My thought is that we don’t have a full canon from any school of early Buddhism other than that of the surviving Theravada school. So, it’s only a partial record of what existed in the past. Very partial. There were at least a half dozen different early Buddhist schools with their own canons, but only some pieces of a three or four of them other than Theravadins still exist. The apparent lack of interest may just be that we’ve lost the texts that would show the interest.
4 Likes
8 verses of the Parayanavagga are found on an old Gandharan scroll. It was a part of a commentary on different verses. So it was definitely known in (somewhat) Northern regions [link].
And it is mentioned in the Mahasanghika Vinaya as a text that “serves for the instruction of novices”.
4 Likes
Yes.
Thanks, yes. It clearly has links to the south, but that doesn’t mean it is restricted there. There are plenty of stories of events in the north in the Pali texts, too. Geography is just one factor. But it does seem that the Parayanavagga served as a conversion text for the south, which eplains its prominence there as evidenced by the existence of the Niddesa.
3 Likes
As “widely accepted” as it is, the idea that the Aṭṭhakavagga and the Pārāyanavagga are old is based on a series of assumptions that resemble a house of cards. I wrote a critical essay on this last year: Some Issues of Pāli Chronology (30 September 2022)
For every line of evidence that has been used to argue for the antiquity of these texts, I can easily show that there is another plausible reading, usually some form of regional variation. The fact that no other versions of those texts exists for other schools is consistent with them being regional rather than chronological variants.
The idea that something is true because it is “widely accepted” is called the bandwagon fallacy. I’ve had too many experiences of this being utterly wrong in Buddhist Studies to give “widely accepted” any credence whatever. Part of why the old pārāyanavagga idea seems plausible is because of European biases with respect to texts. All of our methods derive from Christian or legal textual studies, both of which entail an obsession with origins for similar reasons (since God’s word is law). In this view, the past is defined Whiggishly as “when things were simpler”. And it leads to misconceptions.
For example, the Heart Sutra is a Chinese text composed in the mid 650s CE and the Sanskrit text is a back-translation from Chinese that contains Chinese idioms literally translated into Sanskrit (aka calques). What was widely accepted about the text, over many centuries, was so far off beam that it wasn’t even wrong.
Regarding the Dhammapada, as Roy Norman said in his notes on Dhp, the verses found there were drawn from a common (generic) pool of such verses that was also drawn on by both Jains and Brahmins. That some verses appear here and there is pretty much what we expect for Indian religious texts.
Do you have anything to show that the Pārāyanavagga was of any interest to Theravāda Buddhism before the Pali Text Society was founded in 1881? AIUI, Theravāda as an historical Buddhist school was largely focused on medieval commentaries on Abhidhamma rather than the Nikāyas. Sure they chanted suttas as parittā, but did they actually talk much about them? I can’t find references to Suttanipāta in Visuddhimagga for example (though I didn’t look that hard).
You must be aware that not everything in the Pāli canon is directly related to the Theravāda school. They preserved numerous texts that are clearly not Theravādin in outlook: c.f. the Mahāvedella Sutta (MN 43).
BTW, as I understand the word “consciousness”, it does not cease in the nirodha-samāpati. What ceases is what we would call the content of consciousness, i.e. in Buddhist terms what ceases are the khandhas, āyatanas, and dhātus (and everything associated with them). The mental state following cessation, i.e. suññatā-vihāra (a state of absence [of sensory experience]), is clearly a conscious state rather than an unconscious one. Indeed, that we are conscious during cessation is a crucial feature of the state of absence and what makes it different from mere unconsciousness (or death).
3 Likes
I agree that just because something is widely accepted does not mean that it is true. That said, there are many scholars who do accept that both are very old whether they are right or not.
Possible is not probable. All our arguments here are inductive. We are always arguing about probabilities and explanatory power. You can say that everything can be explained by regional differences, but you need to demonstrate that it is more likely than the other alternatives by showing that it explains the evidence better.
The PTS pulled together documents that existed prior to it. Its not like they were part of a hoax. With regard to modern and medieval Buddhism, I am not interested in them that much. If we are going to be radically skeptical, how do you know what was medieval? Couldn’t it have been all regional differences.
The Pali Canon is a mess. It is like going to a garage sale and paying five cents for a bag that says puzzle and finding out that there are pieces from several puzzles and some of them have missing pieces. That said, there are many places that reference or allude to the Atthakavagga and Parayanavagga which is more than you can say about many suttas or sub collections.
I am saying “cessation of consciousness” because that is how the translator translated it. Consciousness has many different usages. What I think is a hard sell is saying that the consciousness in Snp 5.2 is the same as sanna in Snp 4.2 or Snp 4.11.
