Proto-Mahayana Features in the Agama Sutras?


The issue really is that you’ll have difficulty marrying up the passages Mahayana texts cite in the Agamas with the Nikayas. The two aren’t the same literary tradition. There’s no sutta in the Majjhima Nikaya in which Maitreya receives the assurance of his future buddhahood, for example, but Maitreya is known to the Pali tradition. It’s just not the literary tradition that Asanga et al were reading. So, yes, it would be good for the Mahayana world to have these texts translated. The arguments about the Pali being earlier or redacted, etc are not very fruitful in any case, given how much speculation is involved. And in the meantime, everyone’s impression of what the Agamas must of have said is a bit distorted by reading the Nikayas in their stead.



Unfortunately, I don’t see much said about the actual dating of the Pali canon. I’ve gathered there’s multiple editions and text critical issues. In Chinese, for example, you have the issue of texts being “updated” (or miscopied) over 1500 years of time. So, even though we believe a text was translated in 400CE (for example), that doesn’t mean we’ve received it in pristine form today.



The pivotal paragraph seems to be this one:

And what are the four absorptions—blissful meditations in the present life that belong to the higher mind—that they get when they want, without trouble or difficulty?
Katamesaṃ catunnaṃ jhānānaṃ ābhicetasikānaṃ diṭṭhadhammasukhavihārānaṃ nikāmalābhī hoti akicchalābhī akasiralābhī?

The MA3 translation that relates is this:

“What is the abundance of four foods in the king’s city, which are easy and not difficult to obtain? Namely, the king’s border city provides for a reserve of water, hay, and firewood. Those inside are made secure, and outside enemies are kept at bay. This is called the first abundance of food in the king’s city, which is easy and not difficult to obtain.

Notice that the classic way to break into a city is to starve it during a siege, forcing the city to draw down on its reserves. I think the MA3 translation must be read with this in mind, that the availability of provisions (or jhana by simile) is conditioned by circumstance. This reading of MA3 would lead to the conclusion that well-practiced jhana is effective (due to its ready availability) while unpracticed jhana is not effective (due to its lack of availability). I wouldn’t infer from MA3 that jhana is easy to obtain without practice.


My only point was that it’s just as speculative to assume that the agamas contain later additions.


As a Mahayana Buddhist, I don’t believe the Mahayana sutras invalidate or supersede the previous Buddhist scriptures. The Four Noble Truths and Eight-fold Path don’t apply to the Mahayana any less than the Ten Commandments apply to Christianity.

In the Smaller Amitabha Sutra, for example, the birds in the Pure Land sing praises to the Eightfold Path, and arahants sit side-by-side with bodhisattvas to hear Amitabha Buddha teach.

Mahayana Buddhism might have a more expanded understanding of the Buddhist path, but the agamas/nikayas remain foundational. This point is made by the Dalai Lama.


It is speculative, but we do have plenty of examples of texts expanded over time in Chinese translation. So, there’s good reason to assume a more elaborate text has developed over time and a simpler one might not have.

It’s just that when I compare the multiple versions of a Pali sutta and its Chinese counterparts, it sometimes turns out to be more complicated than that. There may be more than two versions that agree and disagree in different ways. They are clearly related, but they’ve diverged. In other cases, it’s not so complex. So, it’s difficult to make generalizations.



On a related note, the Greek Septuagint, though a translation, might be more faithful in certain ways to an earlier version of the Hebrew Bible than the Masoteric text we have today:

Although the consonants of the Masoretic Text differ little from the text generally accepted in the early 2nd century (and also differ little from some Qumran texts that are even older), it has many differences of both greater and lesser significance when compared to the manuscripts of the Septuagint, a Greek translation (about 1000 years older than the MT made in the 3rd to 2nd centuries BCE) of the Hebrew Scriptures that was in popular use by Jews in Egypt and the Holy Land (and matches the quotations in the New Testament of Christianity, especially by Paul the Apostle).[8]…

The discovery of the Dead Sea Scrolls at Qumran, dating from c. 150 BCE-75 CE, shows that in this period there was not the scrupulous uniformity of text that was so stressed in later centuries. According to Menachem Cohen, the Dead Sea scrolls decided these issues ‘by showing that there was indeed a Hebrew text-type on which the Septuagint-translation was based and which differed substantially from the received MT’.[13] The scrolls show numerous small variations in orthography, both as against the later Masoretic text, and between each other. It is also evident from the notings of corrections and of variant alternatives that scribes felt free to choose according to their personal taste and discretion between different readings.[13]

However, despite these variations, most of the Qumran fragments can be classified as being closer to the Masoretic text than to any other text group that has survived. According to Lawrence Schiffman, 60% can be classed as being of proto-Masoretic type, and a further 20% Qumran style with bases in proto-Masoretic texts, compared to 5% proto-Samaritan type, 5% Septuagintal type, and 10% non-aligned.[14][page needed] Joseph Fitzmyer noted the following regarding the findings at Qumran Cave 4 in particular: "Such ancient recensional forms of Old Testament books bear witness to an unsuspected textual diversity that once existed; these texts merit far greater study and attention than they have been accorded till now. Thus, the differences in the Septuagint are no longer considered the result of a poor or tendentious attempt to translate the Hebrew into the Greek; rather they testify to a different pre-Christian form of the Hebrew text".[15] On the other hand, some of the fragments conforming most accurately to the Masoretic text were found in Cave 4.[16]
Masoretic Text - Wikipedia

While the Chinese agamas are a translation, they might nonetheless be closer to the Buddha’s original teachings in certain parts than the Pali suttas are. Let’s also not forget that the Buddha didn’t speak Pali.


