Chapter 3. The root division between Sthaviras and Mahāsāṃghikas¶

The preceding chapter ended by asking what turned a shared body of teachings into several independent transmissions. The answer was the division into schools. The first and most fundamental division separated the early saṅgha into the Sthaviras, or "elders," and the Mahāsāṃghikas, or "majority of the monastics." All the later "eighteen schools" developed from the two branches.

A difficulty with the sources must be explained first. Accounts of the root division contradict one another about its date, cause, and the rights and wrongs of the participants. We can only set out the principal accounts in turn and then draw cautious inferences.

1. Disagreement among the sources¶

The principal sources for the root division include the Dīpavaṃsa of the southern tradition, Vasumitra's Sarvāstivāda treatise Samayabhedoparacanacakra in Xuanzang's translation, and the Mahāsāṃghikas' own accounts.¹ They disagree on the following matters:

• Date. Some place it about a hundred years after the Buddha's death, others more than a hundred years afterward, and still others around Aśoka's time. Dates can differ by several decades, and the traditions associate the event with different kings.
• Cause. One account gives a disciplinary dispute, another a doctrinal dispute about the status of arhats, and each presents the other side as the instigator.

The disagreements alone show that the accounts were written down later by particular schools seeking to justify themselves and discredit their opponents. The present book does not readily accept any one side's claim that "we alone are orthodox."

One body of evidence is an exception. Aśoka's stone inscriptions, from his reign of approximately 268 to 232 BCE, are independent of the schools' accounts of themselves. Similar inscriptions have been found at present-day Sarnath, Sanchi, and Kosam. Aśoka warns monks and nuns that anyone who divides the saṅgha must wear white and live outside a monastery, while also declaring that "the saṅgha has been united."² White was the color of lay clothing. The penalty therefore used dress to mark expulsion from monastic status, rather than simply requiring a change of clothes. The inscriptions favor neither side, but show that roughly two hundred years after the Buddha's death, a king regarded monastic unity as a political aim requiring legal protection. We may infer that differentiation had become sufficiently pronounced under Aśoka to prompt royal suppression. The inscriptions thus provide external evidence for the latest possible date of the root division. Whether its cause was discipline or doctrine, differentiation was already a fact by the middle of the third century BCE.

2. The cause of division, the five points or discipline¶

The main accounts place the cause respectively in doctrine and discipline.

The first account concerns Mahādeva's five doctrinal points. A monk named Mahādeva proposed five claims which together reduced the status of the arhat. An arhat might still be affected by external causes, resulting in an emission during a dream; might have ignorance unrelated to defilement; might have doubt; might require another person's indication before knowing with certainty that he had attained; and might need an audible exclamation to bring about the noble path.³ The claims qualified the arhat's complete perfection. Tradition, chiefly in the Sarvāstivāda line, attributed the position to the Mahāsāṃghikas. In its account, the Sthaviras upheld the arhat's perfection, while the Mahāsāṃghikas accepted his limitations. The community therefore divided.

The philosophical implications of the five points require explanation. In the early teaching, the arhat (arahant) and the Buddha have equally ended all taints (āsava) and are equal in the cessation of suffering. Scriptural evidence appears in the Sambuddha Sutta of the Saṃyutta Nikāya. The Tathāgata "awakens for himself to teachings not previously heard," penetrates unsurpassed awakening, and "attains what had not been attained, gains what had not been gained, knows, distinguishes, teaches, and masters the path." The disciples "follow the teaching and the path."⁴ A distinction is thus established between the one who opens the path and those who follow it. They are equal in ending suffering, but unequal in their capacities. The ten powers and four grounds of fearlessness belong to the one who opens the path and need not belong to his followers. The ending of taints is the same, and liberation has one taste. The Sutta Piṭaka had therefore already drawn a distinction. The novelty of the five points was not to distinguish Buddha from arhat for the first time, but to relocate the distinction. The discourses placed the difference in capacities, while treating the Buddha and arhat alike in abandonment and realization. The five points placed the difference within abandonment and realization themselves. An arhat had ended defilements but might retain certain "functional deficiencies." The deficiencies could not cause further rebirth, yet left the arhat less perfect than the Buddha even in abandonment and realization. The fourth point, requiring confirmation by another, and the fifth, requiring an external sound to bring about the noble path, are especially consequential. If arhats cannot know their own attainment with certainty, the common scriptural declaration "I have understood; birth is ended and the holy life established" (āsavā me khīṇā) becomes only approximate in their case. Full certainty seems to belong to the Buddha alone. Once that reassessment is accepted, arhatship becomes a secondary form of liberation. The five points may themselves have been a scholastic dispute, but they prepared directly for later Mahāyāna arguments that buddhahood surpasses arhatship and that the bodhisattva path differs from the path of liberation, and this was not accidental.

