Suttas & Agamas, aren't strictly historical?

I was reading many Agama sutras. While the big picture is the same, there are various unsettling (to me) differences on induvidial sutta level.

Some stories are different/contradictory. Sometimes quite significant. For example in Pali Sutta (MN140) Pukkusati after seeing the Buddha was killed by a “cow” and has died as an Anagami. In Agama MA162 sutta, Pukkusati became a stream-enterer and it doesn’t say what happened to him later. When it comes to talking points, there are difference as well. When talking about what can be cognized with consciousness:

Pali sutta says:
19. "Then there remains only consciousness, purified and bright. What does one cognize with that consciousness? One cognizes: ‘[This is] pleasant’; one cognizes: ‘[This is] painful’; one cognizes: '[This is] neither-painful-nor-pleasant.'” - MN140 BB trans.

The Agama sutta say:

There then remains just consciousness. Consciousness of what? Consciousness of pleasure, consciousness of pain, consciousness of mental joy, consciousness of mental sadness, consciousness of equanimity.” - MA162

Often in the place of 4 elements that are found in various Pali suttas, the Agama suttas talk about 6 elements.

The teachings in Agamas aren’t technically wrong, those 5 categories of feelings or 6 elements (in Agama Satipatthana), and other different pericopes like that, are often found elsewhere in Pali suttas.

Even within ONE Canon there are contradictions. Ex: Sariputtas awakening, MN111 vs MN74. Buddha’s exact sequence of Awakening (MN36 vs AN9.41)

It does, however, raise questions about using suttas as historical and verbatum records of historical events. I don’t doubt the spirit of the suttas, but I guess the historical details were less important for Ancient Indians than for modern westerners.

I can understand if the differences were merely due to forgetting some words or suttas, translation ambiguities, etc…

While I have my preferences, should one side with Pali suttas, Agamas, both, neither?

Question: What should one do? How should one handle this?

I believe that pericopes are the most accurately preserved Teachings, if not verbatum, then in spirit. If one compares the technical pericopes from suttas and agamas, IMHO, from what I have seen they look to be practically the same. Good.

It almost seems to me that regarding Sutta Pitaka what was composed during “the 1st Buddhist Council” was the list of pericopes arranged in a “schedule”, something like DN33, DN34, that almost looks like Abhidhamma and basic stories that could have been recollected more or less accurately into which these standardized pericopes were inserted.

Question: After the “1st Buddhist Council”, how and why did many suttas we recompiled with different pericopes for the different schools?


Unfortunately, it’s difficult to tell in the big picture because very little in the way of historical accounts have come down to us.

On a case by case basis, sometimes it appears that the original sutra was brief and then was expanded in different ways as time went on. For instance, sometimes a list is the same in different versions, but the explanations of individual items differ - meaning that the explanations were added by each tradition more or less independently of one another. Or sometimes stories grew in detail. But there must have been an older version that the later canons had inherited. So, the sutra itself isn’t invalidated, but the details represent different opinions about it.

In some cases, which are fairly uncommon, there were genuine sectarian differences that led to the adoption of doctrines intended to contradict those of other Buddhist schools. There were a number of controversies that led to this, such as the position that the past and future exists or not. We do have some historical accounts of what these controversies were, so they can be identified easily.

Sometimes, there was an historical development of ideas that weren’t controversial and that were generally adopted by most Buddhist traditions, though usually in slightly different ways since the idea developed in different canons in parallel.

Some of the disagreements are really nothing more than the random variations that are seen in most ancient literary traditions around the world. It was difficult to maintain an exact wording over centuries of time. It was mainly the invention of the printing press and copyright laws that made it possible to keep the original version intact the way we expect it to be in modern times. There were exceptional scriptures that were very widely read and disseminated which remained consistent over time. But scholars who study ancient texts usually have to contend with divergent versions of any given text they work with. Sometimes the differences are slight, and sometimes they’re quite large.

