俱舍論 三界五趣·四生·中有Abhidharmakośabhāṣya · The Three Realms and Five Destinies · Four Modes of Birth · Intermediate Existence

Before reading this fascicle

The Chapter on the Elements takes stock of the entities of dharmas and the Chapter on the Faculties of their functions; and when the Chapter on the World is reached the author suddenly draws the lens back, and what he asks is no longer “what is a dharma?” but “where do sentient beings dwell, how do they get there, and once there in what class are they reckoned?” The opening sentence of the fascicle is only two questions — “Mind and the rest have been analyzed in dependence upon the three realms. Now it should next be said: what are the three realms? And within each of them, how many separate places are there?” — the former closing the two preceding chapters, the latter opening the whole Chapter on the World.

What this fascicle covers is the first seven of the nine items of “the world of sentient beings” marked out by the three ancient masters. Yuanhui’s Verse Commentary marks here, “from here the third, explaining the chapter on the world, and within it there are two: first, explaining the world of sentient beings; second, explaining the receptacle world,” and beneath that divides it into two further layers — “within the explaining of the world of sentient beings there are two: first, sentient beings treated in general; second, the adjudication of the differences among the classes,” and “within sentient beings treated in general it divides again into three sections: first, explaining the arising of sentient beings; second, their abiding; third, their passing away” — and the nine items hang beneath the lowest layer, “explaining the arising of sentient beings”: explaining the three realms, the five destinies, the seven stations of consciousness, the nine abodes of sentient beings, the four stations of consciousness, the four modes of birth, intermediate existence, dependent origination, and the four existences. What Puguang’s Record marks is word for word the same as Yuanhui’s Verse Commentary, the two intervening layers included; Fabao’s Commentary, on the other hand, does not set out these two layers but divides the nine directly beneath “explaining sentient beings.” So the matter must be stated with a distinction: the names of the nine items agree word for word in all three commentaries, while the outline layers to which the nine items are attached are of two patterns (three layers in Puguang’s Record and Yuanhui’s Verse Commentary, one layer in Fabao’s Commentary). That the names can agree completely across three independent commentaries shows that these nine items are not any one master’s ingenuity but the very joints of the treatise’s own text; the difference of outline layer is recorded without adjudication. This fascicle completes the first seven items, dependent origination and the four existences being left for the next.

One more thing should be said in advance: this fascicle gathers the three longest passages of prose commentary in the whole chapter — the establishing of the three realms (over nine hundred characters), the seven stations of consciousness (over a thousand), and the proof that intermediate existence really exists (over a thousand). Wherever the prose commentary runs long, it is usually because the author is tying a knot: either there is a divergent account within his own school to be cleared away, or he means to bring in Sautrāntika words to press his own school, or he means to set up before an outside party a dharma that is his own school’s signboard. Section by section below, the place of that knot will be pointed out.

I. The Establishing of the Three Realms: Beginning from the Question “How Many Separate Places”

〔1.1〕Why the fascicle opens by asking the number of places. First ask: do the three realms not speak of the height of states — why should places be counted at all? Answer: because it is the rule of Abhidharma that “whenever a dharma is established it must be countable, gatherable, and adjudicable.” If the three realms were only adjectives there would be no way of setting sentient beings in them; and once they are to be set in them, the question “how many positions are there in all?” must be answered. So the stanza gives, as soon as it opens its mouth, three answers in one form: the desire realm is said to have twenty places, the form realm seventeen, the formless realm none. The three answers do not stand abreast but decrease as one goes: the higher one goes the less the concept “place” can stand up, until at the formless realm it is withdrawn altogether. The easy error is to read “the formless realm has no place” as meaning that the formless realm does not exist. It exists; only it cannot be asked after with “where.” In practice: this decrease is itself the hidden thread of the whole fascicle — the seven stations of consciousness, the nine abodes of sentient beings, the four modes of birth, and intermediate existence which follow are each of them asking, “can it be set down, and in which square?”

〔1.2〕How the twenty of the desire realm are counted out. The stanza says, “they are called the desire realm, twenty, by the difference of the hells and of the continents.” First ask: under the name of the desire realm only five classes are listed — the hells, animals, ghosts, human beings, and the six heavens of desire — so how do they become twenty? Answer: because the two classes of the hells and of human beings are each opened out further. The hells open into eight — Equal Reviving, Black Rope, Massed Crushing, Howling, Great Howling, Burning Heat, Great Heat, and the Unremitting; and the destiny of human beings opens by continent into four — Jambudvīpa in the south, Videha in the east, Godānīya in the west, Uttarakuru in the north. Eight and four are twelve, and with the three further items of the six heavens of desire, animals, and hungry ghosts added, they come to exactly twenty. The author at once gives the upper and lower boundaries of the desire realm as well: on the side of sentient beings it runs from the Heaven of Sovereignty over Others’ Transformations down to the Unremitting Hell; on the side of the receptacle world it runs all the way down to the wind circle. The easy error is to read the twenty as twenty storeys. Animals and hungry ghosts are each counted as one, which is not to say that they dwell in only one place; “place” here is a square of classification and not a storey of geography. In practice: hold on to this pairing, “eight hells and four continents,” and when the section on the receptacle world later sets out one by one the positions and measures of these twelve places, one will know that it is filling in the content of these squares.

〔1.3〕Seventeen and sixteen in the form realm — the first place in the fascicle where two accounts are recorded without adjudication. The stanza itself has “above this, seventeen places, named the form realm, within which,” three on each of the first three meditative absorptions and eight on the fourth, seventeen in all. Yet the very same passage of prose commentary at once records another account alongside: “the great masters of the treatises of the land of Kaśmīra all say: the places of the form realm are only sixteen,” the reason being that the Mahābrahmā Heaven is merely a high terraced pavilion above the place of the Brahmapurohita Heaven, “the dwelling of a single lord and not a separate level, as when an honored one sits upon his seat with the fourfold assembly around him” — one more person upon the seat is not one more place. First ask: the stanza and the prose commentary contradict each other; which is to be believed? Answer: the treatise makes no choice here and records the two accounts together. The easy error is to think that recording another’s account is granting it, or that what is listed in the stanza must be the settled conclusion. Here it is precisely the case that the stanza establishes one number and the prose commentary records another, and each account has its own warrant. In practice: when in reading Abhidharma one meets numbers that do not agree, the first step is not to judge right and wrong but to ask, “have these two numbers used different standards for dividing places?” — and here that is exactly so: one side takes the Mahābrahmā Heaven for a separate level, the other for a high spot upon the same level. The two accounts are recorded together, and neither is judged to prevail.

〔1.4〕Why the formless realm has no place, and how the author proves it. The treatise says, “within the formless realm there is no place at all, since dharmas without form have no location,” and adds a corroborating remark: even the nonmanifest and the formless dharmas of past and future do not abide in a location, how much less a realm wholly without dharmas of form. On what, then, does the formless realm divide into four? Answer: on “birth” — the sphere of the infinity of space, the sphere of the infinity of consciousness, the sphere of nothingness, and the sphere of neither perception nor non-perception, four in all; and the author expressly declares that “these four are not high and low by reason of place, but differ in superiority and inferiority by reason of birth alone.” First ask: since there is no location, on what ground do we say that sentient beings are “born in” the formless realm? The treatise gives two reasons in the manner of direct perception: wherever one obtains that level of absorption, there one takes birth when life ends; and when one passes away from the formless realm and is to be born in the desire or the form realm, the intermediate existence becomes present in that very place. The easy error is to read the four formless spheres as four layers of sky. They are four births called forth by four absorptions, not four pieces of ground. In practice: this item is the preparation for reading the chapter on absorption later — the four formless absorptions and the four formless births share names while differing in position, the one the cause that can call forth and the other the result called forth; and failing to distinguish them, one reads the cultivation of absorption as a change of address.

〔1.5〕Upon what the mind of the formless realm continues — an exchange that circles three times. This is the most winding place in the prose commentary of this section. The Abhidharma masters answer, “mind and the rest there continue on in dependence upon the commonality of sentient beings and the life faculty.” The objector at once asks in return: why then do sentient beings possessed of form not likewise depend upon these two dharmas alone? Answer: in those born in the realms possessed of form these two dharmas are weak. He asks again: why are those of the formless realm strong? Answer: they are born of a superior absorption, and that meditative attainment can suppress the perception of form. He objects again: if a superior absorption suffices, why establish a further support at all? The author thereupon turns the question about and asks another round: the commonality of sentient beings and the life faculty of those in the realms possessed of form proceed in dependence upon form; upon what do these two dharmas of the formless realm depend? Answer: upon each other. Question: why do those possessed of form not depend upon each other? Answer: they are weak. Question: why are the formless ones strong? Answer: they are born of a superior absorption. And it was already said that absorption can suppress the perception of form — so that one is back where one began. First ask: why does the author walk this whole circle through? The text here only sets out the exchange of question and answer and does not state his own intent; and when the round is finished he records a third route: “or else mind and mental factors depend upon each other alone, and hence the Sautrāntika masters say that the continuity of mind and the rest in the formless realm has no separate support.” The Sautrāntika account is this: mind and the rest drawn forth by a cause that has not left craving for form arise together with form and proceed in dependence upon form; mind and the rest drawn forth by a cause that has left craving for form do not arise together with form and do not proceed in dependence upon form. The easy error is to read this passage as the author endorsing the Sautrāntikas. The text itself only records that “there is another route” and passes no verdict; this fascicle records Sautrāntika accounts in two places, which is the standing practice of the treatise. In practice: whenever hereafter the prose commentary circles three times and then records another account alongside, one should recognize this manner of writing — and as to whether the author takes the recorded account for his own view, the text does not say, while the ancient masters have their own adjudication, for which see Further Resonances.

〔1.6〕How the names of the three realms are fitted on. The treatise says, “because it can uphold its own intrinsic mark it is called a realm,” and adds, “or it has the sense of lineage, as was explained above” — the two senses already explained in the Chapter on the Elements are here touched upon only in a phrase. Next it explains the formation of the words “desire realm”: the realm that belongs to desire is called the desire realm, the connecting word being dropped, as one says a pepper drink or a vajra ring. For the formless realm a further layer of discrimination is added: the formless does not take “the absence of form” for its entity — “the entity of that realm is not form and so the name formless is established; it is not that it takes the mere absence of form for its entity.” At the end a second reading is given: “this realm has the power to uphold desire” — a reading that takes the desire realm as “the realm that can uphold desire.” First ask: where do the two readings differ? The former takes desire as that to which the realm belongs, the latter as that which the realm upholds. Puguang’s Record explains this as follows: “the second explanation is by dependent-determinative analysis. This realm is the realm of desire, and so is named the desire realm. That is, it is the upholding of desire, since this realm can uphold desire” — the name being established by joining what upholds with what is upheld. The easy error is to think this a game of grammar. It decides whether the word “realm” means a container or a power. In practice: throughout the treatise, wherever one name receives two explanations, one is usually by entity and the other by function; and so it is here.

II. What the Entity of Desire Is: Three Quoted Stanzas and One Exchange

〔2.1〕The first quoted stanza — true desire does not lie in the object. The author first asks himself, “what dharma does the word desire here state?”, and answers himself, “briefly, the greed drawn forth by morsel food and by sexual union,” and thereupon quotes a stanza of scripture as proof. The stanza runs: “The many fine objects of the world are not true desire; true desire is a person’s discriminative greed. Fine objects abide in the world as they were; among them the wise have already removed desire.” First ask: since fine objects are not true desire, what is it that practice must cut off? Answer: what is cut off is the discriminative greed aroused with an object as its condition; the object remains in the world as before, and the wise only remove desire in the midst of it. This stanza is the first quoted stanza of the fascicle, warranted by the preceding words “as a stanza of scripture says,” and is therefore not a root stanza. The easy error is to read it as loose talk of the sort “objects do not matter, only mind matters,” and thereupon to let go of guarding oneself in the presence of objects as well. The stanza’s third line, “fine objects abide in the world as they were,” is exactly meant to say that objects really are there; it is not that they are absent, but that they are not what is called desire. In practice: the criterion for adjudicating the three realms was fixed in the preceding section as “where greed increases by attachment,” and this stanza is the scriptural proof of that criterion — the boundary of a realm lies not in the height of the object but in where the greed falls.

