Ānāpāna in the Saṃyukta Āgama

From the forward to The Fundamental Teachings of Early Buddhism:

Briefly put, Yin Shun’s claim is that the distribution of the sutras into four nikāyas/āgamas did not take place at the First Council; initially the sutras were grouped in a single collection, whose structure is largely preserved in the extant SN and SA. The other three principal nikāyas/āgamas were developed subsequently, probably at the Second Council, in response to a substantial increase in the number and size of the remembered sutras that had taken place during the intervening century. These conclusions are based in large part on a demonstration that the contents of SN/SA fit the first, second, and third categories in the traditional aṅga classification (sūtra, geya, vyākaraṇa), i.e. that SN/SA is structurally archaic. It follows that SN and SA are of special significance both historically and doctrinally—which helps explain why the author of the Yogācārabhūmiśāstra saw fit to include in his great treatise a lengthy commentary on the sūtra-aṅga portion of SA.

There are also some passages from northern sources that state that the SA served as the basis for the other three āgamas. Yin Shun held that not only was the SA the first āgama, but even that the sūtra-aṅga section of it was the earliest and most authoritative portion of that collection. The sūtra-aṅga is entirely based around the framework of the Four Noble Truths. Yin Shun maintained that the MA, DA, and EA were later collections that originated at the Second Council.

Some of this perceived similarity may be because the Pali nikāyas sometimes have extra sūtras that make each nikāya more representative. For example, the large sūtra on the Four Bases of Mindfulness in the DN is considered late by many, and does not exist in the DA. If we compare the topics and themes, though, the collections are not so uniform. For example, the SA has quite a lot about ānāpāna, but the MA has practically nothing about it. The MA has a lot about impurity contemplations, but the SA has practically nothing about it.

I agree it’s best to consider views like this as hypotheses when considering their history. Otherwise preconceptions may dissuade people from studying the particularities of each text or collection.

Not in Pali, but I think amṛta also occurs in the Ṛgveda, is cognate to the Greek ambrosia, and has PIE roots.

This is actually a really interesting point that I had not considered before: leaks can be inward too. In general, though, the Chinese term is used mostly for outward or downward leaks, but literally it does just mean “leak.”

Ok, but as far as the textual history is concerned, the interesting question here is which texts were recited at the first communal recitation, not which text was the first one to be recited at that recitation. Whatever was recited at this time, must be relatively early.

2 Likes

Whatever the truth of Yin Shun’s ideas - and there are varying opinion about that - it is still the case that the suttas were moved around a lot between the different Āgamas. Even if do assume that the Saṃyukta was the first Āgama, it does not necessarily follow that the suttas of the Saṃyukta as we have it now are older than the suttas of the other Āgamas. It is perfectly reasonable to think that as the Saṃyukta expanded it eventually had to be divided into smaller groups, thus forming the four or five Āgamas. This may well have happened after the first saṇgīti, but this late division does not say much about the age of individual suttas, whatever Āgama they belonged to.

There is a lot of overlap, as I am sure you would agree, but I don’t think we should expect uniformity. The Āgamas are clearly different kinds of containers and so we should expect the content to be at least somewhat different. Suttas over a certain length would have found a more natural home in the Dīrgha and the Madhyama, than in the other two Āgamas. And of course length will have some bearing on the content.

1 Like

I disagree to some extent. I don’t think the contents for certain subjects are so similar between collections. For example, the treatment of the Four Bases of Mindfulness in the SA / SN is quite different from that in the MA / MN and DA / DN. In the SA / SN, the Four Bases of Mindfulness are described with extremely short and simple formulas. The dozens of sūtras in their saṃyukta are quite consistent and vary just a tiny bit with slight variations. In the MA / MN, though, they are expanded into huge suites of meditation methods.

When patterns like that appear consistently, with different or more developed formulations appearing in the MA / MN / DA / DN, and much more developed narratives appearing in them too, then it probably follows that they came from a somewhat more developed state or period.

Especially considering that the SA / SN together contain thousands of texts, I see no logical reason why—if they are all from the same period, and differ primarily in the length of their texts—the SA / SN should not clearly contain all the practices described at length in the MA / MN / DA / DN / AN / EA.

Yin Shun’s theories depend to a certain degree upon Sarvāstivāda and Yogācāra sources, but also upon a close examination of the structure and development of the SA, and the position of the SA in the scheme of four āgamas and nine aṅgas. If I understand the matter correctly, he regards the SA / SN as representing the basic starting point not only for the āgamas, but for the compilation of the nine aṅgas. The first three of the nine aṅgas correspond directly to the three aṅgas of the SA.

