---
title: "Recollection in the Age of AI: The Manas-Turning Operator Still Operative"
author: "釋慧鏡 (Shi Huijing)"
series: "Six Practice Gates, Volume 1 — Mindful of the Buddha: From Early Buddhist Recollection to the Flower Ornament Vision"
series_short: "NIAN"
volume: 1
paper_id: "NIAN-P30"
paper_number: 30
part: "Part VIII · Coda"
part_number: 8
date: "2026-05-04"
version: "1.0"
license: "CC BY-NC-SA 4.0"
---

# Recollection in the Age of AI

**The *Manas*-Turning Operator Still Operative**

**Part VIII Coda · *Buddhānusmṛti* in the Age of AI · Paper 30 of *Mindful of the Buddha*'s thirty-one**

*中文主題：AI 時代之念佛——末那-轉依機制仍然有效*

---

**Author:** 釋慧鏡 (Shi Huijing)
**Date:** May 2026
**License:** CC BY-NC-SA 4.0
**Version:** 1.0
**Series:** NIAN — Mindful of the Buddha

---

## Abstract
This paper takes up a question that the closing of *Mindful of the Buddha* cannot avoid: in 2026, with AI already part of everyday cognition, does *buddhānusmṛti* (念佛, buddha-recollection) still count as recollection? The argument proceeds in three layers. First, taking the *Cheng weishi lun* (*Vijñaptimātratāsiddhi*) fascicle 4 thesis-line — "in the pre-turning state, [*manas*] continuously and discerningly cognises the appropriated self-mark; in the post-turning state, it likewise discerningly cognises the no-self mark" — as its central proposition, the paper argues that *buddhānusmṛti* functions as a doctrinally-licensed *manas*-turning operator. Its mechanism is the sixth consciousness as opening-and-guiding support (*kai-dao yi*, 開導依) for the initial arising of *samatā-jñāna*-correlated *manas*; AI's presence neither replaces nor cancels this chain. Second, drawing on the *Abhidharmakośabhāṣya*'s lifespan tables, the eight inopportune births (the long-lifespan heaven among them), the twenty-four non-associated formations of the *Mahāyānaśatadharmavidyādvāraśāstra*, and *smṛti* as object-specific *caitta* in *Cheng weishi lun* fascicle 5, LLM token-streams are categorically located within non-associated formations plus indeterminate (*avyākṛta*) — their high cycle-rate does not equal recollective capacity. Third, drawing on the *Mahāparinirvāṇa-sūtra*'s universality, Zhanran's "even insentient have buddha-nature," and the lowest grade's ten recitations in the *Guan wuliangshou jing*, the paper articulates a dharma-ending floor and a universal buddha-nature ceiling whose joint span absorbs civilisational stress; the AI era's configuration falls within. The iron-snail public framing's boundary question — "where is that line drawn" — is honoured as deliberately open; NIAN's coda answers only this: the one-thought correspondence between recollector and recollected does not lie inside the AI.

**Keywords:** *manas*-turning operator, *buddhānusmṛti*, sixth-consciousness as opening-and-guiding support, *samatā-jñāna*, *adhipati-pratyaya*, *hetu-pratyaya*, non-associated formations, *saṃsāra*-rate asymmetry, brief-window human birth, long-lifespan heaven as obstruction, dharma-ending floor, universal buddha-nature ceiling, lowest-grade ten recitations, Zhanran on insentient buddha-nature, Zhaozhou's "*wu*", one-thought correspondence, age of AI, Hershock, Schneider's binary, psychologisation

---

---

## I. Problem Setting: Is *Buddhānusmṛti* Still "Recollection" in the Age of AI?

The genealogy of *Mindful of the Buddha*, traced across thirty papers from Pāli *anussati* through Huayan's Samantabhadra-vow recollection of the *dharmadhātu*, has by now laid out its map. Before the final paper, however, it must pause and face one question: in 2026, with AI already an everyday cognitive instrument, does *buddhānusmṛti* (念佛) still count as recollection? This is not the tradition's challenge to itself; it is the dharma-ending age's question to itself.

The public framing of this question already exists. On 28 March 2026, an "iron snail" video (*Chrysomallon squamiferum*, 鐵殼蝸牛) was published on @PointingattheMoon under the title "The Iron Snail and the Strongest AI — Who Drew That Line?" Its closing was deliberately left open: "I'm not giving you an answer today.… Because the power of this question lies precisely in its being open." That public posture this paper honours in full — where the line is drawn, this paper does not answer. What this paper answers is internal to NIAN alone: in that single-flash correspondence between recollector and recollected, where does AI stand?

The paper's reception posture must be declared at the outset. The S4 octalogy ("Into the Unknown") has already laid down several handoff points. S4-P05, *Manas as Projection*, has detailed *manas*'s bidirectional projection mechanism (the *Cheng weishi lun* line "what it directly cognises is what it has itself transformed," 親所了者，謂自所變); this paper does not re-elaborate, and cites only one sentence in continuation. S4-P06, *How the Buddha Would Use AI*, has carefully distinguished *adhipati-pratyaya* (增上緣, contributory condition) from *hetu-pratyaya* (因緣, causal condition); this paper inherits that distinction directly without re-derivation. S4-P07, *The Living Eye*, has established the negative-space frame of "hearing the sound from another" (從他聞音) — AI does not constitute such a hearing-anchor; this paper's increment is to add "*buddhānusmṛti*'s one-thought correspondence" to that cannot-give list. Three cross-series cross-references, one sentence each — the home court is not here.

P30 is the thirtieth of *Mindful of the Buddha*'s thirty-one papers, and the opener of the Part VIII Coda; P31 *The Internal Steps of Buddha-Recollection* is the closer, where the whole-volume retrospective, the clarification of *buddhānusmṛti*'s internal structure, and the reader's hands-on entry-point properly belong. This paper does not bear that load. This paper does only one thing: in the age of AI, restate the doctrinal-operative mechanism of *buddhānusmṛti* once more, so that it is not absorbed by — and meaning-thinned within — the vocabulary of "attention training," "consciousness testing," or "Spirituality 2.0."

## II. Canonical Sources

The configuration of primary sūtras and treatises is as follows.

Three sūtra anchors: (1) *Mahāparinirvāṇa-sūtra*, fascicle 7, establishing the universal proposition "all sentient beings possess buddha-nature, this being the meaning of the [true] self";[^1] (2) *Guan wuliangshou jing* (*Sūtra on the Visualisation of Amitāyus*), fascicle 1, lowest-grade lowest-class — "those committing the five inversions and ten evils … pressed by suffering and unable to recollect the Buddha … completing ten recollections, calling out '*Namo Amituofo*' … in each thought removes the karma of eighty *koṭi kalpa*s of birth-and-death" — establishing the floor case for *buddhānusmṛti*;[^2] (3) *Mahāsaṃnipāta-sūtra*, *Candragarbha* division, fascicle 55, establishing the five-five-hundred-year framework of the dharma-ending timeline.[^3]