〔2.2〕The second quoted stanza — the challenge of the non-Buddhist of wrong livelihood. The treatise records that “a non-Buddhist of wrong livelihood then challenged the Venerable Śāriputra,” and his stanza is: “If the fine objects of the world are not true desire, and desire is said to be a person’s discriminative greed, then a monk should be named one who enjoys desire, since he arouses evil discriminative applied thinking.” First ask: where does the force of this objection lie? It seizes the definition of the preceding stanza and reasons backward: if desire is the same as discriminative greed, and a monk too arouses evil discriminative applied thinking, then is a monk not also one who enjoys desire? This is to fit the opponent’s own definition upon the opponent’s own men, and as a form of reduction it is very sharp. Puguang’s Record supplies the scene: “the Venerable Śāriputra once spoke this stanza for others. At that time a certain non-Buddhist of wrong livelihood was standing not far off, and challenged Śāriputra with a stanza” — which shows that this was an exchange on the spot and not an objection set up on paper. The easy error is to read the non-Buddhist’s objection as mere unreasonableness. It does hold logically, and the author records it precisely because it holds, which is what gives weight to the counter-question of the next stanza. In practice: within Abhidharma, wherever an objection of the non-Buddhists is quoted it is mostly in order to force out the precision of one’s own definition; in reading, first grant that the objection has reason, and then watch how the answer narrows the definition.

〔2.3〕The third quoted stanza — Śāriputra’s counter-question. The treatise records that “Śāriputra then questioned him back,” and his stanza is: “If the fine objects of the world just are true desire, and desire is said not to be a person’s discriminative greed, then your own teacher should be named one who enjoys desire, since he constantly gazes upon agreeable and beautiful forms.” First ask: is this an answer or a counter-strike? In form it is a counter-strike — it fits the opponent’s own thesis (that the object just is desire) upon the opponent’s own teacher: your teacher gazes all day upon agreeable and beautiful forms; is he not then one who enjoys desire? Puguang’s Record records the outcome as “at that the non-Buddhist was silent and could not answer, since his teacher does indeed gaze upon lovely forms,” and supplies as well Śāriputra’s own positive answer: “then Śāriputra said to the non-Buddhist: to arouse evil applied thinking is indeed to be named enjoying desire. It is not that monks all arouse unwholesome discriminative applied thinking toward the world’s fine objects, and so you ought not to raise this challenge” — arousing evil applied thinking is indeed called enjoying desire, but it is not that a monk necessarily arouses evil applied thinking as soon as he sees an object. The easy error is to suppose that the two stanzas are mere wrangling and to pass over them. The third stanza is effective precisely because it shows that the route “the object just is desire” cannot be walked — for if the object were desire, no one who sees an object could escape. In practice: the three stanzas taken together provide both the scriptural and the dialectical warrant for the adjudication of affiliation to the three realms throughout the fascicle.

〔2.4〕The adjudication of affiliation to the three realms, and the answer about tying a horse. The author next sets himself a question: is every dharma active in a given realm therefore said to belong to that realm? Answer: no; it must be that within it “what has the greed of the three realms increasing upon it by attachment belongs to the three realms.” The objector asks again: what is the greed of the three realms? Answer: the greed that increases by attachment within each of the three realms. This answer looks circular, and the objector reproaches him with it: “what is now said is like the answer about tying a horse. It is as if someone asked, who is the one tying the horse? Answer: the horse’s owner. And he asked again, who is the horse’s owner? Answer: the one tying it. Two answers such as these both fail to make one understand.” Puguang’s Record fills in the concrete matter of the simile: “the non-Buddhists kill a horse in sacrifice to a god and tie the horse to a post. Someone asks, ‘who is the one tying the horse?’ Answer: ‘the horse’s owner.’ ‘Who is the horse’s owner?’ Answer: ‘the one tying it.’ Two answers such as these both fail to make one understand” — the two answers push each other about, and neither unties anything. First ask: knowing that it looks circular, why does the author answer so? Because he unties it immediately after: for one not yet free from greed toward the various places of the desire realm, that greed is called greed for desire, and what has it increasing upon it by attachment belongs to the desire realm; the form and the formless realms accordingly. Or, taking another route, one may adjudicate by level: the greed of the level that is not a level of absorption is called greed for desire, the greed of the levels of meditative absorption greed for form, and the greed of the formless levels greed for the formless. The easy error is to read “the answer about tying a horse” as the author admitting that he has answered wrongly. That is the reproach the objector adds, and the treatise at once answers with “what is now said is not like that answer,” dividing it by the two explanations, by place and by level. In practice: the two explanations, one by place and one by level, are the very things the chapter on the latent tendencies later uses when adjudicating the realm affiliation of the ninety-eight latent tendencies — and it uses the latter route.

〔2.5〕Three appended discussions — greed toward a transformation mind, the boundlessness of the three realms, and the worlds abiding crosswise. This section gathers three small questions that look scattered, and each in fact hangs upon something. First: how is greed for desire aroused toward a transformation mind of the desire realm? Answer: either by hearing of it from another; or by having oneself fallen back from it and arousing a relishing of it; or by seeing the sovereign power of one who can work transformations and arousing greed. And further, if a mind can bring the two dharmas of odor and taste into being by transformation, that mind able to transform definitely belongs to the desire realm, since a mind of the form realm cannot transform odor and taste into being. Second: are the three realms only this one set? Answer: “the three realms are boundless, as measureless as space,” and so, although there is no sentient being that first begins to be, measureless Buddhas appear in the world and convert countless sentient beings and cause them to enter nirvāṇa without remainder, and the three realms are still not exhausted. Third: how do the worlds abide? Answer: “it should be said that they abide crosswise,” with the sūtra’s simile of raindrops as thick as a cart axle quoted, saying that in the east there are measureless worlds without interval and without break, now dissolving, now forming, and that the south, the west, and the north are likewise, the above and below not being spoken of; and another account is recorded, that above and below there are worlds too, since a sūtra of another school speaks of the ten directions. First ask: why are these three questions crowded in here? Because all three are drawing the boundary of the concept “realm”: greed toward a transformation mind fixes its affiliation, the boundless measure fixes its breadth, and the aspect of abiding crosswise fixes its arrangement. The easy error is to read “the three realms are boundless” as meaning that, since they can never be delivered, there is no need to deliver them. The tone of the text is just the contrary — because they are inexhaustible, the Buddhas who appear in the world are likewise measureless. In practice: the last two sentences, that “if one becomes free from the greed of any one desire realm, all greed of the desire realms is extinguished and left behind,” and that a supernormal power reaches only as far as the Brahmā world of the realm of one’s own birth, are two buckles that pull “realm” back from cosmology to the doctrine of practice and realization.

III. The Five Destinies: A Classification That Looks Simple and Yet Must Be Argued Four Times

〔3.1〕The establishing of the five destinies and “unobscured-indeterminate alone.” The stanza says, “within them, the hells and the rest are called the five destinies by their own names; unobscured-indeterminate alone, sentient beings and not intermediate existence.” First ask: are the five destinies not simply the hells, animals, ghosts, human beings, and gods — what is left to argue? What is argued is their entity. The author adjudicates the entity of the five destinies as “unobscured-indeterminate alone,” and the reason is one sentence: “the entity of the five destinies is unobscured-indeterminate alone; were it otherwise the destinies would be mixed together” — because sentient beings within one and the same destiny carry in themselves the actions and afflictions that lead to all five destinies; and if the wholesome and the defiled were reckoned into the entity of a destiny, one person would then possess all five destinies at once. So the entity of a destiny can take only that portion which is neither wholesome nor evil and is not obscured either — namely the unobscured-indeterminate. The author adds further that within the three realms there are indeed things not gathered into a destiny: “there are, namely the wholesome, the defiled, the outer receptacle world, and intermediate existence, which although of the nature of a realm are not gathered into a destiny.” The easy error is to read “unobscured-indeterminate” as “the maturation of the five destinies is neither painful nor pleasant.” Indeterminate speaks of the ethical quality of wholesome and unwholesome, not of the feeling of pain and pleasure; the hells are extremely painful and yet the entity of their destiny is still indeterminate. In practice: this adjudication is the gate defining the word “destiny” for the whole treatise, and both the chapter on action when it treats the projecting and the completing action, and the chapter on the latent tendencies when it treats realm affiliation and increase by attachment, must come back to this square.

〔3.2〕Why intermediate existence does not count as a destiny — four rounds of exchange. The treatise first quotes the Treatise of Designations, “the four modes of birth gather in the five destinies; the five do not gather in the four modes of birth,” and further quotes the Dharmaskandha and the item of the seven existences in the Sūtra of the Seven Existences (the existences of the five destinies, the existence of action, and intermediate existence) as proof, showing that scripture itself sets intermediate existence out separately from the destinies. Then come four rounds of exchange. First, the objector reproaches him with a treatise of the schools: “if it be so, how is the Treatise on Categories of Dharmas to be reconciled? For it says that the five destinies have all the latent tendencies increasing upon them by attachment” — if the entity of a destiny is unobscured-indeterminate alone, how can that treatise say that the five destinies have all the latent tendencies increasing upon them by attachment? The treatise answers, “that says that within the rebirth-linking mind of the five destinies all the afflictions of the five categories may be present, together with the destiny and the mind that enters it; stating them in sum as a destiny is not contradictory,” with the simile “as a village and the outskirts of a village are together named a village.” Here “the destiny and the mind that enters it” must be read as Puguang’s Record reads it — “at the rebirth-linking of the five destinies the afflictions may arise, and this is named the mind entering a destiny. The destiny and the mind entering it are together named a destiny” — that is, the destiny and the entering mind are stated together, and not two minds, a “destiny mind” and an “entering mind” (Fabao’s Commentary on the same passage likewise has “rebirth-linking is provisionally spoken of as a destiny, as the outskirts of a village are provisionally named a village,” running on the same track). The afflictions of the five categories in the rebirth-linking mind, and that mind which enters a destiny, are the edge of a destiny and not its entity, and when stated in sum may be named a destiny together with it, so that the two accounts do not conflict. Clearing a misreading: reading this round as the author retracting and allowing that the entity of a destiny extends to the defiled — what he allows is the breadth of “stating in sum,” not the breadth of the “entity.” In practice: this is the treatise’s standard handling of a conflict between itself and a treatise of the schools — divide entity from edge, and do not alter the adjudication of the entity. Second, the objector asks: the existence of action is a destiny too, so why is it stated separately? Answer: it is stated separately in order to display the cause of a destiny, as among the five turbidities affliction and view are stated separately. Third, the objector presses on: if so, intermediate existence too should be a destiny. Answer: “a destiny is what is gone to, and one cannot say that intermediate existence is what is gone to, since it arises in the very place of death” — a destiny is the destination, intermediate existence is on the road. Fourth, the objector presses again: then the formless realm too should be no destiny, since there also one takes birth in the very place of death. The author draws back one sentence: since intermediate existence is named intermediate existence it should not be named a destiny, “because it lies between two destinies it is named intermediate existence; and if this were gathered into a destiny it would not be between, and then it should not be named intermediate existence.” First ask: what is really at issue in these four rounds? Whether the word “destiny” takes the sense “what is gone to” or the sense “class.” The easy error is to suppose that the author is denying that intermediate existence exists — quite the contrary, the whole latter half of this fascicle proves that intermediate existence really exists; here he only refuses to let it be entered in the square of the five destinies. In practice: that a thing “exists” and that it is “entered in a given classification” are two matters, and this is one of the most important senses of proportion in Abhidharma.