Bhante Sujato goes into the matter of the three aṅgas in some more detail in A History of Mindfulness, and points out some evidence from the āgamas and nikāyas themselves that seem to indicate an early formulation of the Dharma as consisting of the three aṅgas of the SA / SN.

By combining some of the contents of the SN and KN, it’s quite easy to come up with something approximating the first seven aṅgas of the nine aṅga classification scheme (keeping in mind that some may have been expanded, like the Jātakas, which would have been only a fraction of the size of the current Pali version):

  1. Sutta
  2. Geyya
  3. Veyyākaraṇa
  4. Gāthā (2)
  5. Udāna
  6. Itivuttaka
  7. Jātaka
1 Like

It seems to me that we should expect short summaries to appear frequently and expanded formulations of a topic only rarely. Take dependent origination. The twelve link formula is found a large number of times throughout the suttas, but the detailed explanation of each factor only rarely. It seems to me that something similar is the case for most of the teachings, including satipaṭṭhāna. That is, we should only expect to see the full exposition of satipaṭṭhāna occasionally (which in this case means once), compared to a large number of short expositions. When people heard the shorter expositions, they would be able to fill in the missing parts without having to be told the full content on every occasion. This would make sense in a society that relied on oral transmission. (By this I do not mean to deny that the Satipaṭṭhāna Sutta has undergone later expansion - I think it has, as has been well documented by Bhante Sujato. I just mean that some form of the expanded sutta probably existed at an early stage.)

An important issue is that the classification of suttas into aṅgas, especially the first two, is far from unproblematic. If I remember correctly, Bhante Sujato’s takes a different approach from Yin Shun as to which suttas belong to the sutta aṅga and which to vyākaraṇa aṅga. Having looked at it briefly myself, it seems far from clear-cut how this division should be done. It seems a bit hazardous to me to then derive a whole theory of the primacy of the Saṃyukta Āgama on the basis, at least in part, of something which is so problematic. Although I find the idea of the Saṃyukta Āgama as the first sutta container interesting, and in part persuasive, I am still not fully convinced.

2 Likes

@seniya , if you are still interested in translating the article, I think it is probably in a fairly stable state, covering the main materials available on ānāpāna in the āgamas. Best regards.

2 Likes

Thank you, @llt :anjal:

1 Like

I forgot to reply to this. Yes, the part about focusing on the tip of the nose is unambiguous (繫意鼻頭). However, the EA is a very strange collection in some regards, and sometimes I get the feeling that some occasional parts drift into explanation rather than translation.

The usual translations of parimukha in Chinese mean something like establish mindfulness at the fore (no mention of the nose). But perhaps EA 17.1 is giving us a different interpretation.

Also, the EA is sometimes not even very consistent with itself. It has some unusual history that we don’t see in the SA, MA, and DA. Unlike those other collections, we don’t know which sect the EA comes from. But even if we did know, there would probably still be some big questions.

For example, the SA and SN always teach ānāpāna in 16 steps. But EA 17.1 has just seven steps, some of which are not found at all in the SA or SN. For example, being mindful if the breath is warm or cool, or if the breath is present or absent.

3 Likes

Thank you, llt.

:anjal:

That’s actually why I like the AN. Sometimes there is an un-polishedness about it with inconsistencies and different wordings than DN, MN and SN. Linked to this is an issue I’d like to mention (and I don’t know if it received enough attention): There is the discussion which are the oldest agamas or nikayas, but something that is very unknown to us - I assume - is the edition history. Let me give an example of why this is essential.

Let’s say the SN/SA was really the first and oldest collection, and the KN/AN the youngest. Let’s say again that at the year 50 CE there was a towering scholar monk that was respected and trusted by everyone. That monk would have been a SN/SA scholar. Seeing many inconsistencies - as original source material should have - he would have taken it on himself to clarify the material, remove the obvious ‘mistakes’ and make the texts consistent, so that the concepts and practices would fit neatly and wouldn’t create paradoxes or conceptual conflicts. Henceforth his edition of the SA/SN would have been the authoritative edition, the one copied and used. In the meantime there would not have been a KN/AN authority daring to make a new edition of the KN/AN. What would have been the result? The oldest SA/SN would have got an editorial facelift, would have lost many valuable alternative wordings, while a much younger KN/AN had retained a lot of inconsistent material - that might be in parts poorly remembered but still with more original material.

