俱舍論 緣起別釋·四有·四食Abhidharmakośabhāṣya · Dependent Origination in Detail · The Four Existences · The Four Nutriments

Before reading this fascicle

The preceding fascicle first set out the dependent origination by stages transmitted by the Vaibhāṣikas — the twofold causality over three lives and the four kinds of dependent origination — together with intermediate existence and entry into the womb, and recorded that at this place the author set down the two words “transmitted and allowed”; Saṅghabhadra read those two words as a mark of the author’s disbelief, and the word-for-word citations and the recording together of the two parties are in the apparatus and Further Resonances of the preceding unit. When this fascicle is reached, the manner of writing suddenly changes: the whole chain is no longer discussed, and instead he turns back and takes the links apart one by one. The very first sentence of the fascicle is “what is the sense of ignorance?”, without so much as a transitional formula.

What is to be noticed is this: he does not take all twelve links apart. Fabao’s Commentary counted the divisions of this passage and says, “within it are four: first, explaining ignorance; second, explaining name-and-form; third, explaining the link of contact; fourth, explaining the link of feeling,” while as for the remaining eight links, “they are explained by referring to the remaining text” — only four links are taken apart, and the remaining eight are expressly said to be treated elsewhere. This is not laziness. The link of consciousness and the link of the six sense fields were stated in the Chapter on the Elements; the links of formations and of existence are reserved for the chapter on action; the links of craving and of appropriation are reserved for the chapter on the latent tendencies; and birth, old age, and death are themselves the aspect of the result. So the links that really need a separate explanation within the chapter on dependent origination are exactly these four. A reader who expects this fascicle to be “a detailed exposition of the twelve links” will be disappointed at the first line; one who knows it to be “a supplementary lecture on four links” will read it smoothly.

The four links taken apart, the author goes on to do three things: first he uses eight external similes to gather the twelve links into the three kinds of affliction, action, and the matter of results; next he distinguishes the defiled and the undefiled among the four existences; then he asks one question, “upon what do these sentient beings depend to abide?”, and out of it comes the long passage on the four nutriments. Last he analyzes which consciousness and which feeling belong to death and birth, and closes with the three classes.

So the skeleton of this fascicle is in fact fourfold: the names and senses of the links → the classifying of the links → the defiled and undefiled among the existences → abiding and passing away. The first half distinguishes names, the second half distinguishes things. The passage on the four nutriments is the longest stretch of prose commentary in the fascicle, over fifteen hundred characters; the passage on the simultaneity of contact and feeling comes next, over nine hundred. The doctrinal density of both far exceeds that of the half-stanza to which each is attached — the stanza is only the hook, and the prose commentary is the body itself, and this fascicle displays that trait most plainly of all.

I. Ignorance Is Not “the Absence of Clear Knowing”

First ask: since ignorance is called the “absence” of clear knowing, might it not be simply an “absence” — the empty gap where clear knowing is not? The author sets three objections in a row as the fascicle opens precisely in order to force this question out. The first objection: if it is said that “its entity is not clear knowing,” then the eye, ear, nose, tongue, and body are not clear knowing either — would they not all become ignorance? The second objection: then let it be restated as “the absence of clear knowing.” The third objection: if it is the nonexistence of clear knowing, then “the entity of ignorance would be nonexistent” — and how could something that does not exist act as a condition to give rise to the link of formations?

Answer: the answer the stanza gives is ten characters only, “ignorance is what clear knowing counteracts, as with ‘not a kinsman,’ ‘not true,’ and the rest.” The key lies in the words “what is counteracted.” Ignorance is not “anything whatever other than clear knowing,” nor is it “the nonexistence of clear knowing”; it is the very thing that clear knowing counteracts head-on. For this the author adduces three ready-made usages of language. An enemy is called “not a friend and kinsman” because he “stands opposed” to friends and kinsmen, and not because he falls outside the list of them; hence “it is not other than a friend and kinsman, nor the absence of a friend and kinsman.” Speaking the truth is called “true,” and then “truthful speech is called true, and the false and deceitful speech that it counteracts is called not true” — deceitful speech is the kind that truthful speech counteracts, and not “every utterance other than a true one.” The word “and the rest” in the stanza then takes in “not a dharma,” “not the sense,” and “the nature of not being a real thing,” the usage being wholly the same.

Clearing a misreading: readers most easily read this passage as a game with language, thinking that the author is playing with the word “not.” In truth he is guarding a line — whatever can act as a condition must be a dharma possessed of an entity. The twelve links are a chain of cause and result, and if the first link were an “empty nothing” the whole chain would hang in the air. Hence the author immediately asks himself “how is it known that it is so?” and answers himself, “because it is said that formations have it as condition”: since the sūtra says that formations have ignorance as their condition, ignorance must be possessed of an entity.

Where it lands: the criterion established in this passage can be put to use again and again throughout the treatise. Whenever a name of the form “not such-and-such” or “no such-and-such” is met with, first ask, “is this an exclusion, or does it point to something further?” Yuanhui’s Verse Commentary puts it most plainly here, explaining this stanza thus: “as for ‘ignorance is what clear knowing counteracts,’ it shows that it has a real entity, that is, that this ignorance does not comprehend the four truths,” and pointing out under the simile that “as for ‘not other than a friend and kinsman,’ it means that the enemy is called not a friend and kinsman” — dividing off clearly the two works of the simile (not to be confused with the eye and the rest, and really to have an entity of its own).

II. Four Reasons: Why Ignorance Must Have an Entity of Its Own

First ask: the preceding section only laid down a definition; on what ground is that definition said to be right? The author answers very solidly — “there are, besides, reliable proofs” — and then gives four, all taken from established usages in scripture.

Answer: the first, “because it is stated as fetter and the rest.” The sūtra enters ignorance among such names for afflictions as “fetter,” “bond,” “latent tendency,” and “outflow,” “yoke,” and “flood”; the eye and the other faculties are not afflictions, and “neither the eye and the rest that are other, nor what is wholly without an entity to be found, can be spoken of as fetter, bond, and things of that kind” — an empty gap that is nothing at all cannot be ranked in the list of afflictions. The second, “because it is neither bad discernment nor view.” Some say that ignorance is just bad discernment, and set up their account with a simile from ordinary life: “a bad wife or child is called having no wife or child” — a bad wife, lacking the virtues of a wife, gets the name “no wife”; and by the same reasoning bad discernment, lacking the virtue of clear knowing, “ought likewise to be named ignorance.” The author does not accept the simile and refutes it straight off with one sentence: “that is not ignorance, because there is view in it” — bad discernment contains view within it, and view is of the nature of investigating keenly and deciding, and is not ignorance. Moreover “the several defiled discernments are called bad discernment, and since there is view within them they are not ignorance,” from which it is evident that the two are not equal. The third, “because it is associated with view.” “If ignorance were discernment, it ought not to be associated with view, since two entities of discernment are not associated with each other” — within a single moment of thought two discernments do not arise together, and since the sūtra grants that ignorance is associated with view, ignorance is not discernment. The fourth, “because it is said to defile discernment.” The sūtra says, “greedy desire defiles the mind so that it is not liberated; ignorance defiles discernment so that it is not pure”; discernment cannot defile discernment, just as greed must be unlike mind in order to defile mind.

Clearing a misreading: these four are often taken as four independent little arguments, whereas in fact they are four sentry posts on one line of defense, guarding together the single conclusion that ignorance is neither an “absence” nor “a kind of discernment.” The defense in the middle of the passage — which Puguang’s Record marks as “the intent of the Sautrāntika defense” and Fabao’s Commentary as “this is the Sautrāntikas’ defense” — holds that being interspersed among wholesome discernment is enough for “being able to defile,” without arising in association; the author records it, and also gives his own adjudication: “yet there is a further ignorance of a kind unlike discernment, as greed is unlike mind, and this account is the better one.” It should be noticed that this is the author choosing between two accounts, and not a fifth reason newly made by him.

Where it lands: at the end the author gives the mark of ignorance: “that is, not comprehending the truths, the real, action, and result.” Pressed further, he frankly acknowledges that “dharmas of this kind can by their nature only be stated so,” and that one “can only distinguish their function” — dharmas of this kind can be identified only by their function, and cannot be taken apart any further. This is the honesty of Abhidharma, and it is the one sentence a reader should most learn: when a definition reaches its limit, one must be willing to stop.

III. Name: Why the Four Formless Aggregates Are Called “Name”

First ask: of the link “name-and-form,” form has already been distinguished in the Chapter on the Elements; why does the remaining “name” refer to the four aggregates of feeling, perception, formations, and consciousness? “Name” is plainly a matter of language, while the four aggregates are dharmas of mind — how are the two to be hooked together?

Answer: the stanza has only five characters, “name is the four formless aggregates,” and the whole reason is in the prose commentary: “following the names established for them and the force of faculty and object they turn and change with respect to the sense, and so they are said to be name.” Mind is called name because it turns along with language — the world establishes a name in common, and mind turns with that name to that object. The example the author gives is of the plainest: “as with such names as ox, horse, form, and taste,” for when you call out “ox” the mind falls upon an ox. So “name” here does not mean the word itself; it means that property of mind by which it turns along with a name. The author then supplies two further explanations: first, that the four aggregates “turn and change upon this and that object in taking it as their object”; second, that “moreover they are of a kind like name, being made manifest by following names.” At the end he records also the account of another master: “these four formless aggregates, having cast off this body, turn and proceed to another birth, and turning and changing like a name they are signalized as name” — a name being transferable from one thing to another, and mind too being able to cast off this body and proceed to another birth, the likeness being taken in that respect.

Clearing a misreading: the commonest misreading is to take the “name” here for the name of “name-body, phrase-body, syllable-body” spoken of in the later system. The two share a graph and differ in sense: the name here takes the whole of the four formless aggregates, while the later name-body is a single item among the formations not associated with mind. Yuanhui’s Verse Commentary here says outright, “name means the four non-form aggregates, that is, feeling, perception, formations, and consciousness,” nailing down the extension first and only then explaining why it is called name — an order worth learning.

Where it lands: this section looks small, and yet it is the foundation of everything later discussed under the two words “name-and-form.” Wherever one reads of name-and-form among the twelve links, of consciousness nutriment among the four nutriments, or even of “name-and-form arising together” at the joining of rebirth, the “name” meant is always these four aggregates and not the word.

IV. Contact: Is It Produced by the Meeting of Three, or Is It the Meeting of Three Itself?

First ask: the definition of the link of contact is “contact is of six kinds, produced by the meeting of three” — faculty, object, and consciousness meeting together and producing contact. But the meeting of the three is already one thing; why must a further “contact” be established?

