Sanghabheda vs. Nikayabheda



When writing Sects & Sectarianism, I tried to account for the modern critical assessments of the evidence as best I could. However, since this is a part-time project done amid a busy schedule, it's difficult to keep up with everything. Just recently I came across an article by Heinz Bechert: 'The Importance of Aśoka's so-called Schism Edict.' (Indological and Buddhist Studies. Ed. L. A. Hercus et al. Canberra: Faculty of Asian Studies (ANU), 1982. 61-68.) I had addressed a number of Bechert's articles in S&S, and in addition his ideas had gained a certain currency among academics, but this was the first time I saw his own presentation of some crucial ideas, in particular some that are contrary to the perspectives developed in S&S. In the interests of completeness, I here record a few notes addressing the main aspects of his little essay.

The first paragraph of Bechert's essay repeats the thesis, originally attributed to H. Kern, that 'a review of all the testimonies available leaves no doubt that the assembly at Pāṭaliputra was a party meeting, from which the Mahāsaghikas were excluded'. Kern was writing in 1898, and his claim to have assessed 'all the testimonies available' is no longer true. In particular, as I pointed out in S&S, the Sudassanavinayavibhāsā omits the crucial reference to the Kathāvatthu in the Third Council. Here and below Bechert appears not to be aware of this (despite the fact that he was writing long after the translation of the Sudassanavinayavibhāsā had been published). The overall narrative of this and the Samantapāsādikā makes no reference to the existence of schools of Buddhism at this point. Of course, the commentary to the Kathāvatthu asserts the sectarian nature of this meeting, as one would expect: but this is just comprised of parts of the Vinaya commentary combined with a passage from the Dīpavaṁsa, no doubt compiled for the purpose of authorizing the Kathāvatthu as a sectarian document. In addition, the Mahāsaṅghikas themselves in their Śāriputraparipṛcchā claim that the schism was much later, an inconvenient datum also ignored by Bechert. There is thus considerable doubt as to whether the assembly at Pāṭaliputra was a party meeting: in fact it looks more like a general assembly that purified the Sangha, in the era when schools had not yet arisen.

This then demands a reassessment of Frauwallner's claim that this Council and the schismatic Council described in the northern traditions (e.g. Vasumitra) were two separate events that took place in the time of Aśoka. As I argued in S&S, there is every reason to believe that Vasumitra's Council was ascribed to the reign of Aśoka purely in order to claim Aśoka's authority for the Kaśmīr Vaibhāṣika Sarvāstivādins, and there is no reason to assume that it took place in this reign. (In fact, of the three major sources for this Council - Vasumitra, Mahāvibhāṣā, Bhavya III - only Vasumitra mentions Aśoka). We are thus left with one general Council in the time of Aśoka, the main purpose of which was to expel fraudulent pretenders to the robe.

Bechert explains that saṅghabheda as defined in Vinaya is always a matter of a disagreement between two groups of monks as to holding the uposatha within the same monastic boundary (sīmā). His comments on saṅgharāji and saṅghabheda are not quite correct. He says both of these result if there are two sanghas of minimum four bhikkhus each. In fact, saṅgharāji happens with lesser numbers - up to three on one side and four on the other - while saṅghabheda (or literally, 'saṅgharāji and saṅghabheda') happens with four on each side; in addition, one bhikkhu makes the formal statement, so a total of nine is necessary for saṅghabheda.

Bechert then goes on to try to establish his core thesis that saṅghabheda is always exclusively a matter of Vinaya. The obvious objection to this is that the basic passage says the Sangha may be divided on the basis of Dhamma or Vinaya. But Bechert points out that 'dhamma' in Vinaya contexts often refers to Vinaya. Fair enough, the meaning of these words is variable, and dhamma is often used in the sense of a rule or procedure of Vinaya. But the stock passage quoted here ('This is Dhamma, this is Vinaya, this is the dispensation of the Teacher') surely is used in a general sense to mean the entirety of the sasana. Indeed, just below the passage Bechert quotes, the Vinaya text then explains in more detail the bases for schism: misrepresenting what the Buddha spoke and uttered as not spoken and uttered by the Buddha and vice versa. This obviously refers to the Buddha's words in general. This is confirmed by the Samantapāsādikā in the context of the Third Council, which, harking back to the just-quoted passage from the Vinaya, says that the heretics each declared their own speculative views (diṭṭhigatāni), saying 'This is Dhamma, this is Vinaya'. It is here, surely, in a Vinaya commentary that we should seek support for the idea that schism happens only due to Vinaya, but this is not the case. We are on solid ground, then, in following the majority opinion of Buddhists of all schools and all times and accepting that the basis for schism can be either doctrine or discipline. This is not an unwarranted reading of Christian Church history into Buddhism - I for one have no real knowledge of what caused schisms in the Christian Churches, but I know full well that Buddhists frequently regard doctrinal differences as real and important.

