list of the saṅgha's living costs.

Put the list back into its geography, and a reading emerges. Vaiśālī was a major commercial center on the northern bank of the Ganges. The ten practices prevailed precisely where the monetary economy was most active and where lay support could most readily be converted from goods into money. The ten practices were not ten acts of laxity but an urban saṅgha's adjustment to a new economic form. Beneath the dispute over strictness in the Vinaya lay a change in how the saṅgha fed itself. The fourth strand of external causes—monastic economies—appears here for the first time. The gap between judgment and enforcement bears this out. The seven hundred elders won a clean victory at the meeting: all ten practices were unlawful. No one subsequently changed the rules either. Volume I compared the schools' codes of precepts: their substantive major rules are almost identical. The Mahāsāṃghikas, later labeled “lax,” likewise retained the prohibition on money unchanged in their own Vinaya. Changes occurred at another level: expedients such as having a “lay steward” receive money on one's behalf or transferring it through a “purifying donation” spread across the schools. All kept the rules unchanged; all made accommodations in practice. The elders won at the meeting, but the needs of livelihood did not lose.

The second point is sequence, where Volume I's examination of the evidence is directly useful. The story that the defeated party indignantly convened a separate great council and became the Mahāsāṃghikas was attached afterward in the Dīpavaṃsa line; the Vinaya's own account contains no such episode. The reliable sequence is this: first came the split, whose date and cause cannot be specified; then came the parties' accounts; only last came the lists of doctrinal differences filled out by the treatises. The accounts cannot even agree on which rules were contested at the time. The Southern account speaks of ten relaxations; comparison of the codes shows that the schools' substantive rules are almost identical, with nearly all variation in the training rules on comportment, such as eating and wearing robes; the Mahāsāṃghikas' own account says that the elders wanted to add rules and the majority refused. The monastic household divided first; doctrine was supplied afterward. Even the cause of the accident was written in later by each party for itself. Doctrinal differences are usually an explanation offered after a schism, not its cause. The explanation is always written after the separation, and each school writes its own.

The Chinese image’s three dashed arrows run from the fissure outward to the later accounts; the English image reverses them. The chapter’s sequence is schism first, retrospective accounts afterward, with neither the precise date nor the cause established. The English arrows do not turn those accounts into demonstrated causes. “Island chronicles” refers specifically to the line of the Dīpavaṃsa.

The link to Aśoka’s schism edicts is also reversed between the two images. This is an external chronological anchor: Chapter 3 takes the edicts to show that division already existed by the middle of the third century BCE; they do not supply its cause. The concluding “households” are the communities of the order: the communities divided first, and doctrinal explanations were supplied afterward.

Now look at the retrospective narratives of the schism. The weightiest among them concerns attainment. The Northern Sarvāstivāda line says that the schism began with Mahādeva, a little over a century after the Buddha's passing. Following Volume I's analysis, this archive more likely projects a later internal Sarvāstivāda dispute back onto the schism's starting point; this volume follows that reading. But suspending judgment on the date does not diminish its analytical value. Whatever the year of the dispute, it is the first record of a public examination of whether arhatship matched its name. Mahādeva advanced five propositions: arhats can still be disturbed by Māra and emit semen; still have nondefiling ignorance; still doubt their own realization; need another person to certify their attainment before knowing that they have attained it; and the path arises through sound—a heartfelt cry of “Oh, suffering!” makes the noble path manifest. Compressed into a verse: “Induced by another, ignorant, / doubtful, led in by another; / the path arises because of sound: / this is called the Buddha's true teaching.”³ When a vote was taken, the majority followed Mahādeva, a minority of elders refused, and the two schools separated.

Under the rules, the straightforward explanation comes first, and this time it is powerful enough in itself. The five points can be read quite straightforwardly as an observational report: people in the saṅgha acknowledged as arhats did in fact still have nocturnal emissions, still not know some things, still hesitate, still need confirmation from teachers, and still work at practice through cries and calls. Someone honestly recorded the gap between the ideal arhat and the actual arhat. On this reading, Mahādeva was the first to say publicly that attainment did not match its name, not someone who corrupted attainment; the five points were the first record of arhatship as it actually was. The question of “verification” posed in Volume II's essay on synthetic criticism had already been asked for posterity by someone in the saṅgha before the Common Era. The result of asking was the order's schism, and that result itself supplies half the answer.

