俱舍論 六種種性·七聖·菩提分法Abhidharmakośabhāṣya · The Six Kinds of Lineage · The Seven Noble Persons · The Factors of Awakening

Before Reading This Fascicle

The preceding fascicle carried the four approaches and four fruits as far as the arhat; this fascicle, at its very first word, overturns a habitual impression — that having realized arhatship does not mean that everyone is exactly alike. The first sentence of the treatise takes up what precedes very tightly: “as was said above, an arhat of the unshakable lineage, after the first knowledge of exhaustion, gives rise to the knowledge of non-arising. Do arhats, like the stream-enterer and the rest, have differences among them?” This one question opens the gate of the whole fascicle: among arhats there are six kinds of lineage; some must await the season and its conditions before they can enter absorption, others have it present as they wish; some may fall back, others assuredly do not.

The three ancient commentators let the knife of the outline fall in the same place here, and their numbering is word for word the same. Puguang marks: “below, the fifth, explaining the six kinds of lineage. Within it: first, the six arhats; second, whether the six lineages come earlier or later; third, falling back from lineage and from fruit; fourth, the lineages of those in training and of ordinary beings; fifth, the difference of the three fallings back; sixth, the marks of the time of falling back from a fruit; seventh, the differences in the refining of the faculties; eighth, the nine persons beyond training.” Yuanhui has: “from here the fifth, explaining the six kinds of lineage, and within it eight: first the six arhats, second whether the six lineages come earlier or later, third falling back from lineage and from fruit, fourth the lineages of those in training and of ordinary beings, fifth the difference of the three fallings back, sixth the marks of the time of falling back from a fruit, seventh the differences in the refining of the faculties, eighth the nine persons beyond training” — hardly a character apart from Puguang. Fabao, on the other hand, gathers the first three into one and so has six: “below, the fifth of the main text, explaining the six kinds of lineage. First, the six kinds of arhat; second, the six kinds among those in training and ordinary beings; third, the difference of the three kinds of falling back; fourth, the time of the six fallings back from a fruit and the rest; fifth, the differences in the refining of the faculties; sixth, the nine kinds of those beyond training.” The three houses divided their fascicles independently, and yet the boundary agrees wholly with Xuanzang’s — from which it is plain that this cut was not forced upon the text by later hands.

There is one further thing in this fascicle rarely met with in this book: of the seven fascicles of Shentai’s commentary that survive, only Fascicle Seventeen belongs to the Chapter on the Worthy and the Noble, and it happens to cover the content of this fascicle entire — at its head it marks of itself, “from here below, the fourth, seven and a half lines of stanza explaining the six kinds of lineage. Within it the first one and a half lines of stanza explain the arhat,” and at its end, the four alternatives of disgust and dispassion once explained, it closes. This is therefore the first unit in the whole book for which all four houses are present; wherever below the four are set out together, it is expressly marked.

What must be said first is this: the longest stretch of prose in this fascicle is not about numbered dharmas but is one long exchange — whether the fruit of the arhat can in the end be fallen back from. Within this stretch four houses hold different opinions, the Sautrāntikas, the Elder, the Masters of Distinction, and the Sarvāstivāda school itself; the author too plainly declares where he stands, and yet in the end he still concludes by returning to his own school with “the Vibhāṣā masters state it as settled: the fruit of the arhat too admits of falling back.” The knack of reading this stretch is to keep asking, “who is speaking now?” — and every section below marks it out.

I. Arhats Divide into Six Kinds: Liberation Dependent on Circumstance and Liberation Not Dependent on It

First ask: since all alike have cut off the afflictions entire, what difference can there still be among arhats? The treatise answers directly — the difference lies not in what has been cut off but in the faculties one is born with. “In the sūtras it is said that the arhat, because of difference of lineage, is of six kinds: first the dharma of falling back, second the dharma of deliberation, third the dharma of guarding, fourth the dharma of abiding, fifth the dharma of being able to reach, sixth the dharma of the unshakable.” The first five are all produced from the line of “one resolute through faith” (in whom confidence predominates and the faculties are duller) at the stage of training, and are together named the liberation of mind that is cherished and dependent on circumstance; the sixth is produced from the line of “one attained through view” (in whom discernment predominates and the faculties are keener), and is named the unshakable liberation of mind.

The name “liberation dependent on circumstance” calls for care. It does not mean “a temporary liberation” but “liberation upon awaiting the time” — “they are also called liberation dependent on circumstance, because it must await the time and because there is liberation. The first word being elided, one says it as one says ‘a butter jar.’” Fearing a misreading, the treatise adds an instance of elision on purpose: a jar that holds butter is called for short a butter jar, the first word being dropped. And what is the time awaited? “That is, only when the supports, the freedom from illness, the place, and other such excellent conditions come together can he enter absorption” — supports, body, and place must meet before absorption can be entered at all. On this simile the glosses of the four houses agree wholly: Puguang has “as one says ‘a butter jar’: in full one should say ‘a jar holding butter,’ but the initial character ‘holding’ being elided one says merely ‘a butter jar,’” while Shentai has “as the world says in full ‘this is made a butter jar,’ the initial words ‘this is made’ being elided, one says merely a butter jar.”

The easy error: to take the six as six grades of “higher and lower power.” Higher and lower they do indeed have, but the criterion the treatise hangs them upon is the faculties and preparation, not the amount of affliction cut off — the afflictions cut off by all six kinds of arhat are the same. In practice: when you come to the long stretch on “falling back” further on, remember that what is fallen back from is never the exhaustion of outflows but absorption and lineage — establish this point first, and that dispute becomes legible.

II. Is a Lineage Had from the First, or Refined into Afterward?

First ask: are these six lineages fixed at birth, or cultivated into being? The treatise answers, “it is not fixed,” and adjudges it with one stanza — “some are the lineage had from the first, some are got afterward by refining the faculties.” Taken finely: the dharma of falling back must be had from the first; the five from the dharma of deliberation on may also be got afterward. So there are those who are of the nature of the dharma of deliberation from the start, and there are those who were of the nature of the dharma of falling back and afterward refined the faculties and turned into the dharma of deliberation, and so on up to the unshakable.

All six names are figures of speech, not serial numbers. The treatise gives a gloss for each in turn: the dharma of falling back is “that on meeting the slightest condition he falls back from what he has obtained — not so with the dharma of deliberation and the rest”; the dharma of deliberation is “that, fearing to fall back and lose, he constantly deliberates upon slaying himself”; the dharma of guarding is “that he gladly wards and guards what he has obtained”; the dharma of abiding is “that apart from a strong condition of falling back he can avoid falling back even without warding himself, and apart from strong preparation he also makes no further advance”; the dharma of being able to reach is “that his nature is capable, he is apt at refining the faculties, and he swiftly reaches the unshakable”; the dharma of the unshakable is “that he assuredly does not fall back.” Yuanhui lists the glosses of all six in full, and of the four houses is the handiest for a beginner — “first, the dharma of falling back: on meeting the slightest condition he falls back from what he has obtained. Second, the dharma of deliberation: fearing to fall back and lose, he constantly deliberates upon slaying himself. Third, the dharma of guarding: he wards and guards the dharma he has formerly obtained.”

These six take shape already at the stage of training. The treatise gives a table of their arrangement: the first and the second both lack constant preparation and reverent preparation, and are divided into two only because the faculties differ as keen and dull; the third, guarding, has constant preparation only; the fourth, abiding, has reverent preparation only; the fifth has both and yet is still of dull faculty; the sixth is of keen faculty and has both complete. The easy error: to take the name for the destiny. The treatise expressly blocks it — “one of the lineage that is the dharma of falling back does not assuredly fall back, and so on up to one of the dharma of being able to reach who is not assuredly able to reach; these names are established only with regard to there being such a possibility.” In practice: these names are established with regard to “possibility,” not to necessity; and the sentence below, “therefore the six arhats are found in all three realms,” holds good only on the strength of this point.

III. Four Fall Back from Lineage, Five from the Fruit — and What Is Fallen Back From Is Never the Portion Had from the First

First ask: among the six, who falls back? And is it the lineage or the fruit that is fallen back from? The stanza answers, “four fall back from lineage, five from the fruit, and not from what was first.” The unshakable assuredly does not fall back; the first five all admit of it. Of these, the last four may fall back from lineage; the dharma of falling back alone has no lineage to fall back from — because it is already at the lowest grade. From the fruit, all five may fall back.

The pivot lies in the two words “not from what was first.” The treatise explains it in three layers: for one at the stage beyond training, the lineage in which he abode at the former stage of training will not be fallen back from, because it was made doubly, by the path of training and the path beyond training; for one at the stage of training, the lineage in which he abode at the former stage of an ordinary being will not be fallen back from either, because it was made doubly, by the mundane and the supramundane path; only the four lineages beginning with the dharma of deliberation, newly got afterward by refining the faculties, admit of being fallen back from. From this a hard conclusion is drawn — “so it is settled that there is no falling back from the fruit of the stream-enterer.”

The easy error: to read “the arhat may fall back” as “he may fall back into being an ordinary being.” The treatise’s reckoning runs precisely the other way: one beyond training of the dharma of falling back has three roads before him (advancing his faculties, falling back to abide at the stage of training, abiding at his own station and entering final nirvāṇa), the dharma of deliberation four, and the rest in order five, six, and seven; and moreover “if the four beginning with the dharma of deliberation fall back to abide at a stage of training, they return to abide at the corresponding station and at no other” — even his falling back goes only to the station that answers to himself. In practice: this section is really drawing a map of the boundaries of falling back; and once the boundaries are drawn clearly, one knows what ground the great dispute of the next section is fought over.