Some general remarks:
- Saying what appears as chronology can always be regional differences seems to imply the ideas didn’t migrate implying a timeline. Sometimes ideas are thought of independently by different people, but relying on that for everything strains credulity.
- Ideas do develop over time. Finer distinctions and clarifications are made because over time the need for them does. Ideas collide and debates require everyone be one the same page as imperfectly as that may be done.
The Parayana is a ghostly text in northern sources compared to the Atthakavagga, but the two do get mentioned in the same breath by a couple Sarvastivada sources. So, a text or collection with the same title existed in at least one other Buddhist canon. That’s not the case with the Atthakavagga, however. A good case can be made that it existed in multiple traditions and is indeed very old.
I took the time to summarize a bunch of references to these two texts in Chinese Buddhist sources (i.e., Indic texts translated to Chinese) in another thread. One of that things that’s hard for me to ignore are the Vinaya stories depicting disciples reciting the Atthakavagga parallels during the Buddha’s time. At the very least, it means it was old enough that Buddhists in the 4th c. considered it as old as the Buddha.
5 Likes
Hey Jayarava, welcome, in fine form as always. I must say, I found that your linked page was urgently factual … in that it linked to global atmospheric CO2. The rest, not so much unfortunately. If you don’t have empathy for your subject, you will end up with ashes.
Thanks, that’s handy.
2 Likes
Is much known about how big a part the doctrine of anatta played in those other traditions and, if so, what flavor of anatta were they concerned with?
What I mean by what flavor is were they concerned about not self or not Self. I see this distinction as being the biggest doctrinal difference between the Atthakavagga and the Parayanavagga. The Atthakavagga is interested in the cessation of sanna and, to put it in the words of Ud 1.10, “self in that” where the Parayanavagga is interested in the cessation of vinnana and Self.
The two texts are clearly addressing different types of themes and audiences. But is there a doctrinal difference, or just a difference in emphasis?
First, we can look to the might MN 43 and read:
“Feeling, perception, and consciousness—these things are mixed, not separate. And you can never completely dissect them so as to describe the difference between them.”
But we may not be satisfied if we think that it is a generalization or a less authoritative statement when it comes to these two texts. Where else can we look then?
Well, we know that in the Pārāyanavagga, the whole point is about brahmins wanting to be free from nāmarūpa. The normal way to do this is to free the ātman — say, a sheer mass of consciousness — out of nāmarūpa leaving only eternal bliss outside of the manifest world; brahman.
What the Buddha discovered is that nāmarūpa and viññāna mutually condition one another. The freedom from nāmarūpa, its cessation, is only to be found in the cessation of its condition: consciousness. So to these brahmins and contemplatives in the audience of the Pārāyana, this is the right material to offer them. It is also consistent with the Buddha’s teachings on causality found in every tradition of early Buddhism, all over the nikāyas/āgamas. In fact, the mutual conditionality between nāmarūpa and viññāna seems to perhaps be an older version of dependent origination that is directly connected to or associated with the Buddha’s awakening in some way.
What do I mean? In the standard story, the Buddha realized all 12 links. But interestingly enough, at SN 12.65, the Buddha gives an autobiography of his practice to awakening. Note that this sutta is actually put on the lips of the Buddha, whereas the story of e.g. Ud 1.1-3 is just narrated by someone else. Here, the Buddha says he realized the mutual conditionality and that was his key insight.
At DN 14, a neat little easter egg is hiding in the story about Vipassī Buddha: He too based his core insight on the mutual conditionality between name-&-form and consciousness! Why would tradition specify this chain? Well, we know that the Past Buddhas are all meant to follow the same patterns as our Buddha, with two chief disciples of the same type, etc. etc. in a cosmic universe. So it seems they are pointing to something they knew about the Buddha’s awakening.
Briefly, we can note that this is how the Buddha presents the complex theme of dependent origination at DN 15 — the longest discourse on the subject in the canon, with several parallels, all sharing this feature. Namely, the explanation culminating in the mutual conditionality between name-&-form and consciousness. Note too that in some parallels, such as DA 13 I believe, there are layers to the text: in the introduction, they added a reference to “these 12 dependent arising” for the standard 12 links, and yet the actual sutta only explains the normal 9! So tradition recognized this lack, and in some cases tried to level it out later on. We also have outside discussions of it such as at SN 12.67, and at MN 38 — the other long explanation of dependent arising — the Buddha starts his actual, every-day example of dependent origination with consciousness and name-&-form.
Of course, even in the standard 12 links shared in every tradition, consciousness is conditioned and in relationship to name-&-form. So not only is this an extremely prevalent, common, and core essential doctrine of Early Buddhism, but it is also explicitly put on the lips of the Buddha and in the stories of past Buddhas several times in significant scenarios.
