8. Dharmagupta: Greek Missions



As recorded in the Sri Lankan chronicles, one of the missions traveled to Aparantaka in the west of India (Gujarat). This was led by a monk called Yonaka Dhammarakkhita, a most intriguing individual.

While most of the monks mentioned in the Pali sources for the Third Council come to us with a name and a few details of their missions, Yonaka Dhammarakkhita is singled out for special honor as the teacher of Aśoka’s brother Tissa. It seems that Tissa’s mind was already inclining towards the Dhamma. While roaming in the forest he saw the Elder seated in meditation, being fanned by a magnificent bull-elephant with the branch of a sala tree. A longing to join the Sangha arose in him, and perceiving this, Dhammarakkhita rose into the air and descended at the lotus lake in the Aśokārāma monastery in Pāṭaliputta. He bathed, all the while leaving his robes hanging in mid-air. Seeing this, Tissa was so inspired he asked to join the Sangha immediately, taking Dhammarakkhita as his preceptor.[1]

When the missions were sent out, Dhammarakkhita went to Aparantaka, in the west of India, where he taught the discourse on the Great Mass of Fire and made 37000 converts, with 1000 men and 6000 women ordaining.

 Yonaka is related to ‘Ionia’, and is used in Indic texts for any Westerner, especially the Greeks. Alexander the Great had led his Greek army into north-west India shortly before Aśoka. He built several cities called ‘Alexandria’, one of which was apparently Yonaka Dhammarakkhita’s home town. Although he is said to have been sent to Aparantaka, in the west of India, this is a general term and elsewhere it is clear that Dhammarakkhita stayed in Greek areas.[2]

The second part of his name is just as interesting. The words rakkhita and gupta have exactly the same meaning: ‘guarded’. Thus some modern scholars (Frauwallner, Przyluski), noting that that the names Dhammarakkhita and Dharmagupta could easily be interchanged, have seen a connection between this ‘Dhammarakkhita’ and the ‘Dharmaguptaka’ school: the Dharmaguptakas were a branch of the Vibhajjavāda that developed in the wake of Yonaka Dhammarakkhita’s mission in the west.[3] To verify this theory we must investigate the exact forms of his name a little closer.

Here are the names mentioned in the Pali missions account,[4] together with the names as recorded in the Sudassanavinayavibhāsā. Fortunately the names are phonetically recorded in the Chinese translation and the reconstruction presents no serious difficulties.



Pali sources




末闡提      Majjhantika



摩呵提婆  Mahādeva



勒棄多      Rakkhita


Yonaka Dhammarakkhita

曇無德      Dharmagupta











末示摩       Majjhima


迦葉           Kassapa


提婆            Deva


毘帝須     Dundubhissara


提婆           ‘another’ Deva



須那迦       Soṇaka


多羅       Uttara



摩哂陀       Mahinda


地臾           Iṭṭhiya2


帝夜       Uttiya


參婆樓       Sambala


拔陀           Bhadda


Notes: 1 CBETA, T24, no. 1462, p. 684, c17-p. 685, a4. Hemavata teachers at CBETA, T24, no. 1462, p. 686, a5-9

2 Not found in the first section, but below at CBETA, T24, no. 1462, p. 684, b26


Whereas the Pali has four different ‘Rakkhitas’, the Chinese version has two ‘Rakkhitas’ and two ‘Dharmaguptas’. Sanghabhadra, the Chinese translator, was obviously capable of phonetically differentiating rakkhita from gupta, and we can only conclude that his manuscript contained these forms. [7] On other grounds, we are justified in regarding the Chinese version of this text as being historically more reliable than the Pali,[8] so we conclude that Dharmagupta was the original form. So according to this account, two of the missionaries,[9] including the monk known in Pali as Yonaka Dhammarakkhita, were called Dharmagupta.