IV. Why the Stream-Enterer Assuredly Does Not Fall Back: Resting on No Entity and Resting on an Entity

First ask: on what ground is it said that the afflictions abandoned through seeing, once cut off, do not arise again? The reason the Sarvāstivādins give is a single sentence — “because what is abandoned through seeing rests on no entity”; and Puguang marks who is speaking: “the Sarvāstivādins answer. Because what is abandoned through seeing rests on no entity, there is assuredly no principle of falling back.” Resting on no entity means that the view of self turns in dependence upon the imagined “I,” and that all the confusions abandoned through seeing have this view for their root; “since the entity of a self does not exist, it is called resting on no entity, and since there is no entity there is assuredly no principle of falling back.”

The treatise at once raises an objection against itself: does that not amount to saying that this affliction “takes what is not there as its object”? Answer: no — for the object upon which it rests is still the four truths, only that it rests upon them otherwise than as they really are. Pressed again: which of the afflictions does not rest upon the truths otherwise than as they are? The answer lies in a difference: the afflictions abandoned through cultivation each have their real thing (the agreeable, the disagreeable, and the rest), whereas those abandoned through seeing conjure an “I” upon the four truths out of nothing. The treatise states the same matter again from three angles, and gives at last a simile — “just as, upon a rope or the like, one may on the instant take it for a snake”: in haste one may see a rope as a snake, but never once one has examined; the afflictions abandoned through cultivation are benighted and arise without waiting for examination, and so a noble person may still lose mindfulness and fall back into them; the afflictions abandoned through seeing are of the nature of reckoning, and one examination brings the truths in sight, and so they assuredly do not return.

The easy error: to take “resting on no entity” for “the object is a sheer nothing.” The treatise has itself blocked that reading. In practice: this passage gives a usable criterion — an attachment built upon a fictitious object never returns once it is seen through; an inveterate habit built upon a real thing must still be guarded against after it has been seen through. In practice the work each requires is different from the start.

V. The Sautrāntika Account and the Author’s Declaration of Where He Stands

First ask: in this dispute, on which side does Vasubandhu himself stand? The text of the treatise has only one sentence — “the Sautrāntika masters say: from the state of an arhat too there is no falling back. What they say accords with principle.” The four words “what they say accords with principle” look light, and yet the ancient commentators unanimously take them for the author’s open declaration. Puguang says it most fully: “below, the Sautrāntikas explain falling back and not falling back with regard to the four fruits. The author’s intent sides with the Sautrāntikas, and so he says that what they say accords with principle”; Fabao changes the wording to “the author by his appraisal takes up the Sautrāntika explanation”; and Shentai is terser still, in four words — “the author appraises it, saying: what they say accords with principle.”

The Sautrāntika argument runs along two roads. Puguang marks the speaker sentence by sentence: “‘by scripture and by reasoning’: the Sautrāntikas answer”; “‘how is it by scripture’: the pressing.” Along the road of scriptural testimony he cites three passages: that affliction cut off by noble discernment is called really cut off; that the Buddha says one in training ought to be heedful and does not say so of the arhat; and that the Buddha told Ānanda that gain and honor and the rest obstruct even the arhat, yet “does not say that he falls back from the fruit of the arhat, but only that he falls back from and loses the pleasant abiding in the present dharma.” Along the road of reasoning there is a dilemma: if the counteracting path has already arisen, he ought not to fall back into affliction; if the path has not arisen, he ought not to be called one whose outflows are exhausted — “and if his outflows are not exhausted, on what ground is he called worthy?”

The easy error: to suppose that the author’s declaration amounts to the treatise’s changing its school. It does not — at the end of this fascicle it still concludes by returning to its own school with the account of the Vibhāṣā masters. In practice: whenever in reading this treatise you meet with words like “what they say accords with principle” or “this is best,” hold two things in mind at once: here is where the author’s inclination lies, and the conclusion of his own school is still to come. That the two stand together is precisely the form of this book.

VI. How to Read the Liberation Cherished and Dependent on Circumstance: What Is Fallen Back From Is Absorption, Not the Fruit

First ask: since the sūtra does speak of a “liberation cherished and dependent on circumstance,” is that not an admission that the arhat falls back? The Sautrāntika answer is not to deny the text but to ask one question, “what is it that he falls back from?” — “but one ought to examine what it is that he falls back from: is it the fruit-nature of worthiness, or the meditative absorptions and the rest?” Puguang marks this as the Sautrāntikas citing and explaining through an objection: “‘if it be said that there is falling back, and so it is not called cherished’: the Sautrāntikas cite the objection and explain it through.” The answer is: what is fallen back from is the concentration of the fundamental meditative absorptions, which becomes present only upon awaiting the season, and so is named liberation dependent on circumstance; and because he longs for it again and again for the sake of the pleasant abiding in the present dharma, it is named “cherished.”

The treatise at once records another account besides: “some say: this absorption is a place of savoring that is cherished; whereas the fruit-nature liberation of the arhats accompanies them constantly, and so should not be called ‘dependent on circumstance’; and since he no longer longs for it, neither is it called ‘cherished.’” — which assigns both words, “circumstance” and “cherished,” to the side of absorption with outflows. Then the Sautrāntikas object once more: if the fruit-nature could be fallen back from, why does the World-Honored One speak only of the pleasant abiding in the present dharma as able to be fallen back from? Puguang quotes their words as: “the Sautrāntikas object in return. If it be said that the fruit-nature of worthiness, which is without outflows, admits of being fallen back from and lost even by one of dull faculties, then why does the World-Honored One speak only of the pleasant abiding in the present dharma, the absorptions with outflows that have been realized, as admitting in principle of being fallen back from, and not of what is without outflows? By this it is proved that the fruit-nature of the arhat must be unshakable.”

The easy error: to read the names “the dharma of falling back” and “the dharma of not falling back” as bearing on whether the fruit-station can be fallen back from. The treatise here gives the source of the naming itself — it is established with regard to “there being or not being falling back with regard to the pleasant abiding in the present dharma obtained.” In practice: hereafter, whenever you meet the words “so-and-so fell back and lost,” first ask in three layers: was it the fruit, the absorption, or the present enjoyment that was fallen back from? Confound the three and this whole stretch becomes unreadable.

VII. The Difference among the Three Names: Not Falling Back, Abiding, and Unshakable

First ask: not falling back, abiding, unshakable — they all sound like not falling back; where is the difference? The treatise closes the matter with one question and answer: “what is the difference among not falling back, abiding, and unshakable? What is not got by refining the faculties is called not falling back; what is got by refining the faculties is called unshakable.” Puguang marks both parties to this exchange — “‘what is the difference among not falling back, abiding, and unshakable’: the question”; “‘what is not got by refining the faculties’ down to ‘the difference among the three’: the Sautrāntikas answer. Not falling back is got by one’s own nature and not through refining the faculties; what is got by refining the faculties is called unshakable.”

The dividing line among the three runs thus: not falling back is got by one’s own nature and not through refining the faculties; unshakable is got through refining the faculties. The excellent attainments raised by these two do not fall back even on meeting a condition of falling back. The dharma of abiding, however, can only avoid falling back from the merits in which he already abides, and “cannot draw forth any further excellent merits; and if he should draw them forth, from that portion he may still fall back” — he has enough to hold what he has and not enough to open new ground.

The easy error: to take the dharma of abiding for another name of the dharma of the unshakable. The treatise has already parted them expressly: abiding is “holding fast what one already has,” unshakable is “not losing even what is newly got.” In practice: the three names are in fact three degrees of stability — stable from the first, stable by cultivation, and stable only where one stands. To judge for oneself which of the three one belongs to is more useful than remembering the names.

VIII. The Affair of Godhika: What Was Fallen Back From Was Absorption, Not the Fruit

First ask: Godhika took a knife and slew himself — is that not iron proof that an arhat falls back from the fruit? The treatise explains it through thus: he “formerly at the stage of training savored the liberation dependent on circumstance to excess,” and being besides of dull faculties fell back and lost it again and again, whereupon he deeply reproached himself in disgust and took a knife and slew himself; and because he had no clinging whatever to body and life, at the time of his death he on the contrary obtained arhatship and forthwith entered final nirvāṇa — so “Godhika too did not fall back from and lose the fruit of the arhat.”

The ancient commentators here mark a divergence between two readings. Fabao records it most clearly: “the author explains the sūtra through. The Sarvāstivādins hold that Godhika fell back from the fruit of worthiness and obtained it again. On the Sautrāntika position, Godhika fell back repeatedly at the stage of training and deeply reproached himself and the rest, and at the time of his death obtained the fruit of the arhat.” Puguang likewise marks this passage as “the Sautrāntikas again explain a sūtra through.” That is to say, of one and the same affair the Sarvāstivādins read a falling back from the fruit and a regaining of it, while the Sautrāntikas read repeated falling back at the stage of training and the obtaining of the fruit at death.

The easy error: to suppose that explaining a sūtra through means altering it. The Sautrāntikas altered not one character of the text; what they altered was “which portion it was that he fell back from.” In practice: in reading biographical scriptural testimony, first part “the fact” from “the adjudication of the fact” — and this passage is the clearest demonstration of it in the whole fascicle.

