俱舍論 義疏通講 · 冊9Abhidharmakośabhāṣya · Exposition Throughout · Vol. 9

Before Reading This Chapter

This chapter is an anomaly within the whole treatise. The preceding eight chapters all have root stanzas, and all first set out the stanza and then explain it themselves in prose; when we come to this chapter the root stanzas are gone — the separate edition of the root stanzas does not include the Chapter on Refuting the Grasped Self at all, and the chapter is prose from beginning to end, and moreover it is disputation throughout: one objection follows another, one defense follows another, pressing forward in the rhythm of “again… again… yet…,” with never a transitional sentence in between of the sort “this having been explained, next it should be explained.” This is no oversight of form; it is a choice of the literary form itself: to refute a grasping is not to expound a body of doctrine for someone to hear, but to press the opponent’s words sentence by sentence until he can say no more.

Puguang’s Record states this chapter’s position most clearly — the whole treatise is meant to explain “all dharmas are without self,” one of the three seals of the dharma spoken by the Buddha, and “the preceding eight chapters clarify the facts of dharmas, and the last chapter clarifies the principle of no-self” — what the preceding eight chapters clarify is fact (which dharmas there are, in dependence on which conditions, completed into which results), and what this last chapter clarifies is principle (that within them there is no self). That is to say, the preceding eight chapters expounded a whole precise science of the marks of dharmas in order that this last sentence, “within them there is no self,” might stand firm. The finer the marks of dharmas are expounded, the less room the grasping of a self has to hide — this is the closing momentum of the whole treatise.

There is one thread in reading this chapter that must not be let go, and that is attribution: which sentence is put forward by the Vātsīputrīyas and which is the author’s challenge; which sentence is spoken by the Sāṃkhya non-Buddhists and which by the Vaiśeṣika non-Buddhists. The difficulty of this chapter is not that the wording is deep but that the exchanges are too dense — within a single passage there are often four or five rounds of “he posits · I challenge · he defends · I refute again,” and to mistake the speaker even once is to read the whole passage backwards. Among the three commentaries, Yuanhui’s Verse Commentary marks “the Vātsīputrīyas ask” and “the Vātsīputrīyas object” sentence by sentence in this passage, and is the most finely detailed; Puguang’s Record marks “first stating the tenet of the Vātsīputrīyas” to make clear that this is the Vātsīputrīyas’ own statement and not the author speaking on their behalf; Fabao’s Commentary marks “the Vātsīputrīyas cite a sūtra to object” and “the Vātsīputrīyas altogether deny that the sūtras cited above are the valid word of the noble ones.” The attribution of speakers in this chapter takes just these markings of the three commentaries as its standard.

The skeleton of the whole chapter, as adjudged by the three commentaries, is this — first the general refutation (blocking a self apart from the aggregates by the two means of knowledge, direct perception and inference), next the several refutations of the three parties (the Vātsīputrīyas · the Sāṃkhya masters · the Vaiśeṣika masters), and at the end three stanzas of seven characters to the line exhorting to study and handing the work on. And among the three parties, the Vātsīputrīya passage alone takes up well over three quarters of the chapter, because it is the hardest to refute: the self of the non-Buddhists is easy enough to refute; what is hard to refute is the party that grasps a self while being disciples of the Buddha.

I. The General Refutation: Both Roads, Direct Perception and Inference, Are Blocked

〔Checklist 1〕Why the refutation of the self begins from the “two means of knowledge.” First the question: this chapter opens by saying “if one relies on another beyond this, is there no liberation? In principle there certainly is none” — why should there be no liberation outside this body of teaching? The answer: because apart from this, what people rely on is all “the false grasping of a self”; by the force of that grasping the afflictions arise, and once the afflictions arise there is no term to the transmigration of the three existences. This binds liberation and no-self into one matter, not two.

Then how is it proved that the grasping of a self is false? The road the treatise takes is the road of cognition — it does not argue about the definition of “self” but asks: this self of yours, how could I possibly know it? Cognition has only two roads: direct perception (present and personal witness) and inference (knowledge derived from an attestation). If the entity of the self really were a further real thing, it ought to be like any other existent dharma: with no obstructing condition it should be obtained by direct perception, as the six object-fields are obtained by the mind; and failing that, it should be obtained by inference, as the five form-faculties are inferred.

The passage on “the inference of the five form-faculties” is worth close reading, because it exhibits what qualifies as sound inferential knowledge: the several conditions are all present and yet the result does not arise; lacking a certain condition the result does not arise, and not lacking it, it does — and from this one dropping-out and coming-in one determines that there is a further, invisible condition present. The object before one and attention are both complete, and yet the consciousness of the blind and the deaf does not arise while that of the not-blind and not-deaf does, and thereupon one determines that there are further faculties, the eye and the rest. This is a piece of reasoning that has a “difference to be attested.” Look now at the self apart from the aggregates: it is neither visible in the present nor is any difference of the sort “if it is there, so; if it is not there, not so” to be found — neither means of knowledge yields it, and the entity of the self accordingly cannot be established.

A place easily misread: taking the treatise to be saying that “whatever cannot be seen does not exist.” It is not so — the five form-faculties equally cannot be seen, and yet the treatise grants that they exist, because there is the attestation of inference. What the treatise blocks is not “the invisible” but “the unattestable.” Where this lands: hereafter, whenever some claim of the form “such and such really exists” is met with, first ask the one question, “what difference does it have to be attested by”; if no answer comes, that “existence” is only a name.

II. The Vātsīputrīyas State Their Tenet: The Four Words “Neither the Same nor Different”

〔Checklist 2 · first part〕What it is that the Vātsīputrīyas posit. The treatise gives it in a single sentence — “yet the Vātsīputrīyas grasp that there is a pudgala, whose entity is neither the same as the aggregates nor different from them.” Puguang’s Record marks here expressly “first stating the tenet of the Vātsīputrīyas,” explaining that this passage is the Vātsīputrīyas’ own statement, not the author speaking on their behalf; and he explains further the source of the school’s name: among the eighteen schools there is this one, because in the Buddha’s lifetime there was a non-Buddhist named Vatsa whose position this school’s grasping resembles, whence its label; and since these are disciples of the Buddha who yet perversely reckon a self, their fault is the more conspicuous, and so this school is refuted first among the three parties.

The purpose of those four words must be clearly seen. The Vātsīputrīyas are not stupid — they know that to say “the self differs from the aggregates” is to fall into the view of permanence (the aggregates cease and the self does not), and that to say “the self is the same as the aggregates” is to fall into the view of annihilation (the aggregates cease and the self ceases with them); so they take a middle road: neither the same nor different, and moreover inexpressible. In this way they avoid both the view of permanence and the view of annihilation refuted by the Buddha, while the subject of transmigration, the subject of recollection, and the bearer of the fruit of action all find a lodging.

A place easily misread: reading the Vātsīputrīyas as non-Buddhists. They are one of the eighteen schools of Buddhism — Fabao’s Commentary points out expressly that “among the several schools only the Vātsīputrīyas grasp that there is a real self,” precisely in order to remind the reader that this contest is a quarrel within the household, not a contest between insiders and outsiders. And just because they are of the household, refuting them must be done sentence by sentence and word by word, and cannot be waved away with “the deluded reckoning of non-Buddhists.” Where this lands: hereafter, on meeting the three words “cannot be expressed,” first ask the one question, “is it inexpressible because it surpasses speech, or because to express it is at once to show the flaw” — it is the latter that this chapter is taking apart.