The basic reality, though, is that historically the Theravadins were one school of Buddhism out of many and their canon is one version out of many as well. Buddhists didn’t have a central authority that could enforce one version of their scriptures on all Buddhists the way it was in Catholic Europe or Confucian China. The Buddha apparently disagreed with this way of doing things. It can be disconcerting to confront, but Buddhism was a pluralistic religion, probably since shortly after the Buddha passed away.


This is the reason I don’t read Agamas. Access to Agamas may seem as a bliss for scholars but it is rather doubtful bliss for one who wants a firm trusted standard. I believe that the main problem is that Lord Buddha’s words are difficult to understand, and Pali Canon provides enough Suttas to investigate Dhamma. No need to introduce more material.

1 Like

the explanations were added by each tradition more or less independently of one another. Or sometimes stories grew in detail.

This is potentially upsetting. Why would bhikkhus add explanations (assuming different from proto-suttas) and add detail to stories.
Wouldn’t that alter the Teacher’s message? Wouldn’t this break the precept not to speak falsehood?

Or maybe what we consider to be “falsehood/lying” wouldn’t be considered as such by early bhikkhus and people living in that region?

but Buddhism was a pluralistic religion,

If the Dhamma was supposed to be “open ended” then this would be ok.

“As for the qualities of which you may know, ‘These qualities lead to utter disenchantment, to dispassion, to cessation, to stilling, to direct knowledge, to self-awakening, to unbinding’: You may categorically hold, ‘This is the Dhamma, this is the Vinaya, this is the Teacher’s instruction.’” AN 7:80  Satthusāsana Sutta | The Teacher’s Instruction Ven. TB translation

Thank you for your reply!

a firm trusted standard

I believe in Pali suttas too. However being true to myself, it is a belief. It would have been awesome if the canon of other 17 (or so) early schools survive. Especially the earliest ones. Today we have pali suttas and agamas (large amount) and few Gandhari suttas/treatise.

Thing about pali suttas is that there are some suttas missing. For example, where are the suttas explaining kasina practice in more detail? There are some other meditations mentioned in the pali suttas that do not have explanation.

Interestingly, one Gandhari sutta “the Four concentrations” describes “concentration accompanied by perception of revulsion toward food” in more detail and in different way than what I remember from pali suttas. And it describes “concentration accompanied by perception of displeasure toward the whole world” something mentioned in pali suttas such as Girimananda Sutta AN10.60 but never explained more fully.

Very good point indeed regarding “falsehood/lying”.

I think the gradual expansion from the proto-sutras/suttas (i.e. the early ‘saṃyukta-kathā’ collection) to the development of the four principal Nikayas/Agamas could be both true transmission and lying/falsehood.

Buddhaghosa in the Pali tradition simply considers that the Pali Piṭakas had originated from the first Saṅgha council, and that the Pali language of the texts was identical with Magadhi, the language spoken by the Buddha.

When you say pericope, do you mean stock passage? Because they aren’t the same thing. A pericope is any passage that contains meaning. There are thousands of pericopes in the EBTs that are not stock passages. And there are many stock passages that wouldn’t really stand on their own as a passage with meaning.

1 Like

Buddhists grasping for a fundamentally and theoretically perfect canon regarding events that took place thousands of years ago is inappropriate and unrealistic. No amount of textual analysis or conjecture is going to bring certainty as to events that happened thousands of years ago and whose record has passed down through oral history and redaction from faulty and imperfect sentient beings. How can it be otherwise?

The pali canon and agamas were made by humans and are impermanent. They arose based on causes and conditions and will cease and are subject to change in between.

We are fortunate to live in an age when the dhamma is available and has not totally degenerated and where progress on the path of liberation is still possible. We should rejoice in this fact and take full advantage while we can!

What matters is actual practice and the effects of that practice on the lives of sentient beings. The dhamma jewel can only be realized through the practice it is put to. Rejoice in the tremendous merit and dhamma practice that millions of sentient beings turn to every day for refuge even in this degenerate age. :pray:


It’s upsetting if one is a literalist who believes the exact wording must be maintained for the teaching to be true. Ancient Buddhists did not believe this. Later Buddhists did develop these attitudes when they closed their canons. We can see the closed canon attitudes still in modern Theravada Buddhism, for example, but that was not the attitude of ancient Buddhists. Ancient Buddhists understood early on that wording is less important than meaning, and we see this expressed in their texts. They likely recognized that the exact wordings had sometimes been lost and strove to maintain the meaning as best they could.