IX. The Three Objections from the Sūtra of the Increase by Ten and the Sarvāstivādins’ Challenge

First ask: how is it proved that liberation dependent on circumstance is not the entity of the fruit of worthiness? The Sautrāntikas adduce the Sūtra of the Increase by Ten — “one dharma is to be aroused, namely the liberation of mind that is cherished and dependent on circumstance; one dharma is to be realized, namely the unshakable liberation of mind” — and at once raise three objections: first, if the fruit of worthiness were itself the liberation of mind cherished and dependent on circumstance, why should the fruit of worthiness be spoken of twice over in this sūtra? Second, nowhere at all is the fruit of the arhat called “to be aroused”; it is called only “to be realized.” Third, the sūtra says that the fruit of worthiness comprised under dull faculties is called “to be aroused”: if this is to show that he is able to arouse it, then one of keen faculties is the more able; if it is to show that he ought to arouse it, then one of keen faculties the more ought to. The conclusion is that “liberation dependent on circumstance is not the fruit-nature of worthiness.” Puguang marks its character as “the Sautrāntikas adduce a sūtra to prove that the fruit-nature of worthiness is not liberation dependent on circumstance.”

The Sarvāstivādins at once return an objection: “‘if that is so, then why is liberation dependent on circumstance said to be the fruit of worthiness’: the Sarvāstivādins object. If liberation dependent on circumstance is not the entity of the fruit of worthiness, why is liberation dependent on circumstance said to be the fruit of worthiness?” The Sautrāntikas answer: that name is established with regard to a class of the fruit of worthiness whose faculties are dull and who must await the season before absorption becomes present, and the contrary of it is called liberation not dependent on circumstance; Puguang marks, “‘that is, there is a fruit of worthiness’ down to ‘is called liberation not dependent on circumstance’: the Sautrāntikas answer,” and Fabao marks the same, “the Sautrāntikas answer. The name is established on the ground of awaiting or not awaiting the circumstance for obtaining absorption.”

The easy error: to take “to be aroused” and “to be realized” for synonymous turns of phrase. The whole of the Sautrāntikas’ second objection is built upon the division of labor between these two words. In practice: in reading argument that rests on scriptural testimony, note that what the arguer rests upon is very often a division of labor among terms rather than a doctrinal assertion — this is one of the forms of argument most used in this treatise.

X. The Sūtra of the Simile of the Charcoal and the Return to His Own School

First ask: the Sūtra of the Simile of the Charcoal says expressly that a noble disciple, while walking or standing, may through the losing of mindfulness give rise to evil and unwholesome thoughts — is that not falling back from the fruit of worthiness? This objection is raised by the Sarvāstivādins. Puguang marks it plainly: “‘if that is so, then you must explain’ down to ‘is the arhat’: the Sarvāstivādins again adduce a sūtra in objection,” and Fabao likewise marks, “the Sarvāstivādins adduce a sūtra in defense.” The Sautrāntikas’ explaining through is: the one who “has already forever vomited forth all the dharmas conformable to outflows and has obtained coolness” is the arhat, whereas the losing of mindfulness and the arising of thoughts belong to the stage of training, before he “had penetrated it well” — “the former passage is spoken with regard to the stage of training, and so there is no fault in it.” Puguang marks this as “the Sautrāntikas explain the sūtra through,” and Fabao has, “the Sautrāntikas note that the later text cites ‘he has already forever vomited forth all the dharmas conformable to outflows and has obtained coolness’; by this later text he is the arhat.”

The wide dispute being thus finished, the treatise concludes by returning to its own school in a single sentence: “the Vibhāṣā masters state it as settled: the fruit of the arhat too admits of falling back.” Puguang marks here, “the wide dispute being finished, he concludes by returning to his own school.”

The easy error: to take the return to his own school for the author’s own final judgment. Having said above that “what they say accords with principle,” and here concluding that there is falling back, the two stand together without contradiction — and this is precisely the form of the treatise, “he sets up the Sarvāstivāda sense, and wherever there is a one-sided place refutes it with the Sautrāntika sense.” In practice: meeting an ending of this two-sided kind, what a reader has to do is not to take a side for the ancients but to hold fast the grounds of each of the two parties. The adjudication of the schools belongs to a weighing of doctrine, and is not for a reader’s momentary opinion.

XI. Those in Training and Ordinary Beings Also Have Six · but the Faculties Are Not Refined on the Path of Seeing

First ask: are the six kinds of lineage the arhat’s alone? They are not. The stanza answers, “those in training and ordinary beings likewise have six; refining the faculties is not on the path of seeing,” and the reason is of the simplest: since the six kinds of the fruit of worthiness have these for their antecedents, the antecedents naturally have six kinds too.

The point lies in the second half of the line. At the stage of the path of seeing the faculties assuredly cannot be refined, “because at this stage there is no room to raise the preparation” — the path of seeing is swift, and there is no time to raise the preparation for refining the faculties. So the faculties can be refined only at the stage of one resolute through faith and at the stage of an ordinary being, just as at the stage beyond training. Yuanhui fills it out clearly: “but on the path of seeing there is assuredly no refining of the faculties, because the path of seeing is swift and there is no room to raise the preparation for refining the faculties.” The treatise then adduces a sūtra in evidence: of the pleasant abidings in the present dharma comprised under the four kinds of higher mind, any one whatever admits of falling back, whereas of the unshakable liberation of mind “I say it as settled that there is no cause or condition by which one could fall back from this.”

The easy error: to suppose that refining the faculties is an exercise that may be taken up at any time. It has a limit of station — it cannot be cultivated on the path of seeing. In practice: reading this together with the earlier passage on falling back, one finds that the treatise’s attitude toward the stages of practice is very concrete: what can be done at what station cannot be stated in general terms.

XII. Falling Back Is of Three Kinds: From What Is Obtained, from What Is Not Obtained, from Enjoyment

First ask: since the dharma of the unshakable does not fall back, how is it said that he falls back from the pleasant abiding in the present dharma? The treatise’s solution is first to divide the word “falling back” into three — “first, falling back from what is obtained, that is, falling back from the excellent merits already obtained; second, falling back from what is not obtained, that is, being unable to obtain excellent merits; third, falling back from enjoyment, that is, the excellent merits already obtained not being present.”

The division made, each has his own portion. The World-Honored One has only the falling back from enjoyment, because his merits are so many that they cannot all be present at once; the dharma of the unshakable has two, from enjoyment and from what is not obtained, because there are merits more excellent than his own not yet obtained; the remaining five may have all three. Whereupon “taken in the sense of falling back from enjoyment, to say that one of the dharma of the unshakable falls back from the pleasant abiding in the present dharma involves no fault of contradiction” — the contradiction dissolves in a fine division of the term.

The easy error: to read “the Buddha too falls back” as the Buddha’s losing something. The treatise says it very plainly: his kind is only “not being present.” In practice: the treatise’s way of dissolving contradictions is demonstrated here at its cleanest — first ask how many senses the word has, then ask which sense each person occupies. Hereafter, when two sentences of scripture seem to conflict, this move will often serve.

Here the treatise records besides an account of those who hold that there is no falling back: “all liberations without outflows are called unshakable,” which it closes with “explain it through as above; it should not be made an objection,” recording it without adjudication.

XIII. One Who Falls Back Assuredly Obtains It Again, and Does Not Do What Is Contrary to the Fruit

First ask: having fallen back, will he go on falling back until death? The treatise answers that he will not — “there is no dying in the interval after falling back from a fruit, because within a moment after falling back he assuredly obtains it again.” The scriptural testimony is: the learned noble disciples, though they fall back and lose right mindfulness, “are very soon able again to make what they have fallen into be exhausted, sunk away, ceased, and departed”; were it otherwise, “the fruit of cultivating the holy life would not be a secure place that may be entrusted with confidence” — a fruit of the holy life that might be lost past return would not amount to a place one could entrust oneself to.

The second question is finer: will he at the time of falling back do the deeds he would not do while abiding in the fruit? The treatise answers that he will not, on the ground that conscience predominates — “since conscience predominates, even at a time of temporary falling back he assuredly does not commit them, just as a strong man, though he stumbles, does not fall prostrate.” Yuanhui explains the two sentences together very fully: “in every falling back from a fruit, within a moment he assuredly obtains it again, and there is none who dies without obtaining it… Moreover the deeds he does not do while abiding in the fruit he assuredly does not commit even at the time of falling back from the fruit, since conscience and shame predominate — just as a strong man, though he stumbles, does not fall prostrate.”

The easy error: to read the simile of the strong man as “he will not stumble.” The point of the simile is that even having stumbled he does not go down altogether — tripping and sprawling are two different things. In practice: this section is really adding a safety rail to the great dispute that precedes it — even granting that there is falling back, both its extent and its duration are locked shut.

XIV. How Many Paths Does Refining the Faculties Take: Nine Uninterrupted and Nine of Liberation

First ask: how much labor does it cost to turn a lineage once? The stanza answers, “at the stage beyond training, refining the faculties takes nine uninterrupted paths and nine paths of liberation, because of long habituation; at the stage of training, one.” At the stage beyond training the turning of each nature takes nine uninterrupted paths and nine paths of liberation, as many as when the fruit of the arhat was first obtained; at the stage of training only one uninterrupted and one of liberation, as when the first fruit was first obtained. The reason is that “that dull-faculty nature comes of long-continued habituation, and a small exertion cannot make it turn” — dullness of faculty was habituated over long aeons and is not to be flipped over at one effort.