This finding from the Chinese adds considerable plausibility to Przyluski’s suggestion that Yonaka Dhammarakkhita was the founder of the Dharmaguptakas. Another finding not available to Przyluski and Frauwallner is the recent confirmation of extensive Dharmaguptaka presence in Greek-influenced Gandhāra.[10] This adds further strong support to the notion that the Dharmaguptakas were centered in the very same region that we find Yonaka Dhammarakkhita.

When we see an ancient account, with confirmed historical validity, saying that a monk called Dharmagupta lived in the north-west; and a couple of centuries later there is substantial evidence of the strong presence of a school called Dharmaguptaka in the same region; and the records of that school confirm that they were named after their founding teacher; it would seem overly suspicious, if not actively perverse, to deny that these sources, disparate though they are, are speaking of the same person.

We might speculate why the Samantapāsādikā appears to have replaced Dhammagutta with Dhammarakkhita, while the earlier form is still found in the Sudassanavinayavibhāsā. I suggest that Buddhaghosa removed the references to the Dharmaguptakas when he edited his new Vinaya commentary, the Samantapāsādikā. In this he may have been influenced by the Dīpavaṁsa, which evidently post-dates the Sudassanavinayavibhāsā.[11] The Dīpavaṁsa appears to have been the first text to have fused the account of the schisms with the account of the missions. Having issued a blanket condemnation of the Dhammaguttas,[12] it would suit the Dīpavaṁsa’s polemical purpose to hide the implied connection between this school and the missions.

We might also wonder why the Sudassanavinayavibhāsā doesn’t describe Dhammarakkhita/Dharmagupta as ‘Greek’ (yonaka). Perhaps modern usage might be relevant here. It is still the custom in Sri Lanka for foreign monks to be called by their country of origin, as say ‘Australian Sujata’. But there is, of course, no point in calling the local monks ‘Sri Lankan Sujata’. So the use of the epithet yonaka must derive from a situation where Greek monks were considered foreign, as would have been the case in central India or Sri Lanka. But in a Greek dominated region this would not be used. Perhaps, then, this passage from the Sudassanavinayavibhāsā is an insider’s perspective, stemming from a tradition which regarded Dhammarakkhita/Dharmagupta as a local, that is, in the north-west.

This would imply that the Sudassanavinayavibhāsā has a close connection with the Dharmaguptaka school. And indeed, Bapat lists many Dharmaguptaka features in the Sudassanavinayavibhāsā that were discovered by Hirakawa. For example the text mentions 24 sekhiya rules dealing with the stupa, an outstanding feature of the Dharmaguptaka Vinaya.[13] Where was this Dharmaguptaka flavor mixed into the text? Bapat sees this as stemming from the Dharmaguptaka influence in China when the text was translated. This interpretation is problematic, as it would imply that the translator made wholesale revisions to his text to accord with his sectarian viewpoint, whereas to my knowledge the Chinese translators did not, as a rule, make such extensive alterations. The necessity for this interpretation stems from Bapat’s underlying assumption that the text is a translation of the Samantapāsādikā. If we accept Guruge’s argument that this text is not a translation of the Samantapāsādikā, but stems from either an earlier Sinhalese commentary, or from a commentary used by the Abhayagiri fraternity, then it would seem more likely that the Dharmaguptaka influences were present in the original text.

We know these affinities are there, but much more detailed textual work is required to ascertain exactly how or why they are there. But the conclusion seems inescapable that the Dharmaguptakas had a Vinaya commentary that may well have included a version of the Third Council and the missions, events that are otherwise only known from the Mahāvihāravāsins.

A further suggestive detail is that only the Mahāvihāravāsins and the Dharmaguptakas claim that the Brahmajāla was the first Sutta recited at the First Council.[14] I believe they placed this Sutta in this strategic position so as to form a mythic prototype for the Third Council where the heretics who expound the 62 views of the Brahmajāla are expelled by Aśoka under Moggaliputtatissa’s guidance. The fact that the Dharmaguptakas bestowed pride of place on the Brahamajāla Sutta suggests that they had a similar tradition regarding the Third Council, which conforms with our previous note regarding the Dharmaguptaka affinities of the  Sudassanavinayavibhāsā.


