Are khandhas early or late EBT?

Discussion
121

So just to annotate your list for MN a bit:

MN9 Spoken by Sariputta
MN22 Probably originates in Bu Pc 68
MN23 Agama parallel not in MA
MN28 Spoken by Sariputta
MN35 Agama parallel not in MA
MN44 Spoken by Dhammadinna
MN62 Agama parallel not in MA
MN64 The Agama parallel omits the aggregates and has only vedana
MN72 Agama parallel not in MA
MN75 The Agama parallel omits at least some occurrences of the aggregates
MN102 No Agama parallel
MN109 Agama parallel not in MA
MN112
MN122 focuses on the late attainment suññataṃ
MN131 No Agama parallel
MN132 Original at MN134 does not contain the aggregates
MN138 Spoken by Mahākaccāna
MN141 Spoken by Sariputta
MN143 Spoken by Sariputta
MN147 Agama parallel not in MA
MN148 Agama parallel not in MA
MN149 Agama parallel not in MA
MN151 Agama parallel not in MA

So I guess that my impression from your very helpful list is that about half the occurrences of the khandhas in MN have their Agama counterparts in SA or nowhere, well over half the occurrences occur in the last 50 suttas of MN, and of the occurrences that are in both MN and MA, over half are spoken by persons other than the Buddha.

Metta

2 Likes
122

So just to sum up, MN112 and MN122 are the only MN suttas mentioning the aggregates that are unambiguously

  1. spoken by the Buddha.
  2. whose Agama parallels are in MA and include the aggregates.
    and
  3. don’t originate from the Vinaya or from a version of the same sutta, MN134, omitting them.

and just on MN122 the word suññataṃ occurs at MN122 and nowhere else in the 4 principle Nikayas, occurring only in the Vinaya and Abhidhamma except for this sutta. It is very likely late.

123

I have a strong hunch that you will see a correlation between khandhas and mentions of any of the following: stream-enterer( sotāpanna), once-returner( Sakadāgāmin), non-returner( anāgāmin), and arahant. You may want to try it with and without arahant.

I think you will also see khandhas correlate to mentions of “cessation of consciousness” and “cessation of ignorance”. You may want to split these into different searches. I think it would also correlate with cessation where it is mentioned without being cessation of something, but that would be a more difficult search.

1 Like
124

I’m wondering if the five khandhas is an early Buddhist doctrine borrowed from an unknown sramana sect in which Gotama and his five ascetics trained the self-mortification. We know that the five (clinging) aggregates are mentioned in the first sermon to the first five monks and it seems that they were familiar with the khandhas theory (no explanation the detail of what it is in the first sermon, only in the second sermon the Buddha showed the khandhas are not-self)

125

You can also go through the MA separately and check for suttas on the khandhas. I just went quickly through half of them (I think there are around 222 discourses) and found around 7 suttas where the khandhas are mentioned. So maybe it is mentioned less in MA, but not absent.

1 Like
126

It looks like the aggregates pre-dated the Buddha, since the other ascetics and philosophers at the time seemed to have use them as a way to categorise and ground the self.

1 Like
127

Yes they are definitely there as well, may i ask what method you ised to look for them?

128

And while we’re here, may I just point out a few things about SN with regards to the aggregates?

First of all the khandhavaggasamyutta is the only book of the 5 books of SN that begins with a teaching spoken not by the Buddha but by Sariputta. Sariputta actually gives the first two teachings, then Mahakaccana the next 2. The fifth sutta does not name a speaker, nor does the 6th. The 7th gives;

At Sāvatthī.
Sāvatthinidānaṁ.

“Mendicants, I will teach you how grasping leads to anxiety, and how not grasping leads to freedom from anxiety.
“Upādāparitassanañca vo, bhikkhave, desessāmi anu­pādā­apa­r­itas­sa­na­ñ­ca­.

Listen and pay close attention, I will speak.”
Taṁ suṇātha, sādhukaṁ manasi karotha, bhāsissāmī”ti.

“Yes, sir,” they replied.
“Evaṁ, bhante”ti, kho te bhikkhū bhagavato paccassosuṁ.

The Buddha said this:
Bhagavā etadavoca:

“And how does grasping lead to anxiety?
“Kathañca, bhikkhave, upādāparitassanā hoti?