Its nature and what it rests upon: these paths are “in every case comprised only under the nature without outflows,” because “a noble person assuredly has no principle of turning his faculties by a path with outflows.” The body it rests upon is only the three continents of the human destiny, “because in the other places there is no falling back”; as for the level it rests upon, the stage beyond training extends to nine (access, the intermediate, the four meditative absorptions, and the three formless), and the stage of training only to six (the last three being taken away), because “there is no fruit of one in training comprised under a formless level.”

The easy error: to suppose that refining the faculties is taking up a new set of exercises. The treatise says it is “as when he obtained the fruit of worthiness” — the same path is used, only walked once again. In practice: this section gives a very solid reminder — the faculties one is born with can be turned, but at a cost equal to the original realizing of the fruit. Knowing the cost, one will not think refining the faculties too light a thing.

XV. The Nine Kinds of Noble Person at the Stage beyond Training

First ask: how many kinds of person are there in all at the stage beyond training? The stanza answers, “seven śrāvakas and two Buddhas; the difference is by the nine faculties.” The treatise explains: the five beginning with the dharma of falling back, with the unshakable divided into two (got by refining the faculties and had from the first, there being an earlier and a later), make seven kinds of śrāvaka; the pratyekabuddha and the great awakened one are called the two awakened ones, making nine persons in all. And the reason the nine are nine is “through the difference of the nine grades of faculty, the lowest of the low and the rest.”

Here is a good instance of gathering and dividing. Puguang adduces Correct Principle to prove the nine and the six against each other: “but another sūtra says that those beyond training are nine, from the first, the dharma of falling back, to the last, liberated in both parts. Their dharma of not falling back is comprised in our dharma of the unshakable; their two liberations are comprised within our six; and so the Abhidharma says there are only six” — the sūtra says nine, the treatise says six, and this is no contradiction but a difference of gathering and dividing.

The easy error: to read the nine persons as nine grades of higher and lower. Their difference lies in the faculties, not in the amount of affliction cut off. In practice: in reading numbered dharmas one often meets the same set of facts opened out into different numbers (six, seven, nine); and this passage gives the criterion — first ask by which gate this set has been divided; change the gate and the number changes.

XVI. The Seven Noble Persons: Four Gates Open Out Seven, the Real Items Being Only Six

First ask: how many kinds of noble person are there in all at the stages of training and beyond training, and on what are they divided? The treatise lists seven names — one who follows through faith, one who follows the dharma, one resolute through faith, one attained through view, one who witnesses with the body, one liberated through discernment, one liberated in both parts — and gathers them under four gates: “by difference of preparation the first two kinds are established”; “by difference of faculty the next two kinds are established”; “by the obtaining of the absorption of cessation the name one who witnesses with the body is established”; “by difference of liberation the last two kinds are established.”

Seven names but six items. The treatise at once explains: on the path of seeing there are the two persons, one who follows through faith and one who follows the dharma; on reaching the path of cultivation two other names are established, one resolute through faith and one attained through view; and on reaching the stage beyond training two names again, liberation dependent on circumstance and liberation not dependent on circumstance — so that “though the names are seven, the separate items are only six,” the stanza showing the reason by “since on each of the three paths there are two.” This is one person changing his name three times at three stages, not seven different persons.

The treatise also demonstrates in passing a multiplying of numbered dharmas: one who follows through faith becomes three as reckoned by faculty, five by lineage, fifteen by path, seventy-three by freedom from defilement, and nine by the body he rests upon; and the five multiplied together “come together to one hundred forty-seven thousand eight hundred twenty-five kinds.”

The easy error: to take the seven noble persons for seven classes of people unrelated to one another. In practice: this section together with the one before it is the place where this fascicle teaches most concentratedly “the difference between a name and an item” — names may be established gate by gate, but of items there are only so many. Whenever one reads the numbered dharmas of Abhidharma, first ask whether it is a name or an item.

XVII. Liberated in Both Parts and Liberated through Discernment

First ask: all being arhats alike, why is there a division into both parts and discernment? The criterion is a single one — whether the absorption of cessation has been obtained. “Those arhats who obtain the absorption of cessation are called liberated in both parts, because by the force of discernment and of absorption they are liberated from the afflictions and from the obstruction to liberation”; “the rest, who have not obtained the absorption of cessation, are called liberated through discernment, because by the force of discernment alone they are liberated from the obstruction of the afflictions.” What is meant by “both” is that discernment and absorption are both there; and the obstructions departed from are two as well — the obstruction of the afflictions and the obstruction to liberation (the obstruction of absorption).

Yuanhui puts this layer more penetratingly: “those among the six kinds of arhat who obtain the absorption of cessation are called liberated in both parts, for first by the force of discernment they are liberated from the obstruction of confusion, and further they obtain the absorption of cessation and are liberated from the obstruction of absorption, and hence are called ‘both’; ‘the rest are called liberated through discernment’… because by the force of discernment alone they are liberated from the obstruction of the afflictions, and not being free of the obstruction of absorption they cannot be called ‘both.’”

The easy error: to suppose that liberated through discernment means “liberated but not thoroughly.” As to the obstruction of the afflictions the two are alike; the difference lies in the obstruction of absorption. In practice: this point explains why the treatise, in speaking of merits, again and again singles out the absorption of cessation — it is the measuring rod by which the inside of the stage beyond training is divided.

XVIII. What Is Meant by “Fullness”

First ask: at the same stage of training, why is one man alone called full? The treatise answers, by three causes — faculty, fruit, and absorption. “One in training alone obtains the name of fullness at the stage of training, and that by three causes complete, namely faculty, fruit, and absorption.” The treatise then ranges five combinations one by one: by faculty only (one attained through view, not yet free of the defilement of desire); by fruit only (one resolute through faith who is a non-returner but has not obtained the absorption of cessation); by faculty and fruit (one attained through view who is a non-returner but has not obtained the absorption of cessation); by fruit and absorption (one resolute through faith who has obtained the absorption of cessation); and by all three complete (one attained through view who has obtained the absorption of cessation). And it says expressly that at the stage of training there is no obtaining the name of fullness “by absorption only” or “by faculty and absorption only.”

Beyond training there is one cause less. “Those beyond training alone obtain the name of fullness at the stage beyond training by the two, faculty and absorption,” because at the stage beyond training there is no fullness that is “not a fruit,” and so no separate item of fullness through fruit is established. Yuanhui here gives the most memorable table of correspondences: “first, fullness through faculty, that is, being of keen faculty; second, fullness through fruit, that is, having attained the non-returner; third, fullness through absorption, that is, having obtained the absorption of cessation.”

The easy error: to read “fullness” as the fullness of perfect Buddhahood. The fullness here is completeness as compared with others at the same station, not an ultimate sense. In practice: this section is the first place in the fascicle where the three variables of faculty, fruit, and absorption are set side by side and combined — and the later analyses of the four paths, the four modes of penetrating progress, and even the thirty-seven limbs all use the same manner of thinking.

XIX. All Paths, Spoken of in Brief, Are Only Four

First ask: the road of cultivation has a thousand threads; can it be gathered into a few? The treatise answers, into four: “the path of preparation is that after which the uninterrupted path arises. The uninterrupted path is that which can cut off the obstruction to be cut off. The path of liberation is that which arises first after the obstruction to be cut off has been liberated from. The path of further advance is the paths other than these three.” These four names are established with regard to four positions in the single business of cutting off obstruction — before the cutting, the cutting itself, the first arising after the cutting, and the rest.

The treatise at once glosses the sense of “path”: “what is the sense of ‘path’? It is the road to nirvāṇa, because riding upon it one can go to the city of nirvāṇa,” and a further explanation, “or else the path is what seeking rests upon, because resting upon it one seeks the fruit that is nirvāṇa.” And since a path means what can cut off, how do liberation and further advance also get the name of path? The treatise gives three reasons: that they are of a class with the path and turn into the higher grade; that the force of each earlier one reaches to each later; and that they can lead one to enter nirvāṇa without remaining support.

The easy error: to see the four paths as four stretches of time. They are divided with regard to functional position within one and the same cutting off of affliction, and are not the names of four stages. In practice: with this rod in hand, one may hereafter ask of any passage on the cutting off of affliction, “which of the paths is this?” — and the later parts of the treatise, in analyzing what is with and without outflows and in analyzing the levels, use it again and again.

XX. The Four Modes of Penetrating Progress: Pleasant and Painful by the Level, Slow and Swift by the Faculties

First ask: bound alike for nirvāṇa, why do some walk with labor and some with ease? The treatise sets out four modes of penetrating progress — painful and slow, painful and swift, pleasant and slow, pleasant and swift. There are two criteria: which level one rests upon and which faculties. “A path arising in dependence upon the four fundamental meditative absorptions is called pleasant penetrating progress, because it gathers in the factors, calm and insight are equal, and it turns spontaneously”; one arising in dependence upon the formless, access, or the intermediate absorption is called painful penetrating progress, “because it does not gather in the factors, calm and insight are unequal, and it must turn with labor.” Among these, in the formless insight is less and calm greater, and in access and the intermediate insight is greater and calm less — unbalanced at both ends. As for slow and swift, “dull faculty is named slow, keen faculty swift.”