III. The Question of Real Existence and Conventional Existence: The Simplest of Measures

〔Checklist 2 · second part〕The measure the author hands over first. Next there appears in the treatise a question — “this should be pondered and discriminated: is it real or conventional? What is the difference of the marks of real existence and conventional existence?” The answer given is utterly crisp: to have a separate thing is the mark of real existence, as with form and sound; to be merely an aggregation is the mark of conventional existence, as with milk and curds. Both ends of this measure are clear: what has an entity of its own is real; what is merely a heap of many things is conventional.

Then he closes in from both sides. To allow that it is real brings three faults: first, since its entity is separately existent, it ought to differ from the aggregates, as one aggregate differs from another; second, being a real entity it ought to have a cause producing it, and if it has none it is unconditioned, which is to walk with the view of the non-Buddhists; third, it is of no use whatever, and one has merely grasped a real existent for nothing. To allow that it is conventional “is then to say just what I say” — you and I do not differ, so what is there to argue about?

A place easily misread: taking this for a merely verbal dilemma. In fact this question is the pivot of the whole passage — had the Vātsīputrīyas dared to answer either horn, the debate would have ended on the spot; the reason they can hold out is precisely that they answer neither, and change their wording to “I only establish it in dependence on the aggregates.” And so the whole of the exchange that follows is the author pressing on that word “dependence.”

One honest record: Puguang’s Record marks this question as “the Vātsīputrīyas ask,” whereas read according to the flow of the prose it reads rather as the author challenging the Vātsīputrīyas — the two readings run in exactly opposite directions as to who asks and who answers. Both readings are recorded here, without forcing a change upon Puguang; which is right is left to be settled.

IV. The Simile of Fire in Dependence on Firewood (First Part): How the Simile Is Set Up

〔Checklist 3 · first part〕What is good about this simile of the Vātsīputrīyas. The author presses on with “what is called dependence,” and the Vātsīputrīyas answer handsomely — “this is like establishing fire in dependence on firewood in the world.” The simile is indeed forceful: apart from firewood no fire can indeed be established, and in the ordinary feel of language firewood and fire are indeed hard to call the same (firewood is wood, fire is flame) and hard to call different (apart from firewood there is no fire). It is just this feel of language that the Vātsīputrīyas want: since the world allows that firewood and fire are neither the same nor different, why should it not be allowed that the aggregates and the self are neither the same nor different?

What has to be taught clearly is the argumentative form behind this simile. The Vātsīputrīyas at once add two horns: “if fire differed from firewood, firewood ought not to be hot” — if the fire differed from the firewood, then the firewood ought not to be hot, and yet burning wood is plainly hot, and so they are not different; “if fire and firewood were the same, then what is burned would be what burns” — if the fire just were the firewood, then what is burned would be what burns, and the acting and the acted-upon would not be distinguished, and so they are not the same. Both ends are stopped, and the “neither the same nor different” in the middle then looks as though it must be posited.

A place easily misread: taking the simile for mere rhetoric. In this chapter the simile is a load-bearing wall of the argument — the Vātsīputrīyas build their whole tenet upon this one simile, and so the author too takes this one simile apart passage by passage. Where this lands: whenever someone establishes a tenet by a simile, first ask the one question, “the thing likened and the thing it is likened to — in exactly what respect are they alike”; if the point of likeness cannot be stated, that simile is only atmosphere.

V. The Simile of Fire in Dependence on Firewood (Second Part): The Author’s Six Layers of Analysis

〔Checklist 3 · second part〕How to take a simile apart to the very bottom. The author does not rebut the simile; he requires the opponent first to state the simile itself clearly — “which is the fire? which is the firewood?” This one question is the lever of the whole passage. The Vātsīputrīyas can only follow what the world commonly understands: what is not ablaze and is burned is the firewood; what has brightness, is exceedingly hot, and can burn is the fire. But once the naming is fixed in this way, the six layers of analysis fall in order.

At the first layer the author grants that the two alike have the eight entities for their entity, but points out that “it is in dependence on firewood that fire is able to arise” — as curds and vinegar arise in dependence on milk and wine, from which it is clear that the fire is after the firewood, and their times being separate, the fire just is different from the firewood. The second layer presses on from this: then your pudgala, if it really is like fire in dependence on firewood, must be said to arise in dependence on the aggregates, to be different in entity from the aggregates, and to become impermanent and not abiding — the very three sentences the Vātsīputrīyas are straining to avoid. At the third layer, if the Vātsīputrīyas change their wording to “on blazing wood the warm touch is called fire and the rest is called firewood,” then firewood and fire arise simultaneously and, their marks being different, are of different entities, and thereupon the word “dependence” falls empty again — for the fire does not take the firewood for its cause, each arising simultaneously from its own cause. At the fourth layer, if “dependence” be said to mean co-arising or resting upon, then that is to admit that the pudgala co-arises with the aggregates or rests upon them, and difference of entity is thereby plainly admitted; moreover, where there is no firewood there is no fire, so where there are no aggregates there should be no self, and this he does not allow. At the fifth layer, the objection he set up for himself, “firewood ought not to be hot,” is turned back upon him — if “hot” just is the warm touch, then the firewood indeed is not hot, its entity and mark being different; if “hot” means being conjoined with warmth, then what is of a different entity also gets to be called hot, and the firewood too may be called hot — neither side amounts to an objection. At the sixth layer, if it be said that when the wood is ablaze throughout, both names, firewood and fire, apply to it, then the pudgala and the aggregates are definitely the same, and there is no longer any principle by which this can be blocked.

A place easily misread: reading these six layers as six parallel rebuttals. They are not parallel; they are a staircase forcing a surrender — each layer first accepts the explanation the opponent has just given and then points out where that explanation carries itself. Where this lands: in taking a simile apart, do not deny that it is alike; ask “in exactly what respect is it alike,” and then see where that respect leads when it is pressed.

VI. The Dispute over Being Cognized by the Six Consciousnesses: Can What Is Inexpressible Be Cognized

〔Checklist 4〕How the author uses “consciousness” to press on “the inexpressible.” The Vātsīputrīyas say that the pudgala is cognized by the six consciousnesses — when eye consciousness cognizes form one thereby knows that it is, yet it cannot be said to be the same as or different from form; and so on down to mind consciousness cognizing dharmas. From this the author brings down seven objections in a row.

The first objection is the most important: if so, it becomes just like milk and curds — when eye consciousness cognizes form one thereby knows that there is milk, and when body consciousness feels touch one thereby knows that there is milk, and it likewise cannot be said whether milk is the same as or different from form and touch; on this showing, milk and curds turn out to be “conventional designations,” so on what ground is the pudgala not one? The second objection takes apart the four words “one thereby knows that it is”: is form the cause of knowing it? Or is it, when form is cognized, cognized along with form? On the former, form too does not differ from the eye and the rest (since form too has the eye, light, and attention for its causes); on the latter, one must ask whether form is what cognizes or whether there is a further cognizer — if form is what cognizes, its entity just is form; if there is a further cognizer, then this differs from form as yellow differs from blue. Third objection: to defend by saying “the two cognizers likewise cannot be said to be the same or different” is to make the cognizer not included among the conditioned, which wrecks one’s own tenet. Fourth objection, citing a sūtra: form and so on up to consciousness are all without self. Fifth objection, pressing by way of the object condition: if eye consciousness arises with form for its object, then this is not a condition of eye consciousness, as the sense field of sound is not; if it arises with this for its object, that contradicts the sūtra that “consciousness arises from two conditions,” and moreover it ought to become impermanent (whatever causes and conditions can produce consciousness are impermanent). Sixth objection: to fall back and say that it is not an object is to make it not cognized and not known, and then “existence” cannot be established for it. Seventh objection: being cognized by the six consciousnesses, then since it is cognized by eye consciousness it ought to differ from sound and the rest as form does, and since it is cognized by ear consciousness it ought to differ from form and the rest as sound does.