Dharmaguptaka & ‘Moggallāna’

Thus taking the Mahāvihāravāsin tradition together with the archaeological findings we are justified in seeing a connection between Dhammagutta (= Yonaka Dhammarakkhita) and Moggaliputtatissa, the leading Elder at the time of the missions. A closer look reveals several further independent sources suggesting a link between the Dharmaguptakas and a certain ‘Moggallāna’. The first of these is Vasumitra:


In this third century from the Sarvāstivādins arose another school called Mahīśāsaka. In this third century from the Mahīśāsaka arose another school called Dharmaguptaka. This school declared that Moggallāna is their main teacher. In this third century from the Sarvāstivāda arose another school called the Suvarṣaka, also called Kaśyapīya.[15]


Bhavya[16] and the San-lun-xian-yi,[17] on the other hand, say that the Dharmaguptakas were so-named after their founding teacher. This of course is natural, since the memory of Moggallāna evidently faded with time.

The Śāriputraparipṛcchā, giving a similar account of school derivation, also connects the formation of the Dharmaguptakas with a Moggallāna. The text, which also sets itself in the third century an, reads thus:


‘The Sarvāstivāda school then gave rise to the Mahīśāsaka school.  目揵羅優婆提舍 (mu-qian-luo you-po-ti-she) started the Dharmaguptaka school…’[18]


Although the passage is part of the discussion of the Sarvāstivāda group of schools, the text, unlike Vasumitra, does not literally connect the Dharmaguptakas with either the Sarvāstivādins or the Mahīśāsakas, but with 目揵羅優婆提舍. The first part of this name is clearly ‘Moggalla-’ or similar. The second  part, 優婆提舍, usually renders upadeśa, in which case it would refer to a treatise by Moggallāna; we remember that Moggaliputtatissa is famous for compiling the Kathāvatthu treatise. But I think it is more likely to stand for upatissa, which reminds us of the final part of Moggaliputtatissa’s name. It is possible this phrase refers to the Buddha’s disciple Moggallāna, together with his friend Sāriputta, whose personal name was Upatissa. But the Indic idiom, so far as I know, invariably pairs these two by their family names as ‘Sāriputta and Moggallāna’ or by their personal names as ‘Upatissa and Kolita’, without mixing the personal and family names.[19]

Now of course the question is, do these passages refer to the Buddha’s disciple Mahāmoggallāna or to the Moggaliputtatissa of the Third Council? The traditional view, recently restated by Yin Shun and others,[20] is that these passages refer to Mahāmoggallāna. This is understandable since Moggaliputtatissa is virtually unknown in the northern sources, so any reference to a ‘Moggallāna’ would naturally tend to be traced back to the great disciple.

The forms of the names do not decide the matter. We do not see the prefix ‘mahā-’, which would definitely identify the great disciple; and the absence of a confirmed parallel to the second part of Moggaliputtatissa’s name is not decisive, for the Vijñānakāya is definitely not referring to Mahāmoggallāna and is very likely referring to Moggaliputtatissa, but it just uses the name Moggallāna.

We cannot decide this question with certainty. Nevertheless, I would like to advance some considerations that, in my view, make it probable that the references in Vasumitra and the Śāriputraparipṛcchā refer to the Third Council Elder.