Answer: the author first settles a technical difficulty: the faculty, object, and consciousness of the five consciousnesses may be simultaneous, “and of the arising of the first five contacts it may be said that three meet, since it is granted that faculty, object, and consciousness arise at the same time”; but mind-contact is not so — the mind-faculty is the mind of the preceding moment that has ceased and belongs to the past; the object of dharmas may belong to the future; mental consciousness belongs to the present; and the three not being simultaneous, how is this to count as meeting? He gives two solutions: “this is precisely what is called meeting, that is, the sense of cause and result is established,” or else “or because they produce one and the same result they are called meeting.” Meeting need not be defined as simultaneity: the establishing of cause and result is meeting, and producing one and the same result is meeting too.

Then comes the first exchange between schools in this fascicle. One party holds that the meeting of three is itself contact, citing scripture in proof: “these three dharmas gathered and met together are called contact.” The other party holds that contact is a mental factor of its own, citing in proof the sūtra of the gate of the six sixes — for beyond the six internal sense fields, the six external sense fields, and the six bodies of consciousness the sūtra establishes six bodies of contact besides, from which it is evident that contact has an entity of its own. After several rounds back and forth the author lays down his pen in one sentence: “to object and explain back and forth in this fashion would make the discussion too much, and so it should stop for the present,” and then says, “the Abhidharma masters, however, hold that there is a separate contact.”

Clearing a misreading: readers often read “it should stop for the present” as the author giving up on judgment. It is not. What he stops is the going back and forth of discussion, not the conclusion — the very next sentence states the position of the Abhidharma masters. But neither should the sentence be read the other way, as the author’s own adjudication: the subject he uses is “the Abhidharma masters,” and this is recording a school’s tenet and not declaring his own. Puguang’s Record here marks by name, one by one, the speakers implicit in the treatise’s text, which makes it clearer to read; yet that is a reading the ancient master supplies on the treatise’s behalf, and the text itself does not name them at every place.

Where it lands: this section demonstrates the standard procedure by which Abhidharma handles a dispute over “a separate entity” — first resolve the technical difficulty, then set out the citations of the two parties, then record the school to which each belongs, and last stop. Hereafter, whenever it is asked whether the latent tendencies are associated with mind, or whether nonmanifest form really exists, the procedure used is the same.

V. Resistant and Designation: One Named from the Support, One from the Object

First ask: why must the six contacts be further combined into two classes? Where does the knife come down in combining them?

Answer: the stanza says, “those associated with the five are resistant; the sixth, arising together with it, is of designation.” The first five contacts are called resistant, the reason being that “the five contacts of the eye and the rest are called resistant, because they take resistant faculties as their support” — the knife comes down on the support: the five faculties have resistance, and so they are called resistant. The sixth, mind-contact, is called of designation, the reason being that “designation is name, and name is the long object taken by mind-contact, and so with respect to just this it is named contact of designation” — the knife comes down on the object: within the object taken by mind-contact there is one thing more, namely verbal designation. The author explains what “long” means by an exceedingly handsome contrast: “as it is said that visual consciousness can cognize only blue and cannot cognize that this is blue, while mental consciousness cognizes blue and cognizes also that this is blue, and so it is called long.” Visual consciousness can catch that patch of blue but cannot catch the judgment “this is blue”; mental consciousness catches both. Hence the object of mind-contact is the wider, and so it is named of designation.

Clearing a misreading: the word “designation” — whose Chinese graphs read literally “increased speech” — is most easily read as “added speech.” In truth it is a name established with respect to the dominance of verbal designation. The author himself also records another way of taking it: “some say: for mental consciousness speech is dominant, and only so does it turn upon its object, whereas the five consciousnesses are not so; and therefore mental consciousness alone is named of designation” — on this account the name of designation is established not from the object but from association. One and the same name may have two routes of naming, and the author lets both stand without adjudicating.

Where it lands: this section is a demonstration of the glossing of technical names. Hereafter, on reading any compound technical name, first ask, “is this name established from the entity, from the support, from the object, or from association?” — and the thing will mostly come untied. Puguang’s Record and Yuanhui’s Verse Commentary both here delimit these two terms by the “dependent-determinative analysis” among the six compound analyses of logic; Fabao’s Commentary, for its part, establishes the names severally from support, from object, and from association.

VI. The Sixteen Kinds of Contact: Three Ways of Cutting Laid One upon Another

First ask: the six contacts have already been combined into two classes; how do sixteen kinds now appear? How are these numbers laid one upon another?

Answer: the stanza — “with clear knowing, with ignorance, neither of the two; the without-outflows, the defiled, the rest; two associated with craving and hostility; three conducing to the three feelings, pleasure and the rest” — gives three ways of cutting. The first cut is by clear knowing and ignorance: contact with clear knowing, contact with ignorance, contact neither with clear knowing nor with ignorance, the three matched in order with the without-outflows, the defiled, and the rest — “‘the rest’ means what is other than the without-outflows and the defiled, that is, the wholesome with outflows and the unobscured-indeterminate.” The second cut is by craving and hostility: “within contact with ignorance one portion arises repeatedly, and in dependence on it the two contacts of craving and hostility are further established,” cut out from within contact with ignorance. The third cut is by the feeling conduced to: “first, contact conducing to pleasant feeling; second, contact conducing to painful feeling; third, contact conducing to feeling neither painful nor pleasant,” and this cut gathers in every contact. The three cuts added to the original six contacts and two contacts make “thus sixteen kinds of contact in all.”

Clearing a misreading: what is most easily misunderstood here is taking the sixteen for sixteen mutually unrelated items. They are not sixteen squares standing abreast but one and the same set of contacts measured several times by different measures — six contacts cut by faculty, two contacts cut by support and object, three contacts cut by association with clear knowing and ignorance, two contacts cut by latent tendency, three contacts cut by the feeling conduced to. Hence one concrete contact may belong to several squares at once.

Where it lands: at the end the author sets himself a question — how can contact be received by feeling? — and answers, “because the mode of apprehension of what depends is exceedingly like contact, and because it arises in dependence upon contact.” This sentence in fact already announces the long debate of the next section: whether contact and feeling are simultaneous or successive. If a reader feels it abrupt on reaching this point, that is just the sign that the author’s arrangement is deliberate.

VII. The Six Feelings, and a Long Debate on “Whether Contact and Feeling Are Simultaneous”

The stanza to which this section is attached is only ten characters, while the prose commentary runs over nine hundred, the second longest passage in the fascicle. The author himself says at the end of it, “the incidental discussion is over; the proper sense should be analyzed” — he knows he has strayed, and yet he has recorded the whole debate in full. This passage is worth a little more space.

First ask: the stanza itself is very simple, “from these six feelings are produced; five belong to the body, the rest to the mind.” Six contacts produce six feelings; the first five depend on material faculties and are called bodily feelings; what is produced by mind-contact depends on mind alone and is called mental feeling. The real question is in what follows: are feeling and contact successive or simultaneous?

The Vaibhāṣikas’ position: “the Vaibhāṣikas say they arise simultaneously, because contact and feeling are reciprocally coexistent causes” — simultaneous, and coexistent causes for each other besides.

The objector’s attack runs in four steps in all. The first step asks: when two dharmas arise at the same time, how are producer and produced to be told apart? “Toward a dharma already arisen another dharma can do nothing” — what can another dharma still do to something already arisen? The Vaibhāṣikas answer that this is merely the thesis restated, “this is no different in sense from the thesis laid down” — where Puguang’s Record expressly marks “the Vaibhāṣikas again deny the Sautrāntikas.” The second step points out the fault of producing each other, and the answerer simply grants it: “our school grants that the two are coexistent causes and results for each other as well.” The third step brings out scriptural proof: the sūtra says only that with eye-contact as condition the feeling produced by eye-contact arises, “there has never been a sūtra that says with eye-feeling as condition the contact produced by eye-feeling arises,” and appeals to common sense — “as with first the seed and then the sprout, first milk and then curds, first the striking and then the sound, first the mind-faculty and then consciousness, cause before and result after is not unestablished.” The answerer at once raises instances of simultaneous cause and result: visual consciousness with the eye and visible form, and “the four great elements are coexistent with derived form”; in the fourth step the objector answers upon that very instance — “here too it is granted that the preceding faculty and object as conditions can arouse the following consciousness” — and what reasoning can block it; the answerer then raises one question that seals it: “are a shadow and a body not coexistent?” (For the speakers of these three turns the two commentaries mark them alike sentence by sentence; see the apparatus.)

The establishing and refuting of a second account. Someone establishes another account: only after contact does feeling arise, and it is at the third moment that feeling arises with contact as its object — an account both commentaries mark as the explanation of the Elder among the Sautrāntikas (Puguang’s Record: “the explanation of the Elder among the Sautrāntikas”· Fabao’s Commentary: “this reports the explanation of the Elder of the Sautrāntika school”). From here attack and defense change places: before, the Sautrāntikas were objecting to the Sarvāstivādins’ doctrine of simultaneity; from here the Sarvāstivādins object to the Sautrāntikas’ new account, and in reading one must exchange the two words “objector” and “answerer.” The Sarvāstivādins press their advantage: then there will be consciousness without feeling, and consciousness that is not contact. The Sautrāntikas dissolve it with “because of the contact of the preceding position, the feeling of the following contact-position arises.” The Sarvāstivādins press once more: if the objects of the two successive contacts differ, “how does a feeling taking one object arise from a contact taking another?”, and they draw out for their opponent a conclusion whose cost is very great — “or one must grant that feeling is associated with this mind and yet does not take one and the same object with this mind”; this sentence is a fault drawn out on the Sautrāntikas’ behalf by the Sarvāstivādins, and is not what the Sautrāntikas own (Puguang’s Record marks “the Vaibhāṣikas bring out the fault”· Fabao’s Commentary marks “the Sarvāstivādins bring out the contradiction of reasoning”). The Sautrāntikas thereupon shift their reckoning: they grant a consciousness that constitutes contact but has no feeling, and a consciousness that has feeling whose entity is not contact, “because a contributory condition was lacking. What fault is there in this?”

The real turning point of the debate is here. The Sarvāstivādins say: then the ten factors of the great ground are destroyed, since the ten factors of the great ground must pervade every mind and arise together constantly. The Sautrāntikas ask in return: on the strength of what is this rule established? The Sarvāstivādins answer: on the strength of the root treatise. The Sautrāntikas then utter the sharpest sentence in the whole fascicle — “we take the sūtras alone as our measure; the root treatise is not a measure, and what fault is there in destroying it?” — and cite “therefore the World-Honored One said: one should take the sūtras as measure.” This one sentence changes the character of the whole dispute: what is at issue is no longer whether contact precedes feeling, but whether the Abhidharma root treatise has any standing as a measure at all.