The easy error: to read “painful” as undergoing suffering. The painful here is labor in walking the path, meaning that calm and insight are unequal and one must force oneself; it has nothing to do with painful feeling. In practice: this section really gives a very serviceable table for self-examination — whether calm and insight are in balance decides whether the road goes smoothly; whether the faculties are keen or dull decides whether it goes fast. Neither can stand in for the other.

XXI. The Thirty-Seven Factors of Awakening and the Sense of “Awakening”

First ask: why are the thirty-seven limbs of the path called factors of awakening? The treatise first lists the names — the four foundations of mindfulness, the four right abandonings, the four bases of supernormal power, the five faculties, the five powers, the seven limbs of full awakening, the eight limbs of the noble path — and then glosses the sense: “the knowledge of exhaustion and the knowledge of non-arising are called awakening,” and “the thirty-seven dharmas conduce toward awakening, and therefore all are called factors of awakening.” So “factor” has the sense of conducing toward, not the sense of a part.

Awakening is again divided into three according to the person. The treatise says that “according to the one who awakens, three awakenings are separately established: first the awakening of a śrāvaka, second the awakening of a pratyekabuddha, third unsurpassed awakening”; and that they are called awakening “because ignorance and the sleep of ignorance have both been forever cut off, and because he knows as it really is that what was to be done has been done and is not to be done again” — the one being the cutting off entire, the other the knowing for oneself that the work is finished. These two sentences answer exactly to the knowledge of exhaustion and the knowledge of non-arising.

The easy error: to take the factors of awakening for a term belonging to the Mahāyāna alone. Here the three awakenings expressly extend to all three vehicles, and the awakening meant is just the knowledge of exhaustion and the knowledge of non-arising. In practice: hold fast the point that “awakening = the knowledge of exhaustion and the knowledge of non-arising,” and the treatise will use it once more further on when it speaks of the entity of right knowledge — “as for the entity of right knowledge, it is as was said above of awakening.” Before and after, it is one rod.

XXII. Thirty-Seven in Name, but Only Ten in Real Entity

First ask: are the thirty-seven dharmas thirty-seven different things? The treatise answers that they are not — “though the names of these factors of awakening are thirty-seven, the real entities are only ten, namely discernment, diligence, and the rest.” It then subsumes them item by item: discernment comprises eight (the four foundations of mindfulness, the faculty and the power of discernment, the limb of awakening that is the discernment of dharmas, right view); diligence comprises eight (the four right abandonings, the faculty and the power of vigor, the limb of awakening that is vigor, right vigor); absorption comprises eight (the four bases of supernormal power, the faculty and the power of absorption, the limb of awakening that is absorption, right absorption); confidence comprises two; mindfulness comprises four; and the rest — joy, equanimity, pliancy, moral discipline (right speech, right action, right livelihood), and applied thought (right intention) — one each. The total, then, is the five faculties and five powers beginning with confidence, with joy, equanimity, pliancy, moral discipline, and applied thought added on top.

Yuanhui numbers the ten items one by one and lists what each comprises, the handiest of all for a beginner: “though the names are thirty-seven, the real entities discussed are only ten: first discernment, second diligence, third absorption, fourth confidence, fifth mindfulness, sixth joy, seventh equanimity, eighth pliancy, ninth moral discipline, tenth applied thought.”

The treatise at once records a divergent account alongside: “the Vibhāṣā masters say there are eleven, for bodily action and verbal action are not mingled with each other, and so moral discipline is divided into two, the remaining nine being as before” — opening moral discipline into the discipline of bodily action and of verbal action. The two accounts are recorded together, without adjudication.

The easy error: to read the subsuming as meaning that “the thirty-seven are a fiction.” The names are not fictions — one and the same discernment functions differently at different stations, and so different names are established. In practice: this is the most typical demonstration in the treatise of “many names, few items”; read numbered dharmas without asking after the entities and the numbers will lead you by the nose.

XXIII. Why the Foundations of Mindfulness Are Called Discernment, the Right Abandonings Diligence, and the Bases of Supernormal Power Absorption

First ask: since the entities of the three groups beginning with the four foundations of mindfulness comprise all preparatory wholesome dharmas whatever, why does each hang upon a single name? The stanza answers, “the four foundations of mindfulness, the four right abandonings, and the four bases of supernormal power are said to be discernment, diligence, and absorption according to the portion that predominates; in truth they are all the preparatory wholesome dharmas.” The treatise explains: these three groups of wholesome dharmas “as to real entity comprise all preparatory wholesome dharmas whatever; and yet, according to the wholesome root that predominates within their own class, they are said in order to be discernment, diligence, and absorption” — the name is established upon the portion that predominates, not upon the whole.

Each of the three names has its own gloss. For the name “foundation of mindfulness,” the Vibhāṣā masters say “discernment is held by the force of mindfulness and so abides,” while the author gives another correct explanation — “in truth of principle it is by discernment that mindfulness is made to abide upon its object, because only one who sees as things really are can note them clearly.” Puguang marks the difference between the two: “the answer: discernment is named a foundation of mindfulness, the name taken from its cause… the author’s correct explanation: discernment is named a foundation of mindfulness, the name taken from its result.” The name “right abandoning” is given because this diligence “can cut off indolence,” or it is named “right excelling,” because “among the properly holding and spurring on of body, speech, and mind, this is the most excellent.” And the name “base of supernormal power” is given “because the various wondrous merits rest upon it.”

At the bases of supernormal power the treatise refutes one account with its own hand. Certain other masters say that “the supernormal is precisely absorption, and the base is desire and the rest”; the treatise refutes them with two objections: that then the entities of the factors of awakening would be thirteen (desire and mind being added), and that this contradicts what the sūtra says — the sūtra saying expressly that “here the Buddha calls the result of absorption the supernormal, and calls the concentration produced by desire and the rest the base.”

The easy error: to read this section as a study of terms. What it really demonstrates is a general rule — the name of a numbered dharma is for the most part taken from the portion that predominates. In practice: whenever you meet a numbered dharma whose name and item seem not to answer to each other, first ask “from which portion is this name taken?”, and most often it will come clear.

XXIV. Why Confidence and the Rest Are First Faculties and Then Powers · and Why the Sequence Is What It Is

First ask: the same five dharmas — why called faculties first and powers afterward? The treatise answers with two senses: “because these five dharmas are divided as earlier and later according to the low and the high grade, and again according to whether they can or cannot be overborne.” At the low grade they are named faculties, at the high grade powers; while they can be overborne they are named faculties, when they cannot be overborne, powers.

Ask again: on what ground is the sequence confidence, vigor, mindfulness, absorption, discernment ranged just so? The treatise gives a chain of cause and effect: “with regard to cause and effect he first raises a mind of confidence and, for the sake of the fruit, cultivates the cause; next he raises vigor; through vigor mindfulness abides upon its object; through the holding force of mindfulness the mind thereupon obtains absorption; and because the mind has obtained absorption he can know as things really are” — confidence rises and practice follows, practice endures and mindfulness and absorption follow, mindfulness and absorption come and the real is seen. So this sequence is not a convenience of listing but the actual order of practice.

The easy error: to see faculties and powers as two different sets of dharmas. They are two intensities of the same five. In practice: this point may be used as a self-examination — if there is confidence without vigor, or vigor without mindfulness, the chain breaks just there; the order of the five is itself a checklist for the course of training.

XXV. The Seven Groups of Factors of Awakening Predominate in Turn by Station

First ask: are the thirty-seven limbs cultivated all at once, or in order? The stanza answers, “at the stage of initial undertaking, at the stage of the factors conducive to penetration, and at the stages of the path of cultivation and the path of seeing, the seven groups beginning with the foundations of mindfulness should be known to predominate in order.” The treatise assigns them station by station: at the stage of initial undertaking the function of discernment is excellent, and so the foundations of mindfulness predominate; at the stage of warmth the function of diligence is excellent, and so the right abandonings predominate; at the stage of the summit the function of absorption is excellent, and so the bases of supernormal power predominate; at the stage of receptivity the wholesome roots are firm, and so the faculties predominate; at the stage of the highest worldly dharma he is not overborne by afflictions or by worldly dharmas, and so the powers predominate; at the stage of the path of cultivation he is near to awakening, and so the limbs of awakening predominate; at the stage of the path of seeing he turns swiftly, and so the limbs of the path predominate.

What matters is the reminder the treatise adds next: the sūtras “speak according to the increase of number, seven first and eight after; this is not the sequence of cultivation” — the text ranges them by increasing number (the seven limbs of awakening before, the eight limbs of the path after), and this is not the order of practice. The treatise also distinguishes in passing two pairs of name and item: among the eight limbs of the path, right view is both the path and a limb of the path; among the seven limbs of awakening, the discernment of dharmas is both awakening and a limb of awakening.

Here the treatise records a divergent account alongside: certain other masters, not refuting the sequence of the sūtras, state it in accordance with them, beginning from the foundations of mindfulness that subdue the mind and going on through the faculties, the powers, the limbs of awakening, and the limbs of the path, and adducing two sūtras in evidence. The two accounts are recorded together, and the treatise turns to what follows with “such being the sequence stated according to the stages at which each predominates, as for the truth of the principle it should be as stated.” Yuanhui’s plain explanation of this passage may be consulted — “at this stage of initial undertaking he can examine and illumine in detail the four objects beginning with the body, and since the function of discernment is excellent, the foundations of mindfulness are said to predominate.”