  • Both of our sources explicitly place themselves in the third century after the Buddha. The mention of Moggallāna occurs in the course of this presentation, with no hint that they are skipping back to an earlier time. I think it is more natural to read the passages as if they are referring to contemporary events.
  • The name in the Śāriputraparipṛcchā is, as I argued above, more likely to be a variant reading of Moggaliputtatissa than to represent Moggallāna-Upatissa.
  • There would seem to be no cogent reason for the Dharmaguptakas to claim Mahāmoggallāna as their forebear. Normally we would expect a school to claim a forebear with whom they had some special connection: for example, the Sautrantikas honor Ānanda, the teacher of the Suttas. Mahāmoggallāna is the chief in psychic powers, but I know of no hint that this was specially emphasized by the Dharmaguptakas. On the other hand, the Sri Lankan sources show a straightforward relation between Moggaliputtatissa and Dhammarakkhita (= Dharmagupta).
  • The accounts of Vasumitra and the Śāriputraparipṛcchā are closely related, and both refer to Moggallāna: why then does the Śāriputraparipṛcchā introduce ‘Upatissa’? This is perfectly understandable if we think of the name as just a variant of Moggaliputtatissa.
  • In Vasumitra’s account, the Dharmaguptakas claim ‘Moggallāna’ as their teacher, and it is understandable that a school may look back to one of the historical masters as their chief inspiration. But the Śāriputraparipṛcchā says that Moggalla (-puttatissa or –upatissa or  -upadeśa) ‘started’ (起) the Dharmaguptaka. It is obviously anachronistic to speak of Mahāmoggallāna as the ‘creator’ of a particular school. On the other hand, it would be entirely natural, under our theory, for the Dharmaguptakas to regard Moggaliputtatissa as their founding teacher.
  • As we have seen, there is good reason to believe that the Dharmaguptakas had a tradition of the missions and the Third Council comparable to that of the Mahāvihāra, which emphasized the role of Moggaliputtatissa as the leader of the missionary movement. Concrete textual support for this thesis is found in the Sudassanavinayavibhāsā. As we have seen, the account of the missions found in this text acknowledges Moggaliputtatissa’s role as the instigator of the mission of ‘Dharmagupta’. If a Dharmaguptaka connection for this text is established, it would also explain the specially prominent role that ‘Yonaka Dhammarakkhita’ (=Dharmagupta) plays in the narrative.
  • The ‘Moggallāna’ of the Vijñānakāya is said to hold the view that there is consciousness without object. It is possible that this is a Dharmaguptaka view, for Buddhaghosa ascribes this and related views to the Uttarapāthakas,[21] and the Dharmaguptakas are likely to have been included among the Uttarapāthakas. Bhavya and Vasubandhu attribute to the Vibhajjavādins (including  Dharmaguptakas) the closely related doctrine that past acts that have yielded their fruit do not exist, while past acts that have already yielded their fruit still exist. Buddhaghosa and Vasumitra ascribe this view to the Kaśyapīyas, but Vasumitra says that in most doctrines the Kaśyapīyas are similar to the Dharmaguptakas.[22] More research would need to be done to see if the Dharmaguptakas actually held the view ascribed to Moggallāna in the Vijñānakāya.

I therefore think we have good reason to accept the thesis that the Moggallāna referred to in connection with the Dharmaguptaka is in fact the vibhajjavādin Elder Moggaliputtatissa rather than the great disciple Mahāmoggallāna. This would simply make a more straightforward and reasonable explanation.


Dhammarakkhita: some other stories

The Sri Lankan chronicles record that Yonaka Dhammarakkhita and many of his followers travelled to Sri Lanka for the inaugural blessing ceremony for the Great Stupa.[23] This is not the treatment we would expect for a schismatic heretic, but for a respected Elder of the tradition.

The Abhidhamma commentaries still depict Dhammarakkhita, far off though he is, as a revered Teacher. Here is the paraphrase from the Dictionary of Pali Proper Names:


Punabbasukutumbikaputta Tissa Thera: He was of Ceylon, and crossed over to India, where he studied under Yonaka Dhammarakkhita. On his way home by sea he felt doubtful of one word, and returned all the way, one hundred leagues, to consult his teacher. On the way from the port he mentioned the word to a householder, who was so pleased with him that he gave him a blanket and one hundred thousand. This blanket Tissa gave to his teacher, but the latter cut it up and used it as a spread, as an example to others (not to desire luxuries). Tissa had his doubts set at rest and returned to Jambukola. There, at the Vālīkāvāma, as he was sweeping the courtyard of the cetiya, other monks asked him questions in order to vex him. But he was able to answer all these, having attained the paṭisambhidā. VibhA. 389.