But wait, there’s more! We know that this question of the cessation of name-&-form as well as consciousness is extremely important to the contemplative Brahmanical groups of the time. We have confirmation of this outside the Buddhist texts in the Brāhmanas and Upanisads by the way continuing for centuries. But we also know that these contemplatives were doing practices that the Buddha himself is also recorded as having done! The Pārāyana talks about the attainment of nothingness several times, precisely what the bodhisatta practiced before his awakening under apparently contemplative Brahmanical-based teachers!
The point is this: the teachings of the Pārāyana are confirmed time and again, from independent sources, internal and external, up the wazoo in the Early Texts. It’s actually astonishing.
So what about the cessation of perception? Well, DN 9 is a good place to start. Here we learn that there were a bunch of contemplatives, ascetics, and religious philosophers of the time with all kinds of different opinions on perception. It was an important subject — the refinement, the limits, the happiness or suffering to be found in saññā (perception). Indeed, when we look at Early Buddhist meditation and contemplation practices, they are also all about perception.
Perceiving impermanence, anattā, dukkha, asubha; perceiving the breath; perceiving earth, forest, the elements, space, the immaterial attainments. Even at DN 15, a text linked to the teachings of the Pārāyana and strikingly similar to Snp 4.11 — in the Atthakavagga!! Seriously, compare Snp 4.11 to DN 15 — we see that there are a class of “non-percepient beings” associated with the 4th jhāna, and there is a refinement of perception up to the cessation of perception and feeling.
At DN 9, the golden sutta on perception, it too leads to the cessation of perception and feeling. This speaks nothing of consciousness/name-&-form, and yet it goes to the same place: the cessation of all experience, which, if one does not grasp to anything afterwards, is equivalent to nibbāna and the same goal as the cessation of consciousness.
We can see that the practices and topics of the Atthakavagga are associated with general sramanas and contemplatives. Different experimental practices and observances (sīla/vata), different views, different ideas on perception, notions of purity, debates and arguments. These are classic descriptions of the contemplative scene of the day and we see in the Atthakavagga a text that talks about navigating this scene. The Pārāyana addresses the other half of the story: Brahmanical philosophy that adapted itself into salvific and metaphysical thought. All throughout the discourses, we see the Buddha is a mastermind in managing both of these camps in very detailed, nuanced ways. It is not contradictory or mutually exclusive.
So when talking to contemplatives interested in perception, the Buddha did so. When talking to those interested in consciousness, he did so. He used some of their own terminology and worked with it in both cases, and yet ended up at the same place. Not only that, but the ideas are consistent with the heartwood of Early Buddhism found elsewhere, and we even see overlap between these approaches at DN 15. Perception, views, and contact are all related in EB discourse. And yet the cessation of contact/perception ends up being the same as the cessation of consciousness. The needs of the audience and subject matter seem to call for different teachings, but on closer inspection, they culminate in the same place.
None of this is to even mention that Snp 4 is all about the types of grasping (upādāna) and how it leads to entanglement in the world → rebirth. So many of the texts in the Atthakavagga talk about ending rebirth by not being established anywhere in perceptions, which means the cessation / non-planting of consciousness. The non-establishment and involvement of consciousness in the other aggregates is a major theme, if not the major theme, of Buddhism itself, summarized in the four noble truths.
There’s a lot to say on this, but I think this is sufficient. The post has gotten quite long. People have been making all kinds of claims about the Atthakavagga ever since the West got a hold of it it seems. Most of them are just wishful thinking. I really appreciate your interest and enthusiasm in looking into this topic more! 
Mettā
1 Like
Anatta plays a central role in all the Buddhist traditions that I’ve ever encountered. So, that question is easy enough to answer. Even the Mahayanists who railed against the Sarvastivada’s svabhava (essence) concept were extending the anatta concept to encompass dharmas as well as sentient beings, adopting the word sunya to express this wider idea.
For your second question, I’m not sure how to answer it. I think most early Buddhist schools understood anatta in the same way as Theravadins; i.e., as the lack of a permanent and unchanging thing that underlying sentient beings. The concept of emptiness appears more frequently in other EBT canons vs. the Theravadins, for whatever reason, which makes it difficult to connect these dots if one only reads Pali sources. The early concept of emptiness appears to have a relationship to meditation and cessation instead of the later controversies about the existence of dharmas. So, this may have some tangential relationship to what you are seeing in these texts - the word “sunya” just hasn’t been put to use in them.
I’m not sure if this really what you are asking about, but here’s some further reading on the topic of the early meaning of emptiness in Buddhism:
3 Likes
I think that it is important to note that the concept/doctrine of anatta was being extended after the Buddha because of intersectarian innovations and reactions to them. Extending the concepts/doctrine forces a change in the practice. I believe that is what we are seeing when we compare the Atthakavagga and the Parayanavagga. The attainment required for liberation went from completely understanding sanna to the cessation of vinnana because the Parayanavagga moved the goal posts extending the concept/doctrine to include not consciousness.