〔3.3〕Reconciling two passages of scripture, and the three accounts of the entity of a destiny. At the end the proponent of the earlier view that the destinies extend to the wholesome and defiled turns back to reconcile that passage of Śāriputra’s scripture recited in Kaśmīra. The sūtra says, “once the maturation has arisen he is named a naraka-being,” which he explains as: only when the maturation has arisen is he called a hell-being — not that a hell-being is maturation only. The sūtra says further, “apart from the dharmas that are the five aggregates that naraka-being is not to be found,” which he explains as: this was said “in order to block a really existing person who goes to the several destinies,” and not in order to block away the remaining aggregates. First ask: why must these two sentences be cleared up in particular? Because if they were not, the first would be read as an isolated proof that “the entity of a destiny is maturation only,” and the second as an annihilationist saying that “apart from the five aggregates there is nothing whatever.” At the end the treatise sets three accounts side by side: “the Vaibhāṣikas say: a destiny is unobscured-indeterminate alone,” “there is another account: it is invariably born of maturation,” “and other masters say: it extends to the nourished as well.” The easy error is to read the three accounts as three layers of one matter. The first account concludes by returning to the Vaibhāṣika school itself (the entity of a destiny is unobscured-indeterminate alone), the second narrows further to “invariably born of maturation,” and the third widens to “extends to the nourished as well” — the third is a widening and not a further narrowing. Which of the three is correct the text of the treatise does not say in a single word; the ancient masters have their own adjudication, for which see Further Resonances. In practice: the sentence “a person is not really existent” is here mentioned only in passing, and is the proper topic only in the chapter refuting the self; here a first spadeful is buried.

IV. The Seven Stations of Consciousness: The Seven Places Consciousness Is Willing to Stay In

〔4.1〕The name “station of consciousness” and the number seven. The stanza says, “different in body and different in perception; different in body and one in perception; the reverse of these in body and perception; and the lower three of the formless — hence the stations of consciousness are seven, the rest not, since they injure.” First ask: what is a station of consciousness? The treatise answers, “the five aggregates or the four aggregates affiliated to that place, each as applicable, is what is called a station of consciousness” — five aggregates where there is form and four where there is not, the heap of dharmas in which consciousness delights to abide. How the number seven arises: the human beings and gods of the desire realm together with the first meditative absorption (excluding the beginning of an aeon) are different in body and different in perception, and are the first; the Brahmā multitudes at the beginning of an aeon are different in body and one in perception, and are the second; the Ābhāsvara gods are one in body and different in perception, and are the third; the Śubhakṛtsna gods are one in body and one in perception, and are the fourth; and adding the lower three formless spheres (infinity of space, infinity of consciousness, nothingness) they come to seven. The easy error is to read the “different” of “different in body and different in perception” as a difference of better and worse. Different means only “not the same”: different in body means that color, shape, and appearance are unlike; different in perception means that the perceptions of suffering, of pleasure, and of what is neither differ. In practice: this section is a rare instance in the treatise of classifying by a fourfold analysis on the two axes of body and perception, the four alternatives being set out and three further places then added, whereupon the number closes — and to see this structure is far more useful than memorizing the seven names.

〔4.2〕The second station of consciousness: one phrase, “the cause of their perception is one,” and a question about having once seen the Brahmā king. The second station of consciousness is the Brahmā multitudes at the beginning of an aeon, different in body and one in perception. The author’s reason for “one in perception” is most unusual: it is not that the content of their perception is the same, but that “because the cause of their perception is one they are called one in perception” — the Brahmā multitudes all perceive that “we were all produced by the great Brahmā,” and the great Brahmā perceives that “these Brahmā multitudes were all produced by me,” so that both take one and the same “cause of production” for what they perceive, and hence are called one in perception; while the great Brahmā’s bodily measure is lofty and vast and his appearance, majesty, and dress are all unlike those of the Brahmā multitudes, and hence they are still called different in body. The treatise then quotes a passage of scripture: the Brahmā multitudes think that they once saw this sentient being long-lived and long abiding, and even that he aroused the wish that other sentient beings be born into his commonality. Whereupon a question arises: where had the Brahmā multitudes once seen the Brahmā king? The treatise records two explanations in a row and refutes both — that they saw him while dwelling in the Ābhāsvara Heaven (objection: then why, not obtaining the second meditative absorption now, can they recall the former abodes of that level; and if they had obtained the second absorption, how could they arouse the grasping of rules and observances with the great Brahmā as their object) and that they saw him while dwelling in intermediate existence (objection: intermediate existence has no sense of abiding a long time, so how could they see him “long-lived and long abiding”) — and at the end takes a third explanation: “therefore the Brahmā multitudes, dwelling in their own heaven, recall what they had passed through before this life; that is, having seen him long-lived and long abiding earlier, when they see him again later they arouse this thought.” First ask: why is so much ink spent on this string? Because the establishing of “one in perception” hangs entirely on this passage of scripture; and if the scripture cannot be made out, the second station of consciousness falls to the ground. The easy error is to take the refutation of the two explanations for the author retailing anecdotes of the heavens. It is the guarding of a line of classification. In practice: whenever the author refutes two explanations in a row and then sets up a third himself, the third explanation is the treatise’s own correct sense, and in reading one must be clear which sentence is the conclusion.

〔4.3〕The third station of consciousness: raising the last to gather in the first, and the “difference in perception” of the Ābhāsvara gods. The third station of consciousness raises the Ābhāsvara gods, one in body and different in perception. The author first adds a phrase about the outline: “here the last is raised and the first thereby gathered in; it should be known that the whole second meditative absorption is included” — otherwise the Parīttābha and the Apramāṇābha Heavens would have nowhere to belong. One in body means that color, shape, and appearance are not unlike; different in perception means that the perception of pleasure and the perception of what is neither suffering nor pleasure interweave. The author further records a transmitted account as explanation: “it is transmitted that those gods, having wearied of the faculty of joy of the fundamental level, bring the faculty of equanimity of the access level to presence; and having wearied of the faculty of equanimity of the access level, bring the faculty of joy of the fundamental level to presence,” matched with the simile “as a rich and noble man, having wearied of worldly pleasure, then receives the pleasure of the Dharma, and having wearied of the pleasure of the Dharma, again receives worldly pleasure.” The objector at once asks in return: should the Śubhakṛtsna gods not be so as well? Answer: the Śubhakṛtsna gods have never wearied of pleasure, since that pleasure is quiescent; with joy it is not so, “with joy it is not so, since it stirs and disturbs the mind.” First ask: why can joy be wearied of and pleasure not? Because joy is stirring, and what stirs breeds weariness; pleasure is quiescent, and what is quiescent is never wearied of. The easy error is to take joy and pleasure for one and the same thing differing in depth. In this treatise they are different feelings, differing both in the level they depend on and in their aspect of stirring or stillness. In practice: this discrimination will be used again and again when the chapter on absorption later adjudicates the limbs and merits of the four meditative absorptions, and here it makes its first appearance.

〔4.4〕The Sautrāntikas give another explanation — one “difference in perception,” two provenances. Here the treatise records another account, that of the Sautrāntika masters: there is another sūtra explaining the sense of the Ābhāsvara gods’ difference in perception, which says that there are newly born gods there who do not skillfully know the forming and dissolving of the world and who, seeing the flames of the level below blazing out, are struck with terror and aversion and fear that the fire will burn out the Brahmā palaces and rise to encroach upon their own place; while the gods born long before, who already know the forming and dissolving well, console them, saying, “pure seers, pure seers, do not be afraid, do not be afraid,” since in former times that fire, having burned out the Brahmā palaces, went out there of itself. And so “they have the perception that the flames will come and that they will not come, and the perception of terror and of no terror, and hence are called different in perception,” and it is not that they obtain the name “different in perception” through the interweaving of the perception of pleasure and the perception of what is neither suffering nor pleasure. First ask: where do the two explanations differ? The former establishes difference in perception by the interweaving of feelings, the latter by the perceptions of terror and of no terror. The treatise records both without adjudicating. The easy error is to suppose that the latter account is the author’s settled position. The text says only “the Sautrāntika masters say,” set beside the earlier “it is transmitted.” In practice: the two places in this fascicle that record Sautrāntika accounts (the continuity of mind and the rest in the formless realm, and here the sense of difference in perception) are precisely the two scenes of the manner of composition recorded in the Life of Master Vasubandhu, that “he established the sense of the Sarvāstivādins and, wherever there was a partial position, refuted it with the sense of the Sautrāntikas — this is named the Abhidharmakośa” — whereby the reader may see in concrete shape what that sentence looks like.

〔4.5〕The fourth station of consciousness, and the three places that are not stations. The fourth station of consciousness is the Śubhakṛtsna gods, one in body and one in perception, “since there is only the perception of pleasure they are called one in perception.” The author then sets the perceptions of the four levels above and below into a single line: within the first meditative absorption they are one in perception because the perception is defiled; in the second they are different in perception because there are two wholesome perceptions; in the third they are one in perception because the perception is maturation-born; and the names of the lower three formless spheres are as in the sūtra. He then asks why the remaining places are not stations of consciousness, and answers, “because in the remaining places there are dharmas that injure consciousness.” Which are the remaining places? The evil destinations, the fourth meditative absorption, and the summit of existence. And what are the dharmas that injure consciousness in each? “That is, in the evil destinations there is heavy painful feeling that can injure consciousness; in the fourth meditative absorption there are the absorption without perception and the entity of non-perception; and in the heaven of the summit of existence there is the absorption of cessation that can injure consciousness and cause the continuity to be broken off, and so they are not stations of consciousness.” The treatise adds a second criterion: a station of consciousness must be a place where sentient beings elsewhere are glad in mind to come and stay and, having come, seek to go out no more; in the evil destinations neither sense obtains, while in the fourth meditative absorption the mind constantly seeks to go out (ordinary beings seeking non-perception, noble ones delighting in the Pure Abodes or the formless, and those in the Pure Abodes delighting in the realization of quiescence), and “the summit of existence, being dull and feeble, is not a station of consciousness.” First ask: are the two criteria a repetition or complementary? The former is by dharma (whether there are dharmas that injure consciousness), the latter by mind (whether one delights to abide). The easy error is to read the exclusion of the fourth meditative absorption as meaning that the fourth absorption is inferior. The reason for excluding it is precisely that the mind there constantly seeks to go out — not that one cannot stay, but that one is unwilling to. In practice: the establishing of a station of consciousness turns on the axis of “delighting to abide,” which is the same axis as the establishing of the abodes of sentient beings in the next section though of a different breadth, and the two sections are to be read together.