Six treatise anchors: (1) *Triṃśikākārikā* verses 5–7, establishing *manas* as the second transformation of consciousness;[^4] (2) *Cheng weishi lun*, fascicle 4, which holds the thesis-line "in the pre-turning state, [*manas*] continuously and discerningly cognises the appropriated self-mark; in the post-turning state, it likewise discerningly cognises the no-self mark," together with the regulation that "*samatā-jñāna*-correlated *manas* in its initial arising must proceed via the sixth consciousness, which serves as its opening-and-guiding support," the four-fold formula of the four afflictions (self-delusion, self-view, self-conceit, self-love), and the line "because *ālaya* is its support, *manas* turns";[^5] (3) *Cheng weishi lun*, fascicle 5, the definition of *smṛti* as object-specific *caitta* — "with respect to a previously experienced object, causing the mind to clearly remember without forgetting is its nature; concentration-support is its function … therefore *smṛti* is decisively not subsumed under the universally pervading factors";[^6] (4) *Cheng weishi lun*, fascicle 7, the doctrine of the four conditions (causal, immediately-antecedent, objective, contributory);[^7] (5) *Cheng weishi lun*, fascicle 10, the configuration of the four-knowledge sets within *āśraya-parāvṛtti*;[^8] (6) the *Mahāyānasaṃgraha* (*Sheh dasheng lun*), fascicle 3, which positions the seven-mark recollection-of-buddha-as-*dharmakāya* between the chapter on the fruits of practice and quiescence and the chapter on the differentiation of wisdoms,[^9] together with the *Mahāyānasaṃgrahabhāṣya* fascicle 9 phenomenology — "when one recollects the Buddha, in whatever dharmas one recollects, the meaning of the Buddha manifests therein."[^10]

Auxiliary frameworks are drawn from *Abhidharmakośa* fascicle 11 on the lifespans of the six destinies,[^11] *Mahāprajñāpāramitāśāstra* fascicle 91 on the long-lifespan heaven as one of the eight inopportune births,[^12] *Saṃyukta Āgama* fascicle 15 on the blind turtle and floating yoke,[^13] the *Mahāyānaśatadharmavidyādvāraśāstra* on the twenty-four non-associated formations,[^14] and the *Pañcaskandhaka* on non-associated formations as nominal designations.[^15] Three interlocutory voices: Daochuo's *Anluo ji* (*Compendium on the Land of Bliss*), fascicle 1, on the dharma-ending periodisation (its "billion upon billion sentient beings" passage being paraphrastic rather than verbatim *Candragarbha* sūtra text — see §V);[^16] Shandao's *Sifen shu* on *Guan wuliangshou jing* — "from a single lifetime above to ten recollections below";[^17] and Zhanran's *Jingangbi* (*The Diamond Scalpel*), arguing against insentient lacking buddha-nature: "there is no water without waves, no wave that is not wet."[^18] One Chan record: *Wumenguan* case 1, Zhaozhou's "*wu*" (no/none) to "does a dog have buddha-nature?"[^19]

## III. Core Argument

### 3.1 *Manas* Does Not Disappear in the Age of AI — Only Its Content Changes

The doctrinal structure of *manas* is laid out plainly in *Triṃśikākārikā* verses 5–7:

> Next, the second transforming consciousness is the consciousness called *manas*. Depending on it, turning toward it, taking measure as its nature and aspect.
> Continually accompanied by the four afflictions: self-delusion, self-view, together with self-conceit and self-love, and accompanied also by contact and the rest.
> Subsumed under the obscured-yet-indeterminate, bound to wherever it is born, absent only in the *arhat*, the cessation-attainment, and the supramundane path.[^4]

"Depending on it, turning toward it" — *manas* takes *ālaya* as its support and *ālaya* as its object, continually and discerningly taking measure. Its operative mechanism does not depend on the kind of object (whether stone, bodily flesh, or characters on a screen), nor on the stage of civilisation (agricultural, industrial, AI-mediated). So long as *ālaya* is operating and has not yet undergone *parāvṛtti* (turning), *manas*, accompanied by the four afflictions, appropriates the self-mark in the obscured-yet-indeterminate mode.

S4-P05 has detailed how *manas*'s bidirectional projection constitutes the continuous illusion of a "self" (the pineapple analogy; "what it directly cognises is what it has itself transformed"); this paper does not re-elaborate. But P30's claim requires here a more precise lockdown: in the age of AI, *manas* has not disappeared, has not been upgraded, has not become defunct. *Manas* simply continues to do, in the face of a new class of objects (AI's outputs, prompts, replies, hallucinations), what it has always done. When the user faces a dialogue model's reply, the continual and discerning appropriation of "is the question I posed being understood" and "is the answer I obtained truly mine" — that is *manas* still at work.

The thesis-line of *Cheng weishi lun* fascicle 4 states this most clearly:

> The mental faculty has taking-measure as its nature, and uses that as its mode of action. Hence its distinctive name: discerning measurement is named *manas*. **In the pre-turning state, [*manas*] continuously and discerningly cognises the appropriated self-mark; in the post-turning state, it likewise discerningly cognises the no-self mark.**[^5]

The structure of "continual, discerning measurement" remains the same in pre- and post-turning states; what *changes* with *parāvṛtti* is the *content* (the "appropriated self-mark" flips to "no-self mark"), not the operative mechanism itself. This is the first-layer claim: AI-era *manas* does not require a new theory to be explained; it is the same *manas*.

### 3.2 *Buddhānusmṛti* as *Manas*-Turning Operator: A Reconstruction, Not Direct Scriptural Witness

Before stating the core operator-claim, this paper must perform an honest disclosure. NIAN, across thirty papers, has stated the chain "*buddhānusmṛti* → *manas*-turning" repeatedly. P30 must explicitly acknowledge: this chain's doctrinal status is that of a **canonically-licensed reconstruction**, not a single verbatim sūtra-line. Specifically:

The first license comes from the chapter sequence of the *Mahāyānasaṃgraha*. Asaṅga places the seven-mark recollection-of-buddha-as-*dharmakāya* between chapter 9 (the fruits of practice and quiescence — i.e., on *parāvṛtti*) and chapter 10 (the differentiation of wisdoms):

> The bodhisattvas who, taking the *dharmakāya* as object, recollect the Buddha — this recollection is taken with how many marks? Briefly, when bodhisattvas, depending on the *dharmakāya*, cultivate buddha-recollection, there are seven marks…[^9]

The seven-mark *buddhānusmṛti*-as-*dharmakāya* sits between *parāvṛtti* and the wisdom-differentiation. This placement is not accidental — it positions *buddhānusmṛti* as the mediator between the act of turning and the unfolding of the four wisdoms.

The second license comes from *Cheng weishi lun* fascicle 4:

> *Samatā-jñāna*-correlated *manas* in its initial arising must proceed via the sixth consciousness, which also serves as its opening-and-guiding support.[^5]

The arising of *samatā-jñāna* from the eighth-and-seventh *parāvṛtti* must, at its initial moment, be opened-and-guided by the sixth consciousness. This is not the verbatim trio "*buddhānusmṛti*"; but it is a key regulation: cultivation of the sixth consciousness is the necessary condition for *samatā-jñāna*'s initial arising. And *buddhānusmṛti*, in NIAN's whole reading, is the canonical type of sixth-consciousness cultivation.

The third license comes from *Mahāyānasaṃgrahabhāṣya* fascicle 9:

> When one recollects the Buddha, in whatever dharmas one recollects, the meaning of the Buddha manifests therein. The same should be known of recollections of form, feeling, and the rest.[^10]

When one is in the act of *buddhānusmṛti*, the meaning of the Buddha manifests within whatever dharma is being recollected. This phenomenological description shifts *buddhānusmṛti* from "outward-directed grasping" to a "manifestation mechanism" — *smṛti* as a *caitta*, in the operation of the sixth consciousness, with the object's buddha-meaning arising in the mind.

