Early Buddhist studies are familiar with the term vibhajjavāda, and our understanding has been put on a sounder footing by L. S. Cousins in his paper ‘On the Vibhajjavādins’. He treats the term as twofold, signifying both the teachings of the Buddha in general, and also the name of a specific Buddhist school, or set of closely related schools. The basic position would seem to be that the Vibhajjavādins emerged as one of the major early schools. The first division was between the Sthaviras and Mahāsaṅghikas. Then the disputes on the ‘person’ and ‘all exists’ produced respectively the Puggalavāda and Sarvāstivāda schools (or groups of schools, or philosophical movements). What remains is the Vibhajjavāda, which, due mainly to geographical separation, gradually differentiated into the Mahāvihāravāsins, Dharmaguptakas, Mahīśāsakas, and Kaśyapīyas, and perhaps others.
There is no doubt that certain sources, such as Bhavya II and III, clearly present such a group of Vibhajjavādin schools. It is less clear that this situation is relevant in the early period. And it is not clear at all that such a group was ever imagined by the Sinhalese. So we need to inquire into the use of the word in our Sinhalese commentarial accounts of the Aśokan period.
Cousins acknowledges that one of our earliest sources for the term is in the commentary to the Kathāvatthu. This is a version of the Third Council, where the good monks and Moggaliputtatissa reassure king Aśoka that the Buddha was a vibhajjavādin. There, the context is suggestive of the kind of ambiguity Cousins sees in the term:
‘The whole point of the story is that no-one can deny that the Buddha was a vibhajjavādin, since he is at least sometimes so portrayed in the canonical texts. Nor of course is it surprising if a leading figure of the Vibhajjavādin school asserts that he was a Vibhajjavādin. None of this gives us any reason to suppose that the Buddha would have been referred to in the third person as a vibhajjavādin prior to the adoption of the word as the name of a school.’
Actually, Cousins’ prose is itself ambiguous: the ‘leading figure of the Vibhajjavādin school’ (i.e. Moggaliputtatissa) used the term vibhajjavādin to refer to the Buddha, not to himself. The text as it stands does not preclude the possibility that the Buddha was referred to as a vibhajjavādin before the formation of a school of that name. In fact, I would say that the main thrust of the passage means just that. Indeed, the Buddha is referred to in the third person as a vibhajjavādin in the canonical text that Cousins has already quoted.
What Cousins is getting at, I think, is that the canonical sources are few and fairly minor. They apply only in specific contexts and speak of how the Buddha would respond when presented with certain questions. Thus they are an insufficient basis to form a general characterization of the Buddha as vibhajjavādin. Cousins therefore concludes that when certain texts choose this particular term to characterize the Buddha’s doctrine, this cannot be explained on the basis of the canonical texts, but must have occurred after the formation of a school called vibhajjavāda, which then tried to authorize itself by claiming that the Buddha was a vibhajjavādin.
But this just defers the argument: neither here nor anywhere else does Cousins attempt to explain why, given that vibhajjavāda is such a marginal term in the canon, should any school choose to call itself that. We therefore propose to re-examine the sources.
The Pali account of the Third Council has Aśoka asking the good monks what the Buddha taught (kiṁvādī bhante sammāsambuddhoti?) to which they reply the Buddha was a vibhajjavādin (vibhajjavādī mahārājāti). Notice the same, rather ambiguous suffix -vādī ends both phrases. This spans a spectrum of meaning, from ‘speaks’, to ‘teaches’, to ‘has a doctrine of’, to ‘adheres to the school teaching such a doctrine’. In this case, the king could hardly have meant: ‘What school did the Buddha belong to?’ Nor was he asking for a detailed exposition of the many teachings give by the Buddha in his career. He needed a concise, pithy summary of the Buddha’s key doctrine. The monks at the time would have been familiar with the Buddha’s skill in adjusting the teachings to time, place and person, and so would have chosen a message that was directly targeted to solving the urgent problem confronting the king.
But here the Mahāvihāra’s version of events, as recorded in the Samantapāsādikā and the Kathāvatthu-aṭṭhakathā,  and elsewhere takes another turn: after the settling of the dispute, the leading Elder, Moggaliputtatissa, is said to compose the Kathāvatthu and 1000 monks are chosen to authorize the Third Council adding this work to the Tripitaka. But let us compare these passages with the corresponding one in the Sudassanavinayavibhāsā.  Differences between the versions are highlighted.