Yes, I agree. And, yes, it’s a similar situation as in your quote. The Chinese Agamas are not just a second canon; we have translations of the same sutra from two or more canons in Chinese sometimes. The dating can differ quite a bit, too–some are early and some are late. It’s a complicated picture.


While the Pali suttas are, in certain ways, different from the Chinese agamas, that doesn’t mean Mahayana Buddhists can’t benefit from reading them. Before converting to Buddhism, I preferred the Jewish Publication Society’s translation of the Hebrew Bible, even though I’ve never been Jewish.


The Mahāyānika āgamāḥ are rare (unless you count ones where the influence seems subtle and is possibly an interpretive stance more so than something actually in the text itself) but stick out like a sore thumb when you find them for simple terminological reasons.

[These below amateur translation efforts have been modified after some corrections and suggestions by @cdpatton and @Gene.]

This is EA 27.5

聞如是: 一時,佛在舍衛國祇樹給孤獨 園。
It was heard like this: one time, the Buddha dwelt at Śrāvastī, at Jetavana.

爾時,彌勒菩薩至如來所,頭面禮足,在 一面坐。
At that time, Maitreya bodhisattva came to the Tathāgata’s location, head facing downward and bowing to the foot [of the Buddha], then beside [the Buddha] to one side sat.

At that time, Maitreya bodhisattva addressed the Bhagavān saying:

「菩薩摩訶薩成就幾法,而行檀波羅蜜,具足六 波羅蜜,疾成無上正真之道?」
"Bodhisattvāḥ mahāsattvāḥ accomplish how many dharmāḥ to perform dānapāramitā, to complete six pāramitāḥ, and swiftly accomplish, with nothing higher, correctly and truly, the path? [or “swiftly accomplish anuttarasamyaksaṃbodhi?”]

I can’t really translate anymore and even this likely has some suspect parts. It goes on to enumerate a set of four dharmāḥ that accompanied these six perfections, as far as I can make out. Maybe @cdpatton can be of assistance here, since he is much more qualified to handle these texts than myself. For instance I am not totally sure of the way that the bowing is being spoken about here.

That is an āgama that has a fully fleshed out six perfection schema, not even one of the pleroma of irregular 5, 8, 9, 10, or 12 perfection schemata that characterized early Mahāyāna, i.e. they took a while to settle on six perfections, and I believe tantrists have 12 perfections but I am not sure. I know they believe in nine yānāni/vehicles but this is a digression at this point as the subject matter of this thread was not scholastic inflations of pre-existing path schemata.

EA 20.7 also gets identified as one of these Mahāyānika āgama sūtrāṇi, and the language in it I find even more difficult, but I can give people here the gist of what I’ve heard argued about it, namely that it is proto-Mahāyāna because it allegedly combines śamatha and vipaśyanā. Śamathavipaśyanā, the fusion of the two, for right or wrong, is often believed to be a Mahāyānika innovation by some, mostly Mahāyānists themselves. I don’t know if śrāvaka traditions have this fused meditation.

聞如是: 一時,佛在舍衛國祇樹給孤獨 園。
It was heard like this: one time, Buddha dwelt at Śrāvāstī, at Jetavana.

At that time, the Bhagavān told the myriad monks: “The secluded monk should cultivate two dharmāḥ.

To speak of which two dharmāḥ? So called śamatha and vipaśyanā also.

[skipping quite a bit for brevity… also there are too many things in this section that I do not understand.]

「過去諸多薩阿竭、阿羅訶、三耶 三佛,皆由此二法而得成就。
“In the past, tathāgatāḥ, arhantaḥ, samyaksaṁbuddhāḥ, all of them followed these two dharmāḥ and accomplished the undertaking.

所以然者,猶如菩薩坐樹王下時,先思惟此法止與觀 也。
The reason is this: like the worthy bodhisattva dwelling by the tree of kings, [dwelling] under for a time, he first reflects on these dharmāḥ: śamatha and vipaśyanā also.

If the bodhisattva mahāsattva has attained śamatha already, he is immediately able subdue Māra, the enemy;

If again, the bodhisattva has attained vipaśyanā also, he has accomplished threefold wisdom, has completed that which there is nothing transcending, is an arhat, [attains?] samyaksaṃbodhi.

For this reason, myriad monks, the secluded monk should seek these appropriate methodologies to walk in [i.e. practice] these two dharmāḥ.

Like this, myriad monks, regard this as your study.”