The gathering up. The Sautrāntikas then offer a middle course: the factors of the great ground need not pervade every mind, and the “ground” of “great ground” may mean the three grounds of with applied and sustained thought and the rest, the three grounds of wholesome, unwholesome, and indeterminate, and the three grounds of one in training, one beyond training, and one neither — whatever extends to these grounds being called factors of the great ground, whatever belongs only to the wholesome grounds factors of the great wholesome ground, whatever only to the defiled grounds factors of the great defiled ground; “dharmas such as these arise by turns each as it is applicable, and do not all arise together.” Last, back to scriptural proof: the Sarvāstivādins adduce a sūtra as objection — the sūtra says “the meeting of three is contact; arising together with it are feeling, perception, and volition”; the Sautrāntikas distinguish that “it says only arising together and does not say arising together with contact,” and adduce “proceeding together with loving-kindness, one cultivates the awakening-factor of mindfulness” to prove that the word “together” may also be used of what is without interval. As for the two words “intermixed,” it is fixed by comparison with the instance of vitality and warmth arising together as being “with respect to the moment” — which Puguang’s Record marks as “the Vaibhāṣikas explain.” At the end one rhetorical question seals it: since the sūtra says the meeting of three is contact, “how could there be a consciousness that is not the meeting of three? Or a meeting of three not called contact?” — and therefore “no contact whatever fails to arise together with feeling and the rest.”

Clearing a misreading: readers most often read this passage as the Sautrāntikas winning and the Sarvāstivādins losing, or the reverse. Neither is so. What the author records is a debate that reaches no verdict, and the gathering up at the end is marked by Puguang’s Record as “the Vaibhāṣikas adduce a sūtra as objection” and adjudged by Fabao’s Commentary as “this concludes the Sarvāstivāda school in sum” — it is the conclusion of the Sarvāstivāda side alone and not the author’s own ruling. The sentence about “the root treatise is not a measure” above all is the utterance of the Sautrāntika side (the two commentaries mark the speakers sentence by sentence throughout this passage, and it may be checked sentence by sentence), and not the author’s own declaration — and the way to tell is simple: the whole of that stretch is in the form of paired question and answer, and the speaker stands on the side opposed to the Sarvāstivādins.

Where it lands: the value of this passage lies not in its conclusion but in its demonstrating that perennial question, which comes first, the proof from teaching or the proof from reasoning. Hereafter, in reading the chapter on action and the chapter on the latent tendencies, wherever “established on the root treatise” and “taking the sūtras as measure” stand opposed, one may come back to this passage for comparison.

VIII. The Eighteen Mental Proximate-Activities: Why Three Times Six Is Exactly Eighteen

First ask: the preceding section said feeling is of six kinds; how in a moment does it become eighteen? Where do the twelve extra come from?

Answer: the stanza says, “this becomes eighteen besides, by the differences of mental proximate-activity.” The author first points out that the word “besides” is “to display the carrying forward of what precedes and the raising of what follows” — taking up the text above and raising what follows. The content of the eighteen is “that is, joy, sorrow, and equanimity, each with six proximate activities”: the three feelings of joy, sorrow, and equanimity each matched with the six objects of visible form, sound, odor, taste, the tangible, and dharmas. The real teaching lies in the three questions the author sets himself: reckoned by intrinsic nature there are only three; reckoned by association there is only one (all being associated with mind); reckoned by object there are only six. The three ways of reckoning each reckon on their own, and none of them yields eighteen; “these come to eighteen, complete by three” — the three conditions must be at work at once before the number eighteen holds.

Clearing a misreading: it is easiest here to suppose that eighteen is the product of three times six. The multiplication is only the surface; the author’s point is that not one of the three criteria may be missing. With intrinsic nature alone, joy, sorrow, and equanimity need not be divided by object; with object alone, the six objects need not be divided by feeling; with association alone, nothing can be told apart at all. Another easy error concerns “not taking a mixed object” — the fifteen proximate activities are called not taking a mixed object “since their objects are severally distinct,” while of the proximate activities upon dharmas alone “the three proximate activities upon dharmas extend to both kinds,” because the object of dharmas may be taken singly or taken all together.

Where it lands: the author then explains the sense of the three words “mental proximate-activity,” setting out two accounts: “it is transmitted that joy and the rest take mind as their proximate condition and therefore roam again and again among the several objects” and “some say that joy and the rest can act as the proximate condition, causing mind to roam again and again upon an object” — the former taking mind as the condition, the latter joy and the rest, opposite in direction and standing together. He then answers four objections: why bodily feeling does not count (“because it does not depend upon mind alone it is not called proximate, and because it is without discrimination neither is it activity”); why the pleasure of the third meditative absorption is not gathered in (“it is transmitted that in the first element there is no pleasure faculty associated with mental consciousness,” and there is no pain faculty to stand opposite it); why the sūtra’s “having seen a visible form with the eye” does not obstruct its being on the ground of mind alone (it is said with reference to the ground of mind drawn forth by the five consciousnesses, as the contemplation of impurity is drawn forth in dependence upon visual consciousness); and finally “the words ‘seeing a visible form’ and the rest are said as the case is plain and easy to understand.” All four objections are for the sake of guarding the line of “on the ground of mind alone” — and the backbone of this section is there.

〔Appended〕The fifth question: whether one object can produce only one proximate activity. At the end of the passage the treatise sets a further question: “can there be visible forms and the rest which, with respect to joy and the other three, can conduce to the arising of one proximate activity only? There can, with respect to one continuum and not with respect to the object.” First ask: how could one and the same visible form produce only one proximate activity? Answer: it depends which end you speak from. Spoken of with respect to the continuum of a single sentient being, this person facing this form at this time in fact arouses only one of joy, sorrow, or equanimity, and so it is said that it “can conduce to the arising of one proximate activity only”; spoken of with respect to the object itself, one and the same form may for different people and at different times extend to all three of joy, sorrow, and equanimity, and so it cannot be said to be matched with one alone. Clearing a misreading: this question is not to be read as a general rule that “one object necessarily produces one feeling” — the treatise answers very carefully, and immediately after the word “there can” comes “with respect to the continuum and not with respect to the object,” the two qualifications neither of which may be missing. Where it lands: this is the place in the whole table most easily confused — the numbers eighteen, twelve, and six above are reckoned with respect to the object, while this question is stated with respect to the person, and the two measures may not stand in for each other.

IX. The Mental Proximate-Activities within the Three Realms: A Table of the Objects They Take

First ask: since the eighteen are divided by object, then, the objects of the three realms differing, how many can each realm have, and objects of which realm can each take?

Answer: this section restates three stanzas together, and is numbers throughout. The desire realm has all eighteen, and taking objects of the desire realm eighteen as well; “taking objects of the form realm there are only twelve, the six of odor and taste being excluded, since that realm has not those objects” — the form realm having no objects of odor and taste, the six proximate activities matched with odor and taste fall away. Taking objects of the formless there are only three, since the formless realm has none of the five objects of visible form, sound, odor, taste, and the tangible, and only the three of joy, sorrow, and equanimity in the proximate activity upon dharmas remain. The first two meditative absorptions have only twelve (the six of sorrow excluded), and taking objects of the desire realm twelve as well (the six objects of the desire realm being complete); as for their own level, four matched with odor and taste are excluded, leaving eight, and two take objects of the formless. The third and fourth absorptions have only six (equanimity alone), and taking objects of the desire realm six as well; as for their own level, two matched with odor and taste are excluded, leaving four, and one takes objects of the formless. The preliminary stage of the sphere of space has four kinds, taking only visible form, sound, the tangible, and dharmas; taking the fourth meditative absorption as object it likewise has these four, and taking the formless only one; the four fundamental levels and the upper three preliminary stages “have only one and take their own objects,” because “the three upper edges do not take dharmas of form as object.”

Clearing a misreading: readers are most easily drowned in the numbers and suppose they must be learned by rote. In truth only three rules are at work in the whole table: which faculties of feeling that level has (the first two absorptions having no sorrow, hence twelve; the third and fourth having equanimity alone, hence six); which objects the realm taken as object has (the form realm having no odor and taste, hence four or two excluded when taking one’s own level); and that a defiled mind does not take lower levels as object, while a wholesome mind can. Hold these three, and the numbers eighteen, twelve, three, six, four, and one can all be worked out on the spot.

Where it lands: the usefulness of this table is that it turns the relation between “realm” and “object” into a checkable list. Hereafter, meeting any question of the form “how many of such-and-such a dharma are possessed on such-and-such a level,” the method is always the same — first ask which faculties of feeling that level has, next which objects the realm taken as object has, and last whether that mind can take a lower level. On the side of collation there is one item that may serve as corroboration: the base text here reads “ten of the first meditative absorption” while the Ming edition reads “middle,” and read as “middle” the whole enumerated sequence is broken — the self-consistency of the context may serve as corroboration, while the choice among variants remains recorded together in the apparatus layer and is not closed here.

X. The Eighteen Are with Outflows Alone, and the Measure of What Is Possessed

First ask: since the mental proximate-activities extend to the three realms, do they extend to what is without outflows? If they do, would not the mind of a noble one also be “roaming again and again upon objects”?

Answer: the stanza blocks it in one sentence: “the eighteen are with outflows alone.” The author’s explanation is of the briefest: “there is no mental proximate-activity that extends to what is without outflows, and therefore it is said that the eighteen are with outflows alone.” Then comes the list of the measure of what is possessed. One born in the desire realm who has not obtained the wholesome mind of the form realm possesses all eighteen of the desire realm, eight of the first two absorptions, four of the third and fourth, and one of the formless; but “those of the upper realms which are possessed do not take lower levels as object, because they are defiled” — those few of the upper realms which he possesses are defiled and do not take lower levels as object. If he has obtained the wholesome mind of the form realm but is not free from craving for the desire realm, he possesses all of the desire realm and ten of the first absorption; of these, joy is possessed only to four, because a defiled mind does not take the odor and taste of a lower level as object, while equanimity is possessed in full to six, because the wholesome mind of access concentration can take odor and taste as objects. One born in the form realm possesses only the one proximate activity of equanimity upon dharmas of the desire realm, “that is, the one arising together with the supernormal-result mind.”

Clearing a misreading: “the eighteen are with outflows alone” does not amount to “an arhat has no joy.” The author himself clears this up in the very next passage. If a reader takes “with outflows alone” to mean that a noble one is not granted joy, sorrow, and equanimity, the next passage will be wholly unreadable.

