AN 9.37 and its cluster of Chinese Parallels explained by Agama expert

My current sutta central offline website has this listed as the parallel for AN 9.37:
AN 9.37 Ānanda AN iv 426 SA 557*

A while back brother Chansik did a translation of the chinese parallel for AN 9.37, SA 559. It’s an old thread within the past year you can find on SCD&D. In his English translation, it looked like it contradicted what AN 9.37 said regarding the possibility of first jhāna in a state where the mind is divorced from the 5 bodily faculties. So I asked my friend Dr. Chu to investigate further, here’s his reply.

SA557-SA559 should be treated as a cluster of suttas on the same theme, i.e., that of animitta samadhi.

This is not a controversial observation. It has been pointed out by a few Agama specialists, including Yinshun (see, for example, Kong zhi tanjiu (1985), p. 36). Also, in the Chinese Agamas, Ananda is most often the main interlocutor or expounder of animitta samadhi. And here, in all three suttas, Ananda was the protagonist.

Animitta samadhi is a tricky matter, both in the context of the Pali canon and that of the Chinese Agamas. In both contexts, animitta could refer to a variety of very different attainments: it could be synonymous with the dimension of neither-perception-nor-non-perception; it could be an unskillful, coma-like state of general non-differentiation (an instance of Wrong Concentration which the Sarvastivadins and the Sautrantikas classify as one possible manifestations of the fourth jhana, and the Theravadins classify as unskillful but nonetheless above the realm of the fourth jhana); it could be a characteristic of the Unbinding itself; it could be the cessation [of feeling and perception] attainment; or it could be this unspecified but highly revered attainment where all disturbances cease (c.f. Culasunnata Sutta).

In the Chinese context, animitta is variously translated as wuxiang 無想 and wuxiang 無相. Given the context of SA557-SA559, there’s little doubt that wuxiang 無想 and wuxiang 無相 are indeed treated as the same thing.

With this in mind, there’s nothing particularly controversial about the doctrinal stance of SA559.

Typically, jhanas in the major Nikayas/Agamas are primarily described in somatic terms (related to the body). There’s good continuity on this issue between the “early of the early suttas” (e.g. portions of KN) and the major Nikayas/Agamas. The former talks often of “staying in touch with bodily bliss” as among the primary duties of a monk, that it is perfectly sensible that the same theme is picked up and elaborated in the other Nikayas/Agamas. It also leaves little doubt that, jhanas, as envisioned in the early suttas, entail tactile/corporeal sensory experience.

In contrast, the formless are differentiated from the jhanas by the experience of sensory shutoff: bodily perception is transcended, the mind is no longer sensitive of the dimension [of the five senses], there’s not the perception of multiplicity, feelings and somatic metaphors are absent in their standard descriptions…

But there’s an exception to this general rule. This is where the animitta is shown to be unique. And in fact, SA557-SA559 are precisely about Ananda being asked about the special status of animitta. When first jhana all the way to the dimension of not-a-thing-ness are practiced in the animitta way, the mind can be “noncognizant [of the sensory dimension]” but still perceiving perceptual data; and when neither-perception-or-non-perception is practiced in the animitta way, the mind can be “noncognizant [of the sensory dimension]” AND also not perceiving perceptual data.

In other words, the main point of the SA suttas in question is to point out the unique nature of the animitta attainment, which subverts the norm. The norm is of course that, in jhanas, one is cognizant of the sensory dimension and perceiving perceptual data.

Although AN9.37 is identified as a parallel sutta to SA557 & SA559 taken as a cluster, AN9.37 is, unlike its supposed SA parallel, actually spelling out the norm: first, it makes no mention of the animitta at all (this is significant, and it brings to question whether we are dealing with sister suttas after all). It is simply talking about the formless attainments. Second, it proceeds to describe the formless attainments as having the characteristic of “not being sensitive to the sensory dimension.” And of course you cannot apply that same description to jhanas, which is precisely why AN9.37 didn’t include the jhanas in its list of “not-sensitive to that dimension.”


Interesting post!

I wonder if I have misunderstood something here. Fully absorbed in jhāna, one cannot hear - is that correct? I had also thought that the other 4 of the five non-‘mind’ senses were not active in jhāna. And I had assumed it is those 5 sense inputs which you ae referring to by ‘sensory dimension’.


The essay is by Dr. Chu, not me. Whether one can hear sound in jhāna and feel leg pain is a controversial topic. I’ll share my detailed notes auditing the pali and english passages on this later this year.

1 Like

Thanks for sharing this essay or rather long commentary. It’s interesting, but of course I would have liked to see specific sutta references, but that was probably not the purpose of the text.