V. The Nine Abodes of Sentient Beings: Before Counting In Those Who Do Not Delight to Abide, Ask Who Delights to Abide

〔5.1〕The nine come from the seven, and the two places that are added. The stanza says, “it should be known that with the summit of existence and the non-percipient sentient beings added, these are the nine abodes of sentient beings, the rest not, since they do not delight to abide.” First ask: the stations of consciousness are already seven, so why establish nine? Answer: adding the two places of the summit of existence and the non-percipient sentient beings makes nine. The reason is that “the classes of sentient beings delight to abide in these nine alone, and so abodes of sentient beings are established.” How, then, can the summit of existence and non-perception come in here when in the section on the stations of consciousness they were plainly excluded? Because the criteria of the two sections differ: what the stations of consciousness ask is whether consciousness will be injured, what the abodes of sentient beings ask is whether sentient beings are willing to abide. Though the summit of existence is dull and feeble, sentient beings still delight to abide there; and though the non-percipient sentient beings are without mind, theirs is still a place sentient beings long for, and so it is established as an abode. As for the remaining places, the author speaks very heavily: the evil destinations are “not places where the classes of sentient beings abide of their own delight, but where the rākṣasa of evil action forces them to abide,” and “they are like a prison and are not established as abodes of sentient beings.” The fourth meditative absorption apart from the Heaven of Non-Percipient Beings is likewise not an abode of sentient beings, for the reason explained under the stations of consciousness. The easy error is to read the seven stations of consciousness and the nine abodes of sentient beings as two versions of one table. They are two different measures, and the number of squares they measure out naturally differs. In practice: whenever this treatise meets “one set of places with different numbers,” the first question is always “which two measures did these two numbers use?” — and this is the second demonstration of it in this fascicle (the first being the seventeen and the sixteen of the form realm).

VI. The Four Stations of Consciousness: Why Consciousness Cannot Abide in Itself

〔6.1〕The entity of the four stations of consciousness and “one’s own level alone.” The stanza says, “the four stations of consciousness should be known: the four aggregates, one’s own level alone; consciousness alone is said not to be a station; with outflows, gathered by a fourfold analysis.” The treatise quotes the sūtra, “consciousness abides following form, consciousness abides following feeling, consciousness abides following perception, consciousness abides following formations — these are called the four kinds,” and their entity is “the four aggregates with outflows, in due order.” Two restrictions are to be remembered: first, that they are with outflows alone; second, that “this holds on one’s own level and not on another, since consciousness adheres to them and so they are named stations of consciousness” — among the aggregates of form and the rest of another level, consciousness will not adhere there by the force of craving. First ask: why the restriction to one’s own level? Because the establishing of a station of consciousness hangs upon adherence, and adherence hangs upon craving; and craving increases only toward objects of its own level and does not obtain across levels. The easy error is to read the four stations of consciousness as an abbreviation of the seven. The seven list places and the four list aggregates — the one a map, the other materials. In practice: read with the preceding section, this shows that the treatise has two languages for “abiding,” abiding as in a place and abiding as in an object; and to mix them is to read geography as psychology.

〔6.2〕Consciousness is not a station of consciousness: the two similes of the king’s seat and of the man and the boat. The line of the stanza, “consciousness alone is said not to be a station,” is the hardest to construe. The author answers with the utmost brevity: “because what abides is set apart and that in which it abides is established” — that in which something abides is established over against what abides, and the consciousness that abides may not at the same time serve as that in which it abides. Two similes are matched with it: “as the king himself cannot be called the king’s seat,” a king and a king’s seat not being one thing; and again, if a dharma is mounted and driven by consciousness, as a man rides a boat, only then is it called a station of consciousness, and consciousness cannot mount and drive its own entity. The author expressly marks this as what the Vaibhāṣikas say. The objector at once brings scripture: since another sūtra says that “within the nutriment of consciousness there is joy and there is defilement, and because there is joy and defilement consciousness abides therein and is mounted and driven by consciousness,” is that not consciousness abiding in consciousness? And again, the seven stations of consciousness plainly have the five aggregates for their entity, consciousness among them. The author’s division is this: that was stated in sum, without analysis, of the aggregates gathered into the place of taking birth, as being able to produce joy and defilement; and when consciousness turns on it too is named a station of consciousness, which is not to state consciousness alone. Whereas the aggregates of form and the rest are each of them able to produce the several joys and defilements and make consciousness adhere, while consciousness alone cannot, and so “it is said not to be a station.” First ask: what is the practical gain of this discrimination? It takes consciousness down from the position of the object. The easy error is to suppose that the author is denying that consciousness can be taken as an object. What he denies is the setting up of consciousness in this square of a station of consciousness, not that consciousness can be an object. In practice: afterwards the author gives two further explanations — that the four stations of consciousness are like a good field while consciousness possessed of appropriation is like a seed, and “a seed may not be established as a good field,” adding honestly, “this is a reverent surmise that such is the World-Honored One’s intent in teaching”; and again that a dharma and consciousness can arise at the same time and so are established as a station of consciousness, while the aggregate of consciousness is not so. Wherever the author expressly marks a “reverent surmise,” it is an honest marking that this is conjecture and not settled doctrine, and in reading one should take it as such.

〔6.3〕The fourfold analysis of the seven and the four. At the end the author sets the two tables over against each other: the seven kinds and the four kinds, though the stations of consciousness differ, are alike with outflows; do the seven gather in the four, or the four the seven? Answer: “they do not gather each other in throughout. A fourfold analysis may be made.” The first alternative (gathered in by the seven and not by the four) is the consciousness within the seven stations of consciousness; the second (gathered in by the four and not by the seven) is the remaining aggregates other than consciousness within the evil destinations, the fourth meditative absorption, and the summit of existence; the third (by both) is the four aggregates within the seven stations of consciousness; the fourth (by neither) is all the rest excepting the foregoing. First ask: why must a fourfold analysis be made? Because the difference between the two tables falls exactly at two points — the table of seven includes consciousness while the table of four does not, and the table of four extends to every level with outflows while the table of seven does not extend to the evil destinations, the fourth meditative absorption, or the summit of existence. Once the four alternatives are made, these two differences are plain at a glance. The easy error is to take a fourfold analysis for an exercise in logic. It is in fact Abhidharma’s operation upon difference sets, used to confirm the boundaries of two definitions. In practice: throughout the treatise, whenever two enumerations of dharmas are set over against each other, the fourfold analysis is the standard procedure, and here is its first demonstration in this chapter.

VII. The Four Modes of Birth: Egg, Womb, Moisture, and Transformation, and Their Assignment among the Destinies

〔7.1〕The establishing of the four modes of birth and “birth means a class of birth.” The stanza says, “within them there are four modes of birth; sentient beings are the egg-born and the rest; human beings and animals have all four; the hells and the gods, and intermediate existence, have birth by transformation alone; ghosts extend to the two, womb and transformation.” The treatise first defines: “birth means a class of birth. Among sentient beings, although the other classes are mixed, the class of birth is equal” — sentient beings within one destiny may be of different classes of birth (the destiny of human beings has all four complete), while within one and the same class of birth they are equal. The four modes are each explained: to be born from an eggshell is birth from an egg, as with geese, peacocks, parrots, and wild geese; to be born from a womb is birth from a womb, as with elephants, horses, cattle, pigs, sheep, and donkeys; to be born from moisture is birth from moisture, as with worms, moths, mosquitoes, and centipedes; to arise with nothing to depend upon is birth by transformation, as with naraka-beings, gods, intermediate existence, and the like, “with faculties complete and unimpaired and limbs arising all at once, and because what was not there suddenly is, it is called transformation.” First ask: what is the relation of the four modes of birth to the five destinies? The Treatise of Designations was already quoted above as saying, “the four modes of birth gather in the five destinies; the five do not gather in the four modes of birth” — the four modes of birth are the wider, and what they exceed by is precisely intermediate existence. The easy error is to read the four modes of birth as a biological classification. Their criterion is “in dependence upon what a being arises,” not its shape. In practice: birth by transformation can gather in intermediate existence precisely because its definition is “with nothing to depend upon”; and this square is the place reserved for establishing intermediate existence in the next section.

〔7.2〕The destiny of human beings has all four complete — some stories transmitted by the three commentaries. The treatise gives its examples in the most concrete way: human beings born from an egg are “such as Śaila and Upaśaila, who were born from a crane’s egg,” and again the thirty-two sons borne by the Deer Mother and the five hundred sons of the king of Pañcāla; human beings born from a womb are the people of today; human beings born from moisture are such as Māndhātṛ, Cāru, Upacāru, Dove-Garland, and Āmrapālī; “human beings born by transformation are only the people at the beginning of an aeon.” Three of the modes among animals are seen by all, and animals born by transformation are such as nāgas and garuḍas. “All the hells, the gods, and intermediate existence are birth by transformation alone,” and “the destiny of ghosts extends only to the two, womb and transformation.” First ask: are these names to be investigated one by one? The treatise takes them only as examples and does not go into the concrete matter; the three commentaries, however, each supply it — Yuanhui’s Verse Commentary relates the story of the crane’s egg as “formerly on this continent a merchant went to sea and got a female crane, of splendid form and color, and delighting in the marvel he took pleasure in her. She then bore two eggs, and afterwards the eggs opened and two boys were born, upright and intelligent; and when grown they went forth and obtained the fruit of an arhat”; and Puguang’s Record relates the name Māndhātṛ as “Māndhātṛ is a king’s name, in Chinese ‘I Shall Rear Him.’ He was born from a blister on the crown of King Poṣadha, and his face and form were comely. The king took him into the palace and asked who could rear him; and each of the palace women said, ‘I shall rear him.’ Hence the name was marked. Of old he was called the King Born from the Crown.” The easy error is to take the stories in the commentaries for the text of the treatise. They are concrete matter supplied by the ancient masters to aid reading, and the treatise itself has none of it. In practice: when in reading the treatise one meets a string of personal and place names, first ask “what is the author raising it to prove?” — here it proves only the one sentence “the destiny of human beings has all four modes of birth complete,” and all the rest is branch and leaf.

〔7.3〕The proof of birth from a womb among hungry ghosts: one quoted stanza. The treatise says that the destiny of ghosts extends to the two, womb and transformation, and thereupon quotes a stanza as proof: “At night I bear five children, and as each is born I devour it; by day I bear five and it is the same — and though I eat them all, I am never filled.” It is warranted by the preceding words “as the female hungry ghost said to Maudgalyāyana,” and is a quoted stanza and not a root stanza. First ask: what is this stanza’s function in the treatise? It is the sole scriptural proof that “hungry ghosts have birth from a womb” — without this proof the destiny of ghosts could be matched only with birth by transformation, and the table matching the four modes of birth with the five destinies would have to be altered. The easy error is to read this stanza as no more than a piteous story. Within the context of the treatise it is a piece of evidence for a classification, and its very piteousness is where it is effective: to be able to bear is to have birth from a womb. In practice: the five words of the stanza, “though I eat them all, I am never filled,” are the real sense of the name “hungry ghost” — the hunger lies not in having no food but in eating and not being filled; and this sentence can be picked up again when the chapter on action later treats the cause of hungry ghosts.

〔7.4〕Birth by transformation is the most excellent — why then does the bodhisattva undergo birth from a womb? The author asks, “which of all the modes of birth is the most excellent? It should be said that the most excellent is birth by transformation alone,” and the objector at once seizes upon it: then why does the bodhisattva of the last body, who has obtained sovereignty over birth, still undergo birth from a womb? The author answers, “because undergoing birth from a womb in the present has great benefit,” and lists four benefits: first, to draw in the great men of the Śākya clan, causing them by the connection of kinship to enter the true Dharma; second, to let others know that the bodhisattva is of the lineage of a wheel-turning monarch, so that they arouse reverence and longing and thereby abandon the wrong and turn toward the right; third, to make those he converts arouse a superior mind — “since he is a man and can accomplish so great a matter, we too are such, and why should we not thereby arouse right diligence and cultivate the true Dharma single-mindedly?”; fourth, so that his clan and lineage may be known, saving people from suspecting him a conjured god or ghost, and laying to rest the non-Buddhists’ slander that “when a hundred aeons have passed a great conjuration will appear in the world and devour it.” First ask: which of the four benefits is the weightiest? The third — it pulls the bodhisattva’s taking of a womb back from the miraculous to the learnable. The easy error is to read this passage as a footnote to the life of the Buddha. It is in fact answering a difficulty of classification: birth by transformation having been adjudicated the most excellent, and the most excellent of beings being seen not to take the most excellent birth, a separate reason must be given. In practice: the four words “we too are such” are the thought of self-encouragement which the treatise imagines arising in those he converts once they see that the bodhisattva is a man as they are — “since he is a man and can accomplish so great a matter, we too are such” — and not the author stepping outside his argument to address the reader directly; and what is worth remembering is precisely this handing over.