It’s when an unlearned ordinary person has not seen the noble ones, …

So the first time the Buddha is actually named as the speaker giving the aggregates teaching in the aggregates book of SN, it’s 7 suttas in, and they are mentioned in passing, not introduced at the beginning in the standard way.

The 8th names no speaker, nor do the 9th. 10th, or 11th.
The following chapter gives no speaker for the first 9 suttas, and finishes with the only sutta that mentions the Buddha as the speaker, SN22.21, a text with no recorded parallels.

Compare all this with the nidanavaggasamyutta.

The book opens with sutta 1 giving:

At one time the Buddha was staying near Sāvatthī in Jeta’s Grove, Anāthapiṇḍika’s monastery.
ekaṁ samayaṁ bhagavā sāvatthiyaṁ viharati jetavane anāthapiṇḍikassa ārāme.

There the Buddha addressed the mendicants,
Tatra kho bhagavā bhikkhū āmantesi:

“Mendicants!”
“bhikkhavo”ti.

“Venerable sir,” they replied.
“Bhadante”ti te bhikkhū bhagavato paccassosuṁ.

The Buddha said this:
Bhagavā etadavoca:

“Mendicants, I will teach you dependent origination.
“paṭiccasamuppādaṁ vo, bhikkhave, desessāmi.

Listen and pay close attention, I will speak.”
Taṁ suṇātha, sādhukaṁ manasi karotha, bhāsissāmī”ti.

The next sutta is an analysis of the previous one, and interestingly, it does not name the buddha as the speaker at it’s head, but uses the incidental bhagavā etadavoca.

The third sutta does the same, then the next 6 suttas put the teaching in the mouths of previous buddhas culminating in the final version being given by Gotama.

The next chapter then opens with:

At one time the Buddha was staying near Sāvatthī in Jeta’s Grove, Anāthapiṇḍika’s monastery. …
ekaṁ samayaṁ bhagavā sāvatthiyaṁ viharati jetavane anāthapiṇḍikassa ārāme …

In the first 2 chapters of the nidanavaggasamyutta no teacher other than the Buddha (or past Buddhas) is given, compared with the khandhavaggasamyutta where the first 4 suttas of chapter one are given by persons other than the Buddha. The first teaching in nidanavaggasamyutta not given by the Buddhas is SN12.24.

And unless I missed one, the Buddha is not given a full formal introduction with a named place etc (where they are the ones giving the teaching) in the khandhavaggasamyutta until SN22.49 which gives:

At one time the Buddha was staying near Rājagaha, in the Bamboo Grove, the squirrels’ feeding ground.
ekaṁ samayaṁ bhagavā rājagahe viharati veḷuvane kalandakanivāpe.

Then the householder Soṇa went up to the Buddha …
Atha kho soṇo gahapatiputto yena bhagavā tenupasaṅkami …pe…

The Buddha said to him:
ekamantaṁ nisinnaṁ kho soṇaṁ gahapatiputtaṁ bhagavā etadavoca:

The two Sona suttas are the only suttas in the first fifty that give the Buddha as the speaker and a location and then have the Buddha actually give the teaching.

And hears the kicker; in the SA parallel, SA30, guess what? it is Sariputta who instructs Sona, not the Buddha, and this is also the case with the other sona sutta.

So unless I missed one, there is no account in the first 50 suttas of the khandhavaggasamyutta where the Buddha is introduced at a location as the speaker, and gives the aggregates teaching, that has a parallel in the Agamas.

Metta

1 Like
129

OK, one more, in the remaining suttas of the khandhasamyutta that is the first and longest chapter of the khandhavaggasamyutta:

SN22.59, the famous Anattalakkhaṇasutta is the first time in SN22 that

  1. The Buddha is the speaker
  2. A location is given
  3. The Buddha gives the discourse
    and
  4. There is a parallel in SA with 1-3

Of the remaining suttas in the khandhasamyutta

SN22.60 (SA81 T99 020 b28)
SN22.63 (SA15 T99 003 a06)
SN22.80 (SA272 T99 071 c14)
SN22.81 (SA57 T99 013 c07)
SN22.82 (SA58 T99 014 b12)
SN22.87 (a gravely ill sutta)
SN22.88 (a gravely ill sutta)
SN22.95 (SA265 T99 068 b29)

are the only suttas with the Buddha at a location given as the initial speaker.