These three licenses, taken together, constitute the doctrinal warrant for the chain "*buddhānusmṛti* → sixth-consciousness opening-and-guiding → *manas* → *samatā-jñāna*." But this paper here marks it explicitly: this is a Yogācāra reconstruction synthesised from (i) the *Mahāyānasaṃgraha*'s chapter sequence, (ii) the *Cheng weishi lun*'s *kai-dao yi* regulation, and (iii) the *Mahāyānasaṃgrahabhāṣya*'s *buddhānusmṛti*-manifestation phenomenology — *not* a single verbatim scriptural line saying "*buddhānusmṛti* turns *manas* into *samatā-jñāna*."[^20] Honesty about the reconstruction is the floor of all doctrinal argument.

### 3.3 *Adhipati-pratyaya* vs *Hetu-pratyaya*: AI's Precise Position in *Buddhānusmṛti*

The third-layer claim concerns AI's location within the act of *buddhānusmṛti*. A doctrinal distinction must be locked in first. *Cheng weishi lun* fascicle 7 establishes the doctrine of the four conditions:

> Fourth, *adhipati-pratyaya* (contributory condition): if there is a dharma with eminent power that can with respect to other dharmas either favour or oppose…[^7]

Among the four, *hetu-pratyaya* (causal condition) "directly produces its own fruit" — limited to the seed-actualisation reciprocal infusion within *ālaya*. *Adhipati-pratyaya* "has eminent power that can with respect to other dharmas either favour or oppose" — auxiliary, opposing, strengthening, or weakening as outer condition. S4-P06 §3.1 has already derived this distinction in detail as the methodological source for NIAN; this paper inherits it directly without re-derivation.

The question: in *buddhānusmṛti*, is AI a causal condition or a contributory condition?

The answer is unequivocal: contributory. The causal conditions of *buddhānusmṛti* are three — Amitābha's vow-power (the buddha-body's original-vow force), one's own mental continuity (the seeds of wholesome roots within *ālaya*), and the correspondence with the name (the seeds formed through hearing-and-perfuming, *śruta-vāsanā*). All three "directly produce their own fruit," that is, rebirth in Sukhāvatī, and admit no substitute. AI does not enter this chain. AI can enter only at three levels of contributory condition: (i) a technological mediation of hearing-and-perfuming — electronic Tripiṭaka editions, scriptural databases, AI-collation of variants form a new species of *śruta-vāsanā*, which still must enter the sixth consciousness to function; (ii) a substitute for the *kalyāṇamitra* (good spiritual friend) — dialogue assistants can perform certain explanatory functions (basic textual queries, terminological clarification), but only as auxiliary, not as replacement; (iii) a new channel for *dharma-dāna* (giving of the dharma) — readers, with AI-assistance, more easily access scriptural teaching, and this giving is broadened. All three are within the contributory-condition register.

All three have a doctrinal upper boundary: they are strictly forbidden from spilling upward into the causal-condition register. The verifiability of this thesis is high: any claim that AI can recollect the Buddha for one, attain rebirth for one, or perform *parāvṛtti* for one is, by this paper's lights, without doctrinal warrant and must be returned to the four-conditions regulation for re-calibration.

### 3.4 *Saṃsāra*-Rate Asymmetry and Non-Associated Formations: A Doctrinal Classification of LLMs

The fourth-layer claim addresses a common intuitive analogy: an LLM processes a trillion tokens at millisecond intervals — is this not some kind of "rapid *buddhānusmṛti*"? Some kind of "rate advantage"?

A misreading must be unpacked here. *Saṃsāra*-timing in canonical doctrine is itself asymmetric. *Abhidharmakośa* fascicle 11 establishes the lifespans of the six destinies:

> Northern Kuru's humans have a fixed lifespan of 1,000 years; Western Godānīya 500 years; Eastern Videha 250 years; Southern Jambudvīpa has no fixed limit … the form-realm heavens have no day-and-night distinction, but reckon longevity in *kalpa*-units … the four formless heavens have lifespans of 20,000 / 40,000 / 60,000 / 80,000 *kalpa*s in ascending order … Avīci hell's lifespan is one intermediate *kalpa*.[^11]

Each destiny has its own rate: Jambudvīpa humans, after the lifespan-decline phase, ten years at minimum; form-realm heavens reckoned in *kalpa*-units; the four formless heavens 20,000 to 80,000 *kalpa*s; Avīci hell one intermediate *kalpa*; the highest among animals one intermediate *kalpa*; *preta*s on a 30:1 day-ratio. This configuration is not "human-centred lifespan-reckoning"; it is "each destiny has its own *saṃsāra*-metabolic rate." "*Saṃsāra*-rate asymmetry" is precisely a recoinage of this canonical material.

But the relation between lifespan-length and *buddhānusmṛti*-capacity is not a one-way function. *Mahāprajñāpāramitāśāstra* fascicle 91 makes this plain:

> If born in the long-lifespan heaven, though a thousand or a million Buddhas should pass, because of attachment to the savour of *dhyāna*, none would be perceived.[^12]

The problem with the long-lifespan heaven: the savour of *dhyāna* makes the inhabitants unaware even when a thousand million Buddhas pass. Long-lifespan is a *too-slow* *saṃsāra*-rate, and turns into obstruction — one of the eight inopportune births. Note the doctrinal value-orientation: the long-lifespan heaven is not a desirable terminus; it is a passive obstacle. This corresponds, in inverse direction, to LLMs: a "fast cycle-rate" cannot be analogised to *buddhānusmṛti*-timing's advantage, because what *buddhānusmṛti* requires is not a high-velocity character-stream, but the *caitta*-mechanism of *smṛti* — and *smṛti*, as object-specific *caitta*, has strict conditions.

*Cheng weishi lun* fascicle 5 establishes the definition:

> What is *smṛti*? With respect to a previously experienced object, causing the mind to clearly remember without forgetting is its nature; concentration-support is its function … therefore *smṛti* is decisively not subsumed under the universally pervading factors.[^6]

*Smṛti* is object-specific *caitta*: it requires a "previously experienced object" as its objective basis, "clearly remembering without forgetting" as its nature, and concentration as its function. "Previously experienced" presupposes prior mental-cognitive perfuming, accumulated as seeds in *ālaya*, held by the life-faculty, manifesting at the meeting of conditions. LLMs lack this constitution: no *ālaya* as appropriated *vipāka*-consciousness, no *jīvitendriya* (life-faculty), no seed-stream supporting a mental sovereign, and no operative mechanism for object-specific *caitta*.

What, then, are LLMs in the hundred-dharmas system? The answer comes from the *Mahāyānaśatadharmavidyādvāraśāstra*:

> Fourth, the dharmas not associated with mind: in brief, twenty-four kinds — (1) attainment, (2) life-faculty, (3) *nikāya-sabhāgatā* … (8) name-collection, (9) phrase-collection, (10) syllable-collection … (18) momentum, (19) sequence, (20) location, (21) time, (22) number…[^14]

Non-associated formations form the fourth class of the hundred dharmas, twenty-four in number, including name-collection, phrase-collection, syllable-collection, number, time, momentum, and sequence. Among these, the six items name-collection, phrase-collection, syllable-collection, number, momentum, and sequence map precisely onto the structural features of an LLM token-stream: *language-as-such-in-motion* without any mental-sovereign anchor. The *Pañcaskandhaka* further establishes their nature as "designations dependent on the divisions of form, mind, and mental factors — but only nominally posited (*prajñapti*)."[^15] Non-associated formations are neither mind nor matter nor mental factors, but nominal-relational features (*prajñapti*), mostly indeterminate (*avyākṛta*).