In that gathering Moggaliputtatissa acting as the Elder refuted the wrong doctrines of followers of other religions. The assembly chose those knowledgable in the Tripitaka and the three-fold realization, numbering 1000 bhikkhus.
In that gathering the Elder Moggaliputtatissa, refuting other doctrines, spoke the Kathāvatthu treatise. And then from the bhikkhus reckoned as 6 000 000 were chosen bhikkhus who were memorizers of the Tripitaka, distinguished in the paṭisambhidas, endowed with the three-fold realization, etc., numbering 1000 bhikkhus.
In that gathering the Elder Moggaliputtatissa, regarding those issues that had arisen and those that would arise in the future, for the sake of dispelling all of them, using the method that had been given by the Teacher, the Tathāgata, arranged the matrix distinguishing 500 statements of one’s own school and 500 of the other schools. Having brought together 1000 statements he spoke this Kathāvatthu treatise, of futuristic character, for the sake of refuting other doctrines. And then from the bhikkhus reckoned as 6 000 000 were chosen bhikkhus who were memorizers of the Tripitaka, distinguished in the paṭisambhidas, numbering 1000 bhikkhus.
Notice that the Samantapāsādikā adds three phrases: the mention of the Kathāvatthu, the exaggeration of the number (elsewhere the Sudassanavinayavibhāsā mentions 60 000), and the mention of the paṭisambhidas. The Kathāvatthu commentary adds further details describing the Kathāvatthu itself, which one might expect. This addition refers to the legendary tale that the Buddha had designed the basic framework of the Kathāvatthu in order that Moggaliputtatissa should fill in the details. Interestingly, it says that the orthodox and heterodox views should be ‘divided’ (vibhajanto); as this passage follows immediately after the passage mentioning the vibhajjavāda, perhaps this offers a clue in interpretation. Notice that the Kathāvatthu-aṭṭhakathā loses the statement that the 1000 bhikkhus chosen to perform the Third Council all possessed the three realizations: thus the early Sutta and practice based ideal of an arahant is sidelined in favor of the Mahāvihāravāsin textual ideal.
All of these changes apparent in the Pali versions as compared with the Sudassanavinayavibhāsā are absolutely characteristic of the Mahāvihāra’s perspective. I cannot see any other reasonable conclusion than that the additions to the Samantapāsādikā and Kathāvatthu-aṭṭhakathā are all interpolations at a late date in the Mahāvihāra, presumably made by Buddhaghosa. It would seem that the original version of the Third Council did not mention the Kathāvatthu.
The Kathāvatthu is an extensive refutation of heretical views, but of Buddhist heretical views. Thus there is a decided tension in the story: are we supposed to see this account as a purification of the Sangha from non-Buddhist heresies (eternalism, etc.), or wrong interpretations of Buddhist teachings? Perhaps we are tempted to synthesize these perspectives; after all, the first and main debate in the Kathāvatthu is against the puggala, the ‘person’, who, in a suspiciously Self-like manner, is supposed to somehow exist outside the 5 aggregates and to pass on from one life to the next. No doubt there is something to this, as Buddhists, sometimes justifiably, often suspect ‘innovations’ of practice or doctrine to be ‘Hindu’ influences. This is perhaps suggested when the Kathāvatthu commentary ascribes the puggala controversy to: ‘In the sasana, the Vajjiputtakas and Saṁmitiyas, and many other teachers not belonging to the sasana.’
Yet the debate on the puggala would seem to primarily revolve around a tension within Buddhist doctrine. When the Buddha taught, he was largely surrounded by ‘Self’ religions, and of necessity had to emphasize ‘not-self’; that is, against those who would assert the absolute unity of the person, he emphasized that what we call a ‘self’ is an abstraction inferred from experience, motivated by fear of death and dissolution, but which, when we look for it in experience, cannot be found. Thus, against those who asserted to absolute primacy of unity, he proposed the contemplation of diversity, without, however, reifying that diversity into another absolute.