The connection between Dhammarakkhita and Abhidhamma is also hinted at in a quasi-Abhidhamma post-canonical text, the Milindapañha. This text, which exists in several versions, famously records (or reinvents) a dialogue between the Greek king Milinda (Menander) and the Buddhist monk Nāgasena. The Pali version introduces a certain Dhammarakkhita in a key role. Nāgasena, after his initial training, travelled ‘a long way’ to the East to the Aśokārāma in Pāṭaliputta in order to receive teachings from ‘Dhammarakkhita’. This episode does not appear in the Chinese translation of the Sarvāstivāda version. It is generally agreed that the Pali version has been subject to elaboration, some blatantly unhistorical.[24] One of the purposes of this modification is to reconnect the action of the text with the Buddhist heartland in the east. Thus the text mentions five rivers: in the Chinese, four of these are from the north-west of India, but in the Pali, all are in the eastern districts.[25] Since the Milindapañha is set in the north-west, it seems likely that the Pali editors wanted to bring the action back further east, to lands they were more familiar with, and which had a long association with the Buddhist heartland.

It is no coincidence that this return is to ‘Aśoka’s monastery’, the center of the action in the Third Council story, and that it is here, with Dhammarakkhita as teacher, that Nāgasena becomes an arahant. It appears that the Pali, while celebrating the spread of the Dhamma to foreign lands, still holds the old places dear, and brings its hero back into the heartland for the crucial event of his enlightenment. Thus the insertion of the Dhammarakkhita episode is probably also to make the connection with the ‘Greek Dhammarakkhita’ – who better to teach the teacher of the Greeks, Nāgasena? It is unlikely that the same ‘Dhammarakkhita’ could have been alive in the time of both Aśoka and Milinda, though McEvilley thinks it is just possible.[26] But given the lack of concern for historicity displayed by the Pali editors, this does not affect the identification of the two Dhammarakkhitas.

Thus ‘Dhammarakkhita’ remained a revered elder for the Mahāvihāravāsins for a long time, fondly remembered by them as a distant brother successfully bringing the Dhamma to the Greek areas. This accords perfectly with the existing manuscript and epigraphical references to the Dharmaguptakas, which are concentrated in Gandhāra, long under Greek rule.


Dharmaguptaka texts & doctrines

Examination of the texts and ideas of the Dharmaguptakas confirms their close relation with the Mahāvihāravāsins. First we shall see how they are depicted in the Mahāvihāravāsin sources.

The Mahāvihāravāsin Kathāvatthu lists literally hundreds of points of contention between various schools. The schools, however, are not named in the text, and to find out who held these views – or at least, who the Mahāvihāravāsins believed held these views – we must turn to the commentary. In its introduction, the commentary classes the ‘Dhammaguttikas’ as one of the branches of the Mahīśāsakas, and hence they are reckoned among the 17 ‘schismatic’ or ‘heretical’ schools. But this is merely a sweeping sectarian dismissal of all different schools. In the body of the commentary there is no mention of the Dharmaguptakas. Thus the Mahāvihāravāsins knew of the Dharmaguptakas, but they knew of no dissentient views held by them.

Information about the Dharmaguptaka’s doctrines can be found in Vasumitra:[27]

·         The Buddha, while living, is included in the Sangha.

·         Gifts offered to the Buddha are more meritorious than those offered to the Sangha.

·         Gifts made to a stupa are meritorious.

·         The liberation of the Buddhas and the two vehicles (= sāvaka and paccekabuddha) is the same (this is mentioned in Xuan-zang’s translation only), though the path differs.[28]

·         Those outside Buddhism cannot gain the five special knowledges (abhiññā).