One of the challenges they faced was that the ancient world was chaotic. Wars, famines, displacements of populations, religious persecutions, and so on periodically disrupted religious life. We live in world in which these things have become rare, and we really have no idea how good we have it. Then we look back on ancient texts and have no idea what they went through from one decade to the next. Periods of stability lasted usually for a century or two and then collapse into warfare and anarchy. Empires rose and fell with the coming and going of invading peoples. India went through this as much as Europe and China did.

Another problem was language. Buddhist scriptures were maintained in vernacular languages for much of the early period of Buddhist history. This means that they were translated to local languages and dialects routinely. This is likely one of the things that lead to changes in the texts over time. Words that sound alike were confused for each other quite often. This didn’t stop until Buddhists finally decided to adopt Sanskrit and Pali to preserve their canons, but by that time a lot of confusion had already happened.

Finally, as I mentioned above, there were schisms and doctrinal controversies that divided Buddhism into a bunch of competing regional sects. They all wrote their favored positions on doctrinal disputes into the early scriptures. They of course considered their positions to be correct and represent what the Buddha would have taught. This is one of the major reasons we study EBTs and compare them against each other - to detect these changes.

As far as storytelling goes, stories changed more than doctrinal texts did in ancient times because they were often composed to illustrate morals and ideas. Multiple versions of famous stories circulated routinely. This is a function of human creativity. The stories that we find in the early sutras were probably composed over a period of time, not all at once. Some are older than others. The oldest no doubt do have their roots in actual events. Some are basically the earliest form of fictional literature, which took the form of religious mythology. This is not special to Buddhism. Every ancient mythological tradition was a creative incubator for early literary traditions.

Really, if we recognize that the history of Buddhism spans the history of human civilization from the Iron Age on, it’s remarkable that their early oral tradition has managed to survive. Most of the old oral traditions are only fragmentary today or lost completely.


Keep in mind that you are looking at translations, and at least in the case of the āgama texts, at translations of translations. Translations from Chinese are strongly affected by interpretation. As Jan Nattier says,

“…when reading any given line of a Chinese Buddhist sūtra–excepting perhaps those produced by someone like Xuanzang, who is justifiably famous for his accuracy–we have roughly an equal chance of encountering an accurate reflection of the underlying Indian original or a catastrophic misunderstanding.” – A Few Good Men, p.71.

If you see differences in the English versions, it could just be how different translators interpreted the text. Chinese is very flexible and open to interpretation, but even Pāli has to be interpreted before it can be translated. If you only have access to translations, and if you are only comparing translations, you have to be very cautious about coming to any conclusions at all because every translation is also an interpretation.

Comparative studies really require a working knowledge of the source texts. It doesn’t work well only using translations.

Buddhism is, was, and always will be pluralistic and syncretistic. From the outset the Pāli suttas incorporate ideas, attitudes, and practices from (at the very least) Brahmanism, Jainism, and autochthonous animistic traditions.

If there ever was a unified core of ideas, attitudes, and practices in our literature it is apparent that no Buddhists in history ever found that satisfying; instead they kept adding new ideas, attitudes, and practices. Even the hardcore fundamentalists think the Pāli texts are chronologically stratified and that new ideas replaced older ideas in the process.

As to why this happened, we don’t know. Buddhists are generally not interested in this question at all. Scholars seem content to describe the situation without ever attempting to explain it.


There are different views as to how the texts came into being, how they developed, what their purpose was, what was important to them…

To simplify the scholarship on this: you have one group of scholars who presuppose a sort of ‘original’ text from which changes (intended or unintended) occur, so that by comparing versions you can identify early and late elements, find changes or ‘mistakes’, and sort of get at the original.

Another group of scholars treats the suttas more as literary creations, where the tradition would modify and retell stories within certain parametres, and so different texts are versions of each other rather than point to an original one.