The second account concerns discipline. Yet the southern tradition and modern scholars mean very different things when they speak of a "disciplinary dispute."

The traditional southern account connects the division to the Second Council discussed in 2.2. The Vajjian monks of Vesālī were defeated when their "ten practices" were declared unlawful. In resentment, they held a separate "Great Recitation" (Mahāsaṅgīti) and became the Mahāsāṃghika school. The ten practices were relaxations of discipline:

• Permission to keep salt in a horn for later seasoning, effectively retaining food overnight;
• Permission to eat until the sun's shadow had passed noon by two fingerbreadths;
• Permission to eat another meal after moving to a different village;
• Permission to hold separate uposatha ceremonies within the same boundary;
• Permission to conduct communal business without a quorum and obtain consent afterward;
• Permission to act on the grounds that "it has always been done this way";
• Permission to drink unchurned milk after finishing a meal;
• Permission to drink unfermented palm juice, a substance close to alcohol;
• Permission to use a sitting cloth without a border or larger than the prescribed size;
• Permission to receive and keep gold and silver, meaning money, the most consequential practice.

The account produced the image of "Mahāsāṃghikas as lax monastics." Yet even the southern Vinaya's own account of the Second Council says only that the ten practices were declared unlawful. It does not say that a division occurred then. Later chronicles in the Dīpavaṃsa tradition connected the division to the ten practices.¹

The scholarly account of Janice Nattier and Charles Prebish in 1977 almost reverses the traditional story. Comparing surviving disciplinary codes, they argued that the disagreement did not concern the ten relaxations at all, but the number of detailed rules of deportment. The substantial categories, including the four pārājikas, thirteen saṅghādisesas, and thirty forfeiture offenses containing the rule against holding money, are almost identical across schools. The real variation lies in the final category of "training rules" (śaikṣa), minor details of eating, clothing, and conduct. Examples include wearing robes properly, not looking around while eating, and not teaching someone holding a parasol or staff. By comparison, versions of the pācittiya category differ by only one or two rules. Theravāda has seventy-five training rules and Mahāsāṃghika sixty-six, while northern Sthavira schools have more, with a hundred in the Dharmaguptaka and as many as 113 in the Sarvāstivāda.⁵ The Mahāsāṃghikas' own account in the Śāriputraparipṛcchā likewise attributes the division to certain elders wanting to expand the Vinaya and the majority refusing.⁶ On that evidence, the Sthaviras appear more likely to have added rules. The same scholarly interpretation places "Mahādeva's five points" in a later dispute within the Sarvāstivāda, subsequently projected back onto the root division.

An apparently strong objection needs an answer first. Might the Mahāsāṃghikas have the fewest rules because they removed some, thereby proving their laxity? If we consider totals alone, the objection holds. Addition on one side and deletion on the other can both leave one code shorter. Numbers by themselves cannot establish the direction of change, so a lower total cannot settle the case. We must ask where the differences lie and who retains the especially sensitive rule about money.

As explained above, the differences lie almost entirely in minor training rules of deportment, while the grave rules are shared. The Mahāsāṃghikas' own Vinaya also explicitly prohibits holding money. The prohibition exposes a central weakness in the southern story. The group accused of breaking away to accept and keep gold and silver retained a clear rule against money in its own code. If relaxing that rule had actually caused the division, retaining the prohibition unchanged in its own Vinaya would make no sense.

The present book does not defend the Mahāsāṃghikas on the ground that they had the fewest rules, since deletion could also explain the smaller total. The substantive evidence supports a more specific judgment. First, the schools agree closely on substantial disciplinary rules, including the money rule. The Mahāsāṃghikas did not relax it, so the specific accusation of "lax heterodoxy," especially over money, fails against the texts. Second, the variation concerns training rules of deportment. Their direction of addition or removal cannot be established from totals alone, but training rules are the flexible category most readily expanded case by case. Together with the Mahāsāṃghika account, that consideration slightly favors "additions by the Sthaviras." It remains an inference, rather than a settled conclusion. The accounts even disagree about which rules were at issue, ten relaxations or changes to training rules, further showing that traditional narratives cannot be accepted in full. Taken together, the evidence makes disputes over the extent of discipline and monastic organization more likely causes of the root division than a clear doctrinal opposition. Its date can be fixed only on the scale of one or two hundred years after the Buddha's death.