Now that is just a simplified thought experiment, and maybe we can never uncover the undoubtedly more layered edition history, but I just want to add this aspect, that - under circumstances - the oldest collection might still be of limited value to uncover the original wording of the dhamma, all depending on the venerable editors in history…

1 Like

The term in MN 118 about calming is ‘kaya-sankhara’. In MN 44, the ‘kaya-sankhara’ is identified unambiguously as the breath. Therefore, ‘kaya-sankhara’ obviously does not mean ‘bodily-doing’.

The word ‘sankhara’ does not necessarily mean ‘doing’, ‘fabricating’, ‘fabrication’, '‘conditioning’, ‘condition’, etc. ‘Sankhara’ is an exceptionally broad term in Pali and has a multitude of nuances.

The word ‘sankhara’ here could simply mean ‘doer’, ‘conditioner’, ‘fabricator’, etc. For example, it is the breathing that results in the body being able to do things (such as live, move, exercise, etc) and it is the quality of the breathing (eg. long, smooth, short, agitated, etc) that determines the quality or health of the (physical) body. Thus, breath here can be ‘cause’ & the state of the ‘body’ an effect.

Just as perceptions & feelings (citta-sankhara in MN 44) condition the state of the citta/mind (being greedy, angry, deluded, etc), so does the breathing condition the state of the physical body (being energetic, lethargic, sickly, healthy, stressed, calm, etc).

Perception & feeling (citta-sankhara) are not ‘mental-citta-doing’ since it is defilements (greed, hatred & delusion) & their absence that are always used as the characteristics that describe the citta.

Thus, in MN 118, the phrase ’ calming the citta-sankhara’ means to calm the feelings of rapture & happiness (since feeling/vedana is the citta-sankhara). Similarly, calming the kaya-sankhara, based on the sub-definitions found in MN 44, obviously means to calm the breath.

3 Likes

I am always handy for criticisms. :innocent:

My reading of the article’s conclusion finds no real departure from what I personally regard as the cumbersome standard Pali-English translations. For example, does the Chinese offer any departure from the standard questionable translation of ‘mindfulness OF breathing’ (since ‘mindfulness’, meaning ‘retention’ or ‘remembering’, is not a synonym for ‘awareness’, ‘knowing’, ‘observing’ or ‘contemplating’ - anupassi). There is also the common questionable or vague translation of ‘bodily-formations’? What exactly are ‘bodily-formations?’

Also, the article’s conclusion offers a mix of interpretations of the 3rd stage, which is similar to the diversity found in Theravada. In other words, similar to Theravada interpretations, it is not entirely straightforward & thus helpful in terms of instruction.

However, what I do find interesting :slight_smile: is: "_SA 810 has awareness of all bodily formation_s “. In the Pali, the term is ‘sabba-kaya’. ‘Sabba’ is generally translated as ‘all’ (rather than ‘whole’ or ‘entire’/‘kevala’). MN 118 does state: “in & out breathing is a kaya amongst other kaya” thus the translation: “all bodily sankhara” is consistent with the idea of 'long, short, warm, cold, etc” breathes.

Also, what I find even more interesting :relaxed: is the following verse from EA 17.1 about how the mind affects the breath since if the Buddha was concerned exclusively with the physical body at stage 3, he would have likely have used the term ‘experiencing kaya-sankhara’ (instead of ‘sabba-kaya’), similar to ‘experiencing the citta sankhara’ at stage 7:

If there is an out-breath conditioned by the mind (若息從心出), then he is aware that it is an out-breath conditioned by the mind (知從心出), and if there is an in-breath conditioned by the mind

Note: mentality or ‘nama-kaya’ is also a ‘kaya’ that can be included in scope of ‘sabba-kaya’.

Positively, the article sums up the diversity of Chinese interpretations very well. However, for the skeptic, such diversity of interpretations can be taken as a sign of possible confusion amongst the ancient scholars.

Interpretations of practice

MA 81 describes it as cultivating mindfulness of the body (修習念身). MA 98 calls it observing the body as the body (觀身如身).

The Entire Body: SA 803 has awareness of the entire body (覺知一切身). SA 807 also has awareness of the entire body (一切身覺). SA 810 has awareness of all bodily formations (一切身行覺知). MA 81 and MA 98 have that he trains mindful of the entire body on the in-breath (學一切身息入), and mindful of the entire body on the out-breath (學一切身息出). EA 3.8 has that he completely observes the body (具觀身體), fully observing and knowing it from the head down to the feet (從頭至足皆當觀知).