·         The body of an arahant is without āsavas.

The first four of these would be acceptable to Mahāvihāravāsins; the fifth would not; the last, while being too obscure to actually make much sense to anyone except an abhidhammika, would conflict with the Mahāvihāravāsin interpretation, which holds that the body of an arahant can become the object of defilements for others; but perhaps it was intended rather as a correction to the first of the Mahāsaṅghika’s ‘5 points’.

In addition to these views, Vasubandhu[29] says that the Dharmaguptakas held, in agreement with the Mahāvihāravāsins and against the Sarvāstivādins, that realization of the truths happens all at once (ekābhisamaya).

It will take us too far afield to examine in detail the actual texts of the Dharmaguptaka, but a quick survey is enough to reinforce the impression of their closeness with the Mahāvihāravāsin.

Regarding the Dharmaguptaka Vinaya, Pachow in his survey of the pāṭimokkhas states: ‘the Dharmaguptaka follows very closely the Pali text in most cases, not merely in numbering the series but also in contents, except the [sekhiya] section, in which it adds 26 prohibitory rules regarding the Stupa.’[30]

Regarding the Sutta literature, McQueen studied various versions of the Sāmaññaphala Sutta, and concluded that of all of them, the Mahāvihāravāsin and Dharmaguptaka were the closest and stood nearest the ancient tradition. He also says that this closeness holds good for the Mahāvihāravāsin Dīgha Nikāya in general when compared with the Dharmaguptaka Dīrgha Āgama: ‘These collections are generally quite close; major disagreements are rare. Where discrepancies do occur the [Dharmaguptaka] Dīrgha is more often wrong (late), showing corruption and expansion of the text.’[31]

Finally, Frauwallner in his discussion of the sole surviving Dharmaguptaka Abhidharma work, the Śāriputrābhidharma, shows the deep connections between this work and various Mahāvihāravāsin Abhidhamma books, including the Dhammasaṅgaṇī, Vibhaṅga, Dhātukathā, and Paṭṭhāna. He sums up by saying ‘While mainly based on old transmitted material, even this is organized in a different way as compared with the other schools we have discussed [Mahāvihāravāsin and Sarvāstivāda]. It contains little in the way of innovation or doctrinal evolution.’[32] Thus, while admitting that there are several significant divergences in the field of Abhidhamma, there is clearly a common source. There is no reason why such differences as exist should not have emerged in the long period of Abhidhamma development that took place after the separation of the schools.

The recent manuscript finds from Gandhāra give us a new source of Dharmaguptaka texts, and a new insight into how they developed. According to Richard Salomon, the existing texts, which are in a very bad state of decay, date from shortly after the Common Era, that is, the beginning of the middle period of Indian Buddhism. They lack the textual uniformity we have come to expect from the Pali, and thus Salomon suggests they stem from a time when the canon was not yet fully formed. Alternatively, it could be the case that the Dharmaguptakas did not place as much premium as the Mahāvihāravāsins on textual precision: we have seen that the Dīpavaṁsa ascribes the root schism to bad textuality, and the prominence of the paṭisambhidās in their root-treatise the Paṭisambhidāmagga confirms the centrality of textual analysis for this school. Indeed, the Mahāvihāravāsins, so far as we know, are the only school to produce a complete body of commentaries on the canonical texts. Perhaps we should regard them as the textual exegesis school par excellence.

The Gandhārī texts of the Dharmaguptakas have only been partially studied. Clearly they represent a different textual tradition to that preserved in Pali or the Chinese Āgama literature, with the obvious exception that they agree closely with the existing Chinese Dharmaguptaka texts, in so far as comparisons have been made. But there are no doctrinal differences apparent. The only really new element is the introduction of several avadāna-type stories relating to local celebrities. Thus the Dharmaguptakas adapted their literature to their local culture, without however changing the doctrine.