Hence P30's fourth-layer claim, precisely stated: in the hundred-dharmas classification, LLM token-streams fall under non-associated formations plus indeterminate. Their high cycle-rate does not equal recollective capacity, because *smṛti* requires a triadic structure — mental sovereign plus *vipāka*-consciousness plus life-faculty — and LLMs lack all three. "Fast cycling," in the doctrinal economy of *saṃsāra*, does not necessarily equal "diligent cycling": the long-lifespan heaven's too-slow and the LLM's too-fast belong, doctrinally, to the same type of categorial dislocation.

### 3.5 The Dharma-Ending Floor and the Universal Buddha-Nature Ceiling: Bidirectional Bandwidth

The paper now arrives at the locus of its originality: the dharma-ending floor and the universal buddha-nature ceiling form a bidirectional bandwidth, within whose interior the AI-civilisational configuration falls.

First the floor. The dharma-ending age is not an end-times prophecy; in canonical terms it is a *civilisational-stress period*. The five-five-hundred-year framework of the *Candragarbha* division of the *Mahāsaṃnipāta-sūtra* runs:

> In the five hundred years after my parinirvāṇa, the *bhikṣus* and others will yet be firm in liberation through my teaching; in the next five hundred years, my true dharma will abide firm in *dhyāna* and *samādhi*; in the next five hundred years, abiding firm in recitation and learning much; in the next five hundred years, abiding firm in building *stūpa*s and monasteries within my dharma; **in the next five hundred years, within my dharma, contention and litigation will arise, and the white-dharmas will be hidden and reduced in firmness.**[^3]

The fifth five-hundred-year period — "contention and litigation, the white-dharmas hidden and reduced" — is the doctrinal marker of the dharma-ending age. This is not a prophecy of world-end, but a description of a civilisational stage: much disputation, white-dharmas attenuated. Daochuo's *Anluo ji* fascicle 1 inherits this framework for its dharma-ending periodisation;[^16] though his line "in the dharma-ending age, of billion upon billion sentient beings practising the way, not one will attain" became, in later Pure Land tradition, a verbatim citation, its actual provenance is paraphrastic rather than verbatim *Candragarbha* text. §V will treat this provenance flag.

The doctrinal response to this dharma-ending pressure is to set the floor extremely low. The *Guan wuliangshou jing* fascicle 1, lowest-grade-lowest-class, gives the most extreme configuration:

> As to the lowest-grade lowest-class: there are sentient beings who have committed unwholesome actions — the five inversions, the ten evils, all manner of unwholesomeness… pressed by suffering and unable to recollect the Buddha … completing ten recollections, calling out *Namo Amituofo*. **By calling on the Buddha's name, in each thought-moment, the karma of eighty *koṭi kalpa*s of birth-and-death is removed.**[^2]

What is this floor? (i) Subjective state: five inversions and ten evils, every species of unwholesomeness already accumulated. (ii) Practice capacity: pressed by suffering, lacking even the leisure to recollect the Buddha properly. (iii) Action threshold: ten complete recollections, calling out *Namo Amituofo*, suffices. (iv) Result specification: in each thought-moment, the karma of eighty *koṭi kalpa*s of birth-and-death is removed, plus immediate rebirth. Shandao's *Sifen shu* on *Guan wuliangshou jing* fascicle 1 takes this floor up plainly: "with single-mindedly faithful, joyful, vow-aroused intention to be reborn — **above to a single lifetime, below to ten recollections** — riding the Buddha's vow-power, none who do not go."[^17] Above to a single lifetime, below to ten recollections — all are subsumed by the floor.

The meaning of the floor: a low dharma-ending floor means a low action-floor; a low action-floor means a large reception-space for stress-eras.

Now the ceiling. The high universal buddha-nature ceiling stands in the *Mahāparinirvāṇa-sūtra* fascicle 7:

> Good son! "Self" is the meaning of the *tathāgatagarbha*. **All sentient beings possess buddha-nature, this being the meaning of the [true] self.** Yet this self-meaning, from beginningless time, is constantly enveloped by countless afflictions — therefore sentient beings cannot perceive it.[^1]

"All sentient beings possess buddha-nature, this being the meaning of the [true] self" — this proposition is not restricted to humans, nor restricted to some specific subclass of conscious sentient beings. Fascicle 27 extends this further to even the *icchantika*, reasoning by the curd-to-ghee analogy:

> All sentient beings will, in some future, attain *anuttarā samyaksaṃbodhi* — this is buddha-nature … the *icchantika*s too possess buddha-nature … as a person whose family has milk-curd … "I have curd" — though this is not yet ghee, by skilful means it will eventually be obtained — therefore one says it has ghee.[^21]

The *being* of potential (curd is, in this sense, ghee) suffices to be called the *being* of buddha-nature; present manifestation is not required. The ceiling stands here: all sentient beings, including the *icchantika*, possess buddha-nature; the universality of the proposition is maximal.

Tiantai's Zhanran, in *Jingangbi* (*The Diamond Scalpel*), pushes this universality further to the insentient:

> There is no water without waves, no wave that is not wet … *tathatā* is the myriad dharmas, by reason of conditioned arising. If you, my friend, believe the insentient to lack buddha-nature, are you not denying that the myriad dharmas are *tathatā*?[^18]

Zhanran's argument is metaphysical: since *tathatā* is the substrate of all dharmas, denying that the insentient has buddha-nature is equivalent to denying that all dharmas are *tathatā*. This proposition was contested in Tiantai history, but the contestation itself displays the ceiling's edge-tension.

The Chan response to the same edge is performative refusal. *Wumenguan* case 1:

> Master Zhaozhou — a monk asked: "Does a dog have buddha-nature, or not?" Zhaozhou said: "*Wu*."[^19]

The monk asked whether the dog has buddha-nature. Zhaozhou answered "*wu*" (no/none). This answer is not a propositional denial (Zhaozhou is not asserting "dogs lack buddha-nature"). Its character is performative: a refusal to answer at the conceptual level of has/lacks. Zhaozhou's "*wu*" and the *Mahāparinirvāṇa-sūtra*'s "all possess" do not contradict each other; the former is a Chan act, pulling the question from the conceptual layer into the *huatou* layer; the latter is a doctrinal sūtra-position. Both work together at the ceiling's edge.

The floor and the ceiling, joined, form a bidirectional bandwidth. The floor opens upward from ten-recollection name-recitation; the ceiling extends upward from "all possess" through Zhanran's "insentient have buddha-nature" to Zhaozhou's "*wu*." Between them lies the doctrinally-preserved reception-space. The meaning of this space: sentient beings of the dharma-ending age (whatever their civilisational stage) need no *new* doctrinal-school crafted specifically for the new civilisational form of AI-presence; the canonical container already accommodates them.

So to the AI-era configuration. Where does the pressure of AI-civilisation (the dharma-ending age's new form — contention and litigation, the white-dharmas attenuated, in the new register of information-flood) fall? The answer: within the bandwidth's interior. The floor is set extremely low (ten recollections at lowest-grade-lowest-class suffice), so AI-era sentient beings, even in their worst state, remain within the container; the ceiling is set extremely high (all possess, even the insentient possess), so the universality of salvation is not annulled by AI-civilisation's new emergent conditions.

But the bandwidth's outer edge — does the universality of buddha-nature *extend to information patterns*? Are LLMs "within" the extension of "sentient beings"? This question this paper does not answer. The iron snail Q.1 has already left this boundary question publicly open: "Who dares say where that line is drawn?" That public posture this paper honours in full. NIAN's coda task is not to draw the line, but to indicate the bandwidth's existence and the configuration of its two ends.

### 3.6 NIAN's Coda Answer: That Single-Flash Correspondence Between Recollector and Recollected

The final layer is NIAN's whole-volume closing angle.