Calming Bodily Formations: SA 803 has awareness of the calming of bodily formations (覺知一切身行息). SA 807 and SA 810 have resting or calming bodily formations (身行休息). SA 813 has calming bodily formations (身行止息). MA 81 and MA 98 both have that he trains halting bodily formations (止身行).

2 Likes

@llt - thanks for the great article, I wish there were more expositions like these on the agamas, since we don’t seem to have this tradition of english translations
http://lapislazulitexts.com/articles/anapanasmrti_in_the_agamas

open, or new, questions I have…

  • I’m curious about the smrti illustrations in SA 616, 617, 618, 620 (623 is I think in SN 47.20? - oh, you translated 620 and 623 yourself: http://lapislazulitexts.com/tripitaka/samyukta_agama). Also about the ones in SA 813. Where can I find English agama translations on the web anyway? Apart from Bh. Analyo’s expositions I just found this: https://sites.google.com/site/ekottara/home

  • MN118 has a pretty impressive intro, with all famous disciples and many more advanced disciples around. This emphasizes the importance of the sutta, and like you I guess that by this the editors wanted to show that this is core buddhism. Does one of the agama suttas have a similarly dramatic intro? actually this is an interesting question and I’ll make a seperate topic out of it…

  • Maybe at the beginning or the end of the article you could list all the agama suttas you referenced? From reading the text I got the impression that there are a lot, but actually there aren’t.

  • Is there a possibility to have hyperlinked contents at the top?

  • You write that in SA 807 the Buddha practiced for two months anapanasati. Is that a report from before his enlightenment, or at-that-time as a Buddha? That’s a bit weird. Does the Buddha still need the breath as a vehicle for the mind to converge, or to observe cessation?

  • You often reference first ‘joy’ and then ‘bliss’ in the development of a sutra. In the pali tradition we usually find first piti as a stronger joy and then sukha as a more subtle one [if they are not together in pitisukha that is]. In the agamas from the terms you chose it sounds the opposite?

  • I would love to know how the 10 forms of smrti are described in EA 3! Any chance for a future project here?

  • EA17.1 seems like a sutta worth translating (just realizing there is a translation here, by Aj. Sujato?: http://ekottara.googlepages.com/eaxv - and, silly me, by lapislazuli himself: http://lapislazulitexts.com/tripitaka/T0125_LL_17-01_anapanasmrti)

Of course these are just suggestions and possible questions. I know that time is limited. Thanks again

1 Like

In SN 54.11 it’s said that the Buddha spent three months in seclusion in concentration by mindfulness of breathing and afterwards he comments that even for arahants:

and

Basically it seems to me he did it for the same reason he lived in the wilderness, mentioned in MN 4:

2 Likes

don’t know if you noticed, but i guess the content featured there must by now have found home on SC because

The Ekottara reminds us that the Buddha said: 'a monk should know that i always admire and praise someone who knows how to be grateful. So i express my tremendous gratitude to the translators for making this work available, and to all those who have assisted in typing and other ways, especially: Julie, Denise, Marianna, Teresa, Sooi, Nilushi, and Bhikkhuni Samacitta.

Bhikkhu Sujato

No, I haven’t translated any Chinese suttas. Which is good, because i don’t know Chinese, so that works out well! These translations are by Bhikkhu Pasadika, and are available on SC. The link you gave is to an old, unmaintained site.

2 Likes

Wow, was that your translation? That’s impressive!

Maybe we will learn that DN/DA (continuing into its “second volume” MA/MN) is broadly the first doctrinally, and jhana is the meditation instruction at the heart of it.

Then anapana becomes deriguer after the suicide crisis amongst the impurities practitioners.

SN/SA come later, perhaps amongst a generation where jhana practice had effectively been lost, think larger urban institutional monasteries, (dominating the production of oral texts, away from the more likely out bush jhana practitioners) where brahmaviharas, anapanas, satipathanas etc duked it out for influence.

Will think/search some more.

Love the thread.

This is more or less true of DN and MN as well.

SA 809 appears to be a later interpolation into the collection from a similar story in the Vinaya, which is why it takes a form very different than every other text in the anapana samyukta. Ven. Analayo covers this in detail.

1 Like