Where it lands: this list looks tangled and is in fact a graduated scale of the stages of realization — how far a person has walked upon the ladder of the three realms can be marked by how many mental proximate-activities he possesses. Before and after obtaining the wholesome mind of the form realm, the difference falls exactly upon those few joys and equanimities of the first absorption. That Abhidharma reckons so finely has its reason just here: advance and retreat can be entered in the ledger square by square.

Where the vital point lies: immediately after the list of what is possessed the author says something quite out of the ordinary — “some say: mental proximate-activities such as these are established by the Vaibhāṣikas as suits their sense of the doctrine. Yet the sense of the sūtras as I see it is otherwise” — and establishes a criterion of his own: “only those that are defiled, drawing one another on with mind and roaming again and again upon their objects, are mental proximate-activities,” with the six constant abidings and “it is not that the arhat has no worldly joy taking a wholesome dharma as object; this is said only in order to block the defiled proximate activities” for evidence. This passage is one of the two vital points of the fascicle; the taking and leaving of it and the later refutations of it all enter Further Resonances, and here only the establishing of the treatise’s own text is recorded.

XI. The Remaining Links Have Been Stated or Will Be Stated: Eight Links Dispatched in One Sentence

First ask: the chapter on dependent origination has come this far and has explained only the four links of ignorance, name, contact, and feeling; what of the remaining eight? Are they simply passed over?

Answer: the author expressly sets up a line of stanza to dispatch them: “the remaining have been stated or will be stated.” The prose commentary says it very clearly: “of the remaining links of existence, some have been stated already and some are yet to be stated, and therefore they are not discussed here.” And he points out one by one where they go — the link of consciousness was stated in the Chapter on the Elements (“consciousness is the cognizing of each severally, and this is what is called the mental sense field”), the link of the six sense fields was stated in the Chapter on the Elements (“the subtle form upon which that consciousness depends is what is called the five faculties of eye and the rest”), and “the two links of formations and existence will be stated in the chapter on action; the two links of craving and appropriation will be stated in the chapter on the latent tendencies.”

Clearing a misreading: readers often skip this line of stanza, feeling it has no content. Quite the contrary: it is the self-declaration of the structure of the whole treatise. The Abhidharmakośa does not treat each link in isolation but distributes one and the same topic to the chapter that best suits it. Consciousness and the six sense fields belong to the entities of dharmas, and so are in the Chapter on the Elements; formations and existence belong to doing, and so are in the chapter on action; craving and appropriation belong to affliction, and so are in the chapter on the latent tendencies. The chapter on dependent origination keeps only those that have no home elsewhere.

Where it lands: to read Abhidharma without knowing this rule is to feel it defective everywhere. Knowing it, one forms a habit — when a link is treated too briefly, first ask in which chapter its proper text lies. Puguang’s Record points out a connection of the same type at the link of name, saying, “form is as analyzed above, that is, analyzed above in the aggregate of form of the Chapter on the Elements,” which is exactly the same way of reading.

(Transition): the author then gathers the twelve links into three kinds: “these several dependent originations are established in brief as three, namely affliction, action, and the matter of maturation results,” and says that “external similes should be resorted to in order to display their several capacities” — whence the eight images in nine uses of the next section.

XII. Confusion, Action, and the Matter of Results: Three Kinds and Eight External Images

First ask: once the twelve links have been split into three kinds, how is a person to remember the temper of each? The author’s way is to strike similes — and he strikes eight of them in one breath.

Answer: affliction has five similes. Like a seed: from a seed there arise sprout and leaves, and from affliction there arise the three of affliction, action, and the matter of results. Like a dragon: “as a dragon guards a pool the water is never exhausted; affliction guards action and the continuity of birth is without end” — the dragon being in the pool, the water does not dry; affliction being present, the continuity of action is not cut. Like the root of a grass: “as when the root of a grass is not pulled up the shoot grows again after cutting, so when the root of affliction is not pulled up a destiny arises again after ceasing” — this simile is closest to practice, for cutting the shoot without pulling the root, one cuts and it grows again. Like the stem of a tree: “as from the stem of a tree branches, flowers, and fruit arise again and again,” one stem producing branches again and again. Like the husk wrapping rice: “as rice wrapped in husk can produce a sprout and the rest, while bare rice cannot” — bare rice does not sprout, it needs the husk wrapping it; bare action does not call forth birth, it needs affliction wrapping it.

Action has three similes. Like husked rice: “as rice that has its husk can produce a sprout and the rest, so action having affliction can call forth maturation.” Like a medicinal herb: “as with the several medicinal herbs the ripening of the fruit is the last stage, so once the result of an action has ripened it calls forth no further maturation” — the herb having set its fruit, the business is finished. Like a flower: “as a flower is the proximate cause producing the fruit, so action as proximate cause can produce maturation.”

The matter of results has one simile. Like cooked food: “as cooked food and drink are only to be consumed and cannot be turned into further food and drink” — cooked rice can only be eaten and cannot be turned back into grain. The author then points out the vital point of this simile: if a maturation result could call forth a further maturation, “the further calling forth the further, there would be no liberation.”

Clearing a misreading: the eight images in nine uses are very easily read past as literary ornament. In truth each has a precise logical position: the dragon governs persistence, the grass root regrowth, the husk the helper, and the medicinal herb and the cooked food cessation. The simile of cooked food above all is not there for prettiness — it is the logical guarantee that liberation is possible: only because a result of maturation produces no further result of maturation does the chain have a break in it.

Where it lands: this section may serve as a mnemonic for the twelve links, but its better use is as a map of where practice takes hold. Of the eight images in nine uses, three speak of the stubbornness of affliction (dragon, grass root, tree stem), one of an action calling forth no further result once its result has ripened (the medicinal herb), one of action as the proximate cause producing maturation (the flower), and one of a result producing no further result (the cooked food). Where the knife should come down to cut the continuity, the similes themselves have answered. Yuanhui’s Verse Commentary here divides the eight images in nine uses very clearly, saying, “the first four lines display affliction by simile, the next two lines display the cause that is action, the last two lines display the matter of results,” and a reader may well keep to that frame.

XIII. The Four Existences: The Existence of Birth Alone Is Invariably Defiled

First ask: of the four stretches — intermediate existence, the existence of birth, existence proper, and the existence of death — which is necessarily defiled? And why that one in particular?

Answer: the stanza says, “among the four kinds of existence, the existence of birth is defiled alone, by the afflictions of its own level; the remaining three, in the formless three.” The existence of birth — the first moment of taking birth — is defiled alone. Which affliction defiles it? “The several confusions of its own level,” that is, “taking birth on this level, all the afflictions of this level defile the existence of birth of this level.” The author then cites a ruling of the Abhidharma masters: “among the several afflictions there is not one affliction that lacks the capacity to moisten at the position of joining rebirth” — there is no affliction that stands with folded hands at the joining of rebirth. But one limitation is to be noticed: “yet the joining of rebirth is by the force of the afflictions alone and not by entanglements and stains arising by their own strength,” it being the root afflictions that act, the secondary afflictions merely following along. As for intermediate existence, “it should be known that the first moment of the continuity of intermediate existence is also necessarily defiled, just as the existence of birth is.” The remaining three existences each extend to the three ethical qualities — “yet the remaining three existences each extend to the three, that is, existence proper, death, and intermediate, each of them wholesome, defiled, and indeterminate.” The formless realm has intermediate existence excluded and only three, because there “within that realm there is no separation of places,” and there being no place to go to, intermediate existence has no way of being established.

Clearing a misreading: the commonest misreading is to widen “the existence of birth is defiled alone” into “taking birth necessarily falls into an evil destiny.” It does not. What is defiled is the nature of the mind at that one moment, not the goodness or badness of the destiny born into; the existence of birth of one born among the gods is defiled all the same, since rebirth must be moistened by affliction. Another easy error is the last five characters of the stanza, “the remaining three, in the formless three,” which is easily read as “the remaining three are present in the formless realm as three”; in fact two layers are folded together: the remaining three existences each extend to the three ethical qualities, and the formless realm has three existences only. Yuanhui’s Verse Commentary here takes the two layers apart, saying, “the first three lines and the remaining three characters of the fourth line are the gate analyzing by the three ethical qualities,” which is precisely to remind the reader where the knife comes down.

Where it lands: this section buckles directly onto the chapter on intermediate existence in the preceding fascicle, and is also the foundation for the talk of “joining rebirth” in the chapter on the noble ones later. The author’s last sentence supplies something very fine: “the stanza does not speak of the desire and form realms, and from this it is known that within them all four existences are granted” — what is not spoken of is read by tacit consent. This too is a way of reading stanzas commonly met with in Abhidharma.

XIV. The Four Nutriments: Upon What Do Sentient Beings Abide

The four existences distinguished, the author asks one sentence, “upon what do these sentient beings depend to abide?”, and out of it comes the longest passage in the fascicle — over fifteen hundred characters. This passage must be given its full weight, for it takes the seemingly everyday notion of “nutriment” and makes of it a doctrinal number running through the three realms, extending to intermediate existence, and bearing upon liberation.

First ask: a man lives by eating — who does not know it? Why must this be made into a doctrine?

Answer: because what the author means to establish is not “eating” but “abiding.” He opens by citing scripture: the World-Honored One “awakened of himself to one dharma and rightly awakened to it and rightly stated it,” and this one dharma is that “no sentient being whatever does not abide by nutriment.” The definition of “nutriment” here is not what goes in at the mouth but what can make a sentient being continue and abide. Hence “nutriment is of four kinds: first morsel, second contact, third volition, fourth consciousness.”

The subtle and coarse in morsel nutriment, and its realm affiliation. Morsel nutriment divides into subtle and coarse: “the subtle is the nutriment of intermediate existence, since odor is its nutriment”; and further “and the nutriment of the gods and of the beginning of an aeon, since it turns into no filth, passing into the limbs like oil poured on sand” — the nutriment of the beginning of an aeon leaves no filth behind when eaten, soaking in as oil poured upon sand; and again “or the nutriment of tiny filthy vermin, infants, and the like is called subtle, and what is contrary to this is coarse.” As to realm affiliation, “morsel nutriment such as this is in the desire realm alone, since it is by becoming free from craving for morsel nutriment that one is born in the upper realms.” As to entity, “odor, taste, and the tangible, these three, are all of them the very entity of morsel nutriment, since they can be drunk and eaten as separate morsels, that is, received portion by portion with mouth and nose” — the three sense objects of odor, taste, and the tangible are the entity of morsel nutriment, and not that lump of food. Someone asks whether light, shade, flame, and coolness count as nutriment, and the author answers, “how do light, shade, flame, and coolness become nutriment? It is transmitted that this sentence is put with respect to the majority”; and he adds that what sustains the body without being drunk or eaten is also gathered under subtle nutriment, as with anointing and bathing. Puguang’s Record here explains the treatise’s text by the receiving portion by portion with mouth and nose, saying, “that is, the mouth can drink and eat; the nose can drink, as with the trunk of an elephant and the like; they are received portion by portion, and so it is named morsel nutriment,” and explains the benefit of light and flame by “as when a man suffers from cold and suddenly meets with sunlight or the touch of a fire’s flame, and is at once benefited” — concrete instances supplied by an ancient master for the treatise’s text, and the easiest of all to enter by.