A place easily misread: taking the author to be splitting hairs. In fact these seven objections are seven exits from one and the same principle — whatever enters into a cognitive relation must fall under some adjudicable class; the Vātsīputrīyas want it both to get in (to be cognized) and to be unadjudicable (inexpressible), which is to want it inside the door and outside it at once. Where this lands: on reading any argument of the form “it both is… and cannot be said to be…,” first ask, “can it get through the door of cognition or not?”

VII. The Proof from the Sūtras: The Author Brings Out the Buddha’s Sūtras One by One

〔Checklist 5 · first part〕Why the passage of scriptural proof comes after the refutation. All that precedes is attestation by reasoning; this passage turns to attestation by scripture, one sūtra after another: the sūtra that the courses of the five faculties are each separate (the five faculties each make use of their own object-field, and the mind makes use of the courses of the five faculties besides) · the reconciling of the sūtra of the simile of the six creatures (the sūtra does not speak with regard to the six faculties, the eye and the rest, but with regard to the mind consciousness drawn forth by the dominance of the five faculties) · the sūtra of the dharma-gate of all that is penetrated and known (the whole of what can be known is only the twelve sense fields together with contact and feeling, and within it there is no pudgala) · the Sūtra on the Person (with the eye and form as conditions eye consciousness arises, and with the contact of the three coming together there arise feeling, perception, and volition, and by just this much is the name “person” spoken) · the sūtra of the twelve sense fields (I say that all that exists is only the twelve sense fields) · the sūtra recited by that school itself (the Tathāgata designates everything just so far) · the Sūtra to Bimbisāra (herein there is no selfhood and no what-belongs-to-self).

What has to be taught clearly is the common structure of this string of scriptural proofs: each of them does one and the same thing — it exhausts “all” by enumeration and then points out that within the enumeration there is no place for a pudgala. The Sūtra on the Person is especially crucial, because it not only enumerates but also accounts for the provenance of the name “person”: by just so much as the eye, form, consciousness, contact, feeling, perception, and volition, one is spoken of by the name “person”; and following the differences of sense one conventionally establishes the several names, sentient being, youth, nourisher, life-principle, one who is born, pudgala, all of which “are only a name and an ideation, only a self-designation, existent merely as conventionally designated in accordance with worldly usage.” And since the World-Honored One himself admonished us to rely on sūtras of definitive meaning, this sūtra is precisely of definitive meaning.

A place easily misread: taking the scriptural proof to be making up what the reasoning lacks. It is not so — what the reasoning refutes is “a self can be established”; what the scripture proves is “the Buddha never said any such thing”, and the two answer different questions. Where this lands: a name that comes readily to the tongue does not mean there is a thing behind it; what the Sūtra on the Person teaches is precisely the difference between “name and ideation” and a real thing.

VIII. The Scriptural Proof Continued: The Stanza of Śailā, the Stanza to Batuoli, and the Vātsīputrīyas’ Last Move

〔Checklist 5 · second part〕Two stanzas and the charge of “no valid means.” At the end of the citations two blocks of stanza are attached. The stanza in which the bhikṣuṇī Śailā answers Māra sets “taking the several parts and by conventional supposition establishing them as a chariot” alongside “conventionally the world establishes a sentient being, which should be known as taking the aggregates” — a chariot is not something over and above wheel and axle, and a sentient being is not something over and above the aggregates; this is the plainest simile in the whole chapter. The stanza to Batuoli begins from defilement and purity: in dependence on mind one is defiled, in dependence on mind one is purified, and “I have in truth no selfhood; it is grasped as existent only by inversion,” and at the end even one who can cultivate the contemplation of emptiness is not to be found — the contemplation of emptiness leaves no contemplator behind, and this sentence presses the refutation of the self to its limit.

The treatise then lists the five faults of grasping a self (giving rise to the view of a self and the view of a sentient being and falling into the courses of evil view · walking off the road with the non-Buddhists · not awakening into the mind of emptiness · being unable to have pure confidence and unable to abide firmly · not obtaining liberation), and here the Vātsīputrīyas play their last card — “these are all no valid means. Why so? Because in my school they have never been recited.” Fabao’s Commentary marks here expressly “the Vātsīputrīyas altogether deny that the sūtras cited above are the valid word of the noble ones.” The author’s counterstroke is very heavy: the valid means you allow — is the school the measure, or is the Buddha’s word the measure? If the school is the measure, then “the Buddha is not your master and you are not a son of Śākya”; if the Buddha’s word is the measure, then these all are the Buddha’s word. Moreover these passages are recited by all the schools, contradict neither the nature of dharmas nor the other sūtras, and yet he dares to raise a slander against them — “this is merely to indulge violence and madness, and so it is utterly unreasonable.”

A place easily misread: reading this passage as a quarrel of temper. It in fact establishes a criterion: a passage of scripture recited in common by the several schools, contradicting neither the nature of dharmas nor the other sūtras, may not be dismissed as not the Buddha’s word on the ground that “my school does not recite it.” Where this lands: whenever someone excludes evidence on the ground that “this school of mine does not accept it,” first ask the one question, “is the standard you accept a school, or the Buddha’s word?”

IX. The Objection from Omniscience: The Continuum Has the Capacity; It Is Not Knowing All in a Single Thought

〔Checklist 7 · first part〕The Vātsīputrīyas turn from defense to attack. The Vātsīputrīyas turn out of the preceding passage an objection: if there is no self, the World-Honored One ought not to be omniscient — for mind and mental factors arise and cease moment by moment, and none of them can know all dharmas; to know all thoroughly there must be a permanent self. Yuanhui’s Verse Commentary marks here expressly “from here the second, citing proof. The Vātsīputrīyas ask,” from which it is clear that this is where the Vātsīputrīyas take the initiative.

The author’s answer is the most important turn in the whole chapter: omniscience is not “knowing everything in a single thought” but “the continuum having the capacity” — that stretch of the continuum of the aggregates which gets the name of Buddha has accomplished a surpassing capacity such that, as soon as attention is directed, an unmistaken cognition arises with regard to the object one wishes to know, and it is therefore called omniscience, “not that in a single thought one can know all at once.” A stanza is then cited in proof — “because the continuum has the capacity, as fire devours everything, so is omniscience, and not by knowing all at once.” Fire does not swallow a whole forest in one gulp; it burns portion by portion, and wherever it burns, there all is consumed; the pervasiveness of omniscience is pervasiveness of that kind.

A further proof: it is said that there are Buddhas of the three times (Buddhas past, Buddhas future, and Buddhas present all extinguish the sorrows of beings), and the Vātsīputrīyas themselves allow the three times only to the aggregates and not to the one who repeatedly takes up a destiny — then what has the three times is the continuum of the aggregates, and thorough knowing likewise hangs upon the continuum. This is to use the opponent’s own concession to force the opponent’s conclusion.

A place easily misread: taking “not knowing all at once” as a diminishing of the Buddha’s excellence. Quite the contrary — it rewrites “omniscience” from a static capacity into a capacity that is unmistaken as attention is directed. Where this lands: whenever an attribute of “the limitless” is met with, first ask whether it is “possessing everything simultaneously” or “being unobstructed with regard to everything”; the difference between the two is very great.