The theories overlap and most scholars accept elements of both with varying emphasis. Those two groups argue about the degree to which texts are ‘historical’ as in mainly aiming to relate historical fact, for example, but hardly anyone doubts there is a connection to historical fact.

There has been a recent back-and-forth between Bhikkhu Anālayo, representative of the ‘historicist’ approach, and Eviatar Shulman, representative of the ‘literary’ approach. Shulman’s last contribution is worth reading because it’s quite clear and because people generally know the ‘historicist’ approach better:

Hope this is interesting.


When you say pericope, do you mean stock passage?

Yes, stock phrase. Common phrases/paragraphs that are repeated exactly the same in many different suttas and nikayas.

I guess that what we are left with is that " whatever is well said, was said by the Buddha".
yaṃ kiñci subhāsitaṃ sabbaṃ taṃ tassa Bhagavato vacanaṃ arahato sammāsambuddhassa

The difficulty starts to arise when one tries to determine exactly what subtle points of the Dhamma is really well said and what only sounds profound but lacks essence.

Yes. The way I personally view it is this: Texts are impermanent things that are liable to change or be destroyed like everything else that arises and ceases. They consist of the five aggregates, etc. So, they are themselves included in the principle of dependent origination and emptiness. Those principles exist because we can observe them to be true, not because someone recited them and wrote them down to be memorized. There are profound lessons that Buddhist scriptures can teach which are not in the words they contain. One is that it’s better not to get emotionally attached to a particular set of words or to think there is some permanent essence underlying them. Even the meaning and form of individual words changes over time: They lack a permanent essence. Existence is a state of changing flux. Truth is something we realize in our lives by observing it clearly. Reading texts are just one of many ways to encounter it, and they are just as problematic as everything else if we haven’t done the preliminary work of clearing our thinking and developing wisdom.


I believe we have Richard Gombrich to thank for the use of pericope in this way with respect to stock phrase in Pāli. At least, I first came across the word used in this sense in a lecture he gave back in 2006 and have seldom, if ever, encountered elsewhere.

1 Like

Although to be fair, this “age” has now lasted something like 2000 years and shows no signs of ending any time soon. The “end of the Dhamma” is a bit like the second coming of Jesus… often talked about and even predicted, but never actually observed.

1 Like

So everyone is just guessing? I’ve never come across anyone with actual proof of anything. Am I missing something?

Huh. That’s too bad. Thanks for the background.

1 Like

Sadly it looks like Dhamma Age might be ending. Probably within the next 50-100 years when in traditional Buddhist countries current generation and their parents die out.

More and more people in SE Asia might adopt western culture along western materialistic & secular values. Thus displacing Dhamma values … Climate change (possibly displacing population toward northern countries) and famine. The laity might stop feeding the bhikhus (" why feed those who do not work?") and the whole Sassana will slowly fade away.

1 Like


Shulman’s first words in that article are…

“The play of formulas” offers a new approach to Buddhist scripture… (emphasis added)

Which means that Shulman doesn’t see himself as participating in an existing discussion or as part of an existing “camp” within Buddhist Studies. He says he is trying to take us in a new direction. This is consistent with my overall impressions of his work. He is not representative of any trend in Buddhist Studies, but is very much ploughing his own furrow. Sometimes this is insightful, as when he argues that dependent arising was intended to describe the arising of experience and not to be a theory of everything; but other times it’s less clear to me how valid or applicable his arguments are.

In fact, the arguments over historicity in our field are largely free floating and tend not linked to any common theory or methodology.

The people you describe “historicists” are almost all Theravādin bhikkhus or their academic allies (and Sujato who is “not Theravādin” these days). Their views are too obviously apologetic in flavour and motivated by their religious convictions (c.f. my review of Stefan Karpik and the Oxford clique).

Academic allies of other Buddhisms are less interested in the historicity of texts or the Buddha because these are not assumptions made by other Buddhisms. If you believe in the Dharmakāya Buddha, for example, the historicity of texts is just not that interesting.