So it seems that the split between the Mahāvihāravāsins and the Dharmaguptakas was due to neither Dhamma nor Vinaya, but mere geography. The Dharmaguptakas were a north-western branch of the Vibhajjavāda, and the Mahāvihāravāsins or Theravādins were the southern branch. While the Mahāvihāravāsins in a belligerent mood issued a purely formal denunciation  of the Dharmaguptakas, the texts, doctrines, and history instead reveal a close affinity.



[1] Pali Vinaya 1.55

[2] Thūpavaṁsa 20: yonakaraṭṭhe alasaṇdā nagarato yonaka dhammarakkhitatthero tiṁsa bhikkhu sahassāni (‘… from the city of Alexandria in the Yonaka country, Yonaka Dhammarakkhita and 30 000 monks [came]…’.) This refers to his visit to the opening of the Great Stupa in Sri Lanka.

[3] The commentaries treat the two words together, e.g. Dhammapāda Aṭṭhakathā 257: Dhammassa guttoti so dhammagutto dhammarakkhito.

[4] The Pali sources are fairly consistent in naming this monk, but there are occasional exceptions. In the story we have just told of Dhammarakkhita converting the king’s brother, the monk is referred to as ‘Yonakamahādhammarakkhita’. But the Chinese here just has Dharmagupta (曇無德 CBETA, T24, no. 1462, p. 682, c14) Similarly, at Mahāvaṁsa 29.39 we find Yonamahādhammarakkhita. But is is worth noticing that monk’s names are subject to confusing modifications. The prefix ‘Mahā’ is added or not, as we have seen in the case of Yonaka [Mahā] Dhammarakkhita. There are so many names beginning with ‘Dhamma-’ that it is normal in modern times to drop the Dhamma and just use the second element; thus Dhammarakkhita becomes ‘Rakkhita’. It is also common to name a monk by his country of origin, but again this may be applied quite inconsistently. So, without trying to sort out anything definitive, I wonder whether some of these monks might have been the same person, known by slightly different titles in different lands.

[5] CBETA, T24, no. 1462, p. 684, c17-p. 685, a4. Hemavata teachers at CBETA, T24, no. 1462, p. 686, a5-9

[6] Not found in the first section, but below at CBETA, T24, no. 1462, p. 684, b26

[7] This point is unfortunately obscured in Bapat’s translation, where he renders 曇無德 (tan-wu-de) as if it harked back to an original dhamma[rakkhi]ta (e.g. Bapat, 36). But 曇無德 is the standard rendering of Dharmagupta, used dozens of times in this sense. Since we know that Sanghabhadra was quite capable of phonetically representing rakkhita by勒棄多 (le-qi-duo), why would he use such a misleading combination of renderings within the same context? Bapat’s interpretation entails that Sanghabhadra’s renderings were arbitrarily inconsistent. Even for the identical Indic phonetic ending –ta, Sanghabhadra used two quite different characters: 德 (de) and 多 (duo). This only makes sense if 曇無德 renders Dharmagupta, since in this case the rendering is common usage, even if it is not internally consistent in this passage. I therefore think that it is virtually certain that Sanghabhadra’s text read Dharmagupta (or equivalent) and Bapat’s rendering as Dhammarakkhita stems from his assumption that the Sudassanavinayavibhāsā is a translation of the Samantapāsādikā; despite noting the very many differences between the two texts, he still tends to read the Pali text back into the Chinese.

[8] For example, in each mission account, a number is given recording the conversions and ordinations made. (Lamotte, History of Indian Buddhism, 296) In the two accounts, in 12 cases the numbers agree. In the remaining cases the differences are, mentioning the Pali first: 100000/1000; 37000/7000; 37000/30000; 13000/3000; 170000 (or 137000)/73000; 10000/1000. Thus whenever they differ, the Pali is larger than the Chinese, and this difference is always by a suspiciously artificial amount.

[9] Unless the names are confused and they are to be counted as one.