This is effective as a philosophical counter to self-theories, but leaves us having to seek an explanation for why we feel or experience a sense of ‘identity’: why, if there is no truly eternal core or essence, do we nevertheless feel as if we are a person? Certain indications in the canonical texts suggest ways of approaching this problem, but the schools were left to work out their own definitive solutions. For some schools, such as the Mahāvihāravāsins, the sense of identity was explained in terms of causal relations among disparate elements. But for the Puggalavādins this was not enough, so they attempted to ‘draw out’ certain Sutta passages as implying the existence of a ‘person’ (puggala) in some sense outside the five aggregates, which was, however, not the Self spoken of by the non-Buddhists. For them, this was a ‘middle way’ between the self-theories and the absolute ‘no-self’ of the Abhidhamma theorists.
Thus we are justified in thinking of the Puggalavāda schism as primarily an internal matter among Buddhists, and while not denying any connection with non-Buddhist teachings, would resist an attempt to simply ‘collapse’ the two issues we are presented with at the Third Council: the infiltration of non-Buddhist heretics, and the development of Buddhist philosophical ideas as debated in the Kathāvatthu. Our text makes no attempt at a synthesis of these perspectives, but rather leaves us with an impression of disparate, although perhaps related, agendas.
Given this situation, and given the flow of the text as preserved by the Mahāvihāravāsins, what role was played by the term vibhajjavādin? Why was this term chosen, and how was it useful at this time? How would it have served as a key to solving the king’s dilemma?
Cousins quotes and translates passages from the later Mahāvihāravāsin literature that define what vibhajjavāda means to them. They say, for example, that the Buddha was a vibhajjavādin because he distinguished the various senses in which he could be called ‘one who leads astray’ (i.e. he leads astray from unwholesome things); or he distinguished the kinds of pleasant feeling or the various kinds of robes to be cultivated or not (according to whether they conduce to the arising of wholesome states of mind).
But as Lamotte comments: ‘… that is a state of mind which is fitting for all Buddhist thinkers in general and it could not have served Aśoka in establishing the orthodoxy of the Aśokārāma monks and separating non-believers from the truly-faithful.’ The mere making of rational distinctions is never regarded by Buddhists as a distinguishing feature of their religion, or of their particular school.
For example, the Mahāvibhāṣā depicts Mahādeva, who it sees as the corrupt founder of the Mahāsaṅghika school, making subtle distinctions between the kinds of doubt an arahant might have or not have; or the kinds of ‘outflows’ an arahant might have or not have, and so on. This is exactly the kinds of distinctions meant by the general use of vibhajja, and they are entirely characteristic of the vibhajjavādins’ supposed enemies.
Or in non-Buddhist circles, we need only think of the Jains, whose cardinal philosophy is the anekantavāda, the doctrine of ‘not just one standpoint’. They hold that any truth may be seen from many different perspectives, so no one perspective can be privileged as ultimate. On the contrary, as Cousins points out, the Buddha himself, while sometimes using the method of distinguishing, in other contexts makes unequivocal (ekaṁsa) statements. Since such unequivocal teachings include the four noble truths, it could be seriously argued that the Buddha was an ekaṁsavādin.
The late Pali texts also, as shown by Cousins, use vibhajjavāda to distinguish the Mahāvihāravāsin school from others, claiming to be the only true vibhajjavādins, and specifically mentioning some doctrines of other schools. This perhaps includes the Sarvāstivāda term hetupaccaya, although this is unclear. More clear is the reference to ‘undefiled ignorance’, which was accepted by the Sarvāstivādins and others, and ‘non-communicating materiality’, which was accepted by the Vaibhāṣika Sarvāstivādins, and possibly others. But these doctrines are all advanced Abhidhamma topics, which, even if they were current at that early time, would have had little relevance to the king’s dilemma.
So we conclude that the meanings of the word vibhajjavādin proposed by Cousins based on the Pali canon and commentaries are not adequate to account for its use in the Third Council narrative.