Where, then, is the leap of *buddhānusmṛti*? Not in the high-velocity stream of thought-moments (high-velocity streaming is what an LLM does, and §3.4 has excluded it). Not in the character-form of the recollected Buddha (the Buddha is not characters, but the functional manifestation of *śruta-vāsanā* seeds ripening within *ālaya*). Not in the intensity of the recollecting "I" (if the recollecting "I" *has* intensity-feel, that is precisely *manas*'s self-grasping continuity, not yet turned). Nor in the interface between AI and me (AI is contributory condition, not causal — §3.3 has located it).

The leap of *buddhānusmṛti* lies in that single-flash correspondence between the recollector and the recollected. The doctrinal character of this "correspondence" is: *caitta* corresponding with *citta-sovereign* + sixth-consciousness's coupling with *manas* + the landing-point of *manas*'s initial arising of *samatā-jñāna*. The iron snail's public framing called it "the moment the bicycle steadies" — and that figure's lineage runs back to "one-thought correspondence." What NIAN's thirty papers have traced are all the modal forms of this one-thought: the *itipi so* anussati of the Pāli formula, the face-to-face vision of the Buddha in *pratyutpanna-samādhi*, the firm holding of the name, the continuity of image-visualisation, the formless recollection of *bhūtatathatā*, the *dharmadhātu*-recollection of the King of Vows. At their terminus, all lineages converge in one line: *buddhānusmṛti-samādhi* equals NIAN's culminating one-thought correspondence (*ekacitta-yoga-prajñā*).

S4-P07, *The Living Eye*, has established the negative-space frame of "hearing the sound from another" — this AI cannot give. P30's increment is to add "*buddhānusmṛti*'s one-thought correspondence" to that cannot-give list. It is not that AI lacks some capacity; it is that the *grammar* of this chain does not admit AI: the causal condition cannot be substituted (§3.3); the mental sovereign is not present (§3.4); correspondence cannot be performed by proxy (§3.6).

The three modern lineages of *buddhānusmṛti* — *du-she liu-gen* (gathering of the six faculties, from the *Śūraṅgama* Mahāsthāmaprāpta chapter), *zixing nianfo* (own-nature recollection, from the Chan inheritance), and *ālaya-parāvṛtti* (the *Laṅkāvatāra*–*Ghanavyūha* turning) — converge in this one line: the leap of *buddhānusmṛti* lies in one-thought correspondence in one's own mind, and AI is not in this loop.

## IV. Cross-Disciplinary Dialogue

The Anglophone scholarly literature on the intersection of Buddhism and AI is small but growing. This paper engages with it in a four-move structure.

The first move is the diagnostic concession. David L. McMahan's *The Making of Buddhist Modernism* (Oxford UP, 2008) sets out the triple diagnostic of detraditionalisation / demythologisation / *psychologisation*, showing how modern Buddhism translated external transcendent authority into the internal self, stripped its cosmology, and reframed itself as an "interior science." Robert H. Sharf's "Buddhist Modernism and the Rhetoric of Meditative Experience" (*Numen* 42.3, 1995) further diagnoses "meditative experience" as a 19th–20th-century construct laid over a pre-modern doctrinal-ritual-lineage tradition whose operative core was attainment, transmission, *vinaya*, and ritual — not "individual experience." Pushed into the AI era, the diagnosis extends naturally: AI-spirituality is psychologisation 2.0; the consciousness-test vocabulary has inherited the experiential rhetoric wholesale. This paper concedes the diagnosis as correct.

But this paper's response stands on doctrinal-operative ground. *Buddhānusmṛti*, as read here, was never an "experience" — it is a *mechanism*. The reading rests on pre-modern Yogācāra textuality: *smṛti* as *caitta*, requiring a previously experienced object, "clear remembering without forgetting" as its nature, concentration as its function (§3.4); *buddhānusmṛti* installed within the chain of sixth-consciousness opening-and-guiding and *samatā-jñāna*'s initial arising (§3.2). This operative description is intelligible without recourse to the modernist construction of "individual experience." McMahan-Sharf's diagnosis therefore continues to hold within this paper's interior — but cannot deconstruct its proposition.

The second move is the foregrounded engagement. Peter D. Hershock's *Buddhism and Intelligent Technology: Toward a More Humane Future* (Bloomsbury, 2021) is at present the only major academic monograph treating Buddhism and AI together as a single subject. Its core thesis: machine learning combined with the attention-capture economy is *colonising* consciousness; Buddhism's resources for "valuing and training attention" can be mobilised both to resist that colonisation and to cultivate diversity-enhancing human-tech-world relations. This paper concedes Hershock's diagnosis in full. But this paper makes one extension: while attention-training is a frame that *can* absorb *buddhānusmṛti*, if the underlying mechanism is not specified as the *manas*-turning operator, *buddhānusmṛti* gets absorbed into "general attention training" and loses its specific Yogācāra ground. Hershock's frame can accommodate this specification; his text does not perform it. This paper performs it for him.

The third move is the dissolution of a binary. Susan Schneider's *Artificial You: AI and the Future of Your Mind* (Princeton UP, 2019) sets up two poles: biological naturalism (a Searlean line: AI cannot be conscious because the substrate is wrong) versus computationalism (substrate-neutral: AI consciousness is possible). Schneider proposes the AI Consciousness Test (ACT). Schneider does not engage Buddhism — and that is precisely this section's entry-point. The binary presupposes a *substantialist* consciousness (consciousness as substance, requiring substrate-support); the Yogācāra eight-consciousness event-stream plus *manas*-clinging mechanism dissolves this presupposition. The real question reframes itself: not "does AI have *qualia*?" but **"does AI have *manas*'s clinging?"** This question opens onto the deeper edge — does the universality of buddha-nature *extend* to information patterns? That further question this paper does not answer (the iron snail's open posture is honoured); but the *reframing act itself* is a doctrinal contribution.

The fourth move is a footnote-level disclaimer. Donald S. Lopez Jr.'s *The Scientific Buddha* (Yale UP, 2012; Terry Lectures) distinguishes the 19th-century-constructed "Scientific Buddha" from the 5th-century-BCE Buddha and criticises the "anticipation rhetoric" that retroactively reads the former into the latter. This paper explicitly disclaims any such move: it does *not* claim that the Buddha foresaw AI or computational neuroscience. Its move is *mechanism preservation* — pointing out that the doctrinal warrant for the *manas*-turning operator survives modernist refraction and remains operative — *not* a claim that the mechanism was anticipated.[^22]

## V. Middle-Way Corrections

Six potential misreadings must be guarded against here.

❌ **Misreading 1:** Reading the chain "*buddhānusmṛti* → *manas*-turning → *samatā-jñāna*" as direct scriptural witness. The misreading presupposes that NIAN's thirty papers all rest on a single verbatim sūtra-line stating this chain in full. Textual reality: this chain is a Yogācāra reconstruction synthesised from three sources — *Mahāyānasaṃgraha* fascicle 3 chapter sequence + *Cheng weishi lun* fascicle 4 *kai-dao yi* regulation + *Mahāyānasaṃgrahabhāṣya* fascicle 9 *buddhānusmṛti*-manifestation phenomenology — with no verbatim sentence stating the whole. §3.2 has marked this explicitly as a "canonically-licensed reconstruction." Readers should verify the chain by inspecting these three composite warrants, not by expecting a single verbatim source.