The fact remains, and Bechert is quite right to emphasize this, that the procedure for schism as such is presented in the Vinaya as a Vinaya issue. This is important, but the significance of it for Buddhist history must be carefully considered. Immediately after this is the passage where Upāli asks the Buddha under what circumstances one who splits the Sangha is condemned to hell for an aeon. While western scholars might tend to overlook such matters, they are of great interest to devoted Buddhists, especially if they are wondering whether they themselves might be schismatic. As is well known, the basic statement on this matter was made in reference to Devadatta. The definition here basically generalizes this case, saying that the schism has to be caused by someone deliberately and maliciously misrepresenting Dhamma or Vinaya for the sake of causing a schism. My feeling is that by making schism in this sense legalistic and narrow, this passage functions as an 'out' for those who may be afraid that they or their party could be considered schismatic. The result of this is that in the history of the schools, what we call schism is rarely described as 'saṅghabheda' in this sense (the great exception being the Dīpavaṁsa). Thus the schools probably did not arise because of saṅghabheda. All this means is that the Vinaya passages relating to saṅghabheda are of limited use in determining how the schools actually arose.

We also need to bear in mind the fact that in Buddhist circles generally, Vinaya terms such as saṅghabheda are frequently used in a more general, non-technical way, and that the majority of Buddhists find no problem with this, however much the Vinaya experts might bemoan such confusion. We are thus not warranted to see in Aśoka's usage (or the Dīpavaṁsa, for that matter) a clear indication that saṅghabheda in the strict Vinaya sense has actually occurred. This is in addition to the fact that the Aśoka edicts, of course, never assert that a schism has happened, they assert that a unification has happened.

Bechert goes on to refer to the Dīpavaṁsa as the most archaic version of events during Aśoka. This is unfortunate, since the Dīpavaṁsa never had a very good claim for antiquity, and an even worse one for coherence (meaning not that some early elements are not preserved, but that the existing work is a late patchwork, barely pre-dating Buddhaghosa). But in the light of the Chinese Sudassanavinayavibhāsā, it is clear that the Sinhalese Vinaya commentary should be regarded as an earlier source, updated slightly in a sectarian direction by Buddhaghosa. The opinion of those who have studied the matter is that the Sinhalese commentaries were passed down for centuries, the main work on them appears to have stopped about 100 C.E., and Buddhaghosa was in the main acting as editor and compiler of much older material. As it happens, the comparison between the two Sinhalese Vinaya commentaries is our only way of directly assessing this picture, and by and large it seems to hold good. It is certainly unwarranted to dismiss the commentaries as the fabrication of Buddhaghosa's time - the mere fact of the existence of the Chinese translation tells us the basic source must predate Buddhaghosa, and the nature of such sophisticated texts is to evolve slowly over a long time; in addition the archaeological evidence is in its favor. Bechert argues that the Samantapāsādikā and the Mahāvaṁsa are later than the Dīpavaṁsa, but he ignores the Sudassanavinayavibhāsā. His main objections are in fact levelled against the Mahāvaṁsa, which I would agree is largely a politicised remodelling of the Dīpavaṁsa, and unlikely to yield any independent historical evidence for the early period.

Bechert comes closest to what I believe to be the real key to understanding in the following passage; but then he shies away from the obvious conclusion. My own response is added.

  • 'M. Walleser has already noted that there is an inner contradiction within the Pali records when it is first said that only Tīrthikas [non-Buddhist ascetics] are excluded from the Sangha, and not the least reference is made to the followers of different Buddhist sects.'