〔7.5〕The account of leaving the bodily element behind, and three rounds of exchange. The treatise records the account of a further master: “it is in order to leave the bodily element behind that he undergoes birth from a womb” — for one born by transformation has no outer seed, and the moment his body expires no remaining form is left, “as when a lamp goes out nothing more is seen,” and then there would be no way to leave relics behind for measureless human beings and beings of other classes to make offerings to, whereby they are born in heaven a thousand times over and obtain liberation. The objector asks: if someone believes that the Buddha has the supernormal power of upholding a vow by which he can long preserve his body, then a body born by transformation could also leave a form behind. The author answers that to put it so “does not hold as an explanation, but is a discussion arising from the discussion” — it is no explanation but a growing of branches outside the trunk. The objector asks again: if a body born by transformation leaves no form behind at death, why does a sūtra say that a garuḍa born by transformation seizes a nāga born by transformation for food? Answer: it is because he does not know that it leaves no remaining form that he seizes the nāga for food, and the sūtra does not say that he was thereby filled; or else the nāga has not yet died and serves for the moment to stay his hunger, and when it dies he is hungry again — “what blame is there in eating for a moment?” At the end the author asks which of the four modes of birth is the most numerous, and answers “birth by transformation alone,” since a portion of three destinies, the whole of two destinies, and all intermediate existence are born by transformation. First ask: what have these three rounds of exchange in common? All of them guard the one item, “a body born by transformation leaves no form behind at death.” The easy error is to read the last question as an idle stroke of the brush. It is the close of this section and the lead into the next — all intermediate existence is born by transformation, and hence the establishing of intermediate existence follows directly upon it. In practice: the conclusion that birth by transformation is numerically the greatest is the very step that widens “the world” out beyond the two modes of birth, womb and egg, that we are familiar with.

VIII. Intermediate Existence: The Establishing of Its Name and Sense, and the Discrimination of “Not Birth”

〔8.1〕What intermediate existence is. The stanza says, “between the two existences of death and birth, the five aggregates are named intermediate existence; not having reached the place it should reach, intermediate existence is therefore not birth.” The treatise explains: “after the existence of death and before the existence of birth, in that very interval an entity of its own arises,” and “this body arises in order to reach the place of birth, and because it is between two destinies it is named intermediate existence.” First ask: what is the entity of intermediate existence? The five aggregates — not a thread of soul-consciousness, but an existence complete with five aggregates. This point is of the first importance in this treatise: if intermediate existence were not the five aggregates, then between death and birth there would appear a stretch that is not the five aggregates, and the whole register of conditioned dharmas would be broken. The easy error is to read intermediate existence as a consciousness that has left the body. The text says plainly, “the five aggregates are named intermediate existence.” In practice: the account by five aggregates makes intermediate existence fall within the enumerations of dharmas already established, with no need to set up a new class — and this is the consistent manner of Abhidharma’s legislation: it would rather push a new phenomenon into an old square than lightly add a square.

〔8.2〕Already arisen and yet not named birth — a matter of proportion in a single word. The objector asks: since this body has already arisen, why is it not named birth? The author answers, “birth means the place that is to be reached in time to come; the name birth is established upon the sense of what is reached” — the name “birth” is established with respect to “having arrived”; and the body of intermediate existence, though its entity has arisen, has not yet arrived there, and so is not named birth. He asks again what “the place that is to be reached in time to come” is, and the answer is, “where the maturation drawn forth is finally and clearly displayed is what is called the place to be reached in time to come.” First ask: what exactly does this discrimination discriminate? It discriminates “existing” from “arriving.” Intermediate existence does exist, but it has not yet arrived, and so may not borrow the name “birth.” The easy error is to suppose that not being named birth means that it does not count, that it is a fiction of a transitional state. The position of the text is exactly the contrary — it is a really existing body of five aggregates, only that names may not be borrowed at random. In practice: this proportion is exactly of a type with “intermediate existence is not a destiny” above: there, existing and yet not entering the square of a destiny; here, arising and yet not using the name of birth. Twice in succession — from which one may see that the author’s intent is to peel intermediate existence cleanly out of the names already established, the better to establish it alone in the next section.

IX. That Intermediate Existence Really Exists: The Two Proofs from Principle and Teaching

〔9.1〕The other school’s position, and the outline of the two proofs. The treatise records the divergent account: “some other school says: between death and birth an interval and a break may be allowed, and so there is no intermediate existence.” The author answers, “this cannot be granted,” and, asked why, answers, “why so? By reason of principle and teaching,” gathering the two proofs under one stanza: “As with the continuity of grain and the rest, the places uninterrupted and each arising in succession; an image really existing is not established; not being alike, it is no analogy. In one place no two stand together; not one continuum, and arising from two; the gandharva is spoken of, and the sūtras of five and of seven.” First ask: how are the eight lines of this stanza divided? The first six are the proof from principle and the last two the proof from teaching (what the proof from teaching points to being the Gandharva Sūtra, the Sūtra of the Five Non-Returners, and the Sūtra of the Seven Destinations of Good Men — Puguang’s Record remarking under the single word “sūtra” that “the one word sūtra displays several sūtras together”); and within the first six Puguang’s Record divides further, “the first two lines positively establish the principle, the next four dispel outside doubts,” and within those next four, “the third line explains the first line and the fourth line the second.” The easy error is to read the eight lines flat as eight parallel reasons. It is in fact a structure of “establishing the principle · dispelling doubt · adducing scripture” — which reckoned by doctrinal gates comes to “establishing one, refuting three, and adducing scripture,” and this is not an outline by the number of the stanza’s eight lines. In practice: the longest passage of prose commentary in this fascicle unfolds from here, and using these eight lines as a table of contents while reading, one will not lose the way.

〔9.2〕The positive establishing of the proof from principle: the continuity of grain and the rest. The author says, “for the moment, in accordance with correct principle, intermediate existence is not nonexistent,” and his positive reason is: “we plainly see that dharmas that turn on in continuity in the world must arise in succession moment by moment with their places uninterrupted” — whatever dharma flows on in continuity must arise in succession moment by moment with its place uninterrupted. The example is grain: the seed is here, and the sprout too comes up here. The continuity of sentient beings ought in principle to be the same, and so “arising in succession moment by moment, the place must be uninterrupted.” First ask: is this an analogy or a general rule? The author takes it as a general rule — he first establishes that “every dharma in continuity has its place uninterrupted,” and then adduces grain as an example. Yuanhui’s Verse Commentary sets it out in the form of logic as “the inference runs: death and birth are the subject, and being uninterrupted in place is the predicate of the thesis; the reason is that they arise in succession; the concordant example is the seed and the fruit. The seed and the fruit arise in succession, and the seed and the fruit have no break of interval; death and birth arise in succession, and death and birth have no break of interval”; Puguang’s Record makes a simpler statement of the inference, “the inference runs: between death and birth there must be a connection, because they arise in continuity, as with the example of the seed and the fruit” — the ordering made by the two commentaries can help the reader to see the structure, though the treatise itself does not expressly cast it as the three members of a logical proof. The easy error is to read the simile of grain as the whole of the argument. The simile of grain is only the concordant example; the real thesis is “between death and birth there must be no interval.” In practice: if there could be an interval between death and birth, then the continuity of cause and result would show a break — and this is the reason the author must establish intermediate existence.

〔9.3〕The other school’s objection: a reflected image can have an interval. The objector puts forward a very strong counter-example: it is plainly seen that there are dharmas that arise in succession with an interval of place — in dependence upon a mirror an image is produced from a substance, the substance here and the image there, with an interval between, and the successive arising still holds; then what harm is there if the existence of death and the existence of birth arise in succession though their places have an interval? First ask: where is the strength of this counter-example? It does not deny “continuity”; it denies only that “continuity must have its place uninterrupted” — and that falls exactly upon the author’s major premise. The easy error is to read it as the nonsense of a non-Buddhist. It is a structurally correct counter-example, and the author must take it apart head-on. In practice: the three refutations that follow are three routes for taking apart this one counter-example — first refuting the thing it depends on (an image is not really existent), second refuting its likeness (substance and image are not one continuum), third refuting its conditions of arising (it is produced from two conditions). Were any one of the three lacking, the general rule of the simile of grain could not be held.

〔9.4〕The first refutation: that an image really exists is not established — a refutation in four gates. The author’s first route is: “because images really existing does not hold in principle” — the image in a mirror is simply not a really existent form, and so cannot be taken as an example. The general cord of the reason is “because in one place no two stand together.” The treatise gives four rounds in order: first, that in one place the form of the mirror and the form of the image are present together, and two forms ought not to be together in one place, “since they depend upon different great elements” (the great elements they depend on differ); second, that upon a narrow water the forms and shapes of the two banks display two images at once in the same place, and those on the two banks each see clearly, and yet no one has ever seen two forms together in one place; third, that shadow and light are never in the same place, and yet a mirror has been seen hanging in shadow while the image of the light appears distinctly on the mirror’s face; fourth, that the mirror’s face and the moon’s image are stated as two and are seen separately as near and far, as when one looks into well water, and if they truly arose together how could they be seen separately. After the four rounds the author concludes, “hence it is known that images are in principle really nonexistent,” and adds an explanation of the first importance: “yet the force of the coming together of causes and conditions makes them be seen so, since the differentiations of capability in the nature of dharmas are hard to conceive” — an image is not really existent, and yet it certainly is seen, and that is brought about by the force of the coming together of causes and conditions. Puguang’s Record makes an outline of these four rounds: “within it there are four: first, refuting by mirror and image in the same place; second, by two images in the same place; third, by shadow and light in the same place; fourth, by their being seen separately as near and far.” First ask: what does refuting the real existence of an image do for intermediate existence? Without a really existent image there is no instance of “an interval of place with successive arising.” The easy error is to read “images are in principle really nonexistent” as meaning that all that is seen is nonexistent. The author says plainly that they are still visible, only without a real entity. In practice: this passage is the most concentrated place in the treatise on “visible and yet not really existent,” and may be read back against the later treatment of nonmanifest form and of form that is the result of absorption.

〔9.5〕The second and third refutations: not being alike, it is no analogy. The second route: “not being alike, it is no analogy” — granting that an image were really existent, it is still not alike to the dharma compared with it, and so the analogy still does not hold. The reason is that substance and image are not one continuum: “that is, the substance and the image are not one continuum,” an image appearing only in dependence upon a mirror and the like, and the image and the substance being present at the same time; whereas the existence of death and the existence of birth are one continuum, uninterrupted before and after, and arise in succession in another place. Set over against each other, the two have no such relation of continuity, and “since they are not alike the analogy does not hold.” The third route: “and again, the image that appears arises from two” — an image arises only from two conditions, first the substance and second the mirror and the like, and the image arises in dependence upon whichever of the two is the more excellent; whereas the existence of birth cannot arise from two conditions, “there being only the existence of death and no further and more excellent support,” and so the analogy adduced and the dharma compared with it are not alike. The author blocks one route in advance as well: nor may outer insentient things such as semen and blood be taken as the more excellent support, for one born by transformation arises suddenly in mid-air, and among those nothing can be pointed to as the more excellent support. First ask: are the three refutations successive or parallel? They are successive — first, your example does not exist at all; falling back a step, even if it did exist it is not alike; falling back another step, even the number of conditions for arising differs. The easy error is to read the three refutations as one thing said three times. In practice: this manner of refuting by three descending layers of granting and taking away is a standard procedure of this treatise, and when one meets such words as “granting that it were established” or “allowing it for the moment,” one knows that the author is speaking a step back.