So we are left with the problem: what did vibhajjavāda mean, and why was it relevant in the context of the Third Council? Let us recall the flow of the text. The non-Buddhist heretics assert various doctrines of the ‘self’; Moggaliputtatissa opposes them with the Buddha’s doctrine of vibhajjavāda; then the Mahāvihāra sources depict him as going on to teach the Kathāvatthu. Even if the Kathāvatthu was a later addition, the Mahāvihāra must have added it for some reason. The Kathāvatthu commentary, as we have seen, specifically says that the Kathāvatthu ‘distinguishes’ (vibhajanto) the heterodox and orthodox views, so perhaps it means to make some explicit connection between the Kathāvatthu and the vibhajjavāda.
Now, the Kathāvatthu discusses very many topics, many of which are trivial and are given little space, and far outweighing all other topics in the book is the first section, the discussion of the ‘person’. This is, as we have seen, the only main topic common to the Kathāvatthu and the Vijñānakāya, apart from the opposing positions on the ‘all exists’ thesis. It was clearly a difficult controversy, and despite the cool Abhidhamma dialectic, an emotional one.
In our present context, surely the emerging theme is this self/not-self debate. I would like to suggest that the term vibhajjavāda is used here to imply a critique of the non-Buddhist theory of Self. This would certainly fulfil the criteria we asked for earlier, that the term must evoke a pithy, essential aspect of the Buddha’s teaching in a way that would answer the challenge of the heretics.
The teaching of not-self has always been regarded as a central doctrine of the Buddha. A characteristic method used by the Buddhists to break down the false idea of self was to use analysis. In early Buddhism, the main method was to systematically determine those things which are taken to be the self, hold them up for investigation, and find on scrutiny that they do not possess those features which we ascribe to a self. Thus the five aggregates are described as forming the basis for self theories. But on reflection, they are seen to lead to affliction, which is not how a self is conceived, so they fail to fulfil the criteria of a self. In the Suttas, this method was exemplified by the disciple Kaccāyana, who was known as the foremost of those able to analyse (vibhajjati) in detail what the Buddha taught in brief; the Dīpavaṁsa says that he filled that role in the First Council. 
This analysis, or vibhaṅga, was gathering momentum during the period of the Third Council. Indeed, the basic text is called, in the Mahāvihāravāsin version, the Vibhaṅga; the Sarvāstivāda version is the Dharmaskandha, and the Dharmaguptaka version is the Śāripūtrābhidharmaśāstra. These all stem from an ancient phase of Abhidhamma development, collecting the ‘analytical’ Suttas, primarily arranged according to the topics of the Saṁyutta Nikāya/Āgama, and elaborating them with varying degrees of Abhidhammic exegesis.
So it would make perfect sense in our narrative for vibhajjavāda to represent the Abhidhamma movement as an analytic approach to Dhamma in general, and as a critique of the ‘self’ in particular. It would also seem appropriate to describe the Buddha as a vibhajjavādin, equivalent to saying he was an anattavādin. This interpretation must remain tentative, since it cannot be backed up with a clear statement from the texts. Yet, as we have seen, the definitions of vibhajjavāda that we are offered by the texts are inadequate to explain the usage by the Mahāvihāravāsins in their own texts: they are late, or irrelevant, or derived from a different school. If our speculations have any value, it would seem that the prime target of the polemics in this passage are not the Sarvāstivādins, but the non-Buddhist Self theorists, and perhaps by implication the Puggalavādins.
But there is another, quite different, aspect of the term vibhajjavāda that is suggested by our sources. When the troubles in the Sangha proved intractable, king Aśoka asks his ministers who can resolve the problems. They suggest Moggaliputtatissa, and so the king orders that he be fetched on a boat. Aśoka dreams that a white elephant will arrive and take him by the hand; accordingly, the next morning Moggaliputtatissa arrives, and, wading in the water to help him, the king and the Elder clasp each others’ hand. This is a serious breach of royal taboos, and the guards draw their swords threateningly before being restrained by the king.
All this acts as a significant mythic precursor to the Third Council. With the exception of the king’s dream, these events closely mirror events surrounding Upagupta; Moggaliputtatissa and Upagupta share such a close mythos that several scholars have seriously argued that they are the same monk. The only significant difference between the two in this instance is the dream sequence, which seems to echo the dream of the Buddha’s mother before she was born, suggesting that Moggaliputtatissa, like Upagupta, is a ‘second Buddha’. The white elephant is also one of the seven ‘treasures’ of a Wheel-turning Monarch.