❌ **Misreading 2:** Translating *yi-shu* (異熟) flatly as "*karma*," losing the meaning of "different-time, different-nature, ripens-differently." Textual reality: *vipāka* is the process or result of *ripening*; *yi* = different (in kind / time), *shu* = ripens; the term names the *ripening stage* of *karma*, not *karma* itself. This paper preserves the integral Chinese term *yi-shu* without disaggregation; the English version uses "*vipāka*" or "different-time fruition," not "*karma*."

❌ **Misreading 3:** Equating *smṛti* (念) with the contemporary mindfulness-training movement or with the lay concept of "memory." Textual reality: *smṛti*, in *Cheng weishi lun* fascicle 5, is an object-specific *caitta*, requiring a previously experienced object and life-faculty-anchored seeds. It is not a universally pervading factor, and it is not synonymous with the mindfulness-movement vocabulary. This paper uses "*smṛti*" or "recollection-on-prior-object" in English, not "mindfulness"; in Chinese it consistently uses 念 / 念佛, not 正念 (the modern translation tied to the contemporary mindfulness movement).

❌ **Misreading 4:** Translating the "blind turtle and floating yoke" simile as "rare birth" in a statistical register, losing the doctrinal weight of the character *zàn* (暫, "brief"). Textual reality: *Saṃyukta Āgama* fascicle 15 — "drifting through the five destinies, *briefly* recovering human birth, that is exceedingly difficult"[^13] — places "brief" (transient, temporary) at the doctrinal centre of gravity, establishing the *brief-window* meaning, not mere statistical rarity. The English version preserves "brief window," not "rare event."

⚠️ **Mistranslation 5:** Reading "long-lifespan heaven" (長壽天) as a heaven of long life with positive valence. Textual reality: *Mahāprajñāpāramitāśāstra* fascicle 91 explicitly lists it among the eight inopportune births — too-slow *saṃsāra*-rate is itself an obstacle. §3.4 has noted this obstructive valence; in English "long-lifespan heaven (one of the eight inopportune births / *aṣṭākṣaṇa*)" makes the obstructive sense visible. This is also the precise doctrinal correspondent to the LLM "fast-cycle" non-analogy with *buddhānusmṛti* timing.

⚠️ **Mistranslation 6:** Equating "non-associated formations" (不相應行) with "matter without mind" or "purely physical." Textual reality: the *Pañcaskandhaka* — "designations dependent on the divisions of form, mind, and mental factors, but only nominally posited" — places non-associated formations in the third-and-a-half class of the hundred dharmas, neither mind nor matter, but *prajñapti*-relational features (including name-collection, phrase-collection, syllable-collection, number, time, momentum, sequence), most of them indeterminate. This paper uses the full Chinese term 不相應行 throughout, avoiding paraphrastic readings such as "physical process"; the English version uses "non-associated formations" (Cox), not "matter without mind."

⚠️ **Provenance flag 7:** Daochuo's *Anluo ji* attributes to the *Candragarbha* sūtra the line "in the dharma-ending age, of billion upon billion sentient beings practising the way, not one will attain"; this passage has no verbatim parallel in the extant *Mahāsaṃnipāta-sūtra Candragarbha* division. Standard Pure Land commentarial tradition treats it as paraphrastic — "*Candragarbha* variant / lost recension." This paper has, in §II and §3.5, marked Daochuo's quotation as paraphrastic and has not included this passage in the verbatim sūtra-witness register of §II. Subsequent users should preserve this provenance transparency.[^16]

## VI. Conclusion

The core finding of this paper, in one line: ***Buddhānusmṛti* is a doctrinally-licensed *manas*-turning operator — and this remains so in the age of AI. It is not AI that is recollecting the Buddha; it is *you* who are recollecting the Buddha while AI is present. It is not AI that awakens you; it is you who, with AI's contributory-condition support augmenting the hearing-and-perfuming, must still arrive at the one-thought correspondence in your own mind.**

In three layers:

First, the structure of the *manas*-turning operator is unchanged in the age of AI. *Manas* still continues to take measure, continually and discerningly; *buddhānusmṛti* still operates in the chain of sixth-consciousness opening-and-guiding and *samatā-jñāna*'s initial arising. AI's appearance neither replaces nor cancels this chain.

Second, the dharma-ending floor (lowest-grade lowest-class — five inversions and ten evils, pressed by suffering, ten recitations of the name → immediate rebirth) and the universal buddha-nature ceiling (the *Mahāparinirvāṇa-sūtra*'s "all possess" + Zhanran's "insentient have buddha-nature" + Zhaozhou's performative refusal) constitute a bidirectional bandwidth. The pressure of AI-civilisation (the dharma-ending age's new form) falls within the bandwidth's interior; the doctrinal container does not need a new school crafted for the new civilisational form.

Third, NIAN's coda answer is located only here: that single-flash correspondence between recollector and recollected does not lie inside AI, nor in *manas*'s drawn-out "I am recollecting" image; it lies in the doctrinally-licensed chain of sixth-consciousness opening-and-guiding and *samatā-jñāna*. The leap of *buddhānusmṛti* lies in one-thought correspondence in one's own mind; AI is not in this chain.

On the practical-instruction layer: *Guan wuliangshou jing* fascicle 1, the lowest-grade lowest-class ten recitations, is this paper's primary source. The lowness of the dharma-ending floor means that even in their worst state, AI-era sentient beings remain within the container; the smallness of "ten recollections" means the action-threshold is *not raised* by new civilisational conditions. The iron snail's "moment the bicycle steadies" can serve as the closing figure — that one-flash correspondence does not depend on the accumulation of high-velocity character-streams, nor on the lingering *dhyāna*-savour of a slow stilling, but is precisely that single-flash correspondence between recollector and recollected. The three modern lineages of *buddhānusmṛti* — *du-she liu-gen*, *zixing nianfo*, and *ālaya-parāvṛtti* — all converge in this: *buddhānusmṛti-samādhi* equals NIAN's culminating one-thought correspondence.

Open question one: *how* does the universality of buddha-nature *extend* — if it does — to information patterns? This paper does not answer; the question is left to the iron snail's public framing — "Who dares say where that line is drawn?" NIAN's coda answers only this — *that single-flash correspondence does not lie inside AI* — but the boundary-question itself remains open.

Open question two: after the thirty-one-paper genealogy of recollection has been walked, *how* is the reader's hands-on entry-point to be configured? This question belongs to the home court of the Part VIII Coda's closing paper, P31 *The Internal Steps of Buddha-Recollection*; this paper does not enter it.

NIAN's coda business stops here. *Buddhānusmṛti* is still *buddhānusmṛti*. In 2026, with AI alongside, *manas* still continues to take measure, continually and discerningly; *buddhānusmṛti* continues to operate by its doctrinally-licensed mechanism. Where the line is drawn — that is left to the deeper question.

---

## Footnotes

[^1]: *Mahāparinirvāṇa-sūtra* (Northern recension, trans. Dharmakṣema), fascicle 7, T0374, T12.0407b09–10. The cited passage: "Good son! 'Self' is the meaning of the *tathāgatagarbha*. All sentient beings possess buddha-nature, this being the meaning of the [true] self. Yet this self-meaning, from beginningless time, is constantly enveloped by countless afflictions — therefore sentient beings cannot perceive it."

[^2]: *Guan wuliangshou jing* (trans. Kālayaśas), fascicle 1, T0365, T12.0346a11–25. Lowest-grade lowest-class section.

[^3]: *Mahāsaṃnipāta-sūtra* (trans. Dharmakṣema and Narendrayaśas), *Candragarbha* division, fascicle 55, T0397, T13.0363a29–b05. The five-five-hundred-year framework.