Indeed. This is the overriding characteristic of the whole narrative leading up to the Third Council, and in my opinion the expulsion of tīrthikas is clearly what the Council was about.

  • 'In the record on the Council, on the other hand, the Katthāvatthu [sic] is said to have been published in order to destroy wrong doctrines.'

This ignores, as usual, the decisive fact that the Sudassanavinayavibhāsā omits mention of the Kathāvatthu. Clearly the mention of the Kathāvatthu is merely a late and clumsy interpolation of a Buddhist sectarian agenda into a story that was originally about the purification of the Sangha from non-Buddhists.

  • The Katthāvatthu [sic], however, does not refer to non-Buddhist teachings only, but mainly to views held by other Buddhist sects. (see M. Walleser, Die Sekten des alten Buddhismus, Heidelberg 1927, p. 11)

This is a highly misleading way of putting the matter: the Kathāvatthu in fact does not refer to non-Buddhist heresies at all, but solely to internal disputes among Buddhists. Of course, some of the views might also be held by non-Buddhists, but the terms of the debate are purely Buddhist.

Bechert asserts, in conformity with all other scholars I am aware of, that: 'There can hardly be any doubt, however, that the formation of nikāyas had begun in a period before Aśoka's reign'. As usual, no evidence is offered in support of this claim: I am still trying to find out why scholars believe this so whole-heartedly. The evidence to the contrary is adequately dealt with in S&S, so no need to re-till that soil here. But we may note that Bechert's core thesis - the difference between nikāyabheda and saṅghabheda - is used to explain the incongruity of the assumed fact that although the sects already existed, Aśoka made no reference to them. This thesis is at best a slightly uncomfortable attempt to deal with a seriously disconcerting clash between fact and theory: much better to simply accept that the sects did not exist and watch the problem disappear into the mist.

One further problem with Bechert's work is his claim that the side by side existence of three nikāyas (Mahāvihāra, Abhayagiri, and Jetavana) was not considered a problem in Sri Lanka. This is in support of his belief that the 'schism edict' did not try to unify the Buddhist sects as a whole, merely the individual monasteries. He draws the parallels with later Sri Lankan tradition: Aśoka did not find the existence of several competing nikāyas to be a problem, and so neither did the later Sri Lankan kings. But his claim founders on the statement attributed to Parakkamabāhu, both in the Mahāvaṁsa and in inscription, that the great efforts of kings of old to unify the Sangha had failed due to the bickering of the monks. Clearly, the existence of different sects was felt to be a serious problem. And Bechert continues by asserting the parallels with the current situation in Sri Lanka, where several nikāyas exist side by side. Again, this is highly misleading: the existing Sri Lankan nikāyas are indeed split solely on matters of Vinaya, ordination lineage, administration, etc.; but in the past the three sects clearly had different texts and different doctrines. True, the chronicles mention Vinaya differences, but just as often or more so they mention the doctrinal and textual differences as primary.

So in conclusion we must express severe reservations about much of Bechert's analysis. However, the distinction between saṅghabheda and nikāyabheda remains a valid one. It is true that saṅghabheda is defined in a narrow, technical way in Vinaya. But we are not justified in reading this legalistic meaning of saṅghabheda into every context where the word is used: probably it was used, then as now, in a variety of senses both technical and colloquial. The actual historical emergence of schools (nikāyabheda) probably happened due to a variety of forces, among which Vinaya issues were only one factor. In support of this I would point out that none of the schools is named after a Vinaya topic, whereas most of them are named after a doctrine, a place, or a person. This surely must give some clue as to the basic feature felt by the schools themselves to define their community. We cannot accept Bechert's conclusion that the basic force behind sectarian formation was Vinaya differences and that dogmatic differences, when they arose, came later. The fact that Vinaya texts treat schism as a Vinaya issue is unsurprising, and Bechert is clearly wrong in asserting that 'dhamma' in such contexts must mean Vinaya. We are also not justified in treating the Aśokan edicts as if they ignored the existing nikāyas; rather, the nikāyas had not yet emerged.