Dhamma or Vinaya?


We have examined at some length the more important sources dealing with the first schism, and it is evident that it is not possible to fully resolve the differences. Clearly there were multiple forces acting to break the Sangha apart, and it is not easy to tell which may have been the decisive factor in provoking the first schism. There have been various attempts by modern scholars to interpret this evidence. I will not review these in detail, but will focus on the most recent trend. L. S. Cousins and Charles Prebish have written on this topic, and while their perspective differs somewhat, their overall theses are quite similar. I have addressed several problems with their ideas incidentally above, and here will explain why I cannot accept their explanations.

While both Cousins and Prebish explicitly reject the Dīpavaṁsa’s account,[1] the version of events they come up with is strangely familiar. The schism did not happen immediately after the Second Council, but a few years after. It did not happen because of Vinaya laxity on behalf of the proto-Mahāsaṅghikas, but because of Vinaya strictness by the proto-Sthaviras. It seems to me this is just the Dīpavaṁsa’s theory exhumed.

Many of the ideas on which Cousins and Prebish base their theories are not supported by a careful reading of the texts. Cousins says that the Śāriputraparipṛcchā: ‘… sees the Mahāsaṅghikas as the conservative party which has preserved the original Vinaya unchanged against reformist efforts to create a reorganized and stricter version’.[2] Similarly, Prebish says: ‘… if the Buddhist community was plagued by the genuine threat of saṅghabheda in the aftermath of the council of Vaiśālī… it may well have been both logical and reasonable to tighten the monastic code by the addition of a number of rules…’.[3] In fact the Śāriputraparipṛcchā speaks neither of an increase in the number of rules nor of a stricter discipline.

Neither Cousins nor Prebish considers the narrative context of the Śāriputraparipṛcchā: the texts were endangered under Puṣyamitra; the texts were saved by taking them to Tusita; the texts were retrieved safely; the texts were housed in a pavilion; the texts were revised by a bhikkhu conceited with his learning. Nowhere is there a hint of problems with discipline. And indeed Prebish has already stated: ‘It is certainly not logical to assume brevity equals disciplinary laxity.’[4] Then how can it be that expansion equals strictness?

To illustrate this, compare the following statement from the Buddha to Mahākassapa:

‘So it is, Kassapa, when beings are in decline and the true Dhamma is disappearing, there are more training rules and fewer bhikkhus established in deep knowledge.’ (SN 16.13)

This suggests that the more rules there are, the less spiritual attainment, and one can only presume, more disciplinary laxity there is. This is simple common sense: Vinaya rules are only promulgated in a community with disciplinary problems. If bhikkhus are enlightened, or at least practicing sincerely, there is little or no need for a disciplinary code. For this reason, the Buddha explicitly refused to lay down a Vinaya, even when begged to do so by Sāriputta:

‘Now is the time, Blessed One! Now is the time, Fortunate One! May the Blessed One make known a training rule for disciples and recite the pātimokkha, so that this holy life shall last for a long time!’

‘Wait, Sāriputta! Wait, Sāriputta! The Tathāgata will know the time for that. The Teacher will not make known a training rule for disciples or recite the pātimokkha until certain defiling dhammas manifest here in the Sangha. But Sāriputta, when certain defiling dhammas manifest here in the Sangha, then the Teacher will make known a training rule for disciples and recite the pātimokkha for the resistance of those defiling dhammas…’ (Pali Vinaya 3.9)

We are thus perfectly justified in thinking that a Vinaya with more rules is indicative of a community with more disciplinary problems. This remains the case today. In a small monastery with a few sincere bhikkhus practicing together, there is little need for disciplinary measures or restraints beyond the basic Vinaya. Only in the large monasteries, which attract many monastics of differing motivations, is there a need to promulgate extra controls on conduct.

Cousins and Prebish treat the Dīpavaṁsa and the Śāriputraparipṛcchā as similar in that both attribute the root-schism to Vinaya rather than Dhamma.[5] We have shown that this position is incorrect, and stems from in part a misreading of the sources and in part a failure to distinguish the difference between a dispute in Vinaya practice and the redaction of Vinaya texts. Ironically, while the Dīpavaṁsa and the Śāriputraparipṛcchā are not connected by attributing the schism to Vinaya, they are connected by attributing the schism to textual redaction.