The easy error: to read “predominate” as “only this group is cultivated.” Predominating is a function’s standing out, not the others’ being left uncultivated. In practice: this section gives a map of the shifting center of gravity in training — with the same thirty-seven limbs, the place where effort belongs differs at different stages.

XXVI. Which Extend to What Is with Outflows, and Which Are without Outflows Only

First ask: are all thirty-seven limbs dharmas without outflows? The stanza answers, “the seven limbs of awakening and the eight limbs of the path are without outflows throughout; the three fours and the five faculties and powers all extend to both kinds.” The treatise explains the reason: the seven limbs of awakening and the eight limbs of the noble path are “established only at the stages of the path of cultivation and the path of seeing” — their very establishing is bound to the noble stations; and as for right view and the like being found in the world too, “yet those do not obtain the name of limbs of the noble path.” The rest — the four foundations of mindfulness, the four right abandonings, the four bases of supernormal power, and the five faculties and five powers — extend to what is with and what is without outflows.

Yuanhui unfolds the stanza’s abbreviations one by one: “‘three fours’ means first the four foundations of mindfulness, second the four right abandonings, third the four bases of supernormal power; ‘the five faculties and powers’ means the five faculties and the five powers” — just the place where a reader is apt to stick in reading the stanza.

The easy error: to suppose that whatever is called a factor of awakening is without outflows. Quite the contrary: most of them extend to both. In practice: this point makes one thing clear — one and the same dharma is with outflows when cultivated at an ordinary station and without outflows when raised at a noble one; the name does not change while the nature turns with the station.

XXVII. The Increase and Decrease of the Thirty-Seven Limbs at the Several Levels

First ask: are the thirty-seven limbs complete at every level? The treatise answers that only the first meditative absorption has them complete, the rest each having something taken away: at the level of access the limb of awakening that is joy is excepted, because “at a threshold level one turns by exertion, and one still has misgivings about the dharmas of the level below”; in the second meditative absorption right intention is excepted, “because in that meditative absorption there is no longer applied thought” — these two levels have thirty-six each. In the third and fourth meditative absorptions and the intermediate absorption, joy and applied thought are both excepted, and each has thirty-five. At the first three formless levels the three limbs of moral discipline are excepted together with joy and applied thought, and each has thirty-two. The desire realm and the summit of existence except the limbs of awakening and the limbs of the path, and each has twenty-two, “because there is nothing without outflows there.”

The easy error: to read the subtraction as meaning that the path cannot be accomplished at that level. What is subtracted is the limb that level does not have (a level without applied thought naturally has no right intention), not that the path cannot be accomplished. In practice: in reading this section one should reckon as one goes — thirty-seven less one gives thirty-six, less one again thirty-five, thirty-seven less five gives thirty-two; these numbers are themselves the method of verification. Yuanhui’s explanation of the numbers level by level may be read alongside.

XXVIII. The Four Purities of Realization

First ask: what is meant by a purity of realization? The treatise first lists the four — toward the Buddha, toward the dharma, toward the Saṅgha, and the noble moral discipline; and then glosses the name: “because he awakens to and knows the principle of the four noble truths as it really is, it is called realization,” while “the three jewels in which he has confidence and the excellent śīla are both called pure, because they are apart from the stain of unbelief and the stain of breaking moral discipline.” So a purity of realization is a pure confidence got through realization, not confidence of an ordinary sort.

The order of obtaining has its niceties. At the stage of the path of seeing, on seeing each of the first three truths he obtains only the purity of realization toward the dharma and that of moral discipline; only at the stage of seeing the truth of the path does he obtain besides those toward the Buddha and the Saṅgha. The dharma in which he has confidence is again divided as general and particular — the general extends to all four truths, the particular is only the whole of the three truths together with the path of a bodhisattva or a pratyekabuddha. The real items are only two: the three, Buddha, dharma, and Saṅgha, have confidence for their entity, and the noble moral discipline has moral discipline for its entity; and all four are without outflows only, “because a dharma with outflows is not a purity of realization.”

The treatise also states the order in which they arise on emerging from contemplation: first he has confidence that the World-Honored One is perfectly and fully awakened; next that the true dharma and discipline are well spoken; last that the noble stage is of excellent conduct; and finally, because the mind is pure, pure śīla is given rise to, and so śīla is spoken of as fourth. The treatise gives two similes — the three jewels as a good physician, a good medicine, and one who nurses the sick; or as a guide, a road, fellow merchants, and the vehicle one rides.

The easy error: to take the four purities of realization for four kinds of confidence to be cultivated. They are four portions accomplished of themselves after the truths are seen, and the order too is a sequence on the side of the fruit. In practice: to examine oneself by the four similes is the handiest way — with physician, medicine, and nurse, if any one of the three is lacking the illness is not removed; and moral discipline is the medicine one can take oneself.

XXIX. Right Liberation and Right Knowledge: Why They Are Not Established at the Stage of Training

First ask: the sūtra says that one in training accomplishes eight limbs and one beyond training ten; why are right liberation and right knowledge not established at the stage of training? The treatise answers: “at the stage of training there are still bonds remaining that have not been liberated from, and so there is no limb of liberation” — adding a sentence of blocking: “it is not that one may be called liberated by being apart from some small portion of the bonds; nor that a knowledge of liberation may be established where there is no entity of liberation.” One beyond training has escaped the bonds and can besides give rise to two knowledges that know liberation, and “only through the manifestness of these two may these two limbs be established.”

The entity of liberation is twofold. “The entity of liberation is twofold, conditioned and unconditioned. Conditioned liberation is the resolve of one beyond training; unconditioned liberation is the cessation of all confusion.” The name “limb” must be established upon what is conditioned, and so conditioned liberation is named a limb of one beyond training; and the liberation comprised under a limb is again the liberation of mind and the liberation through discernment spoken of in other sūtras, that is, the aggregates of liberation.

Here the treatise raises an objection against itself and records a divergent account: since the sūtra says “the mind is liberated in becoming free of the defilement of greed, and liberated in becoming free of the defilement of hatred and delusion,” it appears that the aggregate of liberation is not resolve alone. The treatise answers with the portion that is cultivation (what is not complete being made complete, what is complete being cultivated with desire and diligence for the sake of holding it fast); and records besides an account of certain other masters — “by the force of true knowledge greed, hatred, and delusion are dispelled, and this very being of the mind apart from stain is called the aggregate of liberation.” As for the entity of right knowledge, “it is as was said above of awakening, namely the knowledge of exhaustion and the knowledge of non-arising spoken of above.”

The easy error: to see the difference between eight limbs and ten as a difference in the quantity of merit. The difference is in whether a limb can be established at all, and whether it can depends on whether the bonds are exhausted. In practice: this section and section XXI interlock head and tail — awakening is the knowledge of exhaustion and of non-arising, and right knowledge is the knowledge of exhaustion and of non-arising too; the net of terms in this treatise bites into itself at every point.

XXX. The Mind beyond Training Is Liberated from Obstruction at the Very Time of Its Arising

First ask: in which time exactly is the mind liberated? The treatise adduces the account of this treatise itself: “the first mind beyond training, at the time of its arising in the future, is liberated from obstruction.” And the obstruction meant is the “acquisition” of the afflictions — “because that can block this mind from arising.” Whereupon there is a pair of correspondences: at the stage where the vajra-like concentration is properly ceasing, that acquisition is properly being cut off, and the first mind beyond training, at the stage of properly arising, is properly liberated; at the stage where the vajra-like concentration has already ceased, that acquisition has already been cut off, and the first mind beyond training, at the stage of having already arisen, is named already liberated.

The treatise at once adds three questions: minds beyond training not yet arisen, and mundane minds, are at that time also named liberated, only that here it is the one settled to arise that is spoken of, “because at that time it is properly proceeding in the present time.” From what are mundane minds liberated? From that same obstruction which blocks the mind from arising. At the stage where it is not yet liberated, does this mind then not arise? The answer: “although it does arise then, it is not as it is now” — the difference being that then it was “together with the acquisition of the afflictions,” whereas if it arises hereafter there will be “no acquisition of confusion together with it.”

The easy error: to imagine “liberation” as the mind’s escaping out of somewhere. What is liberated from here is the acquisition of the afflictions, that blocking; the obstruction is removed and the mind is able to arise. In practice: this section demonstrates the fineness of the treatise’s analysis of times and stages — properly arising, already arisen, and not yet arisen each have their name, and to confound one with another is to be wrong throughout. Yuanhui’s explanation may be read alongside.

XXXI. The Path Cuts Off Obstruction Only at the Stage of Properly Ceasing

First ask: at which stage does the path cut off obstruction? The stanza answers, “the path, only at the stage of properly ceasing, makes that obstruction be cut off.” The treatise explains the wording: “to say ‘the stage of properly ceasing’ shows that it stands in the present; to say ‘properly arising’ shows that it is in the future time” — so the path can cut off obstruction only at the time of properly ceasing, and at the other stages there is no function of cutting off obstruction.