I can’t speak for the Agamas of course, but some claims about the Pali canon are doubtful. I think I can refer to my nimitta source collection as a reference where I collected all occurrences of animitta samadhi as well (3b).

As the sources show animitta samadhi (whatever it is) can be located rather clearly:
a. it’s beyond the fourth jhana
b. when the arupas are mentioned it’s beyond the highest arupa
c. when the cessation of perception and feelings are mentioned it is below it
d. it appears a few times at the same level together with suññata (emptiness) and appaṇihita (desirelessness)

Nowhere it is mentioned in EBT that it could be unskillful, quite the contrary, it is praised, or recommended, or described as a state where the Buddha chills out.

So maybe in the Agamas it has a different character, but at least in the Pali suttas these are pretty much its features…


Hi Gabriel

There’s an interesting intersection of the "signless " shown in SN 22.80/SA 272 and MN 78. Taking the former’s vitakka to be synonymous with the latter’s saṅkappa, it does look jhanic, does it not?

I guess you mean the following passages?

SN 22.80: And where, bhikkhus, do these three unwholesome thoughts cease without remainder? For one who dwells with a mind well established in the four establishments of mindfulness, or for one who develops the signless concentration.

MN 78: And where do these wholesome intentions (saṅkappā) cease without remainder? Their cessation is stated: here with the stilling of applied and sustained thought, a bhikkhu enters upon and abides in the second jh›na, which has self-confidence and singleness of mind without applied and sustained thought, with rapture and pleasure born of concentration.

You’re right that SN 22.80 is interesting in that it takes animitta samadhi as a tool to get rid of vitakka - it seems like too fancy a tool. The context is not clear but one could guess (since it is usually consistent as a high meditative attainment) that it means to completely nirodha the akusala vitakkas, not just to supress them like in jhana.

I know that sometimes sankappa and vitakka appear in lists of synonyms, but they also appear in very different contexts. Why would you equate the two particularly here?

I’m pretty sure what Dr. Chu means is that the chinese characters used to represent “animitta samadhi” is what can represent those other samadhi states. That is, it’s not the “animitta samadhi” of the Agama itself that can take on the nature of those other samādhis, it’s that the Chinese characters used to represent animitta samadhi is ambiguous in meaning different types of samādhi in different contexts. So if we had access to the original Sanskrit texts from which the Chinese translation is derived, we probably would have more congruence with Pali (just my guess).

In the Pali, DN1 talks about a wrong type of non-perception concentration, a-sanñña, that for a non disciple of the Buddha, would lead to a future rebirth in a formless non-perception realm with an incredibly long life span.

1 Like

There’s also something from Snp 2.11 where it pops up in the context of the 5 cords of sensual pleasure.

Edit - apologies for the very short reply above, as I was on my phone.

There was a very brief discussion about this at - Interpreting the first Jhana

Since that time, I’ve had the opportunity to look at some of the MA usages, which suggest interchangeability between these terms. In the MA parallels to MN 19 and MN 20, we have 念 (saṅkappa) popping up where the Pali texts have vitakka.

What strikes me about these texts is that where the context is meditation and the concern is with sensual desire, ill-will and cruelty, the terms are saṅkappa and vitakka are synonymous. We can see this play out in AN 4.35 -

So yaṃ vitakkaṃ ākaṅkhati vitakketuṃ taṃ vitakkaṃ vitakketi, yaṃ vitakkaṃ nākaṅkhati vitakketuṃ na taṃ vitakkaṃ vitakketi; yaṃ saṅkappaṃ ākaṅkhati saṅkappetuṃ taṃ saṅkappaṃ saṅkappeti, yaṃ saṅkappaṃ nākaṅkhati saṅkappetuṃ na taṃ saṅkappaṃ saṅkappeti. Iti cetovasippatto hoti vitakkapathe.

He thinks whatever he wants to think and does not think what he does not want to think; he intends whatever he wants to intend and does not intend what he does not want to intend; thus he has attained to mental mastery over the ways of thought.

Yes, that’s how it appears to me also. SN 14.12 seemed like it might provide a clue, but it’s unclear where vitakka would fit along the food chain here. The absence of vitakka makes me think (based on this sutta SN 14.12) it’s either equivalent with sankappa, or it comes at the end of the chain after the last element quest (pariyesanā). But “quest” seems like a much more obsessive and complex activity than a mere “thought”, I still think vitakka=sankappa is the best fit, though this sutta unfortunately doesn’t spell it out explicitly for us.