爾時, 諸比丘聞佛所說,歡喜奉行。
At that time, the myriad monks heard from Buddha what was taught, they rejoiced, they propagated it.

If it is not too much trouble I should like @cdpatton to take a look at these amateur translations and give me some feedback. There are more than a few points I am not sure of in the above, especially in the second scripture.

Interestingly, the second scripture, while allegedly containing Mahāyānika meditative cultivation instructions and overtly referencing Mahāyānika pudgalamarga/path-persons (a thread on whether or not the fusion of śamatha and vipaśyanā is legitemately Mahāyānika or is found in the EBTs would be very interesting), it seems to treat bodhisattvāḥ and arhantaḥ as equivalent stations, potentially: 薩阿竭阿羅訶三耶三佛. This, to my amateur eyes, looks like a list of titles in the tradition of bhagavā arahaṃ sammāsambuddho but that is predicated on the assumption that 阿竭 can stand in for “thus gone” by itself.


The 薩 sa that I took as short form for (bodhi)sattva/(pu)sa might actually be sugata or something like that too with the 阿竭 agatha proceeding it. Maybe?


Other than the use of the term “bodhisattva” as a separate class of practitioners, I don’t see how this Agama is inconsistent with Pali suttas. an 4.94 talks about how Samatha and Vipassana both need to be balanced/cultivated. SN 35.245 speak of Samatha and Vipassana as a “swift pair of messengers” that delivers nibbana.

I don’t get how combining samatha and vipassana is proto-mahayana.

Yes, it would.


This, along with an allegedly more refined understanding of emptiness, is one of the dubious claims of the Mahāyāna with regard to exceptionalism, one of the things that allegedly separate it from lower vehicles. It’s more of a self-narrative in my experience. But it’s so pervasive that I, as a bodhisattvayāna practitioner, didn’t know either way that this fusion was also in śrāvaka literature.


Hi, Coemgenu,

I assume 頭面禮足 means he prostrated himself with his head at the Buddha’s feet. There are a few different Chinese compounds that were invented to describe this posture.

幾 is a question word “how many” or “what number.” So, the question is “How many dharmas does a bodhisattva-mahāsattva accomplish to practice the dāna-pāramitā …”

無上正真之道 may be a translation of Skt. anuttarā-samyak-saṃbodhi. Not sure. 之 can function to connect a series of adjectives to the noun they modified, and 道 can translate bodhi as well as marga (Daoist Dao vs. ordinary dao). I hven’t studied the EA closely yet, so you may be correct.

This is one of the things that gives us a difficult time with early Chinese translations. It’s easy to get confused because we are thinking of the wrong source language (and the actual one is nearly lost to us). The language translated is Gandhari, not Sanskrit, and transliterations are the main evidence of that.

The Gandhari equivalent of Tathāgata is Tasāgada. So, the Chinese transliteration 多薩阿竭 = G. Tasa-aga(da) = Skt. Tathāgata. is a great resource for trying to decipher unknown transliteration in early Chinese translations.

多 is one of those characters that’s usually a normal word (many) but also a transliteration syllable (‘ta’), so it’s another pitfall when at the head of a transliteration–it can look like an adjective.

猶如 means “It’s like/for example”–introducing a simile or example. “Just as when the Bodhisattva was still sitting under the king of trees.” 先 is also an adverb “prior/before”. “He first contemplated these dharmas.”

三達智 = Skt. trividyā. 至真 is an old translation of Skt. arhat (“arrived at the truth”), and 等正覺 is a translation of Skt. samyak-saṃbodhi, but the samyak and saṃ- are reversed here. When you see 等 modifying nouns in Chinese translation, it’s often a saṃ- compound in the original.

Hope this helps,



Really, bodhisattva path practioner ? Could you elaborate a bit ? Do you practise bodhisattva precepts ?


Interesting! I think part of the problem is there was indeed a tendency in later Theravada teachings to exalt Vipassana, a tendency which is still seen amongst modern Burmese practitioners (though not in the Thai Forest tradition so much). I guess Mahayana folks are conflating this “Vipassana-only” approach with “Hinayana” or “Theravada” in general, leading to the inaccurate notion that only Mahayanists pioneered the samatha/vipassana-balanced approach.


I certainly try. Am I perfect? Hardly. I could certainly be a much better Buddhist in general, let alone Mahāyānika.

I used to practice with the Buddhist Church of Toronto (Jodo Shinshu) but gravitated away from that when I moved back from the city to the country. The nearest saṁgha to me is a Tendai group that I go to when I can, but they are a 4 hour drive away!

Tbh I have more contact and exposure to the Venerables at the Scarborough Mahāvihāra these days.

I view not disparaging Theravāda as an important part of my practice, being (I think) the 10th (?) root downfall of the beginner bodhisattva precepts.


猶如here read as “just as” or “just like”.

所以然者 ,猶如 菩薩坐樹王下時
And why is that so , just as the Bodhisattva was sitting under the king of trees .


It’s actually 18 & 19, I just looked.


Both about farming and trading . Avoid killing insects in farming and fair measurement (&price) in the trading of goods .