What came last the Agama or the Pali?

For starters, excuse my ignorance this question probably has a simple answer for someone who knows text critical history well. How come something in the Pali Suttas is considered late if it doesn’t appear in the Chinese recension? How is it decided that it wasn’t the Agamas that left something out, intentionally or not?

Do you have some specific text or passage in mind? It’s a really complex subject that can’t really be discussed in the abstract. Chinese sources run the gamut of early-looking and later-looking texts just like Pali or Sanskrit sources do. If someone claims Chinese sources trump Pali sources or vice versa, they are likely oversimplifying things.


Something not appearing in Chinese or Pali is one form of evidence that can be considered when making an inference about what might be late.

In general, it’s better with multiple sources of evidence pointing in the same direction.

We can’t know this for sure. We can only make inferences, there’s not much we can do except weigh the evidence and arguments.

Moreover, something appearing in multiple sources (Pali, Chinese, Sanskrit, Tibetan) is stronger evidence for earliness than an absence is evidence for lateness.

Absences have multiple explanations, like lateness, but also maybe it was just a copying error, or something else.

On the other hand, the presence of identical or near-identical texts in multiple languages separated by time and geography basically only has a single plausible explanation; that it comes from the same source.


If I’m not mistaken the prediction at the end of MN10 is not present in the Chinese right? Is this a good case where a judgment can be made about its lateness?

It’s absent from one of MN 10’s Chinese parallels, which is EA 12.1. It occurs at the end of MA 98 in more or less the same way.

The thing to understand is that the Agamas in Chinese belong to several Buddhist traditions. Some were closely related to the Theravada tradition, like MA (Sarvastivada) and DA (Dharmaguptaka), and some were not, like EA (an unknown Buddhist tradition). So, it’s not a surprise that MA 98 is very close to MN 10, but EA 12.1 isn’t.


Yes, but the meaning of the passage in MN 10 doesn’t contradict the simpler passage in EA 12.1. It added becoming a non-returner as a second type of liberation. And someone expanded that statement to say, “However long it takes, it’ll happen,” by giving lots of possible lengths of time. All of that is implied by saying “and then you’ll attain nirvana.” It’s just more specific about how and when.

The long answer that gets into the weeds:

The basic meaning of the passage that concludes EA 12.1 isn’t invalidated by the expansion we see in MN 10 and MA 98. If we assume that simpler passages are older because Buddhists tended to expand their texts over the centuries,* then EA 12.1’s conclusion is closer to the original version of these sutras. It says:

“Monks, relying on this one entry to the path, sentient beings attain purification, part with grief, and no longer delight in conceptions. They then obtain wisdom and realize nirvāṇa. That is, they cease the five hindrances and cultivate the four stations of mindfulness.”

* Note: The above assumption isn’t always true. It’s a general tendency that we see in traditions that develop over a long period of time. They add new ideas and rewrite passages to make them more detailed or specific. Sometimes, though, people decide to simplify instead of complexify. Then, texts get smaller and simpler. E.g., the Heart Sutra or Diamond Sutra vs. the Large Prajna Paramita Sutra. This may have happened in early Buddhism, too, resulting in lots of little suttas.

The Theravada/Sarvastivada textual lineage represented in MN 10 and MA 98 may have originally mentioned only that a practitioner will attain nirvāṇa (or something equivalent) in the same way EA 12.1 does. But, at some point early in Buddhist history, the conclusion added becoming a non-returner as a second outcome (“one of two fruits”). This kind of statement about attaining one of two fruits wasn’t invented by them, however. It’s found in other EA sutras like EA 21.6, which has this statement at the end:

This monk then achieves (one of) two fruits: Destruction of the contaminants in this life or becoming a non-returner.

When I look at its Theravada parallel, AN 3.16, I notice that it doesn’t include becoming a non-returner at the end. It only mentions destroying the contaminants (P. āsava). In this case, it’s a redactor in EA’s tradition who has added non-returning as a second form of liberation.

So, I think what we are seeing in the bigger picture is the introduction of the idea of becoming a non-returner as the second best outcome of practice. It became the noble birth in heaven for Buddhists because it led directly to liberation afterward rather than more rebirth. I think this is a very old development because it’s found in all Buddhist traditions that I have seen so far. It might have been taught by the Buddha or his direct disciples. We can’t really be sure.


Thank you for this excellent post.
I think it highlights the difficulties and numerous pitfalls when attempting to categorize a text as being ‘early’ or ‘late’.