〔9.6〕The proof from teaching: five collections pressed in order. The proof from principle finished, the author turns to the proof from teaching, and his method is to press back layer by layer: first he quotes the Sūtra of the Seven Existences, in which intermediate existence is among the seven existences; if that school does not recite this sūtra, he asks, “do they not recite even the Gandharva Sūtra?” — the sūtra says that entry into the mother’s womb requires three things present together, that the mother’s body be in fit condition, that father and mother join in mutual love, and that a gandharva be just present; and the author at once asks, “apart from the body of intermediate existence, what gandharva is there? The former aggregates have already perished — what is it that is present?” If this sūtra too is not recited, he asks how the Sūtra of the Horse-Keeping Clan is to be explained — the sūtra asks of this gandharva just present which of the four castes he is and from which quarter he comes, and the author adjudicates, “the former aggregates having already perished, it cannot be said to come; and so the coming spoken of here can be nothing but intermediate existence.” If they still do not recite, he asks how the Sūtra of the Five Non-Returners is to be explained — the first of the five non-returners is one who attains parinirvāṇa in the interval, and “if there were no intermediate existence, whence the name of attaining parinirvāṇa in the interval?” First ask: why press with this manner of one collection after another? Because the party addressed does not necessarily recite the same sūtras, and only by pressing collection by collection can it be shown that whichever collection you recite, intermediate existence cannot be evaded. The easy error is to read it as a piling up of scriptural proofs. It has a sequence, and every step turns upon “if you do not recite this one.” In practice: this is the passage in the treatise where the adducing of scripture as proof is most orderly, and it may be read as a model.

〔9.7〕Refuting two divergent explanations of “attaining parinirvāṇa in the interval,” and one quoted stanza that gathers it in. Upon the name “attaining parinirvāṇa in the interval” the treatise records two divergent positions and refutes them. First, certain masters hold that “there is a heaven named Interval, and one attains parinirvāṇa there, and hence the name of attaining parinirvāṇa in the interval” — which the author refutes: then one should also grant heavens named “Birth” and the rest, “and since these are not granted, such a grasping is not well said.” Second, other masters take attaining parinirvāṇa in the interval to be the cutting off of the remaining afflictions and the becoming an arhat either in the middle of the lifespan or in the middle of the approach to a heaven. This section goes on to list two routes and three grades — and it must first be declared that “two routes and three grades” is said following Puguang’s outline, the text of the treatise itself recording only in sum “other masters say further: either in the middle of the lifespan or in the middle of the approach to a heaven,” without separately setting up two routes. Following Puguang’s outline: the route of the middle of the lifespan divides into reaching the stage of the element, the stage of perception, and the stage of applied thought; while the route of the middle of the approach to a heaven divides into three grades of its own — “either one attains parinirvāṇa as soon as one has taken up the commonality of sentient beings of the form realm, and this is the first; after that, one attains parinirvāṇa only when one has enjoyed the pleasures of the gods, and this is the second; and after that again, one attains parinirvāṇa only on entering the Dharma assembly of the gods, and this is the third.” And under this very outline Puguang’s Record preserves further explanations of his own: “a further explanation: the three kinds spoken of for the middle of the approach to a heaven are again divided as three by the stages of element, perception, and applied thought,” and again “these three kinds extend to the middle of the lifespan as well,” and “these three kinds likewise explain in order the foregoing three of element, perception, and applied thought” — that is, Puguang himself allows the three grades of the route of the heaven-approach to be read back as the three stages of element, perception, and applied thought, and does not read the two routes and three grades as fixed. First ask: why are there two routes and three grades? Because the two words “the middle” admit of two senses — the middle of the lifespan, and the middle of the heaven approached; and each sense has its own gradation, and so each divides into three. The easy error is to read the three grades of the latter route as another name for the former. What the former route divides is the stages of cutting off affliction — Puguang explains these three stages as “first by reaching the stage of the element, the element being the seed of affliction; second by reaching the stage of perception, perception being defiled perception; third by reaching the stage of applied thought, applied thought being seeking” — while what the latter route divides is what is passed through after taking birth (taking up the commonality, enjoying the pleasures of the gods, entering the Dharma assembly). Their criteria differ. In practice: the treatise’s refutation is not delivered across the board — what the simile of the sparks compares is the nearness and farness of place and of time, and the treatise judges these two routes to be “in no way answering to the simile of the sparks”; while the counter-objection that the formless realm too has a middle of the lifespan bears only upon the route of the lifespan. The two commentaries have further outline wording and a further adjudication of the range of the refutation, for which see Further Resonances. The author refutes them with the three similes of the seven destinations of good men in the sūtra: the threefold division of attaining parinirvāṇa in the interval is “by the nearness, middle distance, and farness of place and of time,” as a small spark of a splinter fire is extinguished near at hand as soon as it rises, a small spark of an iron fire is extinguished only at mid-course, and a great spark of an iron fire flies far and is extinguished before it falls; whereas on the latter account, “since there is no difference in the course traversed there,” the three grades become a difference of time only and do not answer to the simile of the sparks. The author adds a further objection: in the formless realm too there are those who attain parinirvāṇa in the middle of the lifespan, and on the latter account the formless realm should also have attaining parinirvāṇa in the interval; yet the sūtra does not say so, the evidence being the stanza of the Udāna verses, “Gathering in sum the noble and the worthy: ten in each of the four meditative absorptions, seven in each of the three formless spheres, and six alone in the sphere of neither perception nor non-perception.” What does this stanza do here? It is a table of the numbers of the noble ones as distributed — ten in each of the four meditative absorptions and seven in each of the three formless spheres, and what is missing is precisely the three kinds of attaining parinirvāṇa; and since the formless does not list attaining parinirvāṇa in the interval, it is plain that this belongs to the form realm alone and that its name must hang upon intermediate existence. One must not read it as a table of numbers with nothing to do with intermediate existence — it uses a gap in the numbers as proof, and is the most winding of the scriptural proofs in this fascicle. Again, it is warranted by the preceding words “as is stated in the Udāna verses,” and is therefore a quoted stanza and not a root stanza; in reading, do not confuse it with the root stanzas.

〔9.8〕The lament over the scattering of the sacred teaching, and the reconciling of two sūtras. Here the author utters a lament rare in the whole treatise: if even these sūtras are not recited, then “the unsurpassed King of the Dharma passed into extinction long ago and the great generals of the Dharma have attained parinirvāṇa; the sacred teaching is scattered and has become many schools, and over its letter and its sense divergent graspings gallop against one another, and taking up and setting aside as one pleases grows stronger to this day. Alas that you cling to folly and confusion, going against principle and refusing the teaching; it is greatly to be lamented” — and he concludes, “for all who take the foregoing principle and teaching as their measure, the real existence of intermediate existence is thoroughly established.” He then turns back to reconcile two passages of scripture that seem to prove the contrary. First, the sūtra says that Dūṣin the Māra, who performed extremely evil action, “fell headlong in his present body into the Unremitting Hell” — which the author explains as: his life was not yet given up when the fierce flames of the hell were already burning his body, and hence, when his life ended, he took up that intermediate existence and fell riding upon it; the force of his evil action being strong, the marks of suffering arrived without waiting for the end of life, so that he first underwent the result to be experienced in the present life and afterwards the result to be experienced in the next life. Second, the sūtra says that the five actions without interval, once performed and increased, are “without interval definitely followed by birth in the naraka” — which the author explains as: the sūtra’s intent is to block his going to another destiny and to display that the action is definitely to be experienced in the next life; and were one to hold to the surface of the text, all five actions would have to be complete before one were born in hell, and lacking any one of them, or by some other action, a great fault would follow; and again, if “without interval” be taken temporally, one ought to be born the instant the action is done without waiting for the body to break up. A further route is given: “the name naraka extends to intermediate existence as well,” and when intermediate existence arises without interval upon the existence of death that too may be named birth, since it is the means to birth. First ask: is the method of reconciling the two passages the same? It is not. What the first passage explains is “fell headlong in his present body” — his life not yet given up while the fierce flames were already burning his body, and only when his life ended did he take up intermediate existence and fall riding upon it; whereas it is the second passage that explains the two words “without interval,” shifting them from a temporal reading to a sequential one, and giving besides the further route that “the name naraka extends to intermediate existence.” The easy error is to take reconciling for forced construal. Each time the author gives a second reading that the text of the sūtra itself can bear, and he does not alter a graph. In practice: this is the treatise’s standard procedure for handling a scripture that seems to conflict with its position — do not alter the scripture, do not abandon the position, alter the reading.

〔9.9〕The last quoted stanza, and the close of the fascicle. At the end the objector raises a stanza of scripture: “Twice-born, you have now passed the stage of your prime; decline draws near and King Yama is not far; you would go forward on the road ahead and have no provisions, and you would seek to stay in between and there is no place to halt.” He asks: if there is intermediate existence, why did the World-Honored One say “in between there is no place for him to stay”? The author answers by two routes: “the sense of this stanza is to display that among human beings he swiftly comes to wearing away with no moment of halt”; or else that that intermediate existence, in order to reach the place where it is to be born, likewise makes no halt, since its going is unobstructed. The objector asks in return: how do you know that the sense of the sūtra is this and not something else? The author questions him back: “and how do you know that it is something else and not this? Since the two reproaches are equal, why single out one side for questioning?” — the reproaches on the two sides being equal, why question my side alone. The last sentence is the close of the fascicle: “neither of the two explanations does violence to the sūtra; how then may it be singled out to prove that intermediate existence does not exist? Whenever words are adduced as proof, their principle may have no other tendency; and since this has another tendency, it does not hold as proof.” First ask: what rule does this sentence establish? It establishes a threshold for adducing proof: whatever words are brought as evidence may have no second tendency in their principle; and since this stanza admits of a second reading, it cannot hold as evidence. The easy error is to read it as the author playing the sophist. It is in fact a rule usable in both directions, and the author is bound by it himself when he adduces scripture. In practice: this fascicle sets out from the first question, “what are the three realms,” and runs through the three realms, the five destinies, the seven stations of consciousness, the nine abodes of sentient beings, the four stations of consciousness, and the four modes of birth, coming to rest at last upon intermediate existence; and of the strength spent in the whole fascicle more than half goes to this last passage. The reason is not hard to see: the first six items take stock of squares already there, whereas intermediate existence must be established anew, by main force, between death and birth — if it is established, all the continuity that went before can be joined up; if it is not, a break appears between one destiny and another. And that this fascicle lays down its brush here and leaves dependent origination to the next has its sequence in just this.

Further Resonances: A Mirror of Three Commentaries

The above is the sense of this fascicle’s own text. What follows is a separate section gathering four matters: the adjudications of the three commentaries and their divergences, Saṅghabhadra’s counter-refutation, the divergences of Paramārtha’s older translation, and coordinates for collateral reading. A mirror is not the treatise, only another way of reading the same text; the items are recorded together without adjudication, each returned to its own place.