SN 14.12 sa-nidāna-suttaṃ
SN 14.12 with-(a)-source-discourse

sāvatthiyaṃ viharati … pe …
At Sāvatthı̄ [the Blessed one] dwelled … [and addressed the monks:]
“sa-nidānaṃ, bhikkhave, uppajjati kāma-vitakko,
“with-(a)-source, *********, arises sensual-thought,
no a-nidānaṃ;
not without-source;
sa-nidānaṃ uppajjati byāpāda-vitakko,
“with-(a)-source arises ill-will-thought,
no a-nidānaṃ;
not without-source;
sa-nidānaṃ uppajjati vihiṃsā-vitakko,
“with-(a)-source arises harming-thought,
no a-nidānaṃ”.
not without-source;

“kathañca, bhikkhave,
"How, monks, [is it the case that]
sa-nidānaṃ uppajjati kāma-vitakko,
“with-(a)-source arises sensual-thought,
no a-nidānaṃ;
not without-source;
sa-nidānaṃ uppajjati byāpāda-vitakko,
“with-(a)-source arises ill-will-thought,
no a-nidānaṃ;
not without-source;
sa-nidānaṃ uppajjati vihiṃsā-vitakko,
“with-(a)-source arises harming-thought,
no a-nidānaṃ?
not without-source;

kāma-dhātuṃ, bhikkhave, paṭicca uppajjati kāma-saññā,
sensuality-element, *********, (is the) dependent-cause (for the) arising (of) sensual-perception.
kāma-saññaṃ paṭicca uppajjati kāma-saṅkappo,
sensual-perception (is the) dependent-cause (for the) arising (of) sensual-intention.
kāma-saṅkappaṃ paṭicca uppajjati kāmac-chando,
sensual-intention (is the) dependent-cause (for the) arising (of) sensual-desire.
kāmac-chandaṃ paṭicca uppajjati kāma-pariḷāho,
sensual-desire (is the) dependent-cause (for the) arising (of) sensual-passion.
kāma-pariḷāhaṃ paṭicca uppajjati kāma-pariyesanā.
sensual-passion (is the) dependent-cause (for the) arising (of) sensual-quest.

kāma-pariyesanaṃ, bhikkhave,
(immersed in a) sensual-quest, *********
pariyesamāno as-sutavā puthujjano
(the questing) un-instructed worldling,
tīhi ṭhānehi micchā paṭipajjati —
(the) three kinds (of) wrong path-(he)-enters -
kāyena, vācāya, manasā.
(by) body, (by) speech, (by) mind.

1 Like

[quote=“frankk, post:1, topic:5810”]
In the Chinese context, animitta is variously translated as wuxiang 無想 and wuxiang 無相. Given the context of SA557-SA559, there’s little doubt that wuxiang 無想 and wuxiang 無相 are indeed treated as the same thing.
[/quote]It seems that 無想 is also occasionally used to represent asaṃjñā, do you know if this usage comes from the same layer of Chinese usage as the āgamāḥ being addressed here?

I would ask more questions, but then I would betray my ignorance as to if asaṃjñāsamāpatti (無想三摩跋提?) & saṃjñāveditanirodhasamāpatti (想滅盡定?) are terms or concepts that appear in EBTs at all as the compound forms above, since obviously nirodha and things in general can be found in abundance.

The topic inspired me to do some reading, and I stumbled on a compilation of essays called India in the Chinese Imagination: Myth, Religion, and Thought that is viewable in preview for free on Google Books. I haven’t read any of the essays in full yet, but Robert H. Sharf has this to say on the subject matter, if anyone is so interested:[quote]Terminological Confusions

The notions of non-conceptualization, cessation, and nirvāṇa are all inextricably tied to the seminal Buddhist doctrine of “non-self” (anātman). It is precisely because there is no permanent, unchanging, ontologically extant self or soul that the temporary cessation of consciousness in nirodha, and the permanent cessation of the aggregates in nirvāṇa, can be construed in positive terms. Buddhist practice, as depicted in the early textual tradition, is directed not towards the realization of some true self or transcendent other, but rather to the end of delusion. The notions of non-self (anātman), non-conception (asaṃjñā), cessation (nirodha), and nirvāṇa are all intertwined in complex and sometimes contentious ways.

The Buddhist tradition employed a host of terms in their technical analyses of self, consciousness, and personhood. Vijñāna, citta, saṃjña, and manas are among the Sanskrit terms commonly used for what we might call mind, consciousness, cognition, or conception, for example, and “self” can be rendered, depending on context and ethical valence, as ātman, pudgala, or sattva. The distinctions among these terms are not always easy to parse, and there are inconsistencies in usage across our sources.