3. The two branches and their developing differences¶

The original cause may have been only the number of disciplinary rules, but the two branches did later develop different intellectual profiles. The tendencies below formed gradually and are inferred from later writings. They cannot be read back wholesale into the moment of division.

The Mahāsāṃghika view that the Buddha transcends the world and his physical body is only a manifestation, together with the originally pure nature of mind, later offered points of connection to Mahāyāna and even tathāgatagarbha thought. The inference drawn in the present book is that the Mahāsāṃghika tradition formed an important part of the background to Mahāyāna's emergence, discussed in Essay 4. The Sthavira branch, meanwhile, concentrated on analyzing dharmas and developing Abhidharma, producing the major Sarvāstivāda and southern scholastic systems discussed in Essay 3. One branch tended to elevate the Buddha and regard the mind's nature as pure, while the other tended toward finer analysis of the characteristics of dharmas. Their directions differed quite clearly.

Geographical distribution

The intellectual division had a rough spatial counterpart, though neither caused the other. The Mahāsāṃghika branch was stronger in southern India, in the Andhra region, and western India, around Caitya Hill. Many surviving textual fragments of its Caitika and Lokottaravāda offshoots come from the south. The Sthavira branch was more widely distributed. Sarvāstivāda was based in the northwest, in Gandhāra and Kashmir. The Vibhajyavāda line spread through Sri Lanka to Southeast Asia. The Dharmaguptaka Vinaya entered northern China in the second to third centuries CE, while the Mūlasarvāstivāda Vinaya entered Tibet with the spread of tantric Buddhism.¹ Geographical separation and intellectual differentiation reinforced each other. Schools developed independently within their regional settings, and doctrinal differences became harder to reconcile. The process resulted from gradual change over centuries, rather than a decree at a single council.

4. The eighteen schools and the misleading implications of "schism"¶

Further differentiation within the two branches produced the traditional "eighteen schools," or twenty. The number is formulaic, and the lists do not agree completely. Their main outlines are shown below:

The text places Sautrāntika within Sarvāstivāda. The diagram's layout does not show that derivation.

The impression conveyed by "schism" needs correction. Under the Pāli Vinaya's definition, monastics commit the grave offense of splitting the saṅgha (saṅghabheda) when they hold separate uposatha, invitation, and communal acts within the same boundary on grounds drawn from eighteen kinds of claim, including reversing what is and is not the teaching. Most "schools" (nikāya), however, did not form through schism in that sense.⁷ Schools were closer to different ordination lineages combined with different doctrinal tendencies. Monks of different schools could often live in the same monastery, practicing according to their own codes and teachers, without necessarily being incompatible in everyday life.⁸ The position taken in the present book is that imagining the history of the schools as a series of bitter, irreconcilable church splits is an effect of the misleading word "schism." The historical pattern was closer to differentiation of lineages during prolonged coexistence.

5. The consequences of differentiation¶

The root division and the differentiation that followed prepared the ground for the later parts of the book.

First, they produced the independent transmissions discussed in the preceding chapter. Separate recitation by the schools gave us the Pāli texts, Chinese Āgamas, and Gāndhārī fragments that can now be compared. Second, the Sthavira branch's scholastic drive developed into the extensive Abhidharma systems examined in Essay 3. It generated different answers to the problem of "continuity without a self," including Sarvāstivāda's dharma entities, Sautrāntika's seeds, and Vātsīputrīya's pudgala. Third, the Mahāsāṃghika tendency toward a transcendent Buddha and an originally pure mind is often regarded as one clue to Mahāyāna's emergence. Essay 4 examines how much explanatory weight it can bear. The third connection remains substantially disputed among scholars, while the first two have documentary support.

2.5 turns to Abhidharma, the background to Essay 3. It examines how the Sthavira branch analyzed and arranged the Buddha's teaching, his "handful of leaves," into a precise and extensive system of dharmas, and what new problems the system itself produced.

References¶