One thing must be recorded first: Shentai’s Commentary on the Abhidharmakośa is a fragmentary text, and its seven surviving fascicles explain parts one and two of the Chapter on the Elements, parts two through five of the Chapter on the Faculties, and part three of the Chapter on the Noble Persons, the Chapter on the World being wholly wanting. Hence for this fascicle only the three commentaries of Puguang, Fabao, and Yuanhui can be attached, and this section is “a mirror of three commentaries” and not of four — not by omission, but by marking the lacuna as it stands.

1. Where the Three Commentaries Diverge

The two divergent explanations of attaining parinirvāṇa in the interval: the two commentaries differ in outline and in the range of the refutation. The treatise records only in sum, “other masters say further: either in the middle of the lifespan or in the middle of the approach to a heaven,” without setting up the masters separately. Puguang’s Record outlines it as “two divergent explanations” within one passage, and under the route of the lifespan explains the names of the three stages — the element being the seed of affliction, perception being defiled perception, applied thought being seeking; Fabao’s Commentary, on the other hand, marks the route of the approach to a heaven as “the divergent explanation of a third master,” and expressly divides the range of the refutation: “this refutes only the middle of the lifespan and does not refute the middle of the approach to a heaven.” Beneath the two routes of his own outline Puguang preserves three further explanations — “a further explanation: the three kinds spoken of for the middle of the approach to a heaven are again divided as three by the stages of element, perception, and applied thought,” “a further explanation: these three kinds extend to the middle of the lifespan as well,” “a further explanation: these three kinds likewise explain in order the foregoing three of element, perception, and applied thought” — from which it appears that “two routes and three grades” is not a settled position even for Puguang, who leaves standing at the same time the readings that gather the three grades of the heaven-approach back into the three stages of element, perception, and applied thought, or extend them to the route of the lifespan. ⇒ The treatise does not distinguish the masters, Puguang makes two explanations while preserving three further ones, and Fabao sets up a third master; and the simile of the sparks refutes both routes while the objection from the formless realm refutes the route of the lifespan alone. The items are recorded together without adjudication.

The three accounts of the entity of a destiny: the two commentaries both adjudicate while the treatise does not. At the end the treatise sets three accounts side by side — the Vaibhāṣikas’ “a destiny is unobscured-indeterminate alone,” “there is another account: it is invariably born of maturation,” and “other masters say: it extends to the nourished as well” — while the text of the treatise itself passes no verdict. Puguang’s Record marks the first account as “concluding by a return to his own school,” and adjudicates the latter two with “of these two masters the former account is correct,” taking as his warrant the question and answer of the Mahāvibhāṣā, fascicle one hundred seventy-two: “Answer: it should be said that the entity of a destiny is maturation alone.” Fabao’s Commentary adjudicates the same but counts differently — he marks it as “setting out a divergent account,” and says, “of these three accounts the second is correct,” adducing separately the ground given in the Correct Principle, “because it is born of maturation alone, and because there is then no fault” — and Fabao’s quotation stops here; what follows, “if it extended to the nourished, then a body in the desire realm that has cultivated and obtained the divine eye is of the nature of the nourished, and the destinies would be mixed and disordered,” is Fabao’s own further argument after the close of his quotation and is not within the text he adduces from the Correct Principle. ⇒ The two commentaries take one and the same account, “invariably born of maturation,” while the text of the treatise does not close in that way; and in reading the main text of this fascicle, “the entity of a destiny is unobscured-indeterminate alone” is what the treatise adjudicates, while “maturation alone” is what the commentaries take.

The round of exchange on the continuity of mind and the rest in the formless realm: Puguang adjudicates it as the author’s expedient. The treatise circles three times here and then records the Sautrāntika account, its own text not stating its intent. Puguang’s Record adjudicates, “objecting in this way is an expedient, meant to display that commonality and the life faculty have no real entity,” and where the treatise records the Sautrāntika account he adjudicates outright, “he adduces an account that agrees with his own.” ⇒ Whether the author takes the recorded Sautrāntika account for his own view the text of the treatise does not say, while Puguang adjudicates it as agreeing with his own; and the choice between them does not belong to this section.

Who holds the seventeen of the form realm: the two commentaries mark the name differently. The stanza of the treatise establishes seventeen while the prose commentary records alongside the sixteen of Kaśmīra. Puguang’s Record sets out four accounts and attaches each to its holder: “in the stanza the author states the sense of the masters of the West, saying that there are seventeen; hence the Mahāvibhāṣā says, ‘the masters of the West say the form realm has seventeen. The intent of those masters is that, since the great Brahmā king’s lifespan, measure, place, and the rest differ from the other two heavens, three heavens are separately established in the first absorption,’” and he further adduces the Correct Principle as having others who separately state seventeen, and the Elder as establishing eighteen for the form realm, concluding, “the masters of this land now establish neither, and since there is no separate level there are only sixteen.” Yuanhui’s Verse Commentary, on the other hand, marks them straight off by school name: “establishing seventeen heavens in the form realm follows the Sautrāntika school; if it be the Sarvāstivāda school, only sixteen heavens are established, the Mahābrahmā Heaven being excluded,” and again, “if it be the Sthavira school, the Mahābrahmā Heaven is also established, since body and lifespan differ; and the Heaven of Non-Percipient Beings is also established, since it differs from the Bṛhatphala in cause and result. Hence the form realm has eighteen heavens.” ⇒ One and the same “seventeen” is assigned by Puguang’s Record to the masters of the West (adducing the words of the Mahāvibhāṣā) and by Yuanhui’s Verse Commentary to the Sautrāntika school. The two commentaries differ in their school labels, and whether these are two names for one school is a doctrinal judgment; the items are recorded here together without adjudication.

Where the Brahmā multitudes had once seen the Brahmā king: the widest disparity of fullness among the three commentaries. Puguang’s Record sets out the several explanations here in full, marking the treatise’s own correct explanation as “having seen him at their first birth, when they see him again later they arouse this thought: we once saw such a sentient being long-lived and long abiding,” and adducing separately one explanation of the Correct Principle: “some say: those Brahmakāyika gods do not undergo the womb or the pain of the severing of the vital points, and thereby have mindfulness without forgetting, and so recall what they saw and the rest in their former birth”; and at the end he makes a summing case of his own, “the foregoing Correct Principle has six explanations in all, the first three the same as here and the latter three different. And the Mahāvibhāṣā, fascicle ninety-eight, has five explanations, the first three the same as in this treatise.” Fabao’s Commentary adduces the explanation of the Correct Principle without making any comparative case, giving “the Treatise Conforming to Correct Principle says: when the Brahmā multitudes first came down to be born they saw the great Brahmā king’s majestic light blazing, and though they cherished reverence and longing and wished to go and attend upon him, they were pressed by his majestic power and did not accomplish the approach, and so with the passing of time much time went by.” Yuanhui’s Verse Commentary gives only one explanation to the same question: “Question: ‘why do the Brahmā king and the Brahmā multitudes arouse the perception of one cause?’ Answer: at the beginning of an aeon the great Brahmā king dwelt alone with no attendants, and so made a vow, saying, ‘how may other sentient beings be born into my commonality?’ At that time the Ābhāsvara gods saw and pitied him, and passing away from that place were born as the Brahmā multitudes. The great Brahmā king, having just made his vow and seeing gods born, aroused this perception: I manifest the power to produce. And those Brahmā multitudes, first seeing the great Brahmā’s majesty and preeminence among the gods, and recalling as well that they had come to be born here through the Brahmā king’s vow, therefore aroused this thought: we were all produced by the great Brahmā king.” The three give six, several, and one explanation to the same question, and the items are recorded together without adjudication.

Where the intermediate existence of one from the formless realm appears: Puguang marks a divergence between the author and the Mahāvibhāṣā’s adjudicators. Puguang adduces the Mahāvibhāṣā, fascicle sixty-eight, with two explanations, one saying that it is in the fourth meditative absorption, which the adjudicators do not take, and one saying, “others say: if one passes away from there and is born in the formless realm, the intermediate existence becomes present in that very quarter and place”; and at the end he offers a remark of his own: “the author takes principle for his tenet and does not take the Mahāvibhāṣā’s adjudicators for his measure; what need is there to labor at reconciling?” ⇒ This is Puguang’s adjudication of the author’s standpoint in choosing, and is not in the text of the treatise, and so it stands in this section.

That one born by transformation leaves no form behind at death: Puguang and Fabao differ in attitude toward the adjudicators’ remark. Puguang adduces the Correct Principle and the Mahāvibhāṣā on this matter, and adduces the remark of the Mahāvibhāṣā’s adjudicators: “the adjudication says: as stated above is the good account. Since the Mahāvibhāṣā’s adjudicators take the former master’s view that one set of four great elements produces only one derived form, it is known that the account that those with many derived forms leave no form behind at death is not the correct sense.” Fabao adds under the same material a remark of his own: “the adjudication says: as stated above is the good account. Now, examining it, the Correct Principle does not take the Mahāvibhāṣā’s adjudicators as correct. If one makes the refutation as a master of the Kośa, then on your reckoning this explanation goes against the correct sense of the Mahāvibhāṣā.” ⇒ Puguang, on the ground of the adjudicators’ remark, judges one account not to be the correct sense, while Fabao turns instead to point out that the Correct Principle and the Mahāvibhāṣā’s adjudicators are at odds, hinting that the Kośa master’s explanation may go against the correct sense of the Mahāvibhāṣā. The two commentaries differ in tone, and the items are recorded together without adjudication.

Whether the entity of a destiny is maturation alone or extends to the nourished: Puguang adduces the Mahāvibhāṣā to settle it. At the end the treatise sets three accounts side by side (unobscured-indeterminate alone, invariably born of maturation, extending to the nourished as well) without adjudicating. Puguang says, “of these two masters the former account is correct. Hence the Mahāvibhāṣā, fascicle one hundred seventy-two, says: Question: it being known that the entity of a destiny is unobscured-indeterminate alone, is it within that maturation alone, or does it extend to the nourished? If it is maturation alone, how is the statement of the Treatise on Categories of Dharmas to be reconciled, which says that the five destinies gather in the five aggregates, the twelve sense fields, and the eighteen elements? And if it extends to the nourished, the entity of a destiny is mixed and disordered, since within the destiny of human beings the nourished faculties and great elements of the form realm are also drawn forth. Answer: it should be said that the entity of a destiny is maturation alone. Question: how then is the statement of the Treatise on Categories of Dharmas to be reconciled? Answer: that text should say that the five destinies gather in the five aggregates, eleven sense fields, and a portion of seventeen elements; and since it does not say so, it should be known that the reciters were in error. There is an account: that treatise extends to the retinue of the five destinies, the action that calls forth the five destinies and what can protect it, and does not state the destiny alone, and so there is no fault. Yet since realms differ by affliction and destinies differ by the aggregates of maturation, therefore the entity of a destiny is maturation alone,” and he adduces its answer, “Answer: it should be said that the entity of a destiny is maturation alone.” ⇒ This is Puguang’s adjudication, while the text of the treatise leaves both standing.

To which destiny the asuras belong: Fabao alone treats it fully. The item of the asuras is not among the five destinies of the treatise. Fabao supplies this question and attaches them to the destiny of ghosts: “the asuras are gathered into the destiny of ghosts. Śakra the lord of gods saying that they were originally gods is a flattering way of speaking,” and again, “it has been heard that King Māndhātṛ’s majesty surpassed that of the gods. Nanda, Upananda, and the rest, though animals, were sovereign in majesty beyond the host of gods. Hence the asuras are gathered under the destiny of ghosts alone.” ⇒ This is not in the text of the treatise, but a question Fabao supplies concerning the number of the five destinies.