X. The Sūtra of the Burden and the Objections That Follow: The String of Challenges Opening Fascicle Thirty

〔Checklist 7 · second part〕Four objections come one after another. The head of Fascicle Thirty is not a transition; it is a question in mid-dispute — “if only the five aggregates of appropriation are called the pudgala, why did the World-Honored One speak thus: I shall now expound for you the burdens, the taking up and laying down of the burden, and the bearer of the burden?” Since he speaks both of the burden and of a bearer, the bearer is evidently not the burden itself. The author’s answer is to put all three back inside the aggregates: the sūtra itself says that craving is named the taker-up of the burden, and craving being included in the aggregates, so is the bearer; the five aggregates of appropriation, oppressing one another by their own mark, get the name of burden, and since each preceding moment draws on each succeeding one, they are called the bearer. And lest anyone mistake this for something permanent and really existent, the Buddha himself explains afterwards in the sūtra that “merely in accordance with worldly usage is this long-lived one said to have such a name.”

The second objection: one who denies apparitionally born sentient beings is included under wrong view, from which it is evident that there is a pudgala. Answer: who says that there are no apparitionally born sentient beings? When a continuum of aggregates can go on to a later existence not by way of womb, egg, or moisture, that is one; and the aggregates of apparitional birth in principle really exist. And a counter-question is pressed further: you say this wrong view slanders a pudgala — then is it abandoned through seeing or abandoned through cultivation? The pudgala is not included among the truths, and wrong view is not abandoned through cultivation — neither end will house it. The third objection: the sūtra says “one pudgala is born into the world.” Answer: this is a conventional speaking of “one” within a totality, as the world speaks of one sesame seed or one grain of rice; and “born” is not a new arising as of the aggregates but is said with regard to taking up other aggregates at this time, as the world says “a master of the exegetical treatises is born” or “an old man is born,” both of which speak with regard to what has been taken up; moreover the Sūtra on Ultimate Emptiness has already blocked it — “there is action and there is maturation, but an agent is not to be found” — and the Sūtra to Phalguna likewise says, “I never say that there is one who appropriates.” The fourth objection takes derived form’s being the great elements as its parallel. Answer: that is a fault of that school’s own; and if one really allowed that derived form is the great elements, one would have to allow that the pudgala is established as just the five aggregates.

A place easily misread: taking these four objections as each independent. In fact they are four performances of one and the same maneuver — picking a noun out of a sentence of the Buddha’s and saying “where there is a name there must be a thing”; and the author’s four answers are likewise one and the same: what the name rests on is a continuum, not a separate entity.

XI. Nāgasena and King Pilinda: Not Declaring Does Not Mean an Answer Is Being Withheld

〔Checklist 6 · first part〕A story that teaches one thing. The Vātsīputrīyas ask: if the pudgala just is the aggregates, why did the World-Honored One not declare that “the life-principle is the body”? The author answers: because he observed the āśaya (the underlying inclination) of the questioner — what the questioner grasps is a life-principle that is “a person acting within whose entity is real and not empty,” and no such thing exists at all, so that neither sameness nor difference can be established, just as the hardness or softness of a tortoise’s hair cannot be declared. Yuanhui’s Verse Commentary marks here expressly “from here the third, the section on setting aside an answer for the benefit of beings, objection first and answer after.”

The treatise then brings out an old case. The Venerable Nāgasena possessed the three knowledges, the six supernormal knowledges, and the eight liberations; King Pilinda came to ask him, requiring first that “the Venerable answer straight.” The king asked whether the life-principle and the body are the same or different, and the Venerable answered, “this should not be declared.” The king reproached him for breaking the agreement, and the Venerable questioned the king in return: as to the mango tree in the palace, is the taste of the fruit it bears sour or sweet? The king said that there is no such tree in the palace at all. The Venerable reproached him again: was there not an agreement beforehand? The king said that since there is no such tree, how could the fruit’s taste be answered as sweet or sour. Whereupon the Venerable instructed him — the life-principle likewise does not exist, so how can it be said to be the same as or different from the body.

What has to be taught clearly is the mechanism of this story: the question itself presupposes a subject that does not exist, and any answer whatever amounts to conceding that subject. Hence not answering is not evasion; it is a refusal to endorse. Among the three commentaries only Yuanhui’s Verse Commentary expressly gives Nāgasena’s name and retells this exchange; Puguang’s Record and Fabao’s Commentary neither of them single out the personal names. Where this lands: when pressed with “you have only to answer yes or no,” first see whether its subject can stand.

XII. Why the Buddha Does Not Simply Say That There Is No Life-Principle at All: Setting Aside an Answer as a Benefit to Beings

〔Checklist 6 · second part〕The reason for not saying “there is none” is not the same as the reason for not saying “there is one.” The Vātsīputrīyas ask again: why does the Buddha not say that there is no life-principle at all? The answer is again that he observes the questioner’s āśaya — this one may be calling the continuum of the aggregates a life-principle, and were he answered “there is none at all,” he would fall into wrong view; he has not been able to penetrate the principle of dependent origination, is no vessel for the true dharma, and so conventional existence is not expounded for him. There is a very fine nicety of measure here: “there is no self” and “not even a conventional self” are two different sentences, and where the hearers differ, the teaching differs.

The treatise at once adduces the question of the non-Buddhist Vatsa in proof: he asked “am I existent or nonexistent in the world,” and the Buddha declared neither — for to declare him existent would contradict the true principle of dharma (all dharmas are without self), while to declare him nonexistent would increase that man’s bewilderment (he would suppose “I formerly was and now am not”). And “compared with the folly of grasping existence, this folly is graver” — the view of annihilation is harder to rescue than the view of permanence. Two blocks of stanza then close the passage. The first takes a tigress carrying her cub for its simile: to grasp a true self as existent is to be wounded by the teeth of view, and to deny the conventional self as nonexistent is to destroy the cub that is wholesome action — grip too tight and it is wounded, hold too loose and it falls; and the Buddha’s speaking of the true dharma keeps just this measure. Puguang’s Record and Yuanhui’s Verse Commentary both expressly attribute this stanza to Kumāralāta of the Sautrāntikas, so that it is not composed by the author. The second is a stanza composed by the author, restating that measure sentence by sentence: because a real life-principle does not exist he does not speak of sameness or difference; fearing that the conventional self be denied he does not say that there is none at all; within the continuum of the aggregates there are action, fruit, and a life-principle, and to say there is none would deny this along with it; that he does not say there is a merely nominal life-principle within the aggregates is because he observes that the questioner has not the strength to understand true emptiness.

A place easily misread: reading “setting aside an answer” as the Buddha’s not knowing either. The treatise says plainly that setting aside an answer is a benefit to beings — it turns on whether this person can receive it at this moment, not on the keeping of a secret.

XIII. Why the World Being Permanent and the Rest Are Not Declared: All Four Declarations Are Unreasonable

〔Checklist 8〕How the first set of the fourteen undeclared matters is analyzed. The author sets himself the question: on what ground are the world being permanent and the rest not declared? Answer: here too he observes the questioner’s āśaya. The analysis divides into two roads: if the questioner grasps “the self” as the world — the entity of the self does not exist at all, and so all four declarations are unreasonable, there being no subject to declare anything of; if he takes “birth and death” together to be called the world — the four declarations are likewise all unreasonable, and are analyzed sentence by sentence: to declare it permanent is to have no attaining of nirvāṇa; to declare it impermanent is to have annihilation come of itself, so that all attain nirvāṇa without effort; to declare it both permanent and impermanent is to have one portion definitely never attain nirvāṇa and another portion realize quiescence of itself; to declare it neither permanent nor impermanent yields “neither attaining nirvāṇa nor not attaining it,” a determinate contradiction, and so an idle proliferation. And since by the noble path one really can attain parinirvāṇa, all four determinate declarations are unreasonable. The Nirgrantha’s question about the death and rebirth of a sparrow is also adduced, where the Buddha knew his mind and made no determinate declaration.