This kind of thing doesn’t happen because someone is talking to a different audience as @Vaddha suggests. That might change the language or metaphors someone uses to get the message across, but it does not change how liberation is achieved despite what later attempts at harmonization say. Complete understanding of sanna is not the same as the cessation of consciousness.
Actually, I don’t need to show it is better. The mainstream view is that only one reading is possible. By showing that another reading is possible and even plausible, I have arrived at a definite result: the hegemonic view is not the only possible view. As open minded students of Buddhism we may now wish to debate the the relative merits, but that is a separate exercise.
This sounds incoherent. If we cannot cite the only evidence we have, then the discussion will be very short, no?
Again, offering an alternative explanation of processes that have only been considered as chronological is an important first step to a broader discussion. In people won’t admit to alternatives then they won’t discuss the issue, or at least not sensibly. Religieux are famously unwilling to consider alternatives and Buddhists are no better in this regard.
Moreover, offering an alternative does not imply that the hegemony is wrong. By highlighting alternative explanations in terms of regional variations, I am not saying that no chronological processes are occuring or possible, and taking that as my message would seem disingenuous. What I’m saying is that implicit bias towards seeing chronological processes as the cause of changes in doctrine limits the discussion unreasonably. History is always complex.
As it turns out, I have ended up with gold. So I think I’ll continue to question your dogma, eh?
Part of plausible is probable. You can separate them if you like, but without going into what you are calling a separate exercise, you are only providing half an argument. So why is your interpretation better? BTW, my views are not mainstream either.
You have stated the conclusion. Now, show your working. If a good case can be made, then let’s see you make it and show your working.
Ironically, @sujato’s explanation was to cite regional differences to account for the difference in how the Pārāyana and Aṭṭḥaka were distributed. Demonstrating that sometimes a regional explanation is better than a chronological one, eh?
The common ancestor came first. This is using regional variation to demonstrate chronology. Yes, innovations happened and innovations happened on top of innovations. These are branches and twigs off the trunk. This is what evolution looks like.
Added later: when you say local variation, what do you think the variation is off of? A common ancestor? That implies a chronology.
Well, if you are interested in pursuing it further (and not merely in trolling people), I’ll have to suggest that you do some homework on your own time because I don’t have time to do it for you. I have a full-time job that requires my attention. And, besides, I’ve noticed over the years that people tend to learn more by doing rather than by being shown. If they have the desire, of course. If they don’t have the desire to learn, then there isn’t much point in either, is there? Chinese sources are woefully underutilized by Indic language academics such as yourself, so nosing around in the Taisho will probably be a good experience, regardless of the conclusion you might reach in the end.
I see you’ve written papers about the Heart Sutra relying on Chinese sources, so I assume you have the language and research skills required. Follow the link to the post where I listed a number of passages mentioning parallels to the Atthakavagga. Those passages will provide you with several different Chinese titles, some translated and some transliterated, which would indicate they were in Prakrit languages that treated S. asta and S. artha as homophones. Sometimes quotes from these parallels are also provided. Both can be the basis for further research and text searches. Be sure to take the time to consider the sectarian affiliations of those passages, dates of the texts, and the types of texts. It’s quite a cross section of sources, really.
2 Likes
Demonstrating, rather, that your depiction of historical criticism has no relation to how it actually works. 
2 Likes
Bhante, I’m curious as to your opinion on its authenticity then. If the Pārāyana is mainly a regional text, not shared as we would expect by all sides of the tradition in the same way the core Nikāyas are, do you find it likely that it is less likely to originate from the Buddha and his lifetime? Or how do you see this regional variation playing out (potentially)?
Mettā
These are all generalizations, rules of thumb; we can’t make strong inferences from isolated cases.
From the time of the First Council (!) there has been an explicit acknowledgement of geographical variation in the canonization process, this is historical criticism 101. And note that not only was there variation acknowledged back then, but it was explicitly said to have been a monk who was in the Dakkhinagiri.
The fundamental sources of all texts are defined geographically, not temporally, which reflects the fact that for early Buddhists, different teachings were given in different places: that is the primary fact. Only later were they brought together; the entire process has been both coming together and moving apart, before finally solidifying in the (relatively) fixed spheres of influence of the later schools. Literally everyone in Buddhist studies knows this and discusses the geographic significance of textual transmission. To say otherwise is just a fever dream. Purely as a random example, here’s Oldenberg in 1879 discussing geographical questions in relation to early text development.
The most regional parts of the Parayanavagga are the opening and closing portions, which everyone agrees were added later. The actual sixteen questions, which are of course set in Magadha, don’t define their geography, but as I have shown, they implicitly agree with the location as stated in the Introduction.
2 Likes