[10] Salomon

[11] Or at least, the Sudassanavinayavibhāsā doesn’t mention the Dīpavaṁsa. Occasionally the Sudassanavinayavibhāsā quotes verses found in the Dīpavaṁsa, but while the Samantapāsādikā mentions the Dīpavaṁsa by name, the Sudassanavinayavibhāsā merely says these verses were spoken by the ancients: 今說往昔偈讚 (CBETA, T24, no. 1462, p. 687, c3, c17-18)

[12] Dīpavaṁsa 4.86

[13] Bapat l-liii; see Guruge, 96.

[14] CBETA, T22, no. 1428, p. 968, b15-16. The Dharmaguptaka version of the Brahmajāla is very close indeed to the Pali, with only trifling variation in the sequence and wording of the 62 heretical views. For a detailed study, see Cheng.

[15] 於此第三百年中。從說一切有部。又出一部。名正地部。於此第三百年中。從正地部。又出一部。名法護部。此部自說勿伽羅是我大師。於此第三百年中。從說一切有部。又出一部。名善歲部。亦名飲光弟子部 (CBETA, T49, no. 2033, p. 20, b14-18). This is Paramārtha’s translation. Xuan-zang’s translation agrees, saying that the Dharmaguptakas followed the teacher Moggallāna (自稱我襲採菽氏師, CBETA, T49, no. 2031, p. 15, b16-17; here Moggallāna is translated as採菽氏, cai-shu-shi. This rendering derives from a story claiming that Moggallāna’s family name (氏) stems from an ancestor who used to pick up (採) beans (菽, Pali mugga). Kumārajīva’s translation says that: ‘The Mahīśāsaka gave rise to another school called Dharmagupta, who followed their main teacher Moggallāna.’(彌沙部中復生異部。因師主因執連名曇無德 (CBETA, T49, no. 2032, p. 18, b1-2) According to Li Ch’ung An there has been a carving mistake here, with 因執連 in place of 目揵連. See http://ccbs.ntu.edu.tw/FULLTEXT/JR-BJ001/03_02.htm#n36)

[16]  Rockhill, 1992, 184

[17] CBETA, T45, no. 1852, p. 9, c13-15

[18] 其薩婆多部。復生彌沙塞部。目揵羅優婆提舍。起曇無屈多迦部 (CBETA, T24, no. 1465, p. 900, c2-4)

[19] E.g. Pali Vinaya 1.42: addasā kho bhagavā sāriputtamoggallāne dūratova āgacchante, disvāna bhikkhū āmantesi – ‘ete, bhikkhave, dve sahāyakā āgacchanti, kolito upatisso ca. etaṁ me sāvakayugaṁ bhavissati aggaṁ bhaddayugan’ti.

[20] http://www.budd.cn/news/budren/news_budren_20030430_9.html

[21] Kv 9.3, 9.4, 9.5, 9.6.

[22] 餘義多同法藏部執 (CBETA, T49, no. 2031, p. 17, b2)

[23] See above note from the Thūpavaṁsa. The event is earlier recorded in Mahāvaṁsa 29.39: Yonanagarā’lasandāso, yonamahādhammarakkhito; thero tiṁsa sahassāni bhikkhū ādāya āgamā.

[24] Such as the mention of Milinda visiting the six heretical teachers who lived in the time of the Buddha.

[25] http://www.saigon.com/~anson/ebud/milinda/ml-01.htm

[26] McEvilley, 378

[27] See Dutt, 172

[28] 佛與二乘解脫雖一。而聖道異 (CBETA, T49, no. 2031, p. 17, a25)

[29] Abhidharmakośa vi. 27

[30] Pachow, 39. For a challenge to the usual interpretation that Dharmaguptakas had a special affinity for stupa worship, see http://sectsandsectarianism.googlepages.com/dharmaguptakasandthestupa

[31] McQueen, 190

[32] Frauwallner, 1995, 116