But next is another episode, which as far as I can see has no parallel with Upagupta. The king asks to see a miracle of psychic power: he wants Moggaliputtatissa to make the earth quake. The Elder asks whether he wants to see the whole earth shake, or only a part of it, saying it is more difficult to make only part shake, just as it is more difficult to make only half a bowl of water tremble. Accordingly, the king asks to see a partial earthquake, and on the Elder’s suggestion, he places at a league’s distance in the four directions a chariot, a horse, a man, and a bowl of water respectively, each half in and half outside the boundary. The Elder, using fourth jhana as a basis, determines that all the earth within a league should tremble: accordingly it does so, with such precision that the inside wheel of the chariot trembles, but not that outside the boundary, and the same for the horse, the man, and even the bowl of water. It was this miracle that convinced Aśoka that Moggaliputtatissa was the right man to stabilize the sāsana.
The crucial value here is the precision with which the Elder can resolve his psychic abilities, dividing the earth as if with a razor. This concern for precision, orderliness, and clean boundaries is a characteristic of the Mahāvihāravāsin school, which evinces a philosophical revulsion for grey areas, graduations, and ambiguities.
For example, while other schools asserted that rebirth took place through a gradual transitional phase called the ‘in-between existence’, the Mahāvihāravāsins would have none of that, declaring that one life ends and the next begins in the following moment. Or while many schools spoke of a gradual penetration to the Dhamma (anupubbābhisamaya), the Mahāvihāravāsins developed the idea that penetration happens all-at-once (ekābhisamaya). Similarly, when explaining the ‘Twin Miracle’ where the Buddha was supposed to simultaneously emit both water and fire: the point of the miracle would seem to be the fusion of opposites, but for the Mahāvihāravāsins there is no fusion, the miracle is an example of how fast the Buddha could advert between a water-kasiṇa and fire-kasiṇa, flashing back and forth to create the illusion of simultaneity.
This notion of a momentary flickering back and forth to explain what the text would appear to present as synthesis is found elsewhere, too. In satipatthana, the meditator is supposed to contemplate ‘internally’ then ‘externally’, then ‘internally/externally’. While the Suttas see the ‘internal/external’ contemplation as the comprehension that there is no fundamental difference between the two, the Mahāvihāravāsins explained it as a rapid flicking back and forth. Similarly, the Suttas speak of ‘samatha and vipassana yoked together’, evidently imagining a concurrent balance of these qualities in a meditator’s consciousness. While the Mahāyāna sources seem to retain this understanding, the Mahāvihāravāsins again speak of a rapid alteration between the two.
So I suggest that this admittedly ill-defined sense of ‘clear-cut-ness’ that we see in the Mahāvihāravāsins may also be implied in the usage of vibhajjavādin.
There is one final implication in the word vibhajjavādin in this account. One of the most dramatic episodes concerns Aśoka’s initial attempt to heal the problems in the Sangha. He instructs a minister to go and order the monks to do uposatha. The minister is told by the good monks that they refuse to do uposatha with the heretics. The minister, misunderstanding Aśoka’s intention, starts beheading the obstinate monks. He only stops when it he realizes that the next monk in line to have his head chopped off is none other that Tissa, the king’s brother. He returns to inform Aśoka, who is understandably seized by remorse, rushes to apologize to the monks, and asks whether he is to be held karmically responsible. The monks tell him different stories: some say he is to blame, some say he and the minister share the blame, while some say that only acts done intentionally reap a karmic result – as he had no intention there is no blame.
But none of them can assuage his doubt. Only the appearance of Moggaliputtatissa can do this. The Elder is then sent for, and after his arrival in the boat and subsequent demonstration of his psychic powers, the king is able to accept his explanation: there is no intention, therefore there is no guilt. This episode reminds us of the spectacular State visit by Ajātasattu to the Buddha, where he similarly confessed to a great crime and was comforted by the Buddha. In both cases the king was unable to find peace of mind until hearing the Dhamma from the right person.
I suggest that in this careful analysis of the distinction between physical and mental acts we see another possible meaning of vibhajjavādin. This was an important doctrinal position that clearly marked off the Buddhists from otherwise similar contemporary groups such as the Jains. We have seen that Mahādeva similarly invokes such a distinction to justify his acts.