A few remarks:

The Agama parallel to SN22.60, SA81 gives a lengthy exchange between the monk and PUrana Kassapa first , and the interaction with the Buddha second.
At SN22.63 it is the monk who gives the aggregates teaching as an explination of the BUddhas breif teaching, this is then repeated and approved by the Buddha.
SN22.80 is overflowing with late motifs, the Buddha dismisses the sangha, regrets it, has a god appear to them, uses psychic powers to assemble the sangha and then gives the aggregates teaching.
SN22.81 is also pretty odd, first the Buddha wants to be alone so goes away without telling anyone, then Ananda says that no one should follow him, then Ananda follows him with a group of monks, Then the Buddha, after giving a talk, reads the mind of one monk, and then gives an aggregates talk.
SN22.82 continues the theme of mind reading with the classic:

Now at that time one of the mendicants had the thought:
Tena kho pana samayena aññatarassa bhikkhuno evaṁ cetaso parivitakko udapādi:

“So it seems, good sir, that form, feeling, perception, choices, and consciousness are not-self.
“iti kira bho rūpaṁ anattā, vedanā … saññā … saṅkhārā … viññāṇaṁ anattā;

Then what self will the deeds done by not-self affect?”
anattakatāni kammāni kathamattānaṁ phusissantī”ti.

Then the Buddha, knowing what that monk was thinking, addressed the mendicants:
Atha kho bhagavā tassa bhikkhuno cetasā ceto parivitakkamaññāya bhikkhū āmantesi:

…

Now, mendicants, you have been educated by me in questioning with regards to all these things in all such cases.
Paṭipucchāvinītā kho me tumhe, bhikkhave, tatra tatra tesu tesu dhammesu.

What do you think, mendicants?
Taṁ kiṁ maññatha, bhikkhave, Is form permanent or impermanent?” rūpaṁ niccaṁ vā aniccaṁ vā”ti?

“Impermanent, sir.”
“Aniccaṁ, bhante”.

“Is feeling …
“Vedanā …

perception …
saññā …

choices …
saṅkhārā …

consciousness permanent or impermanent?”
viññāṇaṁ niccaṁ vā aniccaṁ vā”ti?

“Impermanent, sir.”
“Aniccaṁ, bhante”.

“But if it’s impermanent, is it suffering or happiness?”
“Yaṁ panāniccaṁ dukkhaṁ vā taṁ sukhaṁ vā”ti?

“Suffering, sir.”
“Dukkhaṁ, bhante”.

“But if it’s impermanent, suffering, and perishable, is it fit to be regarded thus:
“Yaṁ panāniccaṁ dukkhaṁ vipariṇāmadhammaṁ, kallaṁ nu taṁ samanupassituṁ:

‘This is mine, I am this, this is my self’?”
‘etaṁ mama, esohamasmi, eso me attā’”ti?

“No, sir.”
“No hetaṁ, bhante”.

“So you should truly see …
Tasmātiha …pe…

Seeing this …
evaṁ passaṁ …pe…

They understand: ‘… there is no return to any state of existence.’”
nāparaṁ itthattāyāti pajānātī”ti.

(At least one way this exchange can be understood is that a monk posits anatta as a metaphysical reality and the Buddha condemns this thesis as being an inappropriate extension of the teaching that phenomena cannot be a self. )

The next two examples are of the “gravely ill” variety, where a monk is gravely ill and the Buddha or another senior monastic visits to give a teaching. I will pass over these as being somewhat less clearly “on the public record” than suttas addressed to the whole of the present sangha.

This leaves SN22.95 a lovely sutta, giving the only occurances of foam pheṇapiṇḍe mirage marīcikā water bubbles udakapubbuḷe , māyākārantevāsī in the early canon…

It is also the only sutta that takes place at ayujjhāyaṃ.

OK. That’s enough form me for a bit.

Metta

130

Great remarks and data :slight_smile:

‘At Sāvatthī’ does not fit the description given in the Vinaya for how to construct an introduction artificially. It is much less complex or elaborate — very bare bones. It assumes the Buddha as the teacher. In many ways it seems to represent what a smaller community would do in compiling doctrinal discourses to memorize from their teacher, rather than other types of discourses we see from a larger more developed community both in number, mythologically, within the larger society, etc.

No complex discussions between disciples, elaborate interactions with deities, complex philosophical arguments against other sects and monastics in elaborate formulaic and narrative content. Just simple doctrinal teachings on essential philosophical concepts for insight with one assumed teacher.

