Are khandhas early or late EBT?

In recent discussions the opinion came up that the khandhas, because they hardly appear in the DN, would be a late dhamma category. I haven’t heard this understanding before and I just assumed that the khandhas are one of the most basic EBT categories.

In more detail the khandhas appear

  • 4x in the DN: DN 14, DN 22, and not surprisingly in DN 33, DN 34
  • 17x in the AN, which is a very low number
  • 20x in the MN, a decent amount
  • many times in the SN obviously, mostly in SN 22-24: —5x in SN 1-11, —7x in SN 12-21, —7x in SN 25-34, —11x in SN 35-44, —5x in DN 45-56
  • 0x in the Snp
  • 2x in the Dhp

I might have missed a few, but that’s more or less the picture.
So while re. the DN we could argue that it’s a collection geared towards general audiences and for conversion. But why don’t the khandhas appear regularly in the AN, why not at all in the Snp?

So maybe you guys have ideas about the distribution or remember some research that’s been done about it.


I think you will find that the aggregates are used mostly to argue against the idea of an unchanging Atman. I have read that the notion of the unchanging Atman came later. The notion of self that was the main concern earlier was that of being a person in the world. The way to combat that was deep stages of samadhi.

If someone is willing to do the work, my guess is you will find more references to “impermanence” or “impermanent” where you find references to the aggregates and you will find more references to “the world” or “namarupa” where you don’t.

The Buddha of the Chapter of Eights seems to be dead set against theorizing and view creation so I suspect he would have been hesitant to introduce them.

1 Like

I would just add that where you find the world/namarupa/name form and the aggregates together would be where the links of dependent arising and immaterial states/jhanas are found. I think the links of dependent arising is later theorizing. Immaterial states, I am not so sure about. Maybe it was expressing namarupa in terms of the five aggregates and a later harmonization.

These are my ideas that need to be tested by looking at the suttas. I do not claim to know for certain.

With regard to DN, 22, 33, and 34 I think we can all agree are Suttas that either (in the case of 22) show significant development over time (for a more in depth discussion of this see @sujato 's HIstory of Mindfullness ) or are simply lists of orthodox Buddhist concepts without any doctrinal content.

With regards to DN 14, I simply note that the discourse is quite ornate compared to much of DN, and that the aggregates are placed immediately after Vipassī elucidates the (10 link) DO.

The sutta then gives;

Some time later Vipassī meditated observing rise and fall in the five grasping aggregates.
Atha kho, bhikkhave, vipassī bodhisatto aparena samayena pañcasu upādānakkhandhesu udayabbayānupassī vihāsi:

‘Such is form, such is the origin of form, such is the ending of form.
‘iti rūpaṁ, iti rūpassa samudayo, iti rūpassa atthaṅgamo;

Such is feeling, such is the origin of feeling, such is the ending of feeling.
iti vedanā, iti vedanāya samudayo, iti vedanāya atthaṅgamo;

Such is perception, such is the origin of perception, such is the ending of perception.
iti saññā, iti saññāya samudayo, iti saññāya atthaṅgamo;

Such are choices, such is the origin of choices, such is the ending of choices.
iti saṅkhārā, iti saṅkhārānaṁ samudayo, iti saṅkhārānaṁ atthaṅgamo;

Such is consciousness, such is the origin of consciousness, such is the ending of consciousness.’
iti viññāṇaṁ, iti viññāṇassa samudayo, iti viññāṇassa atthaṅgamo’ti,

Meditating like this his mind was soon freed from defilements by not grasping.
tassa pañcasu upādānakkhandhesu udayabbayānupassino viharato na cirasseva anupādāya āsavehi cittaṁ vimuccīti.

However in the next section, the appeal of brahma, we have

Then the Blessed One Vipassī, the perfected one, the fully awakened Buddha, thought,
Atha kho, bhikkhave, vipassissa bhagavato arahato sammāsambuddhassa etadahosi:
‘Why don’t I teach the Dhamma?’
‘yannūnāhaṁ dhammaṁ deseyyan’ti.

Then he thought,
Atha kho, bhikkhave, vipassissa bhagavato arahato sammāsambuddhassa etadahosi:

‘This principle I have discovered is deep, hard to see, hard to understand, peaceful, sublime, beyond the scope of logic, subtle, comprehensible to the astute.
‘adhigato kho myāyaṁ dhammo gambhīro duddaso duranubodho santo paṇīto atakkāvacaro nipuṇo paṇḍitavedanīyo.

But people like attachment, they love it and enjoy it.
Ālayarāmā kho panāyaṁ pajā ālayaratā ālayasammuditā.

It’s hard for them to see this thing; that is, specific conditionality, dependent origination.
Ālayarāmāya kho pana pajāya ālayaratāya ālayasammuditāya duddasaṁ idaṁ ṭhānaṁ yadidaṁ idappa­c­ca­yatā­paṭi­c­ca­samu­p­pādo­.

It’s also hard for them to see this thing; that is, the stilling of all activities, the letting go of all attachments, the ending of craving, fading away, cessation, extinguishment.
Idampi kho ṭhānaṁ duddasaṁ yadidaṁ sabbasaṅkhārasamatho sabbū­pa­dhi­­paṭinissaggo taṇhākkhayo virāgo nirodho nibbānaṁ.

Which gives DO as “this thing” and has no mention of 5K.

It seems pretty clear from that that the 5K formula in DN14 is a late substitution for the simpler formula as for example at MN36

When my mind had immersed in samādhi like this—purified, bright, flawless, rid of corruptions, pliable, workable, steady, and imperturbable—I extended it toward knowledge of the ending of defilements.
So evaṁ samāhite citte parisuddhe pariyodāte anaṅgaṇe vigatūpakkilese mudubhūte kammaniye ṭhite āneñjappatte āsavānaṁ khayañāṇāya cittaṁ abhininnāmesiṁ.

I truly understood: ‘This is suffering’ … ‘This is the origin of suffering’ … ‘This is the cessation of suffering’ … ‘This is the practice that leads to the cessation of suffering.’
So ‘idaṁ dukkhan’ti yathābhūtaṁ abbhaññāsiṁ, ‘ayaṁ dukkhasamudayo’ti yathābhūtaṁ abbhaññāsiṁ, ‘ayaṁ dukkhanirodho’ti yathābhūtaṁ abbhaññāsiṁ, ‘ayaṁ dukkhanirodhagāminī paṭipadā’ti yathābhūtaṁ abbhaññāsiṁ.

I truly understood: ‘These are defilements’ … ‘This is the origin of defilements’ … ‘This is the cessation of defilements’ … ‘This is the practice that leads to the cessation of defilements.’
‘Ime āsavā’ti yathābhūtaṁ abbhaññāsiṁ, ‘ayaṁ āsavasamudayo’ti yathābhūtaṁ abbhaññāsiṁ, ‘ayaṁ āsavanirodho’ti yathābhūtaṁ abbhaññāsiṁ, ‘ayaṁ āsavanirodhagāminī paṭipadā’ti yathābhūtaṁ abbhaññāsiṁ.

Knowing and seeing like this, my mind was freed from the defilements of sensuality, desire to be reborn, and ignorance.
Tassa me evaṁ jānato evaṁ passato kāmāsavāpi cittaṁ vimuccittha, bhavāsavāpi cittaṁ vimuccittha, avijjāsavāpi cittaṁ vimuccittha.

When it was freed, I knew it was freed.
Vimuttasmiṁ vimuttamiti ñāṇaṁ ahosi.

I understood: ‘Rebirth is ended; the spiritual journey has been completed; what had to be done has been done; there is no return to any state of existence.’
‘Khīṇā jāti, vusitaṁ brahmacariyaṁ, kataṁ karaṇīyaṁ, nāparaṁ itthattāyā’ti abbhaññāsiṁ.

I think it is therefore a fair assessment to make that in DN the 5A are absent except in places where they appear late.

I haven’t done an analysis of the MN appearances, but I would also make the observation that 40-45 years is a long time to teach and that Northern India is a very big place, so I have no difficulty at all in believing that 5A was a teaching from the Buddha, in their lifetime, perhaps directed specifically towards people with a strong leaning to some pre-existing doctrine of an immortal soul, but it does seem to me, that assuming (as you know I do) that DN is the earliest N then is seems clear that 10DO predates 12DO and DO IMHO predates 5A which is a version of DO for the purpose of analyzing the “person”.


As I say I haven’t really had an opportunity to think too much about MN but for example MN28 is quite dense and seems to combine multiple layers of analysis, mixing the sense bases with the aggregates with the 4 elements, with DO, but even it in the end seems to betray a little anxiety with regards to needing to justify 5A in terms of DO:

But the Buddha has also said:
Vuttaṁ kho panetaṁ bhagavatā:

“One who sees dependent origination sees the teaching.
“yo paṭiccasamuppādaṁ passati so dhammaṁ passati;

One who sees the teaching sees dependent origination.”
yo dhammaṁ passati so paṭiccasamuppādaṁ passatī”ti.

And these five grasping aggregates are indeed dependently originated.
Paṭiccasamuppannā kho panime yadidaṁ pañcupādānakkhandhā.

The desire, adherence, attraction, and attachment for these five grasping aggregates is the origin of suffering.
Yo imesu pañcasu upādānakkhandhesu chando ālayo anunayo ajjhosānaṁ so dukkhasamudayo.

Giving up and getting rid of desire and greed for these five grasping aggregates is the cessation of suffering.’
Yo imesu pañcasu upādānakkhandhesu chandarāgavinayo chandarāgappahānaṁ so dukkhanirodho’ti.

At this point, much has been done by that mendicant.”
Ettāvatāpi kho, āvuso, bhikkhuno bahukataṁ hotī”ti.

That’s what Venerable Sāriputta said.
Idamavoca āyasmā sāriputto.

Satisfied, the mendicants were happy with what Sāriputta said.
Attamanā te bhikkhū āyasmato sāriputtassa bhāsitaṁ abhinandunti.

one further thought; MN28 is a Sariputta sutta, MN23 the riddle is presented to Kassapa Kumara, and MN44 is a conversation with Dhammadinnā.so for at least the first 100 suttas in MN the grasping aggregates are not put directly into the mouth of the Buddha except in the Kassapa Kumara case.

I should mention that my research at this point involves pasting “upādānakkhandhā” into the searchbar here and browsing, so hardly an exhaustive or informed analysis!

(I should say i haven’t mentioned MN10 for the same reason as the DN case)

Post has been edited April 3 as a result of feedback, I am now including the following disclaimer:

“To clarify: the following comments (this and the second comment) are about a general trend, and not necessarily implying that OP is displaying a personally conspiratorial mindset here or supporting this trend.”

Original post:

Why should the number of times a word appears in a collection mean anything in the absence of other evidence? 17 references is actually more than I would have guessed for the AN.

If anything, the numbers shown show a high level of cross referencing for the idea of the khandhas across multiple collections. There are many other concepts which aren’t referenced directly in the Snp too, who said the Snp had to reference everything?

If the Buddha were here now, I would expect to be able to have a discussion with him about the khandhas as if the suttas were written yesterday. Outside of potentially one verse which has been open to interpretation, is there any actual evidence of doctrinal development re the khandhas in the Pali canon? If this happened, why didn’t anyone notice and raise a fuss, as happened for heretics like the Pudgalavadins, who tried to innovate on this point?

These days there seem to be people around who want to try to pick holes in basic Buddhism for no good reason. I think it is conspiracy theory type stuff which has been influenced by Christianity. But Buddhism doesn’t need a Bart Ehrman because unlike Jesus’ teachings in the New Testament, the basic doctrinal categories of the suttas can more or less be reliably be traced to the Buddha. Which is what we would expect, because Buddhism came from a sophisticated culture of preserving oral texts, as opposed to being the post-hoc doctrinal creation of the Council of Nicaea. :thinking:


Sometimes a larger context helps give a better orientation. For example, the five khandhas hardly receive a passing mention in Mahayana sutras, which are obsessed with emptiness, paramitas, and so forth. If the khandhas were to come up, it would just to say they are empty along with all the other basic constituents of Abhidharma like the ayatanas and dhatus. Buddhist thought did go through a process of dropping older concepts and picking up new ones over the arc of its history, and we see some evidence of that in the early Buddhist canons as well, just not in such high contrast.


I think that there is pretty good evidence to suggest that the Pali canon formed over a period of several hundred years and that there is internal evidence of evolution both in terms of doctrinal content and language. I for one find it very helpful to research and examine and think about which explanations might have come first and which later, why and how they developed the way that they did, and which, if any of them might post-date the lifetime of the Buddha.

I for one am not arguing that the 5A come from after the Buddhas time, just from later in the teaching career than the 10 link DO which appears to me to come earlier than the 12 link DO, I have my own theories about why this would have occurred in the Buddhas teaching career; basically because in response to his teaching competing teachers sharpened their soul-theory teachings and ways of educating people needed to adapt.

I think that the the idea that

Is not really true, for example, many on here would reject Tathagatagarba doctrine as being not-buddhist, while many Mahayana followers would claim that it is. To say that it is OK to critique and reject Mahayana literature on the basis of historical and comparative analysis but then say that applying the same principles to Therevadan orthodoxy is to be a “conspiracy theorist” betrays a certain amount of sīlabbataparāmāso if I may be so bold.

I agree that expecting Snp to have too much to say about the doctrinal formulae in prose is probably expecting too much.

Finally I want to say that I am not trying to pick holes, just trying to understand, and I don’t think that any of DO or the 5A are “basic” rather that they are subtle and profound, and worth thinking critically about. Doing that is not to doubt the truth of Buddhism, it is just being open to the idea that the medium in which it has been preserved is imperfect and that secular, critical study of it may help to penetrate it’s mysteries.

1 Like

So the distribution could be the result of a shift in focus, is that what you’re saying? A model would then be that later bhanakas/composers/compilers would keep in mind the contemporary needs their audiences have, drop certain topics (like khandhas), pick up certain topics (like antibrahmanism). And the SN (and MN) could represent a more conservative selection, while AN and DN could focus more on lay needs and thus drop the (maybe by then less evocative) khandhas?

That’s correct, the vast majority references the khandhas either in context of arising and disappearing, impermanence, or as not me/mine/my-atta.

I don’t understand the second part - could you elaborate?

Would you think the same if I wrote that the khandhas appear 17 times in the SN, that it’s actually more than you thought? The logic is simple: if a concept is fundamental it should have been repeated often, in order to make sure that also common monastics in EBT times know it. In relation to the almost 10.000 suttas in the AN, 17 is a meager number.

Please refrain from insinuations. Raising questions is the purpose of this forum.

“Protestant Buddhism” is a well-researched strand of Western engagement with Buddhism.


Sure, but Protestant Buddhism is hardly a conspiracy theory, it’s just not Theravada Buddhism, and it’s also not the case that trying to understand how a doctrine might have evolved or changed over time within a canon of literature is necessarily motivated by the antipathies towards the supernatural that characterizes “Protestant” outlooks on scripture. Saying that there is “no good reason” to examine the corpus from a critical perspective, that to do so is “conspiracy theory stuff” is not charitable and also seems to clearly indicate an attachment to views about the canon that the canon itself explicitly states are bad Buddhism!

“Mendicants, if others criticize me, the teaching, or the Saṅgha, don’t make yourselves resentful, bitter, and exasperated.
“Mamaṁ vā, bhikkhave, pare avaṇṇaṁ bhāseyyuṁ, dhammassa vā avaṇṇaṁ bhāseyyuṁ, saṅghassa vā avaṇṇaṁ bhāseyyuṁ, tatra tumhehi na āghāto na appaccayo na cetaso anabhiraddhi karaṇīyā.

You’ll get angry and upset, which would be an obstacle for you alone.
Mamaṁ vā, bhikkhave, pare avaṇṇaṁ bhāseyyuṁ, dhammassa vā avaṇṇaṁ bhāseyyuṁ, saṅghassa vā avaṇṇaṁ bhāseyyuṁ, tatra ce tumhe assatha kupitā vā anattamanā vā, tumhaṁ yevassa tena antarāyo.

If others were to criticize me, the teaching, or the Saṅgha, and you got angry and upset, would you be able to understand whether they spoke well or poorly?”
Mamaṁ vā, bhikkhave, pare avaṇṇaṁ bhāseyyuṁ, dhammassa vā avaṇṇaṁ bhāseyyuṁ, saṅghassa vā avaṇṇaṁ bhāseyyuṁ, tatra ce tumhe assatha kupitā vā anattamanā vā, api nu tumhe paresaṁ subhāsitaṁ dubbhāsitaṁ ājāneyyāthā”ti?

“No, sir.”
“No hetaṁ, bhante”.


I am personally a practicing Buddhist lay follower and certainly do not want to get into rows with monastics, but I also think it is really important to recognize that the EBT’s are the common heritage of all Buddhists, monastic and lay, and that thy are not the exclusive purview of Theravada practitioners, something I definitely do not profess to be.

So for those of us that are not a member of a pre-exisiting school of buddhism I think it is more or less unavoidable that some evaluation and analysis of the textual material takes place as without a tradition to explain it how else will we work it out?

That Therevadins believe the whole of the tripitaka is the Buddha-vacana is a non starter for me as I think that the evidence that the abbhidhamma is later than the buddha is overwhelming even simply based on the Pali material itself.

Therefore it is not open to me to simply believe that the tripitaka can be “reliably traced to the buddha” and so I have to read and think critically and engage in discussions, and this forum has been by far the most welcoming and stimulating place to do it.

I realize that sometimes these discussions can touch a nerve amongst people who have differing opinions about the extent to which the canon represents the word of the buddha and how, but I hope it remains a place where robust debate and discussion can occur.


1 Like

I have to disagree with you here. Not only are they not mentioned, they are arguable not alluded to either and I think for good reason.

  1. The self the Buddha is not interested in is not an unchanging Ataman, but a person in the world, that is, nama-rupa. The five aggregates appear to have been tailored to target the unchanging Atman view specifically.
  2. The Buddha at this point is dead against theorizing. Why would he then theorize about a person consisting of bundles? Kamma is not even mentioned in “The Chapter of Eights”. Rebirth is just something people cling to because they fear death.
  3. No frameworks are used or alluded to in the Cof8s. The development of frameworks is theorizing.
  4. The Buddha at the Cof8s only uses terms that would have predated his teaching. The experts here will have to fact check this.

I’m not sure that’s completely true. Snp4.11 has a version of dependent origination: SuttaCentral

1 Like

You are correct. That said, these are observations that anyone can make. They are very mundane. We don’t just have to take the Buddha’s word for it. They do not go to the dizzying heights that the 12 links go through to explain rebirth.

Minus some poetic references to gods maybe, I think the lack of unverifiable claims is remarkable. I would go as far to say that the Cof8s could serve as a micro canon for Secular Buddhists. It is a complete Eightfold Path. I wonder which came first.

1 Like

On a tangential note: it’s worth observing that the khandhas, if considered as what a person consists of, is to be taken experientially or phenomenologically.

The khandhas are not presented as ingredients out of which experience arises for an individual (i.e., things prior to experience—which would then make them inaccessible to direct insight), but as facets through which experience is grasped, which gives rise to the notion of selfhood. It is the grasping (upādāna) in pañcupādānakkhandhā that could be pointed out as atta, and not the khandhas themselves.

In other words, the khandha ‘theory’ is not really a theory but another (entirely practical) way of contemplating the puthujjana experience.


I think it can be thought of that way, but I think that would qualify as a view, especially when it is used to “prove” the nonexistence of the unchanging Atman.

For me this description by Bhikkhu Analayo sheds some light on the relationship between the SN/SA and AN/EA.

As a result of the inclusion of discourses with an emphasis on doctrinal matters in the topic-wise arrangement in the saṃyukta collections, the discourses left for assignment according to the numerical principle tend to cover more practical matters and everyday issues, which makes the numerical collection particularly suited to the concerns of the laity, although the collection also has several passages that are related to monastic discipline.
Source: Bh. Analayo, Dirgha Agama Studies

Another thing that might be the case is that the AN/EA was ‘open’ longer. So maybe sutta which came to light a little later were subsequently placed in the AN. At least that is what happened with the EA.

The Ekottarika-āgama extant in Chinese appears to have remained open to later influences for a considerably longer time than the other Āgamas; in fact a whole discourse appears to have
been added and others seem to have been reworked after the collection had reached China.
Source: Bh. Analayo, Dirgha Agama Studies

As for the Snp
In SN 22.3 a passage from the Atthakavagga is quoted and subsequently analyzed by MahaKaccana. In the poetic verses most of us wouldn’t see a link with the khandhas, but MahaKaccana draws in his explanation the relationship with the khandhas. So it could be that the khandhas are not directly named in the Snp, but are being hinted at. So maybe the khandhas are more an ‘insiders concept’, and that is why they dominate the SN/SA.


This could make sense. If we look at Bodhi’s introduction and the “thematic guide to the Anguttara” we see under “X.6 The Domain of Wisdom” only few entries for all categories: DO, khandhas, anatta, 4NT.

As oversimplifying as it is, but maybe nikaya/agama bhanakas were not only specialized on certain texts but on purposes of texts too. And then it would make sense that the doctrinal collections were closed earlier, and that collections with matters concerning life, laity, non-liberation rebirth mechanics (i.e. gods), religious giving, etc. remained open longer, simply because these topics and needs themselves were more in flux in the first centuries post-Buddha.

This could well be, yes. It would fit in the sense that the SN would represent material for dhamma teachers, as I remember Bodhi suggests somewhere in his intro. So ‘khandha’ could have served as a memo for more detailed elaborations, whereas different audiences would have gotten less abstract headlines and more content.

Still, it’s strange to me that the khandhas so rarely appear in verse, be it in SN 1-11 (btw I missed a few, there are 1x in SN 4, 3x in SN 5, 1x in SN 8), or the Snp. It’s not that it doesn’t work - take the Dhp verses for example:

  1. There is no fire like passion, No offence there is like ill will,
    There is no misery like the khandhas, No ease there is higher than peace.
  1. Howsoever one thoroughly knows The rise and demise of the khandhas,
    One attains joy and delight That is ambrosia for those who are discerning.

Or SN 5.9

In the same way the aggregates and elements And these six sense fields
come to be because of a cause, And cease when the cause breaks up.


For me this is all new. This research about developments in texts. I am learning.

It think is indeed not very surprisingly, the Buddha would also show a kind of development in how to bring his message to the world. That must almost be surely so.
But, i have always learned that the first discourse was SN56.11. Is that not true? or only true per Theravada tradition? That sutta already mentions the five khandha’s.