Cousins and Prebish also develop similar arguments to dispose of the idea that the schism was due to Dhamma, i.e. the five points. They both agree that the dispute over the five points was not fundamental to the Mahāsaṅghikas, and was rather a doctrine that was propagated later by Mahādeva II in the southern Andhaka schools. How plausible is this idea?

The support for this thesis is twofold. First, certain sources mention Mahādeva II in association with the formation of Mahāsaṅghika sub schools in the Andhra region. These include the Śāriputraparipṛcchā, Vasumitra, and Bhavya III. This tradition, then, is quite widespread, although we note that these sources are all closely related, and may not constitute independent evidence. In addition, certain sources associate the activities of this Mahādeva II specifically with the five theses. In fact, only the two later Chinese translations of Vasumitra make this association.[6] The consensus position does not associate Mahādeva II with the five theses, and the natural explanation would be that this became a part of Vasumitra’s later translations due to the growing notoriety of Mahādeva. We also notice that Vasumitra’s primary theory is that the five points, taught not by Mahādeva, at the time of Aśoka in Pāṭaliputta, were the cause of the root schism. It is bizarre, if not perverse, to use the later translations of his work to support the theory that the five points, taught by Mahādeva after Aśoka in Andhra, were the cause of subsequent schisms.

This position rests on slim textual grounds, but is further buttressed by the attempt to show a difference in doctrine between the southern and northern Mahāsaṅghikas. If it can be shown that the southern Mahāsaṅghikas held the five points but the northern schools did not, this would lend substantial support to the suggestion by the later translations of Vasumitra that Mahādeva II propagated these theses in the Andhra region.

Prebish does this by geographically analysing a series of theses attributed to the Mahāsaṅghikas, which are supposedly connected with the fifth point, that the path can be aroused by exclaiming ‘Aho, what suffering!’[7] There is some uncertainty as to this point, so we are not surprised that Prebish quotes a series of varying possibilities. But in fact his theses 6-12 have nothing to do with the fifth point, except they include the word ‘suffering’. These are obviously irrelevant and should not have been introduced here. They were apparently mentioned because they are geographically associated with the southern regions, and thus support Prebish’s argument. Leaving them aside, Prebish’s points 1-5, which are connected with the vocal utterance of ‘Aho, what suffering!’ are attributed to the Mahāsaṅghikas generally; only his point 3 is specific to the Andhakas. Thus Prebish’s data establishes clearly that the fifth point is connected with the Mahāsaṅghikas in general, not specifically the southern schools.

Cousins also suggests a connection between the doctrines attributed to the Andhakas and the later introduction of the five points by Mahādeva in Andhra.[8] But wisely he does not make much of this, since the weakness of the argument is obvious. Cousins is primarily working from the Kathāvatthu and its commentary, and these works betray their Sinhalese connection by attributing half the theses to the Andhakas. This tells us only that the Mahāvihāravāsins learnt of such theses from the Andhakas, and tells us nothing of what the other Mahāsaṅghikas believed. The Kathāvatthu commentary does not pretend to give exhaustive lists of schools for each thesis. For example it says that the first of the five points was held by some: ‘such as, these days, the Pubbaseliyas and Aparaseliyas.’[9] Perhaps a detailed examination of these points in conjunction with the corresponding northern sources might yield something of value, but to my knowledge this has not been undertaken.

Cousins’ more important argument is his detailed philosophical reconstruction of the history of the five points. His historical approach is sound: the Kathāvatthu is the earliest source, so we should see what this says, without reading into this work presuppositions deriving from later listings of the five points. The Kathāvatthu, of course, does not present us with a neat list of ‘five points’, although it does mention the points, they are listed one after the other, and are treated in a similar way. We cannot be sure, however, what the original group was.