It is very possible that this is more representative of original introductions and original formatting that was then later replaced by the very common, formulaic nidāna we are all familiar with. After all, if people were composing or compiling a sutta later on, we know that this longer introduction is the type they would almost always give it.

That is to point out that this information does not necessarily point to the lateness of the khandhas. It could point to the earliness of the discourses and the an arrangement (or composition) into the nikāyas based on content, format, length, and style.

It could go many different ways. It may in fact be later, or not. Even if it is, what ‘later’ would mean is not clear.

Mettā

1 Like
131

Yes, I agree with that, it’s entirely possible that many of the Sāvatthinidānaṁ suttas are in fact early, and that the more elaborate locations are late.

Similarly, in my discussion of MN aggregates suttas often having Agama parallels in SA rather than MA, it is entirely possible that the Sarvastivada school moved the suttas to SA from MA precisely because they where collecting all those suttas (or most of them) in one Agama.

For me it is really about trying to define what might be a “gold standard” for taking a sutta or doctrine to be early, and at the moment I am using something like:

  1. The teaching occurs in all 4 principle NIkayas
  2. The Agama parallel has the same content
  3. The Agama parallel occurs in the same collection as the NIkaya parallel
  4. The Buddha is introduced at a location and is directly addressing the sangha
  5. The Agama parallel gives the same location and the same audience

Something like that.

Now it is entirely possible that material meeting literally none of these criteria is in fact early, and material meeting every one of them is in fact late, we can never know for sure sans manuscript or other evidence showing up from the 3rd century BCE or something.

However I am pretty comfortable with my methodology, and would note that there are plenty of other formulaic teachings that would not collapse in the way that the aggregates in MN do with the same approach.

I would also just note that there are in fact several more suttas in the khandhasamyutta that do give a location in the traditional way, it’s just that they are almost always sariputta teaching.

Metta.

132

Not really a method, the first 131 are translated by the BDK group, so it was easy to go through the PDF and look for ‘aggregate’ or ‘volitional formation’.

1 Like
133

It’s great that you put all that work into a reconstruction of the suttas! Any criteria will produce false-positives and false-negatives, this just can’t be avoided.

Personally, I don’t trust locations at all. Also if the speaker is just casually mentioned. Words can be put into the Buddha’s mouth, just as so often he confirms what someone else says - which to me often sounds like legitimizing a commentary. But leaving these methodical questions aside - where are you in your understanding of the khandhas now?

1 Like
134

Khandha shows up only once in Snp, but not as aggregate. It appears to mean trunk based on the translation. I would count that as a zero.

1 Like
135

Here’s a short Q&A topic with attention on Ven. Sujato’s reply that might add something to this discussion.

2 Likes
136

Basically my impression is that the teaching is one that seems to respond to a shift in emphasis from “why am i suffering?” to “why am I?”.

I suspect the reason may be the growth of a proto-Vedanta “atman equals brahman” type of brahmanism that is otherwise absent from the EBT’s.

It takes several of the elements of DO and rearranges them into a comprehensive picture of “phenomenal experience” and shows how we can’t be any of those things, or any combination of those things, or speak of or conceptualize anything outside those things.

It seems to have a particular association with Sariputta, a particular association with SN, a particular association with the gravely ill and dying.

I have mostly spent the last year interrogating it in one way or another because it seems to me to be one of the most often misunderstood doctrines in early Buddhism.

I think it probably obscures the depth and profundity of DO and the undeclared points for many, who end up taking the doctrine to be something like “persons are fictions but there really are five aggregates.” which I think is wrong on just about every conceivable level.

Metta.

2 Likes
137

This I think is very true generally. We know that the Theravādins particularly are infamous for their fire and brimstone no-self anxiety, seemingly born against the puggalavādins. But generally in Buddhism as well we see this. Hardcore momentariness; Ultimate Reality of Dhamma categories vs. Conventional Reality; etc. The whole Abhidhamma project in general seems to have tended towards this, but so did anti-Abhidhamma or essentialist movements like some Mahāyāna responses — in their own round-about ways.

I also agree it changes DO and the four noble truths. I’ve noticed in Western Buddhism, the 4NT are very commonly understood as “we suffer psychologically because we desire things.” Rather than “the craving for existence drives the production of existence and the propelling of consciousness forward via a process of localized appropriation and delight.” Some of this is similar in certain regards, dissimilar in others.