Why form is not morsel nutriment. This is a brilliant question. Form is plainly visible and comes along with food and drink; why is it not counted? The author gives two reasons: “it cannot benefit the faculty it faces; and because of one who is liberated.” First, form cannot benefit the very faculty it faces — the eye sees a form and the eye faculty is not thereby nourished; and “whatever is called nutriment must first assist one’s own faculties and great elements, and only then reach the rest.” Second, “moreover the non-returner and the arhat are liberated from craving for nutriment, and take no benefit though they see all manner of fine food and drink.” As for joy and pleasure arising on seeing a form, that “arises from the contact conditioned by form; that contact is the nutriment, not the form” — what benefits is contact and not form.

The remaining three nutriments, and the excluding of what is without outflows. “Contact means the several contacts produced by the meeting of three; volition means mental action; consciousness means the aggregate of consciousness; these three are with outflows alone and are present throughout the three realms.” Someone asks why the entity of nutriment does not extend to what is without outflows, and the author records the Vaibhāṣikas’ explanation: “to be able to assist the several existences is the sense of nutriment; what is without outflows arises from cultivation and is for the extinguishing of the several existences” — the definition of nutriment is itself “nourishing birth and death,” while what is without outflows is for extinguishing birth and death, opposite in direction, and so cannot count as nutriment. And he cites the sūtra in support: the four nutriments “can cause beings that have come to be to abide, and can assist the several seekers of birth.”

Beings that have come to be, and seekers of birth. “What is called a being that has come to be displays the sense of already born,” all who have taken birth being such. “Seeker of birth” refers to intermediate existence, and from this the author opens out the five names of intermediate existence: “this designates intermediate existence, since the Buddha, the World-Honored One, spoke of intermediate existence under five names” — “first, formed by mind, since it arises from mind,” “second, seeker of birth, since it constantly delights to seek out and examine the place where it is to take birth,” “third, eater of odors, since its body takes odor for its nutriment as it goes to the place of birth,” “fourth, intermediate existence, since it is the aggregates in the interval between the two destinies,” “fifth, named arising, since it arises for a time facing the birth that is to come.” There is besides another explanation: “further, a being that has come to be means the arhat, and the rest, who have craving, are called seekers of birth.”

The four alternatives of the fetter of arising and the fetter of birth. After the five names of intermediate existence the treatise cites a sūtra to make a fourfold distinction: “there is a person who has cut off the fetter of arising and has not cut off the fetter of birth” — “arising” being intermediate existence (the text above having already explained that “arising means intermediate existence”), and “birth” being the existence of birth. The treatise lists four alternatives: “the several who go upstream, being free from craving for two realms, are the first alternative; those who attain parinirvāṇa in the interval are the second; the several arhats are the third; and what is left when the foregoing cases are excluded is the fourth.” First ask: what do these four alternatives divide? They divide whether each of the two fetters, of “arising” and of “birth,” is cut off or not — “arising” meaning intermediate existence and “birth” the existence of birth. Those who go upstream do not take up intermediate existence in the formless realm, and so have cut off the fetter of arising; yet they still take birth, and so have not cut off the fetter of birth. Those who attain parinirvāṇa in the interval are precisely within intermediate existence, and so have not cut off the fetter of arising; yet they take birth no further, and so have cut off the fetter of birth. The arhat has cut off both fetters; the rest have cut off neither. Clearing a misreading: do not read “arising” as a general term for “coming into being.” Here “arising” is an alternative name for intermediate existence, which the treatise has just established — the last of the five names being “named arising, since it arises for a time facing the birth that is to come.” Where it lands: the fourfold alternative is the treatise’s most commonly used instrument of distinction; whenever “four alternatives should be made” is met with, first ask which two axes are being divided, and the four squares will fall into place of themselves.

The four nutriments with respect to what has come to be and what seeks birth. The treatise at once asks, “how many nutriments can make beings that have come to be abide? How many nutriments can assist sentient beings that seek birth?” The answer: “the Vaibhāṣikas say all four in each case.” First ask: morsel nutriment plainly nourishes only the present body — how can it assist a seeker of birth? The reason the treatise gives is: “the several who have craving are assisted by morsel nutriment as condition as well, so that it makes them call forth a later existence, because the World-Honored One said that the four nutriments are conditions of sickness, of the boil, of the arrow, of the root, and of old age and death” — morsel nutriment nourishes this body, and this body, having craving, goes on to make the actions of a later existence, so that morsel nutriment has its share in the seeking of birth too. Clearing a misreading: do not read “all four in each case” as meaning that the four nutriments have no division of labor. Extending to and excelling in are two layers: in function, all four nutriments extend to both what has come to be and what seeks birth; with respect to the excellent capacity, “the first two are like the mother who rears, since they rear what has been born; the latter two are like the mother who bears, since they bring to birth what has not been born.” Where it lands: this question and answer and the later “why only four nutriments are stated” are the two ends of one line — first their extension is established, then their excellence is delimited.

Contact and the other two nutriments likewise each have four alternatives. After the four alternatives of morsel nutriment the treatise passes over it in a sentence: “thus contact and the rest, each as is applicable, may each be known to have four alternatives.” And it sets itself a question: “are there cases where contact and the rest as condition assist the several faculties and great elements and yet are not nutriment?” The answer: “there are, namely contact and the rest without outflows and of another level.” First ask: how can contact without outflows assist the faculties and great elements and yet not count as nutriment? Because what makes a nutriment a nutriment is judged by whether it makes existence grow; what is without outflows does not make a later existence grow, and though it have the use of assisting, it does not get the name of nutriment. Clearing a misreading: do not put an equals sign between “assisting” and “nutriment.” The treatise’s order is first to ask whether it makes existence grow, then whether it assists, and the two may not be reversed. Where it lands: this is precisely the most important boundary in the whole passage on the four nutriments — nutriment is named with respect to “making existence continue,” not with respect to “being good for the body.”

The three matters of volition nutriment. This is the most moving narrative in the whole treatise. First, the father in the famine who put ashes into a bag and hung it on the wall, deceiving his children that it was a bag of parched meal, so that “the two children, cherishing hope, prolonged their lives a good while”; and afterwards someone opened it, and “the children saw that it was ashes, and their hope being cut off they died.” Second, the merchants wrecked at sea who looked far off at the heaped foam and took it for the shore, so that “their hearts hoping to reach it quickly, their lives were prolonged for a time,” and when they touched it and knew it was not the shore, “their hope being cut off they died.” Third, the author cites what the Treatise on the Collected Various Gates says of the great living beings of the sea that bury their eggs in the sand, the eggs not spoiling if the mother constantly thinks of them; and yet he does not adopt its explanation — “this is unsound, since it contradicts the sense of nutriment. Can another’s volition nutriment sustain one’s own body?” — and corrects it thus: “by right reasoning it should be said that the eggs constantly think of the mother and thereby escape rotting and spoiling, and that when they forget, their life ends.”

Clearing a misreading: these three stories are most easily read as parables exhorting to virtue. They are not. They are arguments: what they prove is that “volition” really can sustain the body, and that when a single hope is cut off life is cut off with it. And the author’s correction of the third matter is especially important — the nutriment must be one’s own volition and cannot be another’s; otherwise “nutriment” would become something lendable from outside, and the whole definition would fall apart.

Why only four nutriments are stated. Since “the several dharmas with outflows all make existence grow,” why are only four stated? The author answers, “even so, to state four with respect to what is excellent is no fault,” and gives an exceedingly clear twofold-and-twofold table of assignment: the first two nutriments benefit this present life, the latter two draw forth the existence to come. “What is called the support refers to the body endowed with faculties, and morsel nutriment can assist it”; “what is called the supported refers to mind and mental factors, and contact nutriment can assist them”; and as for the birth of the existence to come, “volition nutriment can draw it forth,” and having drawn it forth, “it is by the force of the seeds of consciousness perfumed by action that the later existence can arise.” One simile at the end gathers the whole passage: “the first two are like the mother who rears, since they rear what has been born; the latter two are like the mother who bears, since they bring to birth what has not been born.”

The four alternatives and the boundaries. The author sets a further fourfold distinction: a morsel and not a nutriment (what is drunk and eaten injuring on the contrary the faculties and great elements), a nutriment and not a morsel (the remaining three nutriments), both a morsel and a nutriment, and neither a morsel nor a nutriment. And he asks whether there is anything that assists the faculties and great elements and is still not nutriment: “there are, namely contact and the rest without outflows and of another level.” He further discusses that “what, having been eaten, harms the eater’s body is called nutriment as well, because at the outset it assisted,” and records the Vaibhāṣikas’ account that nutriment is named nutriment at both times — “first, at the time of first eating, when it can remove hunger and thirst; second, after digestion, when it can assist the faculties and the great elements.”

All five destinies and four modes of birth have all four nutriments. How can there be morsel nutriment in the hells? “Are iron balls and molten copper not morsel nutriment?” The author first relieves the difficulty by citing the definition of the Treatise on Categories of Dharmas — that treatise “states it as named nutriment merely with respect to the side of being able to assist”; and then he gives the hells a place: “yet the hot iron balls and the rest in the hells, though after being eaten they act to injure, can nevertheless remove hunger and thirst for a time and so obtain the character of nutriment, and therefore they too are called nutriment,” and “moreover the morsel nutriment of the solitary hells is just like that of human beings.”

The weighing of the ordinary being in the Jambu grove. Appended at the end of the passage is a piece of collation-like discrimination: giving food to a hundred non-Buddhist seers free from craving is not equal to giving food to one “ordinary being in the Jambu grove.” Who is this ordinary being? The author excludes them one by one — to say it is all who have a belly in Jambudvīpa does not agree with the word “one”; to say it is a bodhisattva near buddhahood does not agree with “giving to him, the merit obtained surpasses giving to a koṭi of arhats”; and the Vaibhāṣika masters’ account that it is one who has obtained the factors conducive to penetration he faults most heavily: “in this, name and sense do not answer to each other either, since there has never been a sūtra or root treatise that says so,” and “it should be known that this is merely a discrimination of their own.” What he takes for himself is: “the bodhisattva in his last body dwelling in the Jambu grove is what is called that ordinary being,” with the World-Honored One’s afterwards weighing a non-Buddhist against one on the way to the fruit of stream-entry as corroboration.