“The sacred teaching is scattered and has become many schools”: the widest disparity of length among the three commentaries. Puguang alone sets out in full the items of the schisms of the schools, beginning at “more than a hundred years after the Buddha’s parinirvāṇa, in the time of King Aśoka, they first divided into two schools, the Mahāsāṃghikas and the Sthaviras,” and listing the names of the nine schools and the eleven schools over some thirty lines. Fabao passes it off in a single sentence: “as for the sacred teaching being scattered and become many schools, from the first hundred years after the Buddha’s parinirvāṇa to the last of the four hundred, root and branch divided into twenty schools, as is set out at length in the Treatise on the Wheel of Tenets.” Yuanhui’s Verse Commentary has no corresponding text whatever for this passage. Material of this length being too great, this section records only where it stands and does not carry the whole passage in.

The name of Dūṣin the Māra’s attendant: a difference of one graph between the two commentaries. Puguang adduces the Mahāvibhāṣā as “named Zhiyuan,” writing the first graph as the one for resolve; Fabao adduces the same passage as “named Zhiyuan,” writing the first graph as the one for reaching. ⇒ A difference of one graph, perhaps of transmission; the items are recorded together without adjudication.

The sentence at the close of the fascicle, “since the two reproaches are equal, why single out one side for questioning”: the two commentaries attach it differently. In that last round of counter-questioning at the end of the fascicle the two commentaries adjudicate the first two sentences entirely alike — “how do you know that the sense of the sūtra is this and not something else” belonging to the party that holds there is no intermediate existence (whom Fabao’s Commentary names the Mahāsāṃghikas), and “and how do you know that it is something else and not this” to the author; only the third sentence diverges. Puguang’s Record attaches it to the party that holds there is no intermediate existence, giving “as to ‘since the two reproaches are equal, why single out one side for questioning,’ the party holding there is no intermediate existence says: since the two reproaches are equal, why single out intermediate existence for questioning as nonexistent? As to ‘neither of the two explanations . . . does not hold as proof,’ the author says”; Fabao’s Commentary, on the other hand, attaches it to the author, giving “on the treatise’s ‘since the two reproaches are equal . . . neither does violence,’ the author shows that the sūtra does not hold as proof, first by the equality of the two reproaches and second by the sameness of the two explanations.” ⇒ One and the same sentence, attached differently by the two commentaries; the vernacular layer of this fascicle follows Fabao’s outline and takes it as the author’s words, and Puguang’s divergent outline is recorded here alongside, without adjudicating which is right.

How the name “twice-born” is obtained: the three commentaries differ in fullness. Puguang explains it as a twofold sense of birth, with an analogy: “a brahmin is called twice-born; the first birth is one birth, and receiving the brahmin dharma afterwards is another birth, and so, having received the dharma, he is called twice-born. So too a monk is called twice-born, that is, by his first birth and by receiving the precepts. So too one born from an egg is called twice-born, that is, by the womb-birth and by the egg-birth.” Fabao is briefer: “a brahmin is called twice-born, since the time of his first birth is called one birth and the time of receiving the dharma is called the second birth.” Yuanhui’s Verse Commentary takes it only as a proper name without glossing its sense, giving “there was a non-Buddhist brahmin named Twice-Born who came in his old age to see the World-Honored One. The World-Honored One reproached him: you are now old and have passed the stage of your prime.” Under the phrase “passed the stage of your prime” the three all give the same number of the five stages outside the womb (infancy, childhood, youth, prime, and old age), and on this point the three commentaries agree.

The phonetic and semantic gloss on gandharva: Yuanhui alone has it. Yuanhui’s Verse Commentary gives, on the sense of intermediate existence feeding on scent, “gandharva means here scent-seeking; intermediate existence feeds on scent, and so is named scent-seeking.” Puguang has no such gloss within the range of this fascicle. Its being his alone is recorded, without adjudication.

2. Saṅghabhadra’s Counter-Refutation

In the passages of Saṅghabhadra’s Treatise Conforming to Correct Principle corresponding to this fascicle he calls Vasubandhu “the Sūtra-master.” His chapter is named the chapter analyzing dependent origination and not the chapter on the world, and its opening sentence carries over from the same transition as the head of the Kośa’s chapter. In the corresponding passages several places expressly name the Sūtra-master in refuting him, and their locations and the points at issue are recorded.

First, upon what the mind of the formless realm continues — the harshest in wording. Saṅghabhadra positively establishes the position of his own school: “by this very sense the Abhidharma masters say that in the formless, there being no body, commonality and the life faculty and the rest depend upon one another”; and of Vasubandhu he says, “the Sūtra-master here plays host and guest by turns and, falsely adding to the correct sense, makes this objection: if so, why do the mind and the rest of sentient beings possessed of form not simply depend upon these two and continue?”, and again, “the Sūtra-master certainly has no inheritance in Abhidharma and states this falsely; or else, out of his own mind’s aversion to Abhidharma, he pretends to say this and confuses the correct tenet. What man of fine understanding of the characters of dharmas would answer the preceding question in such a way?” In the same passage Saṅghabhadra also refutes the Elder’s line and its account that “they depend upon one another as two sheaves of reeds abide leaning together.” ⇒ Three parties are set out here together; one meaning and two schools recorded together, with neither judged to prevail.

Second, the number of the places of the form realm. Saṅghabhadra refutes the Elder’s account of eighteen and concludes, “therefore the great masters of the treatises of the land of Kaśmīra all say that where the great Brahmā king dwells is the Brahmapurohita place, and by this the places of the form realm are sixteen alone.” ⇒ What this passage refutes is the Elder, and it does not use the words “the Sūtra-master”; yet the sixteen he concludes with differs from the seventeen established on the face of the Kośa’s stanza. The two parties only are recorded, without adjudication.

Third, whether the entity of the five destinies is unobscured-indeterminate alone or extends also to the wholesome and the defiled. Saṅghabhadra adduces the Sūtra of the Seven Existences to establish the nature of the entity of a destiny, saying, “this sūtra is in order to display that the entity of a destiny gathers in the unobscured-indeterminate alone, and so distinguishes the cause as different”; and of Vasubandhu he cites the words and refutes them: “yet the Sūtra-master says that, since it is not stated separately it is definitely not gathered in by it, as among the five turbidities affliction and view are stated separately as turbidities; and since they are stated separately, that view is definitely not gathered in by affliction,” and continues in judgment, “hence there are those who say that the entity of a destiny extends to the wholesome and the defiled. Their statement is contrary to principle, since it is stated nowhere,” closing with, “how is it definitely known that action is the entity of a destiny? Because it has a cause, it is stated separately within the existences; and it is not that the existence of action, whose entity is not a destiny, is stated as the existence of action in order to display the cause of a destiny — and so the analogy adduced has no power to prove.” ⇒ What is at issue is the nature of the entity of a destiny; the Sūtra-master takes the separate statement of the five turbidities for his analogy, and Saṅghabhadra refutes his analogy with “it is stated nowhere.” One meaning and two schools recorded together, with neither judged to prevail.

Fourth, that intermediate existence is not a destiny. Saṅghabhadra twice cites the Kośa’s words verbatim and adjudicates them: “the Sūtra-master says further: that being so, since intermediate existence is named intermediate existence it should not be named a destiny. As to ‘because it is between two destinies it is named intermediate existence,’ this is contrary to principle, since the reason is not established”; and again, “and as to the Sūtra-master’s saying that if this were gathered into a destiny it would not be between, and then it should not be named intermediate existence, this too is contrary to principle.” The Elder gives a separate explanation, that one ought to grant an action outside the five destinies able to call forth the arising of intermediate existence, upon which Saṅghabhadra remarks, “yet that Elder’s cognition and discernment are in decline, and where there is no fault he falsely raises the charge of fault.”

Fifth, the ground of the Ābhāsvara gods’ difference in perception. Saṅghabhadra adduces the Sautrāntika explanation recorded in the Kośa and remarks, “the Sūtra-master adduces a sūtra to explain the sense of difference in perception, saying that among the Ābhāsvara the newly and the long born, with the aeon-fire as their object, have the two perceptions of terror and of no terror mixed, and hence are called different in perception, and not that the two perceptions of joy and of equanimity are mixed,” and objects that on that account there would be no separate principle by which the third meditative absorption is called one in perception through the perception of pleasure; while of the Elder he says outright, “now, examining his statement, it does not answer to the stations of consciousness.” The wording differs in weight toward the two.

Sixth, that in which consciousness adheres. Saṅghabhadra cites Vasubandhu’s explanation and adduces his passage “these four stations of consciousness are like a good field,” remarking at the end, “an explanation of this kind merely states his own feeling; sought out with careful examination, it has no deep principle”; and of the Elder’s divergent explanation he remarks, “his explaining it so is not a word of awakening to principle.”

Seventh, why the bodhisattva undergoes birth from a womb. Saṅghabhadra adduces the Kośa’s objection about the supernormal power of upholding a vow, first answering and then concluding: “now I say that this explanation’s principle must hold. A body preserved by supernormal power, not being perfumed as the support of the Buddha’s powers of merit and of the fearlessnesses and the rest, cannot benefit the world broadly”; and at the end he returns to his own school: “therefore the Vaibhāṣika masters from whom I have received my inheritance all say: the bodhisattva of the last body does not undergo birth by transformation, for the sake of benefiting others. This sense is thoroughly established and cannot be overturned.” The tone of this item is more level than that of the items above.

Eighth, whether an image really exists — the most concrete divergence. Saṅghabhadra adduces the Kośa’s words refuting the image and objects, “now I say that his reason cannot dispel the image, and so cannot resolve the objection refuting intermediate existence”; the condensed version of the same argument in the Treatise Revealing the Doctrine runs, “yet the Sūtra-master and a class of masters like him, who grant that since an image is not established it is no analogy, speak contrary to principle. An image is not unestablished, since with respect to a distinct thing there arises just such an image, as this image’s own substance is what it depends upon,” and it records one who, agreeing with the Sūtra-master, proves by many reasons that an image is not existent — “the Venerable Xihui too, by many of the same reasons, proves that an image is not existent, in agreement with the Sūtra-master.” ⇒ The two parties are directly opposed on whether an image really exists, while on the conclusion that intermediate existence really exists they in fact agree; what differs is the route of the argument. One meaning and two schools recorded together, with neither judged to prevail.

Ninth, after the simile of the image there are still eight sets of divergent accounts — the actual end of this fascicle’s section boundary. The argument about the image is not the end of the debate for this unit. After the simile of the image and before the transition to a new topic, Saṅghabhadra records and refutes, one after another, eight sets of divergent accounts: the transmission of sound onward, the example of passing away from the formless, the simile of the inchworm, going by the force of intention, the account that “gandharva” is a mistaken recitation, the gloss on “about to be born,” the strained explanation of a sound formed by mind, and a divergent explanation of the three grades of attaining parinirvāṇa in the interval. At the sentence “the nonexistence of intermediate existence having been refuted at length, one should now consider: to which destiny will it go?” the two treatises change topic together, and this is the true boundary between this unit and Unit 2. ⇒ All eight sets are divergent accounts recorded by Saṅghabhadra himself and are not statements of the Kośa; the words of each are given in the ninth item of section V of this unit’s apparatus (the general table of page locators standing separately in section IX). The items are recorded together without adjudication.

3. Divergences of Paramārtha’s Older Translation

In the passages corresponding to this fascicle, Paramārtha’s Expository Treatise on the Treasury of Abhidharma forms a lineage of renderings of its own, and its format differs as well. The most conspicuous difference of format is the dividing and joining of the stanzas: 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.” The stanza on intermediate existence in this fascicle Xuanzang writes as four lines in a row, while Paramārtha breaