The good reasons for the compilation of Abhidhamma


I am usually upset to see people throwing the baby with the bathwater when it comes to suttas like the MN117.

As per the commentaries, in the early Sangha this was a crucial part of any bhikkhu curriculum and the way it presents a gradual causation / foundation for the path to mature and flourish is very useful for anyone who wants to make sense of his/her practice of the Dhamma.

It seems that some advocate discarding it just because it makes use/contains of Abhidhammic pedagogy / concepts.

The intention of this topic is to gather any material and discuss on what have probably been a genuine and good reasons for Abhidhamma to have been compiled and preserved of at least ~2,000 years.

I would like to not explore the topic of why or whether our current generation has lost our ability to make sense of it. Or make this a place for people to share the reasons they seem to dislike/despise it so much.

Hence the topic question is put: what might have been the good reasons for the creation and preservation of the Abhidhamma?


MN 117 can give the impression of labeling certain ‘right views’ (about kamma & rebirth) as ‘defiled with pollutants (asava)’, ‘siding with attachment’ (‘upadi’) &, essentially, ‘ignoble’. I would guess or speculate that such negative descriptors may be regarded by some as idiosyncratic & thus give rise to doubt about the authenticity of MN 117.

1 Like

This is a really good point. It’s incredibly important that we don’t see the subsequent generations of Buddhists as mere losers who got everything wrong. That raises just too many problems; but more important, it’s not a compassionate or wise way of looking at things.

As one of the few people in the world who works pretty much full time with Pali and related texts, I am acutely conscious that much of what I do is “abhidhamma” in the original meaning of the word. That is, it’s doing stuff that’s “about the teachings”.

As just one example, yesterday we posted a major upgrade to our PTS dictionary, fixing hundreds of thousands of issues. Most of these are petty details, mere conventions and markup. Now, acting as a dictionary is one of the function of the old Abhidhamma (and in a different way, the commentaries). In fact, in their lists of terms and synonyms we could argue that they are the word’s first Thesaurus. The benefit of doing this is obvious; collect various terms, collate similar terms, and clarify the meaning of different terms in various contexts.

Now, these days that’s not very useful. Modern dictionaries are much more useful, and I can easily use search across the whole corpus to find what I want. But I can only do these things because of the “abhidhamma” work done by myself, other SC developers, and the countless scholars and volunteers on whose work we rely.

This is far from denying that there are major issues with the abhidhamma project in general, and the Theravadin Abhidhamma texts in particular. But to understand these we have to start by empathizing with and understanding their goals and methods.

Long ago I wrote The Mystique of the Abhidhamma, which I believe is the most humorous essay on Abhidhamma ever written. Of course, there’s not much competition; none, to be precise. The point of the article, as explained in the final paragraph, is that we can never understand Abhidhamma properly if we insist on worshiping it and making it into something that it quite obviously is not.

I find it frustrating that even today there are so many people following Burmese methods based on the assumption that everything in the Abhidhamma is true. It’s not, even in the canonical Abhidhamma; and almost everything taught as “abhidhamma” in fact stems from much later commentaries. We have thousands of meditators believing with complete conviction that they are seeing kalāpas and mind moments and the rest. But these things just don’t exist. And they were certainly never taught by the Buddha. How is this a path to non-delusion? If we’re unable to rise above denial regarding even the simple facts of history, how can we see subtle truths of the mind?

The problem is not so much the abhidhamma texts as such, it’s the role they’ve been pressed into by the community. If we see them as attempts to write a curriculum, to make dictionaries and concordances, to classify and clarify concepts, then they can be seen as valuable or not in so far as they achieve these goals. To force them to serve as a model of “absolute truth” is to disrespect the texts and their authors. This is not listening to the actual Abhidhamma, it’s worshiping a false idol.


That is very interesting remark Deele, for the way I interpret MN117’s differentiation is not as it putting these views down.

I read the MN117 as telling me that yes, up until the threshold of right insight and liberation has been crossed, the whole cultivation of the path needs to be based on a fabricated set of working assumptions which are “affected by taints, partaking of merit, ripening in the acquisitions”. Once that threshold has been crossed, the path works / operates on a purified mode, hence “noble, taintless, supramundane”.

And these working assumptions of the still unenlightened ones are summarized as the acknowledgment of kamma, rebirth and the importance of generosity as well the constant cultivation of thoughts of loving kindness, renunciation/detachment and compassion/non-violence.

This is I think that is why the MN117, and the “abhidhammic” pedagogy it embeds, was so important in the early Sangha’s curriculum. It helps people solving the conundrum of how right view is the forerunner of the whole natural unfolding of the path and its goal.

1 Like

Very interesting point Bhante. Yes, Abhidhamma probably started as an index, or a prehistoric Google of the teachings! :grinning:


Exactly. I mean, the Google search index must make for some pretty boring reading, right? But at the same time, it’s incredibly useful.


Interesting Topic!

This is a very important point. I find it frustrating that many people, even those who think of themselves as Abhidhamma followers, can’t, or won’t, distinguish between the Canonical Abhidhamma and texts from hundreds or thousands of years later, such as the Abhidhammatthasangaha
I have not made a detailed study, but most of the Canonical Abhidhamma appears to me to be collections of lists and synonyms compiled from the Suttas, as Bhante @sujato notes. Concepts such as kalapas, billions of mind moments per second, and so on seem to be much later developments.

I agree. On the other hand, it would be hasty to conclude that the interpretations of the ancient commentators were necessarily any worse than the interpretations of modern commentators. My impression is that much of the ancient commentary is based not only on analysis, but also the experience of ancient practitioners, just as modern teachers make use of their experience, and the experience of their students, to inform what they teach.

In fact in some commentaries, and particularly in texts as the Vissudhimagga (which does contain many of the post-Abhidhamma concepts) there is an almost schizophrenic feel, with some parts containing very practical advice and others very academic-sounding analysis.

My inclination is to think of such works as a collection of Dhamma talks from ancient scholars and practitioners. Their usefulness to me varies, just as the usefulness of talks by modern teachers and scholars varies…


This may be the case but what grounds are there for the ‘Abhidhamma’ allegation or taint that MN 117 is accused of? At least most of the constituents of the various path factors in MN 117 I personally am aware of are not exactly alien to the suttas, such as paññindriyaṃ, paññābalaṃ, dhammavicaya sambojjhaṅgo, etc. I do recall browsing Bhikkhu Anālayo’s paper once and was not convinced of the arguments presented. There does appear to be an actual division in the suttas (atlhough I have not read them all) where the teachings about “There is nothing given, etc” are generally spoken to householders & Brahmans rather than in the context where a bhikkhu attains enlightenment.

I think this is a very important point. As unenlightened beings, we inevitably conceptualize the Dhamma in terms conditioned by our cultural, social, linguistic, and religious environment. The dominant religion at least in Western countries seems to be democratic secularism (my own view that is in no way binding on anyone), the dominant social stratum is service-sector middle class, the major languages are Romance or Germanic and the dominant culture is post-Christian European. Compare it to the Ancient Indic society speaking in terms of castes, agriculture, and Brahmanism in mostly Indo-Arian languages. The context will always intefere with the phenomenal contents of the teachings, papanca always sets in whether one wants it or not.

The overwhelming concern of the Indic culture with lists, numbers, mattikas, etc., seen even in the cosmological treatises of such religious traditions as Jainism, could not but lead to the creation of a Googlesque system like the traditional Abhidhamma. The more pragmatically minded Chinese culture, while preserving the Abhidhamma treatises as an object of worshi, couldn’t make much practical sense out of them. I think we Westerners will also not be able to make much use of them unless they are ‘translated’, ‘transformed’ into the context that is more understandable to us, and this is pretty much what the Pali scholars and Buddhist community like SC have been attempting to do for the last 150 years. Even then, this new Abhidhamma, just as any puthujjana interpretation of the Dhamma, will not be quite correct. Paradoxically, the more work will be done the clearer and more precise our understanding of the Dhamma will be, while at the same time the tiny mistakes or misconceptions we introduced at an earlier stage will be further exacerbated and will have a more drastic effect on the practice. The history teaches us this is the destiny of any religious tradition, just look at Christianity, Islam, Hinduism, Buddhism in the Far East, etc.

One can compare abhidhammic projects with attempts to describe a beautiful flower or chocolate flavour to a person who whas never seen this flower or tasted the flavour. As time passes, the descriptions will become rigid, their venerable history will become much more important than their - albeit inevitably deficient - adequacy. Still, an acute mind will always be able to appreciate the beauty of what they describe and will make an effort to see the Truth as it is, without any intermediaries.


I would be cautious about making assumptions regarding this. What we know of as abhidhamma is very much a product of the Theravada school. Of course there were many abhidhammas on the mainland, and they shared much in common. However the preoccupation with lists, matikas, and spiralling permutations of factors—which is virtually the entirety of the central Pali Abhidhamma texts—is not found in the Sanskritic Abhidhammas at all, so far as i know.

They are, it seems to me, much more sane, and much more like you’d expect an abhidhamma to be. They take up important dhamma topics, define terms, bring in sutta quotes, and discuss interpretation; more like, in fact, the later Theravada commentarial literature.

Check out, for example, the Dharmaskandha in Sanskrit (which we should really have on SC! @vimala, if you have a moment …) as compared to its Pali equivalent, the Vibhanga. The Dharmaskandha is much more discursive, it has a much wider range of sutta quotes, it entirely lacks the “cycles” of repetition, and is, overall, a far more useful text. In form, it is more comparable to the Pali Kathavatthu.

I suspect it is the very alienness of the Theravada abhidhamma texts, their magisterial inscrutability and disdain of readability, that led to their fetishization as holy objects of absolute truth.

For your interest, here’s a random passage from the Dharmaskandha, with my very rough attempt at a translation. It is part of a passage discussing the dependence of consciousness with name and form.

api khalv evam uktaṃ bhagavatā svātiṃ bhikṣuṃ kaivartapūrviṇam ārabhya /
And this was said by the Buddha concerning Sati the former fisherman.

trayāṇāṃ bhikṣavaḥ sannipātān (DhskD 6v5) mātuḥ kukṣau garbhasyāvakrāntir bhavati / vistareṇa yāvad /
With the coming together of the three, the embryo is conceived in the mother’s womb. (Expand in detail up to:)

ity evam ayoniśomanasikāreṇa sahajaṃ nāmarūpaṃ pratītya mātuḥ kukṣau vijñānasyāvakrāntir bhavati prādurbhāvas / tad ucyate nāmarūpapratyayaṃ (DhskD 6v6) vinjñānaṃ /
So because of irrational focus, consciousness is conceived in the mother’s womb, with name and form as simultaneous condition. Note: this line reads as if it were a continuation of the previous, but the passage is not found in the sutta as we have it, perhaps it is unique to the Sanskrit.

api khalv evam uktaṃ bhagavatā mahānidānaparyāye /
And this was said by the Buddha in the exposition on the Great Cause:

vistareṇa yāvad / vijñānaṃ ced ānanda nāmarūpapratiṣṭhāṃ na labheta tathāpratiṣṭhite vijñāne anatirūḍhe + + + + + (DhskD 6v7) bhavasamudayasamutthāpakaṃ jātijarāmaraṇam abhinirvarteta / no bhadanta / sarvaśo vā punar ānanda nāmarūpe asati api nu vijñānaṃ prajñāyeta / no bhadanta / tasmād ānanda (DhskD 6v8) etan nidānaṃ vistareṇa yathā pūrvoktaṃ /
(Expand in detail up to:) “Ananda, if name and form were not to be planted in consciousness, and once planted, to grow … The proceeding of old age and death, originated by the process of existence?” Very unsure about this! “No, sir.” “And furthermore, if all name and form did not exist, would consciousness be found?” “No, sir.” “Therefore, Ananda, this is the source …” (Expand in detail as said earlier).

Note not only the frequency of sutta quotes—which are entirely absent from the Vibhanga parallel at Vb 6—but how they are regularly abbreviated and referenced (also absent from the Pali Abhidhamma). The author assumes that the student knows the suttas, or has access to them. The text reads as a guide to the topic, stringing together interesting and relevant sutta passages with an explanatory framework. It would serve very nicely as a lesson framework for a teacher. It is, in fact, not all that different from the discussions we have here.

It’s abhidhamma, but not as we know it.


Sure, tomorrow. How about the other one that is mentioned in the menu: Jnānaprasthāna? Is that to be found somewhere? I had a brief look though Gretil but couldn’ t see it.


Just as a note on the text. It is divided into 19 sections, each of which should be a file, I guess. These parallel the pali Vibhanga but not one to one. To be precise:

  • Ds 1-14 = Vb 6
  • Ds 15-17 = Vb 14
  • Ds 18-19 = Vb 13

These are all partial. As in, sectional (obviously) but also resembling, in that the text itself, while dealing with the same set of subjects, is vastly different.

In addition to this, there are multiple parallels between the these two texts, and two in Chinese: the Dharmaskandha (which is a complete translation, not just the few fragments in Sanskrit) and the Sariputrabhidharma. In addition, there are many Sutta quotes in these texts, which also have their parallels.

Anyway, I don’t expect you to do all this, just letting you know the situation. Perhaps we’ll find the parallels for these in an essay or book at some point.

Yeah, lol. The Dharmaskandha, while a complex reconstructed text, is pretty short. The Jnānaprasthāna is like this huge compendium. It basically takes all of the Sarvastivada abhidhamma up to that point and arranges it into what may well have been the first example in Buddhism—or anywhere, for that matter—of a, well, an encyclopedia or something. Like seriously, it’s huge. There are two versions in Chinese, and a partial Sanskrit text, which, however, seems to be on the 2-do list for digitization.

If/when the Sanskrit text becomes available, we could look at it.

1 Like

I’ll go out on a limb here and say that while abhidharma is kind of boring (which I suspect is the real reason why most people dislike it), it is also kind of awesome. The EBT’s are not systematic, and likely contain some contradictions, or otherwise do not form one clearly-defined system.

Abhidharma was an attempt to put the Dharma under a microscope, using all sorts of methods including canonical texts, formal logic, and (perhaps most significantly) meditation. The Mahāvibhāṣā, for example, preserves much of the history of Sarvāstivāda Buddhist ideas, including all sorts of very valuable information about meditation and their dhyāna traditions.

By contrast, much of modern scholarship excludes meditation and personal experience, even when including it might be useful and appropriate. As a result, sometimes their theories may make sense from the perspective of an academic, but complete nonsense to someone who is actually applying these theories to practice and expecting real results.


Again, it depends what we’re talking about. If we use “abhidhamma” in the sense of “later teachings about the Dhamma”, then sure. the Mahavibhasa is probably more comparable to the Visuddhimagga than to the canonical Pali abhidhamma.

I would distinguish a strictly ideological sense of the word Abhidhamma. This is used in Theravada circles in a way that not dissimilar to “Mahayana” in Mahayana circles. That is, on the one hand, it’s just the word for the thing. At the same time, though, it stakes out a definitive ideological claim: it’s better!

The Abhidhamma ideology claims that the abhidhamma is the true, accurate version of the Buddha’s teachings, expressed throughout in terms of absolute truth. It dispenses entirely with “conventional” speech in terms of “persons” and so on. As such, it effectively supersedes the Suttas (again, much like the Mahayana suttas). This is no empty ideology; in much of the Theravada world, the suttas are in fact almost completely ignored in favor of the Abhidhamma. In many monasteries, in point of fact, there is only Abhidhamma and no suttas.

A moment’s reflection shows that this claim cannot be true. Heck, there’s even an entire book in the Pali Abhidhamma called “the concept of a person”. Like any absolutist ideological claim, it serves not to illuminate, but to stake out territory. The real goal is to set the Abhidhamma scholars up on higher ground than students of the suttas, a purpose which, to all intents and purposes, has been accomplished in modern Theravada.

This is why I said that the problem is not the Abhidhamma as such, but the modern ideology. It’s simply impossible to discuss the actual historical and doctrinal contributions of abhidhamma within modern Theravada. To even try is to become a heretic. This is no exaggeration: in Burma, the Sangha bans books that contradict this orthodoxy.


I’ve added the Dharmaskandha but with a few issues (see below) and the Chinese I don’t have.
The parallels I will do tomorrow.
Because I feel this discussion does not really fit here, I’ve moved it to Sanskrit texts markup


Parallels added to staging also.


This is very interesting to me. As I think has been mentioned already, one interpretation of the prefix “abhi” is that it suggests the notion of discussion/s (historically, perhaps, among sangha during the early Buddhist period) “about” dhamma and, seen in that light, softens some of the tendency to accord the Abhidhamma much of its sanctified status that it seems to have accrued.


This is very true. As it happens, I have just translated one of the passages in the Nikayas where the word abhidhamma is clearly used in this sense. From MN 32:

dve bhikkhū abhidhammakathaṃ kathenti, te aññamaññaṃ pañhaṃ pucchanti, aññamaññassa pañhaṃ puṭṭhā vissajjenti, no ca saṃsādenti, dhammī ca nesaṃ kathā pavattinī hoti.
two mendicants engage in discussion about the teaching. They question each other and answer each other’s questions without faltering, and their discussion on the teaching flows on.

This is where the whole abhidhamma tradition begins, by discussing the Dhamma. Ironically enough, what we do on Discourse is a direct continuation of this, so we are all Abhidhammists!

Note how this passage slips easily from abhidhamma in the first part to simply dhamma in the second. These were not different things.


Thank you Bhante. I’ll look at MN32. Bhikkhu Bodhi in his 2013 online lectures also suggests MN28 and 111 as well as the Vibhanga suttas, some of which I’ve quickly looked at here in the Vibhanga. I’m sure the list of suttas that mirror the abhidhamma is extensive and needs to be explored and given due contemplation.

1 Like

Back a couple of years, attending a series of teaching-discussion groups led by Noah Ronkin (author of “Early Buddhist Metaphysics…”, and student of Rupert Gethin), it occurred to me:

The Buddha said his knowledge – from awakening and especially of the Buddha-sort – is like the umpteen (huge) number of leaves throughout the forest, but that he chose to teach only “a handful of leaves”, extracts of that knowledge that pertain to and are sufficient to lead others along the Path.

The Abhidhammika-s could be thought of as analytically reverse-engineering that handful-of-leaves, mapping-out, scoping-out what the whole forest of leaves might be like; or at least the principles for doing so.

Noah commented that that’s not an unreasonable view.

1 Like