SuttaCentral
Sarvastivada and Theravada traditions respectively have their historical view on the four principal Agamas/Nikayas (see below discussions). Which tradition do you support?
Sarvastivada view:
https://discourse.suttacentral.net/t/eb … ts/41671/3
Theravada view:
https://discourse.suttacentral.net/t/eb … s/41671/21
--
As I stated previously, Ven. Yin Shun supports the Sarvastivada view.
For the discussion of this topic, I repeat here what I have stated:
In his work, The Formation of Early Buddhist Texts 原始佛教聖典之集成, Ven. Yin Shun suggests that there were two phases in Early Buddhism (Pre-sectarian Buddhism):
‘Samyukta/Samyutta Buddhism’ based on saṃyukta-kathā 相應教 (had its origin in the first Saṅgha council, shortly after the death of the Buddha)
‘Agama/Nikaya Buddhism’ based on the four principal Agamas/Nikayas (i.e. SA/SN, MA/MN, DA/DN, EA/AN) (originated at the second council, one hundred years after the death of the Buddha)
However, the extant Agamas/Nikayas are sectarian texts. One can seek an understanding of early Buddhist teachings by studying them comparatively.
The connection between SA/SN and the other three Agamas/Nikayas in historical context are:
• SA/SN has three-anga structure (sutra, geya, vyakarana angas), whose structure had its origin in the first council.
• These three angas formed in sequence. The sutra-anga was the earliest of the three.
• MA/MN was expanded mainly from vyakarana (弟子所說) of SA/SN.
• DA/DN was expanded mainly from geya (祇夜) of SA/SN.
• EA/AN was edited mainly for the promotion of Buddhist teachings for the general public.
• Saṃyukta-āgama/Saṃyutta-nikāya (SA/SN) was not at first being termed as āgama or nikāya, but generally named the ‘Connected Discourses’ 相應教 Saṃyukta-kathā.
• Calling the Saṃyukta/Saṃyutta as āgama/nikāya ‘collection’ was until when the other three āgamas/nikāyas (MA/MN, DA/DN, EA/AN) were gradually developed and expanded from it (相應教 Saṃyukta-kathā)
• MA/MN, DA/DN, and EA/AN originated at the second council.
• However, the extant four principal Agamas/Nikayas are sectarian texts. One can seek an understanding of early Budhist teachings by studying them comparatively.
The following is a quotation, in Chapter 10, Section 4, from the book The Formation of Early Buddhist Texts by Yin Shun:
第四節 結說
經上來的比對研究，「四阿含」（「四部」）的成立，可得到幾點明確的認識。1.佛法的結集，起初是「修多羅」，次為「祇夜」、「記說」——「弟子所說」、「如來所說」。這三部分，為組成「雜阿含」（起初應泛稱「相應教」）的組成部分。「弟子所說」與「如來所說」，是附編於「蘊」、「處」、「因緣」、「菩提分法」——四類以下的。這是第一結集階段。
在「雜阿含」三部分的集成過程中，集成以後，都可能因經文的傳出而編入，文句也逐漸長起來了。佛教界稟承佛法的宗本——「修多羅」，經 「弟子所說」的學風，而展開法義的分別、抉擇、闡發、論定，形成了好多經典。結集者結集起來，就是 「中阿含」；這是以僧伽、比丘為重的，對內的。
將分別抉擇的成果，對外道、婆羅門，而表揚佛是正等覺者，法是善說者，適應天、魔、梵——世俗的宗教意識，與「祇夜」精神相呼應的，集為「長阿含」。
「雜」、「中」、「長」，依文句的長短而得名。
以（弟子所說）「如來所說」為主，以增一法而進行類集，《如是語》與《本事經》的形成，成為「九分教」之一，還在「中」、「長」——二部成立以前。但為了便於誦持，著重於一般信眾的教化，廢去「傳說」及「重頌」的形式，而進行擴大的「增壹阿含」的編集，應該比「長阿含」更遲一些。
以「雜阿含」為本而次第形成四部阿含，《瑜伽師地論》的傳說，不失為正確的說明！近代的研究者，過分重視巴利文（Pāli）；依巴利文聖典，不能發見四部阿含集成的真相。
即使以「雜阿含」的原形為最古，而不能理解為三部分（「修多羅」、「祇夜」、「記說」）的合成；不知三部分的特性，與三部阿含形成的關係，也就不能理解依「雜阿含」而次第形成四部的過程。
–-
Cf.: Choong Mun-keat:
The Fundamental Teachings of Early Buddhism: A Comparative Study Based on the Sūtrāṅga portion of the Pāli Saṃyutta-Nikāya and the Chinese Saṃyuktāgama (Series: Beitrage zur Indologie Band 32; Harrassowitz Verlag, Wiesbaden, 2000).