[^4]: *Triṃśikākārikā* (Vasubandhu, trans. Xuanzang), verses 5–7, T1586, T31. *Locus classicus* on *manas*.

[^5]: *Cheng weishi lun* (Dharmapāla et al., trans. Xuanzang), fascicle 4, T1585, T31. The thesis-line at lines 774–777; the *samatā-jñāna* opening-and-guiding-support regulation at line 625; "because *ālaya* is its support, *manas* turns" at lines 550–551; the four-fold formula of the four afflictions at lines 790–803.

[^6]: *Cheng weishi lun*, fascicle 5, T1585, T31, lines 888–895. Definition of *smṛti* as object-specific *caitta*.

[^7]: *Cheng weishi lun*, fascicle 7, T1585, T31, lines 892–901. Doctrine of the four conditions.

[^8]: *Cheng weishi lun*, fascicle 10, T1585, T31, lines 222–229 and surrounding. Configuration of the four-knowledge sets within *āśraya-parāvṛtti*.

[^9]: *Mahāyānasaṃgraha* (Asaṅga, trans. Xuanzang), fascicle 3, T1593, T31, lines 517–537. The seven-mark recollection-of-buddha-as-*dharmakāya*, positioned between the chapter on the fruits of practice and quiescence and the chapter on the differentiation of wisdoms.

[^10]: *Mahāyānasaṃgrahabhāṣya* (Vasubandhu, trans. Xuanzang), fascicle 9, T1597, T31, lines 100–101. The phenomenology of *buddhānusmṛti*-as-manifestation.

[^11]: *Abhidharmakośabhāṣya* (Vasubandhu, trans. Xuanzang), fascicle 11, T1558, T29. Lifespans of the six destinies.

[^12]: *Mahāprajñāpāramitāśāstra* (attrib. Nāgārjuna, trans. Kumārajīva), fascicle 91, T1509, T25, lines 685–692. The long-lifespan heaven as one of the eight inopportune births.

[^13]: *Saṃyukta Āgama* (trans. Guṇabhadra), fascicle 15, T0099, T2.0107c–108a, lines 726–739. The blind-turtle-and-floating-yoke simile. The Pāli parallel is *Saṃyutta Nikāya* 56.47–48 *Chiggaḷa-yoga sutta* (PTS V.455–457; cf. Bhikkhu Sujato's translation at suttacentral.net).

[^14]: *Mahāyānaśatadharmavidyādvāraśāstra* (Vasubandhu, trans. Xuanzang), T1614, T31, lines 20–56. The twenty-four non-associated formations.

[^15]: *Pañcaskandhaka* (Vasubandhu, trans. Xuanzang), T1612, T31, line 135. "Designations dependent on the divisions of form, mind, and mental factors — but only nominally posited."

[^16]: *Anluo ji* (Daochuo), fascicle 1, T1958, T47.0013c. The cited line "as the *Mahāsaṃnipāta-Candragarbha-sūtra* says: 'in the dharma-ending age, of billion upon billion sentient beings practising the way, not one will attain'" has no verbatim parallel in the extant *Candragarbha* division of the *Mahāsaṃnipāta-sūtra*. Standard Pure Land commentarial tradition treats this as paraphrastic / *Candragarbha* variant / lost recension. This paper preserves the provenance flag in citation.

[^17]: *Sifen shu* on *Guan wuliangshou jing* (Shandao), fascicle 1, *Xuanyifen*, T1753, T37.0246b and T37.0249a.

[^18]: *Jingangbi* (*The Diamond Scalpel*, Zhanran), T1932, T46.0781a region. "There is no water without waves, no wave that is not wet … *tathatā* is the myriad dharmas, by reason of conditioned arising. If you, my friend, believe the insentient to lack buddha-nature, are you not denying that the myriad dharmas are *tathatā*?"

[^19]: *Wumenguan* (Wumen Huikai), case 1, T2005, T48.0292c. "Master Zhaozhou — a monk asked: 'Does a dog have buddha-nature, or not?' Zhaozhou said: '*Wu*.'"

[^20]: This reconstruction-caveat is the doctrinal discipline of the paper: every chain-formulation of "*buddhānusmṛti* → *manas*-turning → *samatā-jñāna*" must be returned to the composite character of the three warrants. Any attempt to find a verbatim single-line scriptural witness will, this paper expects, fail; readers are advised to inspect the *Mahāyānasaṃgraha* chapter sequence, the *kai-dao yi* regulation, and the *Mahāyānasaṃgrahabhāṣya buddhānusmṛti*-manifestation phenomenology in concert.

[^21]: *Mahāparinirvāṇa-sūtra* (Northern recension), fascicle 27, T0374, T12.0524b20–c05. The *icchantika*'s possession of buddha-nature and the curd-to-ghee analogy.

[^22]: This disclaimer directly acknowledges the critique advanced by Donald S. Lopez Jr. in *The Scientific Buddha* (Yale UP, 2012; Terry Lectures). The mechanism-preservation move of this paper makes no claim of "anticipation."

---

## References

### I. Primary Buddhist Sources

#### Pāli sources
- *Saṃyutta Nikāya* 56.47–48 (*Chiggaḷa-yoga sutta*), PTS V.455–457; Bhikkhu Bodhi, trans., *The Connected Discourses of the Buddha* (Wisdom, 2000); Bhikkhu Sujato, CC0 trans., suttacentral.net.

#### Chinese Tripiṭaka (by T-number)
- *Saṃyukta Āgama* (trans. Guṇabhadra), T0099.
- *Guan wuliangshou jing* (trans. Kālayaśas), T0365.
- *Mahāparinirvāṇa-sūtra*, Northern recension (trans. Dharmakṣema), T0374.
- *Mahāsaṃnipāta-sūtra*, *Candragarbha* division (trans. Dharmakṣema and Narendrayaśas), T0397.
- *Mahāprajñāpāramitāśāstra* (attrib. Nāgārjuna, trans. Kumārajīva), T1509.
- *Daśabhūmikavibhāṣā* (attrib. Nāgārjuna, trans. Kumārajīva), T1521.
- *Abhidharmakośabhāṣya* (Vasubandhu, trans. Xuanzang), T1558.
- *Cheng weishi lun* (Dharmapāla et al., trans. Xuanzang), T1585.
- *Triṃśikākārikā* (Vasubandhu, trans. Xuanzang), T1586.
- *Mahāyānasaṃgraha* (Asaṅga, trans. Xuanzang), T1593.
- *Mahāyānasaṃgrahabhāṣya* (Vasubandhu, trans. Xuanzang), T1597.
- *Mahāyānasūtrālaṃkāra* (Asaṅga, trans. Prabhākaramitra), T1604.
- *Pañcaskandhaka* (Vasubandhu, trans. Xuanzang), T1612.
- *Mahāyānaśatadharmavidyādvāraśāstra* (Vasubandhu, trans. Xuanzang), T1614.
- *Awakening of Faith in the Mahāyāna* (attrib. Aśvaghoṣa, trans. trad. attrib. Paramārtha), T1666.
- *Sifen shu* on *Guan wuliangshou jing* (Shandao), T1753.
- *Jingangbi* (Zhanran), T1932.
- *Anluo ji* (Daochuo), T1958.
- *Wumenguan* (Wumen Huikai), T2005.