But the straightforward reading cannot explain the distribution of the targets. This volume's conjecture begins here, explicitly marked as usual. This reading does not depend on the date of the five points: whether they record the root schism or a later internal Sarvāstivāda dispute, the distribution of the targets belongs to the archive itself. Consider what each point targets. The first four all cut away at the arhat's authority: freedom from the influxes, omniscience, freedom from doubt, and realization known for oneself. These four were precisely the grounds on which the elders judged others. As the previous chapter showed, eligibility reviews, enumerations of faults, and rulings of unlawfulness all rested on “our attainment is guaranteed genuine.” The fifth point, conversely, allowed ordinary practitioners' daily practice to lead to the noble path. Together, the five points form a redistribution plan: lower the standing of attainment and ease the requirements for entering the path. The Mahāvibhāṣā's own account of the occasion gives away a third layer. Mahādeva had first certified his disciples' attainments; only when they questioned him point by point—if they were arhats, why did they still have emissions, ignorance, and doubt?—did he answer with the five points. The hostile biography intended to portray him as glossing over his faults, yet in passing it recorded the actual form of certification at the time: attainment was certified by the teacher, not evident from the attainment itself. On this reading, the five points are an apologia for the certification system. They place the authority to certify firmly in the teacher's hands (the fourth point, “led in by another,” outright makes external certification the norm), while lowering the standard of attainment to a level that could be reached. Lower standards and exclusive certification together made a complete plan. Who benefits? The majority, together with every teacher who has disciples whose attainment he can certify. That is why the majority sided with Mahādeva in the vote. The name “Mahāsāṃghika” itself records that vote, just as “Sthavira” names the other side of the same record. The circumstantial evidence: the accounts agree that the dispute ended by counting heads, not by settling doctrine. If the earliest layer of accounts shows that the five points emerged from self-examination among the elders rather than from adversarial propositions, this reading fails. One more admission must be made in advance under the rule of “reading the opponent's archive at half weight”: the most widespread version of Mahādeva's life says that he committed three heinous offenses before ordination. This comes from the side that judged him a demon (the Mahāvibhāṣā). This volume records it without accepting it. An archive that makes a doctrinal opponent into a murderer of his own kin has already declared its provenance.

What settled out of the schism was two ways the order imagined itself, not two lists of rules. The status of these self-images must first be made clear: they took shape later and need not reflect the alignments at the moment of schism. On the Mahāsāṃghikas' own account and the comparison of the codes, it may have been the elders who wanted to add rules at the outset, while the majority defended the old text. “Mahāsāṃghika equals laxity” is a label attached by the victors' narrative that fails against the textual evidence. But after several centuries, the roles each branch claimed were indeed distinct. The Sthavira line: the order is the custodian of memory, what it values most is orthodoxy, its basic strategy is to remain unchanged—better small than wrong. The Mahāsāṃghika line: the order adapts to the present, what it values most is accessibility, a path in which everyone has a share, its basic strategy is to respond to change—better broad than cut off. These roles may not have begun at the moment of separation, but they did settle into two actual Buddhisms. The custodians later moved south and sealed the Tripiṭaka into the island's palm leaves; Essay 3 will examine them. The adapters remained on the mainland, and over the next few centuries their imagination would give rise to the greatest rewriting in Buddhist history.

One concurrent shift after the schism deserves the closest attention. The schools of the Mahāsāṃghika line downgraded the arhat while upgrading the Buddha: his body was free from the influxes, his lifespan and power without limit, and he taught all dharmas in a single utterance.⁴ The two movements went in the same direction: once the credibility of attainment became a problem, the ideal had to be housed elsewhere. Calling them two halves of the same thing ties them a little too tightly, because the elevation of omniscience also followed an independent stratigraphic line within the sūtra collections, outside the schools' votes, with textual coordinates for each of its three layers. In the sūtras' own account, the Tevijjavacchagotta-sutta treats the claim that the omniscient, all-seeing one has knowledge and vision constantly present while walking, standing, sitting, and lying down as “falsely slandering me”; the proper designation goes only as far as the three knowledges. The Kaṇṇakatthala-sutta rules out knowing everything all at once. In a late layer of the sūtra collections, the Paṭisambhidāmagga includes omniscient knowledge in its list of knowledges. At the layer of commentarial reconciliation, the Milindapañha establishes omniscience through attention: knowing depends on turning the mind toward its object. Volume I, Part II, Essay 13, Chapter 6 gives the judgment: reconciliation always comes after what is reconciled. Once the saints in the real world could no longer be relied on, sanctity was pushed beyond the reach of the real world. The arhat stepped down from one pedestal; the Buddha stepped up onto another. Between the two pedestals a stretch of road opened: human beings remained on this shore, the ideal had moved far away, and a role capable of traversing the space between was needed. A little over a century later, a new ideal came to claim that road: the bodhisattva. The Mahāyāna of Essay 4 already had its place prepared here.

References¶