“Ācāriya Buddhaghosa and Master Yinshun 印順 on the Three-aṅga Structure of Early Buddhist Texts”, Research on the Saṃyukta-āgama (Dharma Drum Institute of Liberal Arts, Research Series 8; edited by Dhammadinnā), Taiwan: Dharma Drum Corporation, August 2020, pp. 883-932.
E.g.: pp. 7-11, notes 32, 34, 36 (on the nine angas), in Choong Mun-keat’s The Fundamental Teachings of Early Buddhism.
Early Buddhism resources (section/page 10), Dhamma Wheel site.
Just to clarify, what you’re referring to as a Sarvastivada view is not the Sarvastivada school’s view of the history of the Agamas. They held that the four Agamas were recited by Ananda at the first council like the other traditions of Buddhism. The idea that the Agamas formed later in history is a theory developed by Yinshun using a variety of Buddhist sources. You should probably call it Yinshun’s view, otherwise people may think Sarvastivadins actually had this idea when they didn’t.
The main difference that I know of between Sarvastivadins and Theravadins is the order in which the Agamas were recited by Ananda. Sarvastivadins have him reciting the Samyukta Agama first, while Theravadins have him reciting the Digha Nikaya first. The account of the first council in the Mulasarvastivada Vinaya depicts the arhats deciding to gather sutras on certain topics into samyuktas. After they had done that, they gathered the remaining sutras based on their length to create the Dirgha and Madhyama Agamas. And then the remainder were put into a collection organized by numbers of items that was then called the Ekottarika Agama.
There is actually an interesting view found among later Sarvastivadins that the early canon was originally much larger but that people lost the ability to remember all of it as time went by after the Buddha’s Parinirvana. This was used to the explain why the 98 tendencies were not found in a sutra. The assertion is that they were originally in the Ekottarika Agama, but the sutra was lost. They believed that the Ekottarika originally had a hundred books instead of only ten or eleven. I’m not sure if it’s only Sarvastivadins to held this kind of view about the early canon, but it’s basically the opposite of the idea that it started small and gradually grew in size.
It should be the view ‘according to the Sarvāstivāda tradition of the Vastusaṅgrahaṇī of the Yogācārabhūmi.’ (pp. 898-9, note 21 in Choong’s “Ācāriya Buddhaghosa and Master Yinshun 印順 on the Three-aṅga Structure of Early Buddhist Texts”)
Yes, correct.
YinShun supports ‘the view according to the Sarvāstivāda tradition of the Vastusaṅgrahaṇī of the Yogācārabhūmi.’
One of the main reasons for YinShun to support the view is his further extensive research on that topic (i.e. saṃyukta-kathā 相應教 and Sutra-matrka), demonstrating in detail on the gradual development of angas and their connection with the four main agamas/nikayas (pp. 9-11 in Choong’s Fundamental Teachings of Early Buddhism; pp. 898-9, note 21 in Choong’s “Ācāriya Buddhaghosa and Master Yinshun 印順 on the Three-aṅga Structure of Early Buddhist Texts”).
YinShun disagrees with this view according to his study on the formation of early Buddhist texts.
The Yogacarabhumi is not a Sarvastivada text. Asanga was not a Sarvastivadin, nor was Vasubandhu. They were actually critics of Sarvastivada Abhidharma. Sarvastivada views are found in Sarvastivada texts written by Sarvastivadins - i.e., their Abhidharma, Vinaya, Sutras, and texts like the Mahavibhasa and Sanghabhadra’s replies to Vasubandhu, etc.
Exactly. His view is not the Sarvastivada view; they had very different ideas.