Where it lands: the real use of the passage on the four nutriments is that it translates the most everyday matter of “keeping alive” into a doctrinal number, letting one see that birth and death continue because four supply lines feed them — morsel nutriment for the body, contact nutriment for the mind, volition and consciousness nutriment for the life to come. Cut one line and the continuity is by so much the less; the one who is liberated, being “liberated from craving for nutriment,” lets go of the first line first. Yuanhui’s Verse Commentary here sets forth the reasoning that form is not morsel nutriment by the instance of one already free from craving, saying, “the fruit of non-returning and the arhat have no craving for morsel nutriment, and this is called liberation,” which points that layer very clearly.

XV. Death and Birth: Six Positions in Mental Consciousness Alone, Two Times with Equanimous Feeling Alone

First ask: at the moment a man’s breath stops and at the moment he takes birth, which consciousness after all is at work? And in what state of heart?

Answer: the stanza governs six matters in one breath, and the author lists them very neatly: “cutting off wholesome roots and restoring them, becoming free from defilement of a realm or level, falling back from freedom from defilement, dying, and taking birth — at these six positions it is by the nature of things granted that there is mental consciousness alone and not the others.” These six — cutting off wholesome roots, restoring them, becoming free from defilement, falling back, dying, and taking birth — have this in common, that they are all too subtle for the five consciousnesses to catch, and mental consciousness alone can. And “the word ‘birth’ spoken of should be known to include also the first joining of intermediate existence,” the first moment of intermediate existence being reckoned in.

And the feeling? “Death and birth are granted to be associated with equanimous feeling alone, because the mind associated with equanimity is not sharp and clear; the remaining feelings are sharp and clear and do not conduce to death and birth.” At the moment of death and of birth the mind is blunt, and so only equanimous feeling can be matched with it.

There are two further limitations: “further, these two times are in a distracted mind alone and not in absorption; they must be at a position where there is mind and not without mind.” The reasons for not being in absorption are three — the realms and levels are severally distinct, absorption arises by preparation, and absorption can render assistance, whereas the mind of death and birth is precisely of its own level, spontaneous, and feeble. The reason for not being without mind is that “at a position without mind life is certainly not injured,” and one without mind “cannot take birth, because the cause is lacking; apart from the arising of the afflictions there is no taking birth.”

The mind entering nirvāṇa has a square of its own: “although it is said that the existence of death extends to minds of the three ethical qualities, yet at entering nirvāṇa there are only two indeterminate minds”; and as to whether the desire realm has a maturation of equanimous feeling, the author records the two parties without adjudicating. Why must it be indeterminate? “Because the force of the indeterminate is feeble and conduces to the cutting off of mind” — for mind to be cut off, it must first grow weak.

Where consciousness ceases. In one who dies all at once, “mental consciousness and the body faculty cease abruptly and together”; in one who dies gradually it differs by destiny: “falling into an evil destiny is called going downward, and his consciousness ceases last at the place of the feet,” “if he goes to the destiny of human beings consciousness ceases at the navel; if he goes to be born among the gods consciousness ceases at the place of the heart,” and of the arhat “his last mind too ceases at the place of the heart,” while another master says “his ceasing is at the crown.” The author strikes an exceedingly vivid simile for the gradual ceasing: “like a little water placed upon a red-hot stone, gradually lessening, gradually vanishing, and at one place wholly exhausted.”

The severing of the vital points. At the end of life one is for the most part pressed by the pain of the severing of the vital points. A vital point is not an organ — “there is no separate thing called a vital point; yet within the body there are particular joints which, once touched, bring death, and these are called the vital points.” The condition that brings death is “if any one among water, fire, and wind grows dominant, then, as a sharp blade’s edge touching those vital points.” And “it is not called severing as chopping firewood is; it is because after the severing there is no awareness that it gets the name severing” — this “severing” is the severing of losing awareness and not the severing of chopping through. Why has the element of earth no such function? “Because there is no fourth inward calamity,” the inward calamities being only three — “the three inward calamities are wind, heat, and phlegm.”

The five signs of decay in the sons of the gods. Among the gods there is no severing of the vital points, and in its place are the five signs of decay. Five small signs: their gear gives out an unwanted sound, the light of their bodies becomes dim and feeble, “third, at the time of bathing drops of water cling to the body,” “fourth, whereas by nature they were restless and swift-moving, they now stagnate upon a single object,” “fifth, whereas their eyes were fixed and still, they now blink again and again” — and when these five appear, “these five signs appearing, it is not certain that they will die.” Five great signs: their garments are stained with dust, their flower garlands wither, sweat comes from the armpits, a foul odor clings to them, and they take no delight in their own seats — “these five signs appearing, they will certainly die.”

Clearing a misreading: readers most easily misunderstand “death and birth with equanimous feeling alone” as “dying is not painful.” Equanimous feeling is said of the mind of that one moment of joining rebirth and of ending life, not of the whole process of dying; the pain of the severing of the vital points is a real pain, and the author himself says that “thereby an exceedingly heavy painful feeling arises.” The two are not at the same moment and do not contradict each other. Another easy error is “mental consciousness alone,” which is not to say that there is no bodily sensation at the end of life but that the decisive moment of thought is in mental consciousness. On the side of collation one item in this passage is worth noting: “the several arhats are called not born again,” where the Yuan and Ming editions read “not going”; the two readings differ in emphasis, and this collection records them without adjudicating.

Where it lands: this section is the most concentrated place in the whole treatise on the end of life, and hereafter, when the chapter on the noble ones treats of attaining parinirvāṇa in the interval and the chapter on action of action to be experienced in the next life, one must come back here to take the measure. The twofold distinction among the signs of decay in the sons of the gods is especially useful — the small signs may be turned, the great signs mean certain death — and this distinction is itself a concrete place to take hold of the contemplation of impermanence.

XVI. The Three Classes: The Threefold Division of the World

First ask: with what does a chapter on the world close? What the author chooses is the three classes — the drawing up of all sentient beings into three heaps according to whether their direction is fixed.

Answer: “within the birth, abiding, and passing away of this world of sentient beings the World-Honored One established the three classes,” namely “first the class fixed in rightness, second the class fixed in wrongness, third the class of indeterminate nature.”

Rightness takes cutting off: the sūtra says, “greed cut off without remainder, hatred cut off without remainder, delusion cut off without remainder,” and “all the afflictions cut off without remainder is named rightness.” And “fixed” takes the noble: “fixed refers to the noble; noble refers to those in whom the path without outflows has already arisen, and being far from the several evil dharmas they are called noble, having obtained the acquisition of final release from bondage,” and since they will certainly exhaust delusion they are called fixed in rightness. The author here sets a question of great weight: those who have obtained the factors conducive to liberation will also certainly obtain nirvāṇa; why are they not fixed in rightness? Three answers: “because they may afterwards fall into the class fixed in wrongness,” “and because the time of obtaining nirvāṇa is not yet fixed, unlike the stream-enterer with at most seven further existences,” “and because they are not yet able to cast off wrongness, and so are not called fixed in rightness.”

Wrongness takes the destiny: “what is named wrongness? It is the several hells, animals, and hungry ghosts, and this is named wrongness”; and “fixed” takes the deeds of immediate retribution: “fixed refers to the deeds of immediate retribution; one who commits a deed of immediate retribution will certainly fall into the hells, and so it is called fixed in wrongness.”

Indeterminate is what is left: “what is other than fixed in rightness and fixed in wrongness is named of indeterminate nature, since these await two conditions and may become either of the two” — fixed upon a condition, and either way possible.

Clearing a misreading: the three classes are most easily read as a moral grading of three kinds of person. They are not. They are divided with respect to whether one’s direction is already fixed, and do not correspond directly to present good and evil. A man who practices very well but has not yet seen the path is still in the indeterminate class; and that is precisely the point of the question the author set. Another easy error is to read “indeterminate” as a middling state, whereas the real sense of “indeterminate” is still under the working of two conditions — which is on the contrary the most promising position of all.

Where it lands: that the whole fascicle closes here has a deep intent. Dependent origination, the four existences, the four nutriments, death and birth have all been speaking of how sentient beings are pushed along by action and confusion; and this last section draws the whole world up once more in ranks according to whether it can still be changed. Most people are in the indeterminate class, and the very definition of the indeterminate class is “may become either of the two” — this one sentence is the way out onto practice for the whole treatment of the world in this fascicle. Yuanhui’s Verse Commentary fixes the skeleton of this stanza in four characters, saying, “the upper line states it, the lower line explains it,” which is plain and may well be followed.

Further Resonances: A Mirror of Three Commentaries

The foregoing is the sense of the treatise’s own text in this fascicle. Below, a further section gathers five things: where the three commentaries diverge, Saṅghabhadra’s counter-refutations, the divergences of Paramārtha’s older translation, two cases recorded together without adjudication, and coordinates for collateral reading. A mirror is not the treatise; it is only another way of reading one and the same text. All are recorded together without adjudication, each restored to its place.

1. Where the Three Commentaries Diverge

Divergence in the outline. The divisions the three mark at the head of this fascicle differ from one another. Puguang’s Record has, “as for ‘what is the sense of ignorance?’, from here the second major part separately explains names and senses”; Fabao’s Commentary has, “the treatise’s ‘what is the sense of ignorance?’: from here the second major gate explains the names and senses of the twelve links,” where the base text’s own graph is “gate” and not “part,” copied as it stands; Yuanhui’s Verse Commentary has, “from here the second major part, the separate explanation, within which are two: first, separately explaining four dharmas; second, separately referring to the remaining text.” The three point to one and the same passage in differing words, and Fabao’s setting up besides a list of the four links is the most distinct of all.

The handling of the Sautrāntika defense. Puguang’s Record, at the stretch “from ‘why should it not be granted’ down to ‘what contradiction in reasoning’,” expressly marks “the intent of the Sautrāntika defense,” and unfolds of himself the two layers of the argument, seeds and manifest activity of defiled discernment, and at the end of the passage lays down a further ruling: “the author is not a Sautrāntika master, and in assessing takes the Sarvāstivādins.” Fabao’s Commentary marks the Sautrāntikas likewise without unfolding it, and adduces instead the scriptural text “the sūtra says that ignorance’s defiling of discernment is not necessarily by association, the several defiled discernments interspersed among wholesome discernment making it impure being what is called defiling.” Yuanhui’s Verse Commentary gathers it all under a fourfold outline, saying, “the first line adduces the proof, and the three lines below refute the divergent account,” without touching on seeds at all. The three differ widely in fullness and brevity, and are recorded as divergent without adjudication.