Why are the four propositions about having a limit and the rest likewise not declared? Because they carry the same faults as the propositions about permanence and the rest. How is their sense known to be the same? Because the non-Buddhist Uttika first asked the four about having a limit and the rest, and then asked crookedly whether “the world is delivered wholly by the noble path, or only a portion of it,” whereupon Ānanda told him outright, “you have already asked the World-Honored One about this matter; on what ground do you now change the name and ask again” — to change the manner of asking and ask about the same matter is just the attestation of sameness of sense. The four propositions about the Tathāgata existing after death and the rest are likewise not declared, because the questioner asks by falsely reckoning a self already liberated and calling it the Tathāgata.

The author presses the grasper of a self still further: since the Buddha declares that there is a present pudgala, why does he not declare that the Tathāgata too exists after death? The answer given is: for fear of the fault of falling into permanence — then when the Buddha declares that Maitreya will become a Buddha in the future, and declares that a disciple has been born in such and such a place, is there not likewise a fault of falling into permanence? And if it be said that the Buddha sees and does not declare, there is the fault of a permanent self apart from the aggregates; and if it be said that both his seeing and his not seeing are inexpressible, then even “the Buddha is omniscient or is not omniscient” becomes inexpressible.

First, the proof from “because it is true, because it abides” can be cited on either side. If it be maintained that a pudgala really exists on the ground that the sūtra says “because it is true, because it abides,” and that one who determinately grasps at no-self falls into the place of evil views — the author answers that this constitutes no proof: the same sūtra also says that one who determinately grasps at a self falls into the place of evil views, and a passage of scripture that stands equally well on both sides cannot be taken on one side only as proof. The author at once cites the masters of the Abhidharma treatises: “to grasp the self as existent and as nonexistent are both included under extreme view, falling in order into the extreme of permanence and the extreme of annihilation,” and comments that “what those masters say is deeply reasonable” — for to grasp a self as existent is to fall into the extreme of permanence, and to grasp at no-self is to fall into the extreme of annihilation, as the Sūtra to Vatsa above said plainly. What the sūtra blocks is evidently determinate grasping at either end, not a proof for one of them.

Second, who transmigrates — the simile of a fire over a plain. The objection: if there is definitely no pudgala, who transmigrates through birth and death? Birth and death surely cannot transmigrate of themselves. The author first points out that establishing transmigration by “laying down the former aggregates and taking up the later” is a sense already challenged and dismissed above. The proper answer is a simile: “as a fire over a plain, though ceasing moment by moment, is said by reason of its continuum to have transmigration” — a fire over a plain arises and ceases thought by thought, and no unchanging grain of fire travels from here to there, and yet everyone says that the fire is spreading, and what this rests on is precisely the continuum. A heap of aggregates is conventionally called a sentient being, and with craving and appropriation as conditions it transmigrates through birth and death, on just this principle. Where the Buddha himself says in a sūtra “I who am now was in the past a guide of the world named Sublime Eye,” to read this as “the former I just is the present I” would make the entity permanent; the author adjudges that what this saying displays is that “the former and the present are one continuum,” just as the world says “this fire once burned that thing.”

Third, if there definitely is a true self, where is the fault — and the gathering up of three classes of view. If there definitely is a real self, then this self ought to be clearly contemplated by the Buddha alone; and being clearly contemplated, a firm grasping of self ought to arise, from that grasping a grasping of what belongs to self, and from these a craving for self and for what belongs to self. Hence the Buddha says: to grasp a self is to grasp what belongs to self, and within the aggregates there arise the craving for self and for what belongs to self and the view of a real person, and “one bound by craving for self thereby slanders the Buddha and is far from liberation.” If it be pleaded that with regard to the self no craving for self arises, the author adjudges this a saying without sense — it is where what is not the self is perversely reckoned as the self that craving for self can arise; that with regard to a real self it should on the contrary not arise is a saying for which there is no principle to serve as proof. Hence one who grasps a true self “has within the Buddha’s true and noble teaching no cause or condition for raising the boil of view,” having blocked for himself even the causes and conditions of raising right view. At the end the passage gathers what is refuted into three classes: one class grasps an inexpressible pudgala, one class denies altogether that any dharma has an entity, and the non-Buddhists grasp a further real selfhood — “all these views are unreasonable, and none of them can escape the fault of having no liberation.” The three classes occupy the two ends and the middle: to deny altogether is annihilation, to grasp a true self is permanence, and the inexpressible touches neither and yet is still a grasping, and so it too is without liberation.

A place easily misread: taking the fourteen undeclared matters as the Buddha’s modesty. What the treatise shows is the exact opposite — the four declarations are not declared because each of the four, when worked out, yields an absurdity; not declaring is the outcome of an analysis, not an attitude. Another place easily misread is to read “refuting the self” as “denying everything.” Among the three classes of view with which this passage closes, denying altogether that any dharma has an entity is itself one of the classes refuted — what no-self dismisses is a permanent sovereign, not the continuum and the fruit of action themselves.

XIV. Recollection and Recognition: With No Self, Who Is It That Remembers

〔Checklist 9〕The first appearance of the chapter’s positive doctrine. All that precedes is refutation; this section begins the positive exposition. The objection: if the entity of the self does not exist at all, on what ground can a mind that ceases moment by moment remember a long-past and similar object? Answer: recollective knowing “arises from a difference of mind within the continuum, of the class of the object and the perception of it” — there must be attention that takes that object as its object, together with the conditions of similarity, relatedness, perception, and the rest, and there must also be a difference of mind of that class for its cause, and only when the two are complete can recollection be cultivated into arising, neither being dispensable.

The second objection is sharper: what one mind saw, a later and different mind can remember? Surely what God-given’s mind saw is not what Sacrifice-given’s mind can remember. The answer is that this objection is not to the point — those two minds are not related to one another, unlike a single continuum which has causality; we do not say that one mind sees and a different mind remembers, because the continuum is one. And so “from the past mind that took that object as its object there is drawn forth the present consciousness that remembers,” recollection arising by the force of the continuum’s transforming and differentiating.

The third objection asks “whose recollection” — with no self, on what ground is the sixth case-ending (the genitive) said? Answer: the sixth case-ending is said with regard to the sense of belonging to an owner, as an ox belongs to Caitra, and what is meant by an owner is “one who attains mastery over riding, milking, employing, and the like.” The author here presses back with a reversal: where do you propose to employ this recollection? Answer: upon the object recollected. To what end employ it? Answer: to make recollection arise. Then — recollection has not yet arisen, so with what will you employ it? And since it can only be to make recollection arise, the cause is what is called the owner and the result is what is called the belonging, so that “the cause producing recollection suffices to be the owner of recollection; what need to set up a self as the owner of recollection?” Moreover, in what the world calls the owner of an ox, there is within it neither a real Caitra nor a real ox, but only a conventional designation. As it is with recollection, so it is with recognition.

A place easily misread: supposing that “with no self there is no memory.” The treatise’s answer is not to cancel memory but to replace the guarantor of memory, from a permanent self to a stretch of continuum possessed of causality. Where this lands: whenever it seems that some matter cannot do without a subject, first ask the one question, “what this matter really requires — is it ‘one and the same thing,’ or ‘one and the same continuum’?”

XV. Refuting the Sāṃkhya Masters: Activity and Agent, the Cognizing and the Cognizer

〔Checklist 10〕The refutation of the non-Buddhists begins here. The Sāṃkhya tenet in one sentence — “there definitely is a self, since an activity must await one who is active,” as God-given’s walking must await God-given, and as the activity of consciousness and the rest must await a cognizer on which it depends. Puguang’s Record marks here expressly “the text’s second, refuting the Sāṃkhya masters,” his outline still keeping the threefold pattern of stating the tenet · properly refuting · resolving objections.