Yes, Asanga was a Mahayana Yogācārabhūmi monk, not any Hinayana monk at that time.
Nevertheless, as stated previously, he wrote the sūtra-mātṛkā (sūtra matrix, 契經, 摩呾理迦 or 本母), essentially a commentary on a portion of the Sarvastivada’s Saṃyukta-āgama, in the Vastusaṅgrahaṇī of the Yogācārabhūmi (T 1579 at T XXX 772c9–868b22), follows the sequence of the Sarvastivada’s Saṃyukta-āgama.
This suggests that Asanga critically and respectably studied EBTs, the four principal Agamas, particularly Samyukta-agama based on the Sarvastivada tradition.
However, Asanga did not state clearly that the sūtra-mātṛkā was in fact essentially a commentary on a portion of the Sarvastiva’s Saṃyukta-āgama followed the sequence of the Saṃyukta-āgama. (The sūtra-mātṛkā (sūtra matrix, 契經, 摩呾理迦 or 本母) - Discussion - Discuss & Discover)
Yes, Yin Shun is a Chinese Mahayana monk to critically study the formation of EBTs.
He supports the view according to the Sarvāstivāda tradition of the Vastusaṅgrahaṇī of the Yogācārabhūmi in his research on the formation of EBTs.
–-
By the way, I now consider because both Asanga and Yin Shun are Mahayana, therefore, they are able to critically and objectively to study EBTs, the four principal Agamas.
We’ve had this conversation before, I believe. Asanga doesn’t tell us which version of the Samyukta Agama he is using, nor does he quote from it very much. I have some doubts about whether he was reading a Sarvastivada version of SA, mainly because the other Chinese translation (Taisho 100) also follows the order of Yinshun’s SA remarkably well inside the individual samyuktas. But it isn’t the same version as Taisho 99. You may be right … it’s just not as obvious to me.
But what does this have to do with what the Sarvastivadins believed about the First Council? Does Asanga tell us anything about that? My understanding is that Yinshun used the passages in the Yogacarabhumi to help confirm that his rearrangement of SA was correct.
To get back to your original post, I believe you were asking us if we accept a modern approach to the formation of the early Buddhist canon or the traditional Buddhist history. At least that was how I understood it.
My answer is that I think it’s very possible that the First Council story is essentially accurate, though some of the finer details were forgotten and filled in by each tradition’s storytellers. The four collections may have taken longer to create than the First Council stories suggest, but it may have happened in the first years after the Buddha passed away.
There are good comparative studies of all the First Council stories that exist, by the way, if anyone is curious about them. There is this old study done by DT Suzuki back in 1904 that compiled together an account based on all the versions he could locate in Chinese translation. I’m sure there are more recent studies, but his translation based on Chinese sources is pretty comprehensive.
On the other hand, I personally don’t have a strong opinion about it because there’s also reason to believe the Agamas may have come into existence at different times. The parallels of the Madhyama and Ekottarika Agamas are difficult to explain. The versions that still exist have much less in common with each other than do the extant versions of the Dirgha and Samyukta Agamas.
The Samyukta, Dirgha, and Ekottarika Agamas were considered to be the “first” Agama compiled by different traditions, but the Madhyama to my knowledge is never placed at the start. The Madhyama and Majjhima collections have little in common with each other beyond the overlap of sutra parallels. They are organized in entirely different ways without a hint of an older version that I can see. And they contain quite a few parallels that appear in other collections, suggesting that MA sutras were sourced from the other Agamas. So, then, the Madhyama is the strongest case to me for a later compilation of an Agama. It was at least expanded quite a bit with sutras moved into it from other Agamas.
And that’s the difficulty. Not having older versions of these collections and only the traditional history of how they were created, it’s difficult for me to come to a conclusion about it. All I can say for certain is that they changed quite a bit over time.
I don’t get this sentence. 
Are you saying that MA ↔ MN (and EA ↔ AN) parallels are more divergent than SA ↔ SN and DA ↔ DN?
I consider both the Asanga’s SA version and the Mūlasarvāstivāda Vinaya’s SA version are later than the Taisho 99’s SA version.
This is because in the Taishō edition the Vyakarana portion of SA is marked off by the editors with the heading 弟子所說誦, but 佛/如來所說誦 is not so marked as indicated in the Asanga’s SA version and the Mūlasarvāstivāda Vinaya’s SA version.