Dharmatrāta’s account. Fabao’s Commentary supplies an explanation under “the class-nature of relying on a self,” saying, “the ignorance this Venerable One speaks of is just the class-nature of relying on a self spoken of in another sūtra,” and explains “differing from the pride ‘I am’” by the nine classes of pride. Yuanhui’s Verse Commentary cites outright as its conclusion, “the Venerable Dharmatrāta says: the entity of ignorance is the class-nature of relying on a self which sentient beings have.” Puguang’s Record sets up no separate section for this account.

The school-label on the entity of contact. Yuanhui’s Verse Commentary, at the end of the passage on the meeting of three, expressly marks where the tenet belongs: “according to the Sarvāstivādins, apart from the meeting of three there is a separate entity of contact, which is a mental factor.” Puguang’s Record, for its part, explains the same question by the division “meeting is of two kinds,” without naming a school. To mark or not to mark is a difference of practice between the two commentaries.

The supplement on contact of designation. Puguang’s Record cites the Treatise Conforming to Correct Principle to supply one further sense — “because it takes resistant faculties as its support; because resistant dharmas alone are its object field” — adding a further layer of “object field” beyond “support,” which is not in the treatise’s text. Fabao’s Commentary and Yuanhui’s Verse Commentary have no such supplement.

The argument for the number eighteen. Puguang’s Record twice cites the Treatise Conforming to Correct Principle to supply an explanation of why the pleasure of the third meditative absorption is not established as a proximate activity — “if so, there ought to be no mental proximate-activity of equanimity, since it has nothing to stand opposite it,” and “moreover the pleasure of that level stagnates upon its object, whereas a proximate activity shifts again and again upon its object.” Fabao’s Commentary cites instead the Mahāvibhāṣā’s third explanation to supply a further gloss: “moreover, because it depends on mind it approaches the object and moves, and so is named mental proximate-activity.” Yuanhui’s Verse Commentary cites neither, glossing only as the text runs, “these come to eighteen, complete by three.” Puguang’s Record and Fabao’s Commentary both like to bring in outside treatises for support, while Yuanhui’s Verse Commentary is markedly sparing — this is the standing condition of the three commentaries as seen in this fascicle.

The taking and leaving of the passage on Vasubandhu’s divergent account. This is the most conspicuous divergence of the three commentaries in this fascicle. Fabao’s Commentary not only records it but fixes its character — “third, the author provisionally puts forward a divergent account in order to refute the Sarvāstivādins” — and brings Saṅghabhadra’s counter-refutation whole into his own commentary, as with “how is it certainly known? Only craving, or hating, or setting aside without discriminating among the several objects is a proximate activity.” Puguang’s Record and Yuanhui’s Verse Commentary also explain it sense by sense — Puguang marking where it belongs with “from here the divergent account is reported. It may be an account of the Sautrāntika masters. A proximate activity is defiled alone, drawn on together with mind, roaming again and again upon its object,” and going on to match the three poisons with joy, sorrow, and equanimity and to cite the six constant abidings in proof; while Yuanhui adjudicates the school outright with “the eighteen proximate activities discussed above all extend to what is with outflows, which is the Vibhāṣā school’s position. Yet according to the Sautrāntikas, within what is with outflows only the defiled is taken and named a proximate activity,” and goes on to explain the thirty-six teacher’s sentences based on relish and on renunciation. Fabao’s Commentary alone takes in Saṅghabhadra’s refutation as well, so that the full text of the two-party debate survives in Fabao’s Commentary alone. Among the three commentaries only Fabao’s preserves this complete case, the other two passing straight over it.

The choice of material on the four nutriments. Fabao’s Commentary alone cites the Treatise Conforming to Correct Principle’s objection about Uttarakuru — “if a man is born in Uttarakuru, will his lifespan be broken off for want of swallowing morsels?” — pushing to the extreme the question whether not eating means not being able to abide; Puguang’s Record lays weight on the glossing of sensory instances (receiving portion by portion with mouth and nose, sunlight and the flame of fire); Yuanhui’s Verse Commentary lays weight on the reasoning about one already free from craving who sees fine food and takes no benefit. On one and the same question of morsel nutriment the three authorities each take a different road.

The supplement on the three classes. Fabao’s Commentary cites the Treatise Conforming to Correct Principle to set forth why “rightness” takes cutting off without remainder alone for its sense — “why is cutting off alone called rightness? Because it permanently exhausts the false and counterfeit dharmas” — and expressly marks the several grounds under “they may afterwards fall into the class fixed in wrongness” as four explanations. Yuanhui’s Verse Commentary gathers it merely with “the upper line states it, the lower line explains it.” Puguang’s Record, for its part, establishes a division for this passage as “the second major part showing the distinction of the three classes,” and offers on the three words “birth, abiding, passing away” a separate explanation by the four existences of the existence of birth, existence proper, the existence of death, and intermediate existence, and at the end cites the three-class question and answer of the Treatise on the Collected Various Gates, Fascicle Four, in proof.

The lacuna in Shentai’s Commentary. Shentai’s Commentary is an incomplete work, the Chapter on the World wholly wanting, and this fascicle has no line-by-line explanation of his to attach — the lacuna is marked as it stands, and the text of another fascicle is not made to serve for it. Only the stanza count of the Chapter on the World recorded in the general conspectus of the whole treatise in his Fascicle One still serves as a witness to this chapter’s stanza count.

2. Saṅghabhadra’s Counter-Refutation

In the Treatise Conforming to Correct Principle Saṅghabhadra calls Vasubandhu “the Sūtra-master.” Within the passage corresponding to this fascicle the Sūtra-master is expressly named in two places, both given in full below.

First, the defense of ignorance’s defiling of discernment. Saṅghabhadra first marks, “here the Sūtra-master provisionally makes a statement in defense,” restates the Kośa’s original words, “the several defiled discernments, interspersed among wholesome discernment, make it impure, and this is said to be able to defile,” and at once rules, “examining his words now, they are not a good analysis, since apart from the class of what is associated there can be no defiling,” bringing out two reductions — “it would follow that one who is not an arhat would have no undefiled mind” and “moreover, undefiled discernment being mixed with defiled discernment, it would follow that the defiled turns into the undefiled.” The Treatise Revealing the Doctrine has the same passage in an abridged version, marking likewise “here the Sūtra-master provisionally makes a statement in defense,” and keeping only the one sentence, “this defense will not do, since the several discernments without outflows would be defiled.” One meaning and two schools: Vasubandhu records one defense — inferring from the ready-made scriptural statement that greed defiles the mind, that the force of the interspersing is enough for “defiling” to be spoken of, and afterwards adjudicating for himself, “this account is the better one,” and letting the defense go; Saṅghabhadra objects by comparing the forces of what counteracts and what is counteracted, holding that to speak of defiling without waiting upon association must lead to several faults. Recorded together without adjudication.

Second, the divergent account of the eighteen mental proximate-activities. Saṅghabhadra marks, “here the Sūtra-master provisionally puts forward a divergent account,” and having restated Vasubandhu’s words objects, “it is not clear by what sense of a mark the Sūtra-master takes a mental proximate-activity to be harbored within the mind,” and again, “that the arhat has no mental proximate-activity is somewhat contrary to correct principle and to the sūtras,” ruling at the end, “therefore what the Mahāvibhāṣā says is the better.” The Treatise Revealing the Doctrine wholly lacks this exchange — its text runs from “there is no proximate activity that extends to what is without outflows,” stating in order what is possessed, what is acquired, and the differences of the objects taken, and passing straight to the thirty-six teacher’s sentences and to what is based on relish and on renunciation, and at the end turning directly to “on what ground are the remaining links of existence not stated?”, so that Vasubandhu’s declaration of a divergent account and Saṅghabhadra’s refutation have no place at all within the scene of this passage. On one and the same type of debate the Treatise Revealing the Doctrine keeps an abridged version in the one case and omits the other wholly, taking and leaving inconsistently, which is honestly recorded and not made into a generalized judgment.

Third, the three-sided situation within the passage on the four nutriments. Within the four passages on the four nutriments, the cutting off of wholesome roots, death and birth, and the three classes, no direct counter-refutation marked by the name “the Sūtra-master” is found in the Treatise Conforming to Correct Principle. Only there is one further passage refuting “the Elder” — “yet the Elder says: whatever aggregate is drunk and eaten is an entity of nutriment, since nothing is stated apart” — which Saṅghabhadra refutes with “his words are unsound. Why so? Because although nothing is stated apart, it should be taken apart,” and establishes “moreover only the sense field of the tangible is the true entity of nutriment.” Here what Saṅghabhadra refutes is not Vasubandhu, and his conclusion runs in the same direction as the Kośa’s “form is not morsel nutriment” — it is another configuration, of the three sides of the Sarvāstivādins, Vasubandhu, and the Elder, and not a two-party debate, and it is therefore set down as a separate item and recorded.

Fourth, under the single link of ignorance there are besides three accounts of the Elder and one of the Venerable Rāma. Besides the two places of “the Sūtra-master” in this fascicle, Saṅghabhadra restates and refutes one by one three ways in which the Elder explains “ignorance”: first, “it is because of the presence of this that clear knowing is not present, and this is called ignorance” — explaining ignorance by the not being present of clear knowing; second, “or else mind and mental factors of such a kind are together called ignorance” — calling a kind of mind and mental factors by that general name; third, “or else inverted clear knowing is just what is called ignorance” — explaining it by inverted clear knowing. And he records one account of the Venerable Rāma: “the entity of wrong view is not itself ignorance.” The speaker of these four accounts is in no case Vasubandhu, and so they do not enter the roll of “the Sūtra-master”; yet they are divergent explanations of the same topic, and in reading the link of ignorance in this unit one should know that besides this treatise there are these further roads. Recorded together without adjudication.