The author’s refutation begins from one challenge: what is this God-given of yours? If a real self, it has been refuted above; if a conventional person, then its entity is not one thing but only a name conventionally established upon a continuum of formations — in which case “as God-given can walk, so consciousness can cognize,” both sides alike becoming conventional, and your simile helping me instead. The author then explains why the world says “God-given can walk”: upon formations arising and ceasing moment by moment, and not apart from the continuum, the name God-given is established, which the foolish grasp as one entity and treat as the cause of their own continuum’s arising in another place; arising in another place is called walking, and the cause is called the walker — as with flame and sound, which continue in another place, the world too says that a flame can travel and a sound can travel.

A further step of the questioning: what does consciousness do to its object? Answer: “it does nothing at all, but arises merely by way of the object” — as a result requites its cause and, though it does nothing, arises resembling the cause and so is called requiting the cause, so consciousness, though it does nothing in arising, resembles the object and so is called cognizing the object. And what “resembles the object” means is that it bears the aspect of the object; consciousness arises in dependence on a faculty as well, and yet is not called cognizing the faculty, precisely because what it bears is the aspect of the object. Or again, to say that consciousness cognizes with regard to a former consciousness drawing forth a later is likewise faultless, for the world speaks of an agent with regard to a cause — as when it says that a bell or a drum can sound, or that a lamp can travel; within a continuum of flame the name of lamp is conventionally established, and when the lamp arises in continuum in another place it is said that the lamp travels, and there is no separate traveler.

A place easily misread: supposing that “consciousness cognizes its object” means that consciousness does something to the object. The treatise says plainly: it does nothing at all, but merely arises bearing the aspect of the object. Where this lands: the subject in language is not a really existent agent — “the lamp travels,” “the bell sounds,” “consciousness cognizes” are all cases of speaking of an agent with regard to a cause.

XVI. Refuting the Vaiśeṣika Masters: Where the Conjunction of Self and Mind Fails

〔Checklist 11〕The third party’s grasping and its defenses layer by layer. The Vaiśeṣika tenet — “a certain class of non-Buddhists grasps thus: that all minds, when they arise, arise from the self.” Puguang’s Record marks “the text’s third, refuting the Vaiśeṣika masters.” The author first presses with the two preceding objections: if all minds arise from the self, on what ground is a later consciousness not always like the former, and not arising in a determinate order like sprout, stalk, and leaf — the self being permanent and one, why should its result have differences?

They defend with “it awaits a difference in the conjunction with mind.” The author takes this apart layer by layer: first, that the self is conjoined with anything else is not commonly allowed, and the conjunction of two things has bounds — they explain the mark of conjunction themselves as “not having arrived beforehand and arriving afterwards is called conjunction,” so that the conjunction of self and mind should likewise have bounds, and if the mind shifts the self should shift, or should perish together with the mind; second, if a partial conjunction be alleged, the entity of the self has no parts and there are no separate parts to be conjoined; third, granting a conjunction, since the entity of the self is permanent and the mind has no differences, whence should the conjunction have differences? Fourth, if it awaits a further cognition, then it must be asked on what ground that cognition has differences; fifth, if the conjunction of self and mind awaits a difference of formations, then “merely by mind’s awaiting a difference of formations different consciousnesses can arise — what use then is the self” — the self being of no use at all in the arising of consciousness, and yet all consciousnesses being said to arise from it, the author mocks it with a simile: as when the business of the medicine is already done and can remove a chronic disease, and yet the physician is deceived with the false claim that it was the merit of reciting a spell.

They defend again with “these two have the self for their support.” The author asks: in the sense of what supporting what? If mind and formations are like a picture or a fruit and the self is like a wall or a vessel, then there are the faults of mutual obstruction and of separate abiding. They accordingly change the simile: “this is only like the earth, which can be the support of scent and the other three things” — whereupon the author speaks the most unhurried sentence in the whole chapter: “what he says in this way establishes no-self, and so I here am deeply gladdened and consoled.” For earth is precisely what cannot be found apart from scent and the rest: who can cognize an earth apart from scent and the rest? It is only upon the differences of an aggregation of scent and the rest that the world conventionally establishes the name of earth; and the self is likewise, being only a name conventionally established upon the differences of mind and the other aggregates. They finally adduce as proof “because there are the quality-categories of recollection and the rest, and a quality must rest upon a substance-category,” and the author answers that this is not commonly allowed — whatever is granted a separate entity is called a substance, since the sūtra speaks of six real things and calls them the fruits of the ascetic life; and the sense of “resting upon” has been blocked above, so that what is set up is nothing but empty words.

A place easily misread: taking the author to be mocking his opponent. What he is really doing is to take over the opponent’s best simile and let it walk of itself over to the side of no-self — this is the handsomest stroke in all the chapter’s refutations of the non-Buddhists.

XVII. The Continuum, Transformation, and Differentiation of Action and Its Fruit: The Gathering Up of the Treatise’s Positive Doctrine of No-Self

〔Checklist 12〕The last question, and the most concrete. The objection: if there really is no self, then action having perished, how can it further produce a future fruit? The author first returns the same words — granting a real self, action having perished, how then can it produce a future fruit? From which it is evident that this question has nothing to do with whether there is a self or not; it is one that anyone whatever must answer. They answer that dharma and non-dharma arise in dependence on the self, and the author dismisses this by the sense of “dependence” already refuted.

The proper answer is this: the fruit is not produced from an action already perished but “from the continuum, transformation, and differentiation of action, as a fruit is produced from a seed.” The world says that a fruit is produced from a seed, and yet the fruit does not arise from a perished seed, nor does it arise immediately upon the seed, but from the seed’s continuum, transformation, and differentiation — the seed produces in succession sprout, stalk, leaf, and the rest, and the flower is the last and draws forth the arising of the fruit; so that what is called “a fruit produced from a seed” is by the seed’s drawing forth successively the capacity for producing fruit in the flower. The three names the treatise defines for itself: that action comes first and form and mind arise thereafter with no interval between is called the continuum; that in this continuum each succeeding moment arises differently from each preceding one is called transformation; that this transformation at the last time has a surpassing capacity that produces its fruit with nothing intervening, surpassing the other transformations, is called differentiation.

Two supplementary explanations follow. First, the order of taking up a fruit at the end of life — when the appropriating consciousness is at the very point of death, though it bears the perfuming of many actions for a later existence, what is “heavy, near, arisen, and repeatedly practiced among what was done before” is clear and the rest are not, and hence the stanza that says “action that is gravest, nearest, arisen, repeatedly practiced, done before — the earlier matures earlier and the later later, and so one turns about in birth and death.” Second, maturation does not produce maturation — the capacity drawn forth by a maturation cause perishes as soon as the fruit is had; asked on what ground a maturation fruit cannot call forth a maturation while a fruit from a seed does have a further fruit arising, the answer is that the simile is not on all fours to begin with, and that a fruit from a seed likewise has no further fruit arising, the later fruit arising rather from a later difference of ripening and change (the fruit of the former seed, meeting the conditions of water, soil, and the rest for ripening and change, draws forth a difference of ripening and change); and so it is here — the former maturation, meeting the conditions of hearing the true or the false and the rest, draws forth a wholesome or unwholesome mind possessed of maturation, from which is drawn forth a continuum that transforms, drawing forth successively a difference of transformation, from which difference the later maturation arises. It is likened further to a citron flower smeared with lac juice, whose fruit afterwards has red pulp and produces no further color. It closes with “only the Buddha realizes and knows it; it is not the object-field of any other,” attaching the stanza “this action, this perfuming, the giving of fruit at this time, the determinate principle of all these kinds — apart from the Buddha none can know it.”