Another way of talking about this all is that I notice a great shift in Buddhism: A shift that goes from practicing to see and realize the Four Noble Truths, to practice for seeing the Three Characteristics, and/or emptiness. It’s not that either one of these was ever absent or completely separate, but the philosophical and methodological emphasis makes a big impact. The ‘three characteristics +/- emptiness’ ideology certainly won out and continues to reign supreme in pretty much every domain. In much of Mahāyāna insight it seems the 4NT are basically an irrelevant side point and seeing the lack of inherent svabhāva in all dhammas is what Buddhism is about.

Mettā

3 Likes
138

This could imply a competitive “Who has the best description of reality?” - ontological arguments between sects.

That is unlikely in this form. See, Brahmins were all about the bandhus, equating microcosm and macrocosm - this part of the body is the West, that one the East, the up-breath is this, the down-breath is that, etc… The khandhas are decidedly microcosmic internal qualities that are not equated with an outside. (A dhamma aspect which would fit the bandhus better are the dhatus: inner-water and outer-water, etc.)

Right, the DO is vertical in either time or logic, while the khandhas are confusingly horizontal - just a list of qualities without detailing their interrelation but obviously on different abstraction levels - form and consciousness just don’t belong in the same party.

Is this connected to the khandhas?

I think I understand what you mean. For a while now I find it annoying how the suttas present a simplified pseudo-logic of anicca–>dukkha–>anatta without sufficient explanation, and good parts of Theravada followers act as if there’s nothing more self-evident than this deduction, which seemingly catapults them to the understanding of the true anatta-doctrine. No need for a humble practice of sila-samadhi-panna or gradual path, you just need to walk around with a mantra that everything is obviously anatta.

2 Likes
139

continuing our ongoing analysis of the aggregates as a doctrinal category, move from the khandhavaggasamyutta to the other books of SN:

rūpaṃ niccaṃ vā aniccaṃ (is form constant or inconstant) occurs

VN: 1
DN: 0
MN: 3
SN: 45
AN: 0
KN: 0
AB: 0
VM: 0

of the 45 occurrences in SN, all bar 3 occur in SN22 and SN24.

these three occurrences are at SN12.70, SN18.10, and SN44.2.

The parallel to SN12.70 is SA347.

Guess what?

Yeah you guessed already didn’t you? It omits any mention of the aggregates.

( i should note I am relying on machine translations here, so those with real knowledge please feel free to correct me :slight_smile: ).

The Agama parallel just starts off with the Buddha giving the DO, without the aggregates

Also, quite mind blowing, the Agama version does not give the psychic powers as what is lacking in the enlightened by wisdom monks, but the Jhanas!

This is all pretty wild IMO.

Next up is SN18.10 Agama parallel SA897. I’ll let you guess again…

yep, SA897 gives the sense bases, not the aggregates formula.

Finally we have SN44.2, SA106 is the parallel, a sutta that qoutes another sutta! (another interest of mine) and it DOES give the aggregates formula.

HOWEVER, SN44.2 is more or less an exact copy of SN22.86. and the Agama version refrences SN22.85 by name, so that’s also in SN22, and of course, it’s a Sariputta sutta.

So in conclusion, outside of SN22 and SN24, the is from permanent or impermanent pericope is once again rare, and often absent from the Agama parallels.

Metta.

3 Likes
140

One more while I am here. SN22.59 has two parallels in SA, SA33 and SA34.

SA33 is the first chapter of the second volume of the samyukta.

It gives the location of the teaching as at Anāthapiṇḍada Park in Jeta Grove in Śrāvastī and gives the audience as just the monks.

This is then followed by SA34, the second sutra in the chapter, that is identical to the first, except it changes the audience to Deer Preserve in the Deer Preserve Park among the sages of the country of Benares and the audience to the five monks, and adds at the end;

“translation”: “After the Buddha taught this sūtra, the other five monks didn’t give rise to the contaminants, and their minds were liberated.”,

but leaves the now redundant line form the previous sutta;

“translation”: “After the Buddha taught this sutra, the monks who heard what he taught rejoiced and carried it out.”

after it.

I am really pretty amazed no-one has ever mentioned this to me before, I guess you learn something new every day.

Metta.

1 Like