Fifth, on the two topics of “name” and of “consciousness and name-and-form as conditions for each other” Saṅghabhadra again refutes the Elder, and records the Venerable Rāma’s model. The Elder explains “name” as “it conduces to the completing of this and that continuum of sentient beings, and so it is said to be name”; and at the passage on consciousness and name-and-form as conditions for each other he twice “frames a hypothetical objection.” The Venerable Rāma, for his part, sets up a model of his own, in which name-and-form in intermediate existence draws forth the consciousness that joins rebirth, and the ceasing of consciousness draws forth in turn the name-and-form of the existence of birth. Saṅghabhadra concludes severally with “therefore his questions and answers are all unsound,” “there are but empty words, contrary to reasoning and to the teaching,” and “examining the Elder’s school on the sense of consciousness and name-and-form being conditions for each other, in reasoning it certainly does not hold.” The Treatise Revealing the Doctrine writes here simply, “for the rest of the extended determination, see the Treatise Conforming to Correct Principle” — the abridged version sending the reader back to the extended one, which shows that this is a whole scene and not a scattering of incidental citations. Recorded together without adjudication.

3. Divergences of Paramārtha’s Older Translation

Paramārtha’s Expository Treatise on the Treasury forms a system of its own in the passage corresponding to this fascicle. In format, Xuanzang gives the whole stanza first under “the stanza says / the treatise says,” while Paramārtha embeds his explanation line by line under “the verse says / the explanation says,” often breaking one stanza into two to four separately cited places; hence Xuanzang’s sixteen blocks appear in Paramārtha as a number of short verse-lines scattered about.

In rendering, the greatest systematic difference is mental proximate-activity — Paramārtha has throughout “mind-discriminating activity” or “mental discriminating activity,” running through the whole passage. The rest — resistant / obstructive, designation / dependent on speech, object / dust, outflow / flow, affliction / confusion, aggregate / skandha, restoring wholesome roots / joining wholesome roots, Dharmatrāta rendered by sense / Dharmatrāta transcribed — each forms a system of its own. “Dependent on speech” takes directly the verbal designation as the object taken, while “designation” takes the dominance of speech, the two renderings taking different angles upon one and the same thing.

Nor should the convergences go unrecorded. The five characters of the stanza “as with ‘not a kinsman,’ ‘not true,’ and the rest” agree word for word in the two versions; the ten characters of “from these six feelings are produced” and “five belong to the body, the rest to the mind” likewise agree word for word, only that Paramārtha breaks them into two verse-lines; the six characters of “among the four kinds of existence, the existence of birth” agree wholly, only that Xuanzang has “defiled alone” and Paramārtha “necessarily defiled.” It is not that the whole is divergent throughout: in basic declarative sentences the wording of the two generations of translators can coincide to a high degree — convergence and divergence standing together, recorded without adjudication.

There are differences of dividing and joining as well. At the vital points Paramārtha adds a quoted verse, “within the body there are such places, and touching them accordingly brings death,” to which Xuanzang’s version has nothing corresponding; while for the delimitation of morsel nutriment as being in the desire realm alone, Paramārtha’s passage unfolds by question and answer layer by layer, and no single sentence closely matching Xuanzang word for word is found — this being a difference of format and not of doctrine, and no correspondence is forced. As for the difference of the one word in “not in absorption, not without mind, the two” and “not one, not without mind, the two,” only the two versions are recorded word for word, and no adjudication is made on their behalf.

Fascicle boundaries: Paramārtha divides the Chapter on the World into four fascicles and Xuanzang into five; the title of Paramārtha’s “Fascicle Eight” falls exactly inside this fascicle (the discussion of the eighteen mental proximate-activities finished and the five similes about to begin), and does not correspond at all to Xuanzang’s fascicle boundary. The two translations’ divisions each have their reason, and this section only records them together.

4. Two Cases Recorded Together without Adjudication

First, the entity and realm affiliation of the four nutriments. On the matter of the four nutriments this fascicle really has several layers of accounts standing together which the author does not adjudicate one by one, and they are recorded here layer by layer.

The first layer, the breadth or narrowness of the entity of nutriment. On the one hand the author says “the several dharmas with outflows all make existence grow” — every dharma with outflows nourishes birth and death; on the other he says “even so, to state four with respect to what is excellent is no fault,” establishing four only with respect to what is excellent. “The several dharmas with outflows all make existence grow” is the wording of an objection set, which the treatise delimits by “to state four with respect to what is excellent is no fault” — it does not raise every dharma with outflows into an entity of nutriment.

The second layer, whether nutriment extends to what is without outflows. This question the author does not answer for himself but records the Vaibhāṣikas’ explanation: “to be able to assist the several existences is the sense of nutriment; what is without outflows arises from cultivation and is for the extinguishing of the several existences.” This is a definitional solution of the Sarvāstivāda line — define “nutriment” first as nourishing existence, and what is without outflows is excluded of itself. Should a reader ask whether what is without outflows can assist body and mind, the author does not answer directly here; what is established here is a boundary of naming and not an adjudication of the thing itself.

The third layer, what consciousness nutriment gathers. The stanza fixes the division of labor of volition nutriment and consciousness nutriment by “the latter two, with respect to the existence to come, draw forth and raise up, in that order”: volition nutriment draws forth the existence to come, consciousness nutriment raises up the existence to come — “it is by the force of the seeds of consciousness perfumed by action that the later existence can arise.” Note that the office of consciousness nutriment here is raising up, not drawing forth; while under the author’s question “how many nutriments can make beings that have come to be abide? how many can assist sentient beings that seek birth?” the answer is “all four in each case” — the four nutriments having a share in both what is born and what is not yet born. “All four in each case” and the twofold-and-twofold division of labor are not two unadjudicated accounts: the four nutriments act upon both what has come to be and what seeks birth, and so it is said all four in each case; while morsel and contact excel in assisting what has come to be, and volition and consciousness in drawing forth and raising up the existence to come, and so with respect to the excellent capacity only these four nutriments are stated. Function extends, and excellence has its own province; the two statements hold at once.

The fourth layer, the entity and realm affiliation of morsel nutriment. Its entity is the three sense objects of odor, taste, and the tangible and not the lump of food; its realm affiliation is the desire realm alone, “since it is by becoming free from craving for morsel nutriment that one is born in the upper realms.” On the side of collation there happens to be a variant just here — the base text reads “odor, taste, and the tangible, these three” while the Palace edition reads “two”; the number three agrees with “received portion by portion with mouth and nose” below, and the number two does not. This may serve as corroboration from the self-consistency of the context, and not as a choice among collated editions.

The fifth layer, the difference from other treatises. To the Treatise on the Collected Various Gates’ account of the great living beings of the sea the author says expressly, “this is unsound, since it contradicts the sense of nutriment,” and gives a correct interpretation of his own; to the Treatise on Categories of Dharmas’ definition of morsel nutriment he reconciles by “it states it as named nutriment merely with respect to the side of being able to assist.” Both belong to the treatises of this school, and the one is refuted while the other is reconciled; the author’s taking and leaving is plain in the text, and this section only records the fact and does not assess its rightness.

The sixth layer, the dispute between Saṅghabhadra and the Elder. Within the passage on the four nutriments Saṅghabhadra separately refutes the Elder’s account that “whatever aggregate is drunk and eaten is an entity of nutriment,” and establishes “only the sense field of the tangible is the true entity of nutriment,” which is narrower than the Kośa’s “odor, taste, and the tangible, these three.” The Kośa, Saṅghabhadra, and the Elder differ each from each in the breadth of the entity of morsel nutriment, and they are recorded together without adjudication.

Second, the eighteen mental proximate-activities. After the list of what is possessed Vasubandhu declares of himself, “yet the sense of the sūtras as I see it is otherwise,” establishing that only what is defiled is named a proximate activity, with the six constant abidings and “it is not that the arhat has no worldly joy taking a wholesome dharma as object” for evidence; Saṅghabhadra counter-refutes, ruling at the end, “therefore what the Mahāvibhāṣā says is the better”; the Treatise Revealing the Doctrine passes from what is possessed, what is acquired, and the differences of objects taken straight to the thirty-six teacher’s sentences and the remaining links of existence, and does not bring out this divergent account or Saṅghabhadra’s refutation; Paramārtha’s version alone carries the corresponding text of this passage, which proves it to be text that the original treatise of Vasubandhu had and not an addition peculiar to the transmission of Xuanzang’s line; and among the three Han commentaries only Fabao’s Commentary takes in Saṅghabhadra’s refutation as well, expressly marking “the author provisionally puts forward a divergent account in order to refute the Sarvāstivādins,” while Puguang’s Record and Yuanhui’s Verse Commentary each explain the divergent account without touching Saṅghabhadra’s refutation. One case, and four authorities each transmitting a choice of his own — this section records only the shape of them all in full, and does not adjudicate the victory or defeat of the two parties.

5. Coordinates for Collateral Reading

The four nutriments, the five names of intermediate existence, and the three classes established in this fascicle are all fixed doctrinal numbers of school Abhidharma, taken up by the later systems each with its own dividing and joining, and they may be read against one another.

As to the names of the four nutriments, the Yogācārabhūmi establishes four nutriments likewise under the one sense of sustaining, and for volition nutriment it has mental-volition nutriment, one character added while the number four is unchanged; and in explaining the sense of nutriment it takes the four nutriments’ nourishing of the five material faculties, the mental faculty, and the great elements upon which the faculties depend, running on the same track as this treatise’s “whatever is called nutriment must first assist one’s own faculties and great elements, and only then reach the rest.” Assisting the faculties and the great elements is a sense of nutriment granted by both systems. Only that the Yogācārabhūmi has one further place where, under the sense of sustaining, it establishes five, adding the life faculty to the four nutriments — this treatise assigning the life faculty separately to the formations not associated with mind and not entering it among the nutriments, while there it stands beside the four nutriments, and the dividing and joining may be read against each other. (The word-for-word text of those two places and the page anchors attached to them are in Section VII of this unit’s apparatus.)

As to the five names of intermediate existence, this treatise has formed by mind, seeker of birth, eater of odors, intermediate existence, and named arising; the Yogācārabhūmi on the same matter gives three names — intermediate existence, taking the arising in the interval between the two existences; gandharva, taking the going in search of odors and being assisted by odor; and proceeding by mind, taking the going to the place of birth with mind for its support. The three names intermediate existence, eater of odors, and formed by mind agree, while the two names seeker of birth and named arising are not found in that passage, and the dividing and joining of the names and their number may be read against each other.

As to the three classes, this treatise has the class fixed in rightness, the class fixed in wrongness, and the class of indeterminate nature; the three classes established in the Yogācārabhūmi have wholly the same names but put wrongness first and rightness second, the order inverted from this treatise’s, and beneath the three classes it opens each further into the two kinds of the intrinsic and the contrived.

As for the single word “name,” this treatise takes the four formless aggregates throughout, while the later learning of dharma-characteristics establishes name-body, phrase-body, and syllable-body separately among the formations not associated with mind — one and the same graph whose extension is wholly different, which is the place a beginner most easily confuses, and it is listed here as a coordinate.