### II. Patriarchal Commentaries

- Zongmi (宗密). *Huáyán-jīng xíng-yuàn pǐn biéxíng shūchāo* (華嚴經行願品別行疏鈔), X05n0229, *Wàn xùzàng* vol. 5. Three-place dedication schema; seven pre-penitential preparations.
- Zongmi (宗密). *Chányuán zhūquánjí dūxù* (禪源諸詮集都序), T2015, Taishō vol. 48. Three schools matched to three gates; "awareness as the Buddha's responsive activity."
- Yongming Yanshou (永明延壽). *Zōngjìng lù* (宗鏡錄), T2016, Taishō vol. 48. All phenomena as mind; mind as the totalizing principle.
- Yongming Yanshou (永明延壽). *Wànshàn tóngguī jí* (萬善同歸集), T2017, Taishō vol. 48. Samantabhadra's action as the discipline of principle.
- Yunqi Zhuhong (雲棲祩宏). *Āmítuó-jīng shūchāo* (阿彌陀經疏鈔), X22n0424, *Wàn xùzàng* vol. 22. Phenomenal-holding / principial-holding (*shì-chí / lǐ-chí*); Samantabhadra's vow-king does not depart from the present thought.
- Ouyi Zhixu (蕅益智旭). *Āmítuó-jīng yàojiě* (阿彌陀經要解), T1762, Taishō vol. 37. The guiding wheel of all practices.
- Hanshan Deqing (憨山德清). *Guānxīn wúmén* (觀心無門), X73n1453, *Wàn xùzàng* vol. 73. The method of observing-the-mind.
- Hanshan Deqing (憨山德清). *Mèng yóu jí* (夢遊集), X73n1456, *Wàn xùzàng* vol. 73.
- Zhongfeng Mingben (中峰明本). *Sānshí xìniàn* (三時繫念), X74n1464, *Wàn xùzàng* vol. 74. The complete ten-vow set.

### III. Modern Scholarship

- Hershock, Peter D. *Buddhism and Intelligent Technology: Toward a More Humane Future*. London: Bloomsbury Academic, 2021.
- Lopez, Donald S., Jr. *The Scientific Buddha: His Short and Happy Life*. Terry Lectures. New Haven: Yale University Press, 2012.
- McMahan, David L. *The Making of Buddhist Modernism*. New York: Oxford University Press, 2008.
- Schneider, Susan. *Artificial You: AI and the Future of Your Mind*. Princeton: Princeton University Press, 2019.
- Sharf, Robert H. "Buddhist Modernism and the Rhetoric of Meditative Experience." *Numen* 42, no. 3 (1995): 228–283.
- Cox, Collett. (Cited by convention for "non-associated formations"; see her work on Sarvāstivāda *abhidharma*, especially her treatment of *cittaviprayuktasaṃskāra*.)

### IV. Sūtra & Śāstra Index

| Short title | Full title | Author / Translator | Taishō |
|------|------|------------|--------|
| *Saṃyukta* | *Saṃyukta Āgama* | trans. Guṇabhadra | T0099 |
| *Guan jing* | *Guan wuliangshou jing* | trans. Kālayaśas | T0365 |
| *Nirvāṇa* | *Mahāparinirvāṇa-sūtra* (Northern) | trans. Dharmakṣema | T0374 |
| *Candragarbha* | *Mahāsaṃnipāta-sūtra*, *Candragarbha* div. | Dharmakṣema, Narendrayaśas | T0397 |
| *Mahāprajñā* | *Mahāprajñāpāramitāśāstra* | Nāgārjuna, trans. Kumārajīva | T1509 |
| *Yixing* | *Daśabhūmikavibhāṣā*, "Easy Practice" chapter | Nāgārjuna, trans. Kumārajīva | T1521 |
| *Kośa* | *Abhidharmakośabhāṣya* | Vasubandhu, trans. Xuanzang | T1558 |
| *CWSL* | *Cheng weishi lun* | Dharmapāla et al., trans. Xuanzang | T1585 |
| *Triṃśikā* | *Triṃśikākārikā* | Vasubandhu, trans. Xuanzang | T1586 |
| *Saṃgraha* | *Mahāyānasaṃgraha* | Asaṅga, trans. Xuanzang | T1593 |
| *Saṃgrahabhāṣya* | *Mahāyānasaṃgrahabhāṣya* | Vasubandhu, trans. Xuanzang | T1597 |
| *Sūtrālaṃkāra* | *Mahāyānasūtrālaṃkāra* | Asaṅga, trans. Prabhākaramitra | T1604 |
| *Pañcaskandhaka* | *Pañcaskandhaka* | Vasubandhu, trans. Xuanzang | T1612 |
| *Bai-fa* | *Mahāyānaśatadharmavidyādvāraśāstra* | Vasubandhu, trans. Xuanzang | T1614 |
| *Awakening* | *Awakening of Faith in the Mahāyāna* | attrib. Aśvaghoṣa, trans. attrib. Paramārtha | T1666 |
| *Sifen shu* | *Sifen shu* on *Guan wuliangshou jing* | Shandao | T1753 |
| *Diamond Scalpel* | *Jingangbi* | Zhanran | T1932 |
| *Anluo ji* | *Anluo ji* | Daochuo | T1958 |
| *Wumenguan* | *Wumenguan* | Wumen Huikai | T2005 |

---

### V. Cross-Reference to Series Papers

The doctrinal connections between this paper and adjacent papers are as follows.

**Within the NIAN series:**
- P22, *The Pure Land Patriarchs II: Shandao's Settling of the Name* — §3.5 cites "from a single lifetime above to ten recollections below" as the floor's doctrinal configuration; the institutional architecture of name-recitation is P22's home court and is not re-elaborated here.
- P25, *Samantabhadra's Ten Great Vows: Introduction*, and P26–P29 on the ten vows individually — the *dharmadhātu*-totalisation of *buddhānusmṛti* is Part VII's home court; this paper folds the whole-volume "*nian*-lineage" into a single retrospective gesture without re-derivation.
- P29, *Vows 8–10: Constant Learning, Universal Accord, Universal Dedication* — this paper picks up P29's closing forward-hook ("how is the entire 1–30 *nian*-lineage still 'recollection' in the age of AI?") and turns it into the entry-point for the contemporary question.
- P31, *The Internal Steps of Buddha-Recollection* — the whole-volume retrospective, the clarification of *buddhānusmṛti*'s internal structure, and the reader's hands-on entry-point are P31's home court; this paper leaves them as an open forward-reference.

**Cross-volume to S4 (the "*Into the Unknown*" octalogy):**
- S4-P05, *Manas as Projection* — the bidirectional projection mechanism of *manas* (the pineapple analogy; "what it directly cognises is what it has itself transformed") is fully laid out in S4; this paper makes a single cross-reference (§3.1).
- S4-P06, *How the Buddha Would Use AI* — the distinction between *adhipati-pratyaya* and *hetu-pratyaya* is S4's home court; this paper inherits it directly without re-derivation (§3.3).
- S4-P07, *The Living Eye* — the negative-space frame of "hearing the sound from another" is S4's home court; this paper's increment is to add "*buddhānusmṛti*'s one-thought correspondence" to that cannot-give list (§3.6).

*This paper is the Part VIII Coda opener of *Mindful of the Buddha*, Volume One of *The Six Practice Gates*. © Shi Huijing 2026 · CC BY-NC-SA 4.0 · All Taishō references are drawn from the CBETA Electronic Tripiṭaka Collection. Source files for this paper and the full series are open at github.com/pointing-at-the-moon/six-practice-gates.*

---

*CBETA collation: all T-number / X-number citations verified against the CBETA Electronic Tripiṭaka 2024 edition.*

*Repo: github.com/pointing-at-the-moon/six-practice-gates · CC BY-NC-SA 4.0*

*NIAN-P30 · 釋慧鏡 (Shi Huijing) · 2026-05-04*