Yin Shun entails two main issues in his work on the structure and content of SA. One is the reconstruction of the sequence of SA in T 99 version. The other is the three-aṅga structure of SA that is connected with first council.
The two main issues are not directly connected, but helpful to understand the structure and content of SA.
Identifying the three-aṅga structure of SA is not based on the reconstruction of the sequence of SA in T 99 version.
The above-mentioned 相應教 (Saṃyukta-kathā, ldan pa’i gtam) is one of the main sources from the Sarvāstivāda tradition of the Vastusaṅgrahaṇī of the Yogācārabhūmi to identify the three-aṅga structure of SA by Yin Shun.
I think one of the main reasons is EA belongs to Mahasamghika version, whereas MA is Sarvastivada version.
Cf. pp. 695-792 (on the four Agamas/Nikayas) in Yin Shun’s The Formation of Early Buddhist Texts.
This reminds me of Jainist account of their lost part (or all parts) of their Agamas….
I think these collections (Nikayas or Agamas) is later compilation of the canon and the First Council was not about compiling the canon, but just discussing about Vinaya, i.e minor rules to be abolished because the account of canon compilation is different each another among the Buddhist sects. The Buddhist canon was not fixed around the First Council, no collections for grouping the individual suttas (individual suttas are mentioned in Asoka’s inscription), and the collections was formulated hundreds years later….
I do agree that the Buddhist ‘four principal Agamas/Nikayas’ were not established at once around the First Council.
Ven. YinShun’s interpretation of the Sarvāstivāda structure is about the gradual development of the four principal Agamas/Nikayas in Early Buddhism, of which Saṃyukta-kathā (相應教, ldan pa’i gtam) was the foundation, although the extant Agamas/Nikayas are sectarian texts.
It just makes so much sense when the texts are structured like that.
Yes. As collections they are much more divergent. They share far fewer parallels in the case of EA and AN, and for MN and MA, there is no parallel organization of any kind to be found between them. When I compared them closely, all I found was a couple sutras that were in the same order. The three Dirgha parallels that exist show evidence of an early core that then was expanded and reorganized in each tradition. (I wrote an essay about it: Thoughts on the Dirgha Agama's History) For the Samyukta Agama, the parallels are strong, but SN has been shuffled up like a deck of cards like SA was in the Taisho … So the Samyukta and Dirgha the two collections that appear to be very old.
I am unable to see clearly the reasons why you have such a conclusion here?
Well, if individual DN and SN collections share more parallels, but MN and AN collections diverge more, it seems the schools have more or less closed the DN/SN books while kept editing & composing MN and AN, no?
But “The three Dirgha parallels that exist show evidence of an early core that then was expanded and reorganized in each tradition.”
So, to conclude the Dirgha collection that appears to be very old is not clear and correct.
The Mahayana Nagarjuna’s emptiness teaching also did not state clearly that the prototype of the notion of emptiness of the Mahāyāna Mādhyamika tradition was in fact based on the sūtra-aṅga portion of SA/SN.
For example, the middle way of emptiness, such as “neither existence (arising) nor non-existence (ceasing), neither eternalism nor annihilationism, neither sameness nor difference, neither coming nor going” of the Mādhyamika tradition, is found in the texts of SA and SN:
Choong Mun-keat:
The Notion of Emptiness in Early Buddhism (1999), pp. 32-40;
The Fundamental Teachings of Ealy Buddhism (2000), pp. 60-66, 192-199, 239.
“neither existence (arising) nor non-existence (ceasing), neither eternalism nor annihilationism, neither sameness nor difference, neither coming nor going”:
不生亦不滅 不常亦不斷
不一亦不異 不來亦不出
anirodham anutpādam anucchedam aśāśvatam/
anekārtham anānārtham anāgamam anirgamam// (Taishō vol. 30, no. 1564, p. 1a and note 16).
p.62 Problems and Prospects of the Chinese Saṃyuktāgama (2010) Choong Mun-keat.pdf (670.0 KB)
Copy from: Mahākaccāna and “Proto-Madhyamaka”? - Discussion - Discuss & Discover