Thus vibhajjavāda might have a variety of meanings in this context. Perhaps we should not seek for a definitive answer. As a mythic text, the passage is evoking a style, an atmosphere for the school, not laying down definitions. It may be that we can go no further than to explore various possibilities. After all, the school itself did not try to close off the specific denotation of the word. But the important conclusion of this discussion is that we can find plenty of implications in the term vibhajjavāda, whether those explicitly offered by the tradition, or those speculatively inferred from context, that do not involve sectarian differences. This stands in marked contrast to the often-assumed conception of vibhajjavāda as the opposite of sarvāstivāda, which we examine next.
 Cousins uses the term Tambapaṇṇiya (Those from the Isle of Tambapaṇṇa) to refer to the Sinhalese school that today we call ‘Theravāda’. I prefer to use Mahāvihāravāsin, as it more clearly differentiates the ‘Theravādins’ from the later Sinhalese schools, who might equally be called ‘Tambapaṇṇiya’. The views accepted as ‘Theravādin’ are those authorized by the Elders of the Mahāvihāra; the large number of dissenting voices recorded in the commentaries show that the ‘orthodox’ views were at no time accepted in toto by all the monks of Sri Lanka.
 Kathāvatthu Aṭṭhakathā, pg. 7; Law, 6
 Cousins, On the Vibhajjavādins, 138
 AN 10.94 at AN v.189f
 Cousins (On the Vibhajjavādins, 171 note 73), on the contrary, believes that this is exactly what the the ‘underlying reference’ to the question was. Hence he does not translate the phrase according to what he admits is the meaning in the Sutta passages: ‘What does the Buddha teach?’ In such remarks we see the distorting effects of reading sectarian agendas into Aśokan passages.
 Samantapāsādikā 1.61: tasmiṃ samāgame moggaliputtatissatthero parappavādaṃ maddamāno kathāvatthuppakaraṇaṃ abhāsi. tato saṭṭhisatasahassasaṅkhyesu bhikkhūsu uccinitvā tipiṭakapariyattidharānaṃ pabhinnapaṭisambhidānaṃ tevijjādibhedānaṃ bhikkhūnaṃ.
 Kathāvatthu-aṭṭhakathā 7: tasmiṃ samāgame moggaliputtatissatthero yāni ca tadā uppannāni vatthūni, yāni ca āyatiṃ uppajjissanti, sabbesampi tesaṃ paṭibāhanatthaṃ satthārā dinnanayavaseneva tathāgatena ṭhapitamātikaṃ vibhajanto sakavāde pañca suttasatāni paravāde pañcāti suttasahassaṃ āharitvā imaṃ parappavādamathanaṃ āyatilakkhaṇaṃ kathāvatthuppakaraṇaṃ abhāsi. tato saṭṭhisatasahassasaṅkhyesu bhikkhū uccinitvā tipiṭakapariyattidharānaṃ pabhinnapaṭisambhidānaṃ. Translation at Law, 7.
 E.g. Dīpavaṁsa 6.40: Maddivā nānāvādāni nīharivā alajjino, Sāsanaṁ jotayivāna kathāvathuṁ pakāsayi. Also see Dv 6.55, 56.
 目乾連蘊 (CBETA, T26, no. 1539, p. 531, c29)
 於集眾中。目揵連子帝須為上座。能破外道邪見徒眾。眾中選擇知三藏得三達智者一千比丘 (CBETA, T24, no. 1462, p. 684, b9-11)
 Cp. pg. 65, note 104 above.
 Kathāvatthu Aṭṭhakathā 9
 Cf. SN 12.48: ‘All is oneness: that is the third cosmological speculation… All is diversity: that is the fourth cosmological speculation…’
 Lamotte, History of Indian Buddhism, 274
 Cousins, On the Vibhajjavādins, 134
 Undefiled ignorance would also seem to relate one of the five points.
 Dīpavaṁsa 4.9
 Sudassanavinayavibhāsā: CBETA, T24, no. 1462, p. 683, b21-c18
 Sudassanavinayavibhāsā: CBETA, T24, no. 1462, p. 683, c22-p. 684, a10. This follows the Pali on every detail, except the distance is 4 yojanas. But 1 yojana at CBETA, T53, no. 2121, p. 179, a24.