A place easily misread: taking “continuum, transformation, and differentiation” for a recondite term. It is in fact only the common sense of a seed producing a fruit broken into three sentences — it joins up (continuum) · before and after differ (transformation) · at a certain moment it has a special force (differentiation). Where this lands: action and its fruit await no permanent bearer; they await only that these three hold.

XVIII. Exhorting to Study and Handing the Work On: The Last Three Stanzas of the Whole Treatise

〔Checklist 13〕How the author brings a book of thirty fascicles to an end. The last three stanzas are of seven characters to the line; the form changes and the tone changes with it — all that precedes is challenge and refutation, and here all is exhortation and entreaty. Fabao’s Commentary marks here “below, three stanzas, the second of the main text, exhorting to study and handing on within this chapter,” and adjudges that the three stanzas each exhort to something: the first exhorts to the study of the discernment born of hearing, the second to the discernment born of reflection, the third to the discernment born of cultivation.

The first stanza sets up what is to be relied on — “this pure road of causes has been well spoken, namely the Buddha’s ultimate word and the true nature of dharmas,” what is well spoken being the road of causes that is purity, whose entity is just the Buddha’s ultimate word and the true nature of dharmas; “the dark and blind graspings of outsiders should be cast off, and those held by evil views should seek the eye of discernment” — those hemmed in by evil views are just the ones who ought to come seeking the eye of discernment. The second stanza displays where it goes — “this is the one broad road to the palace of nirvāṇa, the nature of no-self on which a thousand noble ones travel,” no-self being not a conclusion but a road, and the very road a thousand noble ones travel in common; “where the light of the Buddhas’ sunlike word shines, though eyes of distinction be opened they cannot see,” the sunlight shining everywhere and yet those unwilling to rely on it still unable to see. The third stanza hands the matter to the reader — “within this corner it has been briefly expounded, in order to open for the wise the poison-gate of discernment; may each, as his own strength is able, comprehend all that is knowable and accomplish the surpassing work.”

The three words “poison-gate of discernment” are the eye in which the whole treatise’s brush comes to rest: as poison entering the body spreads from one place through the whole of it, so does discernment, entering by one gate and pervading all dharmas. The author says of himself that what he has spoken is only “a corner,” whose use lies not in exhaustiveness but in opening a gate for the reader — after which the comprehending of all that is knowable and the accomplishing of the surpassing work are each according to his own strength. The last word of the whole treatise is “work,” and this “work” is no longer the work of transmigration but the surpassing work of comprehending all that is knowable. This turn is where the thirty fascicles of the book come to rest.

Further Resonances: The Mirror of the Three Commentaries

The above is the sense of this chapter’s own text. Below, a separate section gathers four matters: this chapter’s position within the whole treatise, the adjudications of the three commentaries and their divergences, the total absence of any dispute within the lineage, and the divergences of Paramārtha’s earlier translation. A mirror is not the treatise; it is only another way of reading one and the same text, recorded together without adjudication, each left in its own place.

1. This Chapter’s Position within the Whole Treatise (Two Readings Recorded Together)

Puguang’s Record has two readings here. The first follows the usual course: this chapter is itself the last of the treatise’s nine chapters and has its own proper sense as main doctrine. The second holds that this chapter belongs wholly to the handing on — that it arises as a question upon the momentum of the exhortation and handing on at the end of the Chapter on Absorption, with the sense of refuting the self remaining to be clarified, so that the question at the chapter’s head, “if one relies on another beyond this, is there no liberation,” is not a fresh beginning but comes down from the lamentation and exhortation of the preceding chapter. Both readings are recorded together, without adjudging which is right.

Fabao’s Commentary puts forward here another standard — “judging by the pattern of the sūtras, a question is not a preface section and cannot be divided into three” — saying expressly that this chapter should not be forced into the pattern of preface section, main doctrine section, and handing-on section. This runs in a different direction from Puguang’s second reading, and each has its own reason; both are recorded without adjudication.

One further matter is recorded to mark this chapter’s singularity. This chapter has no root stanzas, and the separate edition of the root stanzas does not include it, so that Saṅghabhadra’s two treatises do not structurally set up this chapter at all (see below); while Yuanhui’s Verse Commentary, which has always “explained the stanzas only,” here explains the text continuously passage by passage, changing to the form of a passage-by-passage outline plus citation of the treatise’s own sentences followed by prose explanation — the form changes and the coverage does not. The one hard constraint on the stanza count of this chapter (seventeen lines of proper refutation · three lines of handing on) likewise comes from Yuanhui’s Verse Commentary alone. The relative value of the three commentaries for this chapter is here exactly the reverse of what it is elsewhere.

2. The Adjudications of the Three Commentaries and Their Divergences

One commentary wholly absent. Shentai’s Commentary is a fragmentary text and is wholly lacking for this chapter — the terms “chapter on refuting the self,” “refuting the grasped self,” “Vātsīputrīyas,” “pudgala,” “Śailā,” and “Nāgasena” all have zero occurrences in his book. What must further be recorded is this: for other chapters, a lack can still be attested by a stanza count in the general conspectus of his Fascicle One, whereas for this chapter not even the attestation of the conspectus exists, since the conspectus stops at the Chapter on Absorption and this chapter has no root stanzas whose lines could be counted. The lack is marked as it stands, and the text of other chapters is not made to stand in for it.

The three take three different approaches to glossing the chapter name. Puguang’s Record glosses the name with “the entity of the self really does not exist, and beings grasp it perversely, and this chapter refutes the grasping of a self at length,” and sets this chapter within the structure of the whole treatise — a single treatise explains “all dharmas are without self” among the three seals of the dharma, the preceding eight chapters clarifying that dharmas exist and this one chapter refuting the self and clarifying that it does not. Fabao’s Commentary is more compressed in its wording, going straight to this chapter’s refuting the self and clarifying its tenet, but supplies one further sense: among the several schools only the Vātsīputrīyas grasp that there is a real self — pointing out that the Vātsīputrīyas are a school of Buddhism and not non-Buddhists, so that refuting them differs from refuting the Sāṃkhya and the Vaiśeṣika. Yuanhui’s Verse Commentary opens with the summarizing phrase “when the clouds part the moon appears; when grasping is gone emptiness is bright,” glosses the source of the school’s name word by word (in the Buddha’s lifetime there was a non-Buddhist named Vatsa, and this school’s grasping being the same as that non-Buddhist’s, it is labeled accordingly), and alone supplies the phonetic-and-semantic correspondence “Vatsa here means Calf” — of the three commentaries this one alone gives a phonetic-and-semantic note.

All three adjudge a twofold chapter-level outline, and their finer divisions diverge. Puguang’s Record makes the main text twofold (the extended refutation of divergent graspings · exhorting to study and handing on), with the extended refutation divided within into general refutation and several refutations, the several refutations covering three parties, and each of the three refutations using the same pattern of “stating the tenet · properly refuting · resolving objections,” recurring three times. Fabao’s Commentary makes it twofold (properly refuting the self · exhorting to no-self), the proper refutation dividing into three. Yuanhui’s Verse Commentary likewise makes it twofold, and his division hangs upon the stanzas — the seventeen lines of proper refutation divide into six (properly displaying true emptiness, six lines · clarifying the Buddha’s thorough knowing, two lines · setting aside an answer for the benefit of beings, six lines · the mark of the cause being unfathomable, one line · the priority and posteriority of the fruit’s arising, one line · the difficulty of knowing the retribution of action, one line), and the handing on, three lines. All three stop at a twofold division, and not one of them sets up a third-level main outline within the chapter, which is precisely why this chapter should not be further divided into units.