The treatise also draws a contrast on purpose: in this it is unlike liberation. Liberation may extend to what has not yet arisen, “because what has arisen and what has not are alike in point of being apart from obstruction”; but the cutting off of obstruction is not so, and lasts a single moment only. Fabao adduces Correct Principle to add a further layer: “although the minds of one in training are also liberated from obstruction at the stage of arising, the treatise speaks only of the first mind beyond training being liberated at the time of its arising, because it goes by the liberation realized in cutting off without remainder.”

The easy error: to see the cutting off of obstruction and liberation as two ways of stating one and the same thing. The treatise here parts them expressly — the cutting off of obstruction is confined to the stage of properly ceasing, whereas liberation extends to what has arisen and what has not. In practice: this is the most finely analyzed section in the fascicle, and its value lies not in its conclusion but in its method: the different sides of one and the same matter each have their own time and stage.

XXXII. The Element of Abandoning, the Element of Dispassion, the Element of Cessation

First ask: the sūtra speaks of three elements — abandoning, dispassion, cessation; what are their entities? The treatise answers, “the three elements beginning with abandoning are the unconditioned liberation spoken of above, divided out to be severally their entities” — that is, the unconditioned liberation spoken of before divided into three portions. The division is: “as for the element of dispassion, it is dispassion toward craving only. As for the element of abandoning, it is the abandoning of the remaining fetters. As for the element of cessation, it is the cessation of the remaining things in which the latent tendencies of craving and the rest increase by attachment.”

So the three are not three different unconditioned things, but one and the same unconditioned liberation with three names established according to what is counteracted — dispassion said with regard to craving, abandoning with regard to the remaining fetters, cessation with regard to the things in which the latent tendencies increase by attachment.

The easy error: to read the three elements as three places or three stages. They are the names of three divisions of one unconditioned dharma. In practice: this section carries on the line of “many names, few items” from before — one entity, with different names established according to what is counteracted. This is the third of the treatise’s standing ways of establishing names (the first two being by what predominates and by the station in the sequence).

XXXIII. The Four Alternatives of Disgust and Dispassion (the End of the Chapter)

First ask: if one can be disgusted, must one be able to become dispassionate? The treatise answers that one need not, and adjudges it by four alternatives. There are two criteria: disgust takes only suffering and origin as its object — “only the receptivities and knowledges arising with suffering and origin as their object are called disgust; the rest are not”; dispassion extends to all four truths and is measured by the ability to cut off confusion — “among the receptivities and knowledges arising with the four truths as their object, all that can cut off confusion obtain the name of dispassion.” The one limits the object and the other limits the function, and their widths and narrownesses cross, and so four alternatives are made.

The four alternatives are: taking suffering and origin as object without cutting off confusion, disgust and not dispassion; taking cessation and the path as object and cutting off confusion, dispassion and not disgust; taking suffering and origin as object and cutting off confusion, both disgust and dispassion; taking cessation and the path as object without cutting off confusion, neither disgust nor dispassion.

At the end the treatise adds an instance: for one who becomes free of the defilement of desire first and sees the truths afterward, among his dharma-receptivities and knowledges all that are comprised in the path of preparation, the path of liberation, and the path of further advance do not cause confusion to be cut off — “because the confusion has already been cut off, and because they are not the counteragent that cuts off.”

The easy error: to read disgust and dispassion as a relation of feeling to result. What the treatise sets up is two different criteria (object and function), and the four alternatives are only the exhaustive crossing of the two. In practice: the fourfold method is the tool most used in Abhidharma — when two concepts are tangled, first establish the criterion of each and then set out four alternatives, and the relation declares itself. That the whole chapter closes with it hands the reader over to the gate of knowledge in the next chapter.

Further Resonances: The Mirror of the Four Commentaries

The above is the sense of the text of this fascicle itself. Below a separate section is set up, gathering four matters: the adjudications of the four houses and where they differ; Saṅghabhadra’s refutations and restatements; the divergences of Paramārtha’s earlier translation; and coordinates for parallel reading. A mirror is not the treatise, but only another way of reading the same passage; both are recorded without adjudication, and each is left in its own place.

1. Where the Four Houses Differ

This unit is the first in the whole book for which all four houses are present. Shentai’s commentary is a fragment: of the four fascicles on the Chapter of the Worthy and the Noble only Fascicle Seventeen survives, and it happens to cover this fascicle entire; for the other three units Shentai is wholly wanting, and the want is marked as it stands and not filled out with text from another fascicle. Shentai’s fascicle heads itself “Chapter Six, Analysis of the Worthy and the Noble, part three” while marking what it explains as “Fascicle Twenty-Five of the treatise,” and at its end bears the title “Commentary on the Abhidharmakośa, Fascicle Twenty-Five” — the fascicle numbers at head and end differ in three ways, and are recorded as they stand without adjudication.

The manner of the outline divides in two. Puguang’s eight subsections and Yuanhui’s are almost word for word the same; Fabao gathers the first three into one and so has six, his layers being the wider; Shentai sets up no general outline table but establishes an outline of question and answer passage by passage (citing what precedes, question, answer, pressing, adducing the stanza in answer, explaining the stanza). On the extent of “the eight matters of the chapter of the six lineages” itself the four houses do not differ at all; what differs is the coarseness or fineness of the outline. Shentai’s form is the oldest, his outline not yet settled into eight matters ranged side by side — this is a difference of form, not an omission.

Shentai leaves an independent piece of evidence for the stanza count. He marks of himself “from here below, the fourth, seven and a half lines of stanza explaining the six kinds of lineage” — seven and a half lines being exactly one hundred fifty characters, answering exactly to Root 58 through the first half of Root 65. This is a further piece of evidence for the stanza count passage by passage, lying outside the interlinear notes of the root-stanza edition.

On the author’s declaration the four houses use three forms of words and mean one thing. Puguang has “the author’s intent sides with the Sautrāntikas, and so he says that what they say accords with principle”; Fabao has “the author by his appraisal takes up the Sautrāntika explanation”; Shentai has “the author appraises it, saying: what they say accords with principle.” The three differ in fullness of wording and agree wholly in the criterion. Shentai has besides one turn of phrase worth noting: he cites Correct Principle directly and calls Vasubandhu “the Sūtra-master,” where Puguang and Fabao say “the Sautrāntikas” and “the author” — a slight difference in the angle of designation, one and the same person being meant. The same happens at the refutation of the entity of the bases of supernormal power: Puguang has “the author refutes the other masters,” Shentai “the Sūtra-master refutes them, saying.”

Yuanhui does not touch the schools at all in the dispute over falling back from a fruit. In this passage he explains only the doctrinal sense of the stanza text itself — “the lineage that is the dharma of the unshakable assuredly has no principle of falling back; and among the five before it, the one lineage of the dharma of falling back alone assuredly does not fall back, since it stands lowest” — neither marking the names of Sautrāntika and Sarvāstivādin nor adducing an assignment of schools. Set against Puguang’s marking of the school sentence by sentence and Fabao’s tabulating of three schools, it is in this passage that the characters of the three houses are most clearly parted. Recorded without adjudication.

Fabao alone sets out a table of three schools on falling back from a fruit. “He sets forth the Sarvāstivāda position. According to the Treatise on the Wheel of Tenets: for the Mahāsāṃghikas and the rest, the fruit of the stream-enterer admits of falling back and the fruit of worthiness does not. For the Sautrāntikas, neither the stream-enterer nor the fruit of worthiness falls back. For the Sarvāstivādins, the stream-enterer does not fall back and the fruit of worthiness does.” Puguang at the same place also adduces the Treatise on the Wheel of Tenets, in a single line of narrative. The assignments of the three schools in the two commentaries agree wholly, and Fabao’s table shows the outline the better. This belongs to the ancient masters’ adducing another treatise to settle an assignment of schools, and is not in this treatise, and so is recorded in this section.

Two instances of a one-character difference. In the simile of the butter jar, Puguang and Fabao make the character elided “holding,” while Shentai makes it two words, “this is made” — “as the world says in full ‘this is made a butter jar.’” And at the four alternatives of disgust and dispassion, Shentai reads “are called ‘condition’” where Puguang and Yuanhui read “disgust.” Both are recorded as they stand without adjudication, and not emended by conjecture.

On the three accounts of the Mahāvibhāṣā the three commentaries agree wholly. At the passage on the entities of the thirty-seven limbs, Fabao adduces: “the Mahāvibhāṣā, Fascicle Ninety-Five, has three accounts: one account agrees with this treatise; one account opens moral discipline into two and takes eleven dharmas as the entities; one master opens it into three, namely right speech, right action, and right livelihood. This treatise states them gathered, taking ten dharmas as the entities,” and what Yuanhui adduces runs the same way. The three houses have no divergence, differing only in fullness.

At the four purities of realization and at the three elements, all three commentaries adduce Correct Principle by way of supplement. Fabao adduces “the Treatise Conforming to Correct Principle says: with regard to what dharmas, and how, does one obtain the purity of realization toward the dharma? With regard to suffering alone: penetrating that there are only dharmas and no real sentient being, one gives rise to settled confidence”; and at the three elements, “Correct Principle, Fascicle Seventy-Two, says: as to the entity of the three elements, taken provisionally they differ, but taken as real things there is no difference at all.” This belongs to the commentators’ supplementing with another treatise what this treatise leaves unstated; there is no divergence of doctrine, and it is recorded only.