As we move to China, the terminological complexity is exacerbated by the problem of translation and profusion of alternative Sinitic renderings of key Indic terms. Some equivalences become somewhat standardized in mature Chinese translation of South Asian sources: wú wǒ (無我) for anātman; miè jìn dìng (滅盡定) for nirodhasamāpatti; wú xiǎng (無想) for asaṃjñā; and so on. But at the same time we find vijñāna, citta, saṃjña, and manas all rendered, in different contexts, as xīn (心), for example, xīn (心), xiǎng (想), and even shí (識) are not consistently or clearly distinguished in commentarial materials.[/quote]He paints a bit of a bleak picture, but may well have been writing without knowledge of some of the key insights and work that has been done on decoding the Chinese tradition of rendering Buddhist language.

I’ll ask my friend and post a response from him if he has one. No guarantees, he’s a busy guy.

想(受)滅盡定 。

a response from Dr. Chu on animitta

Yes, part of the problem that exacerbated a streamlined definition for animitta comes from that Chinese translators used different words to translate the same Pali/Sanskrit terms, or same words to translate different Pali/Sanskrit terms. But it appears that since the earliest days of Buddhism, animitta has been shrouded in ambiguity and obscurity.

A passage in the SA (T2.146b) explains how an attainer of animtta-cetosamādhi wanted to find out what it was that he attained. He followed Ven. Ananda for as long as six years, waiting for another person to ask about it or for Ananda to talk about it on his own accord, but neither happened with the passage of such a long time. He finally asked about it himself. This was an indication of how little-known and little-discussed the attainment was even in the early Sangha.

The Abhidhamma-kośa corroborates the ambiguous and loaded nature of the attainment (T27.541): “As for the teaching on animitta, it can be expounded in a great variety of ways: it could be in reference to suñña-samādhi; it could be in reference to [the state of mind when one] sees the Path; it could be in reference to the Imperturbable Deliverance; it could be [the same as] the base of neither-perception-nor-non-perception; and it could be [its own thing, i.e.] the animitta-samādhi.”

As with regards to the issue of wuxiang 無相 and wuxiang 無想: the asaṃjñā/asaññā is only expressed in Chinese as wuxiang 無想, but animitta is rendered in Chinese both as wuxiang 無相 and wuxiang 無想—hence the compounded confusion. This discrepancy might be attributable to a historical reason: before the advent of xylographic technology (woodblock printing) in China, scriptural manuscripts that were hand copied would allow for variant logographs for certain characters (zhengzhi 正知 & zhengzhi 正智 is an example); and the hand-copied manuscripts were the templates from which the printed versions would emerge. In that process, sometimes the variant characters would be made uniform by editors, and sometimes they would be preserved.

But even prior to the transmission of Buddhist texts into China, the connections between asaññā & animitta might have been complex and multi-layered. You can detect in the suttas that the two are sometimes conflated. In these instances, asaññā is not an unskillful state, but a skillful state that is variously tied to different kinds of mundane and supramundane release.

An example is from the DA (T1.15b), where the Buddha is said to find alleviation from physical pain by going into assañña. And there, the description for that state is identical with what you typically find with animitta (c.f. SA at T2.149c): “not giving attention/manasikāra to any sign.”

Another example, Ud6.7, probably predated the mature meditation schemes we see in the major Nikayas. See how it treats what in the mature scheme would typically take to be second jhāna:

“The Blessed One saw Ven. Subhūti sitting not far away, his legs crossed, his body held erect, having attained a concentration free from directed thought.”

“[He] exclaimed: Whose thoughts are vaporized, well-dealt-with within, without trace — going beyond that tie, perceiving the formless, overcoming four yokes (sensuality, becoming, views, ignorance), one doesn’t go to birth.”

Still another example is from the MA (T1.543a). Here, the animitta is said to be attainable by examining/reflecting upon the drawbacks of saṃjñā/saññā (Interestingly, the parallel sutta of this sutta, MN106, speaks of neither-perception-nor-non-perception, rather than of animitta/asaññā—still another indication that animitta/asaññā is, in some places, mixed up or related to neither-perception-nor-non-perception). In other words, nimitta IS saññā in this and many other cases. In all of these examples, the boundary between animitta and asaññā is blurry, complex, or nonexistent.


This must be a sutta made in the time the sangha got doubt in Arahants. And the start of the 3 paths. Entering the stream was already the way to explain a sort of temporal liberation state. But these sutta as seen against the pali parallel. Is attaining temporal nirvana vs attaining nirvana. It’s obviously when sangha divided. The fight about the state of Arahants. Less Arahants. Doubts. True Dhamma was lost.