Cousins’ analysis is insightful, its main merit being to display the inner logic of these points, otherwise presented as bare axioms. But I am cautious about using the results of such philosophical inquiries as the basis for historical inferences. It seems to me that such reconstructions can proceed along many different lines, and it is not easy to extrapolate from logical to historical development.

I also cannot accept his conclusion that the five points must have originated as startling paradoxes to stimulate Abhidhammic discussions. It seems to me much more likely that the Mahāvibhāṣa’s account is realistic here, and the five points arose due to the disparity between a teacher’s assumed attainment and his conduct.

Further, Cousins’ main argument rests on the evident close connections in form between the five theses as presented in the Kathāvatthu and the preceding thesis, that an arahant is subject to falling away from his attainment.[10] Cousins takes this as evidence that these were originally part of the same discourse. While the formal coincidence is striking, I don’t think this tells us anything about the origins of the five points. Schematic formalism is a universal characteristic of the Abhidhamma. There are countless examples of doctrinal terms or sets that originated from quite distinct sources, yet become grist for the same Abhidhammic mill. So, while clarifying the philosophical logic of the five points, I don’t think Cousins has established a clear case for the proposition that the five points were developed later among the Andhaka schools.

In conclusion, then, we can say that the theory that the five points are not at the root of the Mahāsaṅghika schism is supported only by the later translations of Vasumitra. The geographical evidence presented by Prebish in fact supports the opposite view, and Cousins’ arguments are too textually and philosophically speculative to be conclusive.

Contrary to all these views is the evidence which we have reviewed earlier that the first of the five points is clearly implied in the Mahāsaṅghika Vinaya. This Vinaya was obtained by Fa-xiang in Pāṭaliputta, so must represent the central Mahāsaṅghikas, not the southern schools. Of course, this is only one of the five points, but as the others have little to do with Vinaya it is unlikely we will find anything relevant there.



Works Cited

Cousins, L. S. "The 'Five Points' and the Origins of the Buddhist Schools." Buddhism: Critical Concepts in Religious Studies. Ed. Paul Williams. London: Routledge, 2005. 52-83.

Cousins, L. S. "Pali Oral Literature." Buddhism: Critical Concepts in Religious Studies. Ed. Paul Williams. London: Routledge, 2005. 96-104.

Nattier, Jan and Charles S. Prebish. "Mahasanghika Origins." Buddhism: Critical Concepts in Religious Studies. Ed. Paul Williams. London: Routledge, 2005. 199-228.

Prebish, Charles S. "Saiksa-Dharmas Revisited." Buddhism: Critical Concepts in Religious Studies. Ed. Paul Williams. London: Routledge, 2005. 186-198.

Warder, A. K. Indian Buddhism. Delhi: Motilal Barnasidass, 2004.


[1] (Cousins, Pali Oral Literature 104), (Nattier and Prebish 204-205). Note, however, that Nattier and Prebish quote Lamotte 288 (315 in French edition) in saying that the later Mahāvihāra works the Mahāvaṁsa and Nikāyasaṅgraha do not follow the Dīpavaṁsa’s theory that the Vajjiputtakas = Mahāsaṅghikas. Lamotte’s mistake was already noted in (Warder 207).

[2] (Cousins, The 'Five Points' and the Origins of the Buddhist Schools 56)

[3] (Prebish 194)

[4] (Nattier and Prebish 204)

[5] (Nattier and Prebish 201)(Cousins, The 'Five Points' and the Origins of the Buddhist Schools 57, 67)

[6] Prebish appears to be mistaken in asserting that Bhavya III also attributes the five points to Mahādeva II. (Nattier and Prebish 216).

[7] (Nattier and Prebish 217-218). They rely on Bareau’s Les Sectes bouddhiques, which I do not have access to, so unfortunately I cannot check the original sources, but simply treat the data as they have presented.

[8] (Cousins, The 'Five Points' and the Origins of the Buddhist Schools 65)

[9] KvA 0.54: seyyathāpi  etarahi pubbaseliyā  ca  aparaseliyā  ca

[10] (Cousins, The 'Five Points' and the Origins of the Buddhist Schools 59)