2. Saṅghabhadra’s Refutations and Restatements

Neither of Saṅghabhadra’s two treatises gives this chapter the name “Chapter, Analysis of the Worthy and the Noble” — the Treatise Conforming to Correct Principle has “Chapter Six, Adjudication of the Worthy and the Noble,” and the Treatise Revealing the Doctrine “Chapter Seven, Adjudication of the Worthy and the Noble,” the chapter number moved back by one. To search these two treatises by the Kośa’s chapter name must yield a zero.

The dispute over falling back from a fruit is the one place in this unit where three parties stand as a tripod. Saṅghabhadra cites Vasubandhu word for word and examines him: “now, upon examination, the Sūtra-master has not established his position well. One ought to press the inquiry: as to the two middle fruits obtained by a mundane path, has the seed of the confusion that obstructs them really been uprooted, or not? If it really has been uprooted and yet falling back is admitted, then falling back from the state of an arhat is established.” He separately marks two further houses: “therefore the Elder maintains that there is no falling back from and losing the fruit of the arhat, though he has neither reasoning nor scripture for it”; “the Masters of Distinction say this: on no noble path whatever is there falling back, and so the confusion cut off never arises again.” ⇒ The Sūtra-master, the Elder, and the Masters of Distinction all three maintain that there is no falling back, and only the Sarvāstivāda school itself maintains that there is. The lines of reasoning of the three are not the same (Vasubandhu appraising and taking up the Sautrāntikas’ two proofs from scripture and reason, the Elder saying flatly that there is neither reasoning nor scripture, and the Masters of Distinction pushing the range to every noble path), and Saṅghabhadra refutes them one by one and so returns to his own school. Recorded only, without adjudication.

In one place what Saṅghabhadra refutes is not Vasubandhu. On the reason “resting on no entity,” Saṅghabhadra adduces a passage of “certain other masters” who extend the reason of there being no falling back from the confusion abandoned through seeing to the confusion abandoned through cultivation and bound to the summit of existence, and then adjudges, “what they say is contrary to principle, since the force of the paths differs.” What is refuted is that extension and not Vasubandhu’s own sense. That one and the same gloss should be extended by different persons to different conclusions is here an instance.

In two places he only restates without setting up a divergent view. In the passage of the Sūtra of the Increase by Ten, Saṅghabhadra cites Vasubandhu’s own sentences word for word with no word of direct refutation to be found; and so likewise in the passage refuting the entity of the bases of supernormal power. This belongs to restatement of the recording kind, and is noted.

In one place what he appraises is the angle of interpretation and not the conclusion. At the Sūtra of the Simile of the Charcoal, Saṅghabhadra says, “examining the Sūtra-master’s intent, he takes it that in this sūtra the stage of training is spoken of first and the stage beyond training afterward; now one ought to examine carefully whether this may be relied on as settled — whether it is the word of the World-Honored One or the Sūtra-master’s own intent.” What he questions is whether the outline order “training first, beyond training afterward” issues from the sense of the sūtra itself or from Vasubandhu’s arrangement. What is contested is the reading, not the conclusion.

The entity of the aggregate of liberation is also restated. Saṅghabhadra reports Vasubandhu’s intent as “it is not resolve alone that obtains the name of this aggregate. If so, what is it? By the force of true knowledge greed and hatred and delusion can be forever dispelled, and the mind’s being apart from stain is called the aggregate of liberation,” agreeing with the same passage as recorded by Fabao and by Shentai.

One honest negative proposition. Within the range corresponding to this unit, the Treatise Revealing the Doctrine shows no sentence citing “the Sūtra-master” word for word. But all that this zero can prove is one thing — that within this range that treatise does not use the designation “the Sūtra-master”; it cannot be inferred from it that the treatise has no divergent view here. The Treatise Revealing the Doctrine is by form the terser, and word-for-word citation of the Sūtra-master is not what it cares for.

3. The Divergences of Paramārtha’s Earlier Translation

Paramārtha’s Expository Treatise on the Treasury forms a system of its own over the passage answering to this fascicle, and is besides the most evenly matched stretch of scene among all the fascicles yet collated — the six kinds of arhat, the earlier and later of the lineages, falling back from lineage and from fruit, the six lineages of those in training and of ordinary beings, the three kinds of falling back, the refining of the faculties, the nine kinds of those beyond training, the seven noble persons, fullness, the four paths, the thirty-seven limbs, the four purities of realization, right liberation and right knowledge, the three elements, and the four alternatives of disgust and dispassion — passage by passage the order answers.

Not one of the six lineage-names agrees wholly. Paramārtha has “first the dharma of falling and sinking, second the dharma of self-slaying, third the dharma of watching and guarding, fourth the dharma of abiding unshaken, fifth the dharma of being fit to penetrate, sixth the dharma of the indestructible.” What calls for care is the fourth, “the dharma of abiding unshaken” — what it answers to is Xuanzang’s dharma of abiding, not the dharma of the unshakable; Xuanzang’s dharma of the unshakable appears in Paramārtha as “the dharma of the indestructible.” The glosses of the two texts make it clear: Paramārtha’s fourth reads, “if a person is apart from the strongest cause and condition of falling and sinking, then although he does not constantly watch, he will assuredly abide in what he has obtained”; his sixth reads, “as for the dharma of the indestructible, such a person assuredly does not fall and sink as the former do.” To go by the look of the words and equate Paramārtha’s “dharma of abiding unshaken” with Xuanzang’s “dharma of the unshakable” is the pitfall most easily stepped into in this stretch.

One difference of glossing that runs through the whole passage. Xuanzang’s “possessed of an entity / resting on no entity” is throughout Paramārtha’s “possessed of kind / without kind” — “because the confusion abandoned through seeing is without kind,” expanded as “this self upon which they rest is assuredly nowhere at all, and so it is said to be without kind.” The doctrinal content indicated is wholly the same, the characters used wholly different.

The author’s declaration is plainer in Paramārtha. Xuanzang has “what they say accords with principle”; Paramārtha has “this sense is truly admissible” — which corroborates exactly Puguang’s marking, “his intent sides with the Sautrāntikas.”

The crossing of the names of the three elements is the one place in this fascicle that most calls for care. Xuanzang establishes the element of dispassion, the element of abandoning, and the element of cessation; Paramārtha establishes “the element of cessation, the element of freedom from desire, and the element of utter removal.” Compared by the context of the glosses, the correspondence is crossed: Xuanzang’s element of dispassion (dispassion toward craving) answers to Paramārtha’s element of freedom from desire (the cessation of desire); Xuanzang’s element of abandoning (abandoning the remaining fetters) answers to Paramārtha’s element of cessation (the cessation of the remaining confusions); Xuanzang’s element of cessation (the cessation of those things) answers to Paramārtha’s element of utter removal (the cessation of the other kind). Both texts have a name reading “the element of cessation,” and the two indicate different things — pair them by the look of the words and they must be mismatched.

The order of the list of ten things differs in the two texts. Xuanzang begins with discernment, Paramārtha with “confidence, vigor, mindfulness, concentration, discernment, joy, equanimity, and pliancy, moral discipline, and awakening,” beginning with confidence; and again Xuanzang’s diligence, absorption, and applied thought appear in Paramārtha as vigor, concentration, and awakening. The difference in order may reflect a difference in the recitational transmission upon which the two texts rested; this is not forced to an explanation here, but honestly recorded as a difference of transmission between the two.

There are besides several smaller divergences that may be read alongside: the personal name Godhika appears in Paramārtha as “the pure-lived Qutijia”; the sūtra title Sūtra of the Increase by Ten as Sūtra of the Ten Increases; the path of further advance as the path of further increase; the jewel of the Saṅgha as “the assembly of disciples in whom one has confidence.” In form, Xuanzang gives the whole stanza first and the prose after, while Paramārtha embeds his explanation sentence by sentence, so that one stanza of Xuanzang’s often appears in Paramārtha as two verses stated separately — as at all three of “those in training and ordinary beings likewise have six,” “the fullness of the one in training and the one beyond training,” and “the four alternatives of disgust and dispassion.” Both are recorded without adjudication.

4. Coordinates for Parallel Reading

What this fascicle establishes is a map of the differences within the noble persons. Its frame has only two variables: the keenness or dullness of the faculties, and whether absorption has been obtained. By these the six kinds of lineage are divided, the nine kinds of those beyond training are divided, the four gates of the seven noble persons are opened out, the fullness of the one in training and the one beyond training is adjudged, and the slowness or swiftness of the four modes of penetrating progress is settled. Having read the fascicle through and looking back, one finds that the treatise has not ranked arhats high and low at all — the afflictions cut off are one and the same portion, and what differs is only the supports and the speed with which one arrived there.

The second line for parallel reading is name and item. Thirty-seven names have only ten entities; seven noble persons have only six items; the three elements are only three portions of one unconditioned liberation — three times over this fascicle demonstrates one and the same technique: establishing a name by the portion that predominates, by the station in the sequence, and by what is counteracted. Whoever reads the numbered dharmas of Abhidharma asking first “is this a name or an item?” will most often not be led by the nose by the numbers.

The third line is left for the next chapter. The whole chapter closes with the four alternatives of disgust and dispassion, and the receptivities and knowledges those four alternatives employ are exactly what the next chapter, “Analysis of Knowledge,” will take up as its own subject. The last sentence of this fascicle says of “one who becomes free of the defilement of desire first and sees the truths afterward” that his dharma-receptivities and knowledges do not cause confusion to be cut off — and by that it has already drawn the reader’s eye away to the gate of knowledge.