The difference of density in the labeling of speakers is the most conspicuous. Puguang’s Record marks “the Vātsīputrīyas ask” and “first stating the tenet of the Vātsīputrīyas”; Fabao’s Commentary marks “the Vātsīputrīyas cite a sūtra to object” and “altogether denying the sūtras cited above”; Yuanhui’s Verse Commentary marks sentence by sentence “the Vātsīputrīyas ask,” “the Vātsīputrīyas object,” “again the Vātsīputrīyas ask,” and is the only one of the three to mark question and answer sentence by sentence. ⇒ The attribution of speakers in this unit’s vernacular layer and exposition layer takes just these markings of the three commentaries as its standard. One honest record: Puguang’s Record marks the question “what is the difference of the marks of real existence and conventional existence” as “the Vātsīputrīyas ask,” whereas by the flow of the prose it can be read as the author challenging the Vātsīputrīyas, the two readings running in opposite directions, and both are recorded without forcing a change; again, for the author’s refutation after the simile of fire in dependence on firewood, and for the questioner in the passage on the four declarations about permanence and the rest, no express marking is found in any of the three commentaries, and “no marking” is recorded as it stands, the attribution being made by the tone of the surrounding text without setting up markings on the commentators’ behalf.

The Nāgasena passage: only Yuanhui’s Verse Commentary gives the name. Puguang’s Record and Fabao’s Commentary both have zero occurrences of the two names “Nāgasena” and “Pilinda” in this passage — this is not a failure to explain, but the habit of the two in handling narrative interludes, which is not to single out personal names but to carry the explanation straight on from the surrounding text; “zero occurrence” does not assert “no explanation.” Yuanhui’s Verse Commentary retells the gist of this exchange sentence by sentence, and its teaching value is highest of all in this passage. Each of the three has its own strength; no adjudication of merit is made.

The provenance of the stanza of the tigress carrying her cub: two commentaries adjudge alike. Although the preceding text reads “hence there is a stanza that says,” Puguang’s Record expressly attributes it with “Kumāralāta among the Sautrāntikas has this stanza,” and Yuanhui’s Verse Commentary likewise gives “first, the stanza of Kumāralāta of the Sautrāntika masters; second, the author’s stanza” — hence this stanza is not composed by the author, and it is the block that follows, at “again he spoke a stanza saying,” that is the author’s own composition. The two commentaries adjudge alike word for word, and the reader must not judge from the two words “a stanza” alone that this is the author’s speech.

3. Dispute within the Lineage: A Structural Zero

Saṅghabhadra’s two treatises have no counterpart at all for this chapter, and not in the sense of “the chapter exists but the text is lacking”: the chapter item itself does not exist — the Treatise Conforming to Correct Principle has eight chapters in all and stops at the chapter analyzing absorption; the Treatise Revealing the Doctrine has nine chapters in all and likewise stops at the chapter analyzing absorption. The opening lines of this chapter’s nine blocks of stanza have zero occurrences throughout the two treatises, and the two terms “refuting the grasped self” and “chapter on refuting the self” likewise have zero occurrences (the phrase “refuting the tenet of a self” occurs three times scattered through that treatise, all of them formulaic references to another school’s position, not a chapter name, and they may not be made to stand in).

The reason for this lack and this chapter’s “having no root stanzas” are two faces of one thing. The catalogue of the teachings concludes of the two treatises that “the two treatises above share the stanzas of the Abhidharmakośa and differ in explanation, and both are composed by Saṅghabhadra” — the two treatises take the Kośa’s root stanzas and compose a separate prose commentary upon them; this chapter having no root stanzas, the two treatises had no stanzas to hang an explanation upon. ⇒ The lack is marked as it stands, and the text of other chapters is not made to stand in for it.

One caution about searching is appended. The Treatise Revealing the Doctrine sets up a preface chapter at its head, which the Treatise Conforming to Correct Principle does not, and so its chapter numbers are all shifted one place later, and its “ninth” is the chapter on absorption; in citing its chapter titles the chapter name must be checked first. Again, in the use of the two near-identical graphs for “adjudge” and “discriminate,” the base text and the three commentaries run in opposite directions, and to search the three commentaries with the base text’s graph is bound to yield a false zero; both forms are to be tried in every case.

4. The Divergences of Paramārtha’s Earlier Translation

The chapter names are close but not the same. Xuanzang gives “the chapter refuting the grasped self,” Paramārtha “the chapter refuting the self that is spoken of”; the glossing graphs both convey the sense of refuting the grasping of a self, and the division is entirely the same (a single chapter in both, with no subdivision by fascicle). Both translations likewise take this chapter as the last, and their closing places are the same.

One instance of divergence in the assignment of the chapter boundary. The set of ten stanzas at the end of the Chapter on Absorption (from “the eye of the world, the great teacher, has closed” down to “seek deliverance and be not heedless”) Xuanzang attaches to the end of the Chapter on Absorption, whereas Paramārtha uses them to open the chapter refuting the self that is spoken of. One and the same set of stanzas is attached by one edition to the tail of the preceding chapter and by the other to the head of this one — a matter purely of the assignment of a chapter boundary, with no doctrinal difference. This corpus follows the three commentaries in attaching them to the end of the Chapter on Absorption, and records the divergence without adjudication.

The names of the school and of person and dharma differ systematically. Xuanzang renders “the Vātsīputrīya school” by sense, Paramārtha transcribes it as “Bosifuduoluo school,” and in the later passage citing the Sūtra to Vatsa he gives “Vatsa” again — so that one school’s divergent names are not uniform even within Paramārtha’s own text; this is recorded without adjudication. Yuanhui’s note “Vatsa here means Calf” serves as collateral attestation that the two names refer to the same thing. Xuanzang transcribes “pudgala,” while Paramārtha renders it straightforwardly as “person.” ⇒ To search Paramārtha’s text with Xuanzang’s words is bound to yield a false zero, and matching passage to passage must in every case be done by the structure of the scene.

Two places of complete agreement sentence by sentence. The first is the adjudication of real and conventional existence — Paramārtha gives “if it is separately existent like form and the rest, it is called a really existent thing; if it exists only as an aggregation like milk and the rest, it is called conventionally named existence,” agreeing entirely with Xuanzang’s form-and-sound simile for real existence and milk-and-curd simile for conventional existence, save that Paramārtha’s text sets out two questions side by side while Xuanzang’s has one question and two answers. The second is the simile of fire in dependence on firewood — Paramārtha gives “as fire is spoken of with regard to firewood, so is the person spoken of with regard to the aggregates… if fire differed from firewood, firewood ought not to be hot; if fire did not differ from firewood, what is burned would be just what burns,” the thing likened, the thing it is likened to, and the form of the two horns agreeing with Xuanzang sentence by sentence. Further, in the one stanza on omniscience, Paramārtha gives “because the continuum has the capacity, it is called fire devouring everything; thorough knowing is said to be so too, and not by understanding all at once,” the sense of the stanza agreeing entirely while the system of rendering differs (Xuanzang’s “omniscience” · Paramārtha’s “thorough knowing”).

An honest record. In the passage of Nāgasena and King Pilinda, the scene in Paramārtha’s text agrees entirely (the king asks about the sameness or difference of the life-principle, and the Venerable challenges him in return with the simile that where there is no tree in the palace the taste of the fruit cannot be answered for), save that the rendered names “Nāgasena” and “Pilinda” are not found expressly set out word for word, and the lack is recorded as it stands — the scene is certainly there, and the detail of the names awaits re-verification; not having found it is not the same as its not being in that text. Again, the word-for-word collation of the stanza of Śailā, the stanza of the peacock’s tail, and the last four stanzas has not yet been completed, and this is recorded as it stands as still to be supplied, without filling it in with Xuanzang’s words. No adjudication of merit is made between the two translations.