Worse, the other “camp” are not really linked by any common theory or methodology. Buddhist Studies scholars almost never discuss questions of methodology or theory; the whole field is deeply averse to discussing theory. Gombrich’s rants against “methodology” are legendary. Also most Buddhist Studies scholars prefer to stick to description and actively avoid trying to explain anything, so there is little work for theory or methodology to do. There are few genuine intellectual “camps” and great deal of individualism.

For example, another new direction is reflected in Jonathan Silk’s Open Philology project. Which is more like your description of a “literary” approach. In this view, authenticity is always with respect to the views of a particular community at a particular time and ur-texts, if they exist at all, are not that interesting or valuable.

I’m involved in another approach, which was proposed by (the late) Sue Hamilton. This is the reading of Buddhist texts as primarily concerned with epistemic issues, especially as they relate to meditation and the cessation of sensory experience as the acme of Buddhist meditation. This is neither “historicist” nor “literary” as far as I can see: one could start from either position and still benefit from our epistemic reading. Shulman made a valuable early contribution to this approach, but has since moved on to pursue his own thing.

I think also of Jonathan S Walter’s approach to “Suttas as History” which seems to me to sit somewhere between your “historicist” and “literary” poles.

Joseph Walser has another quite different and distinctive approach. One of his observations, for example, is that amongst living Buddhists in traditional Buddhist countries, ātman views are ubiquitous. The idea that Buddhists “don’t believe in ātman” based on normative texts is contradicted (and might argue refuted) by what actual living Buddhists in Buddhist cultures say they believe. This calls into question the idea that we can study ancient normative texts to find out what Buddhists believed back then. An analogy would be trying to understand how people in the 1970s used computers by reading contemporary computer manuals.

Walser’s approach echoes Schopen’s observations about discrepancies between archaeology and normative texts. For example, archaeological digs of Indian Buddhist monasteries invariably find money, which is notionally forbidden to Buddhist monks by the Vinaya (at least in Theravāda, this restriction is not always observed in other lineages or ordination). At least one Indian Buddhist monastery had the equipment to mint their own coins. Donative inscriptions across India often mention monks and nuns paying for monuments and memorials. So the Vinaya is the ideal and not how things worked in practice.

David Drewes takes what I think of as a strictly historicist position. But he concludes that the texts are not evidence for historicity of the people in the texts. None of the people mentioned in Pāli, including the Buddha, have ever been connected to any historical facts or events. There is no “archaeology of early Buddhism” that corresponds to the Early Buddhist texts, though there is archaeology of those places. A genuinely historicist reading tells us that, at best, the texts as we have them reflect the beliefs of people who lived some centuries later than the events depicted in the texts. And the events could be fictions set in real places. On this basis, I would question labelling the Theravādin faction as “historicist”. They are more like historical apologists (using this word in a descriptive rather than pejorative sense).

I don’t think this idea that there are two camps is accurate or that it leads to interesting reflections. It seems like an example of the common cognitive bias towards conceptualising differences of opinion as binary oppositions.

In the field of Buddhist Studies there is no common body of theory or methodology. Rather there are a plurality of approaches, often with a strongly individualist flavour, that eschew explicit theories and methodologies and prefer to rely on the individual “genius” of the scholars, who genuinely seem to make it up as they go. The discussions of methodology that we do see, are often in service to a particular goal. There’s no enunciation of hermeneutic principles that is not closely tied to a particular outcome, that I’m aware of.

If you want to make a binary opposition, then it might be more apropos to talk about it in terms of what each group takes as given.

There is one, quite small, camp which accepts the historicity of the Buddha and the authenticity of (their subset of) Buddhist texts as articles of faith. And who (surprise) find that the methods they adopt confirm their beliefs.

There is another much larger camp, that is anything but uniform or unified, which does not accept these axioms and whose individualised methods lead to a variety of individualised opinions on historicity and authenticity, including positions that treat historicity as irrelevant or unknowable.


Have you read Richard Carrier’s “On the Historicity of Jesus” and do you think a Bayesian approach like he takes would be fruitful in Buddhist studies to move forward on the question of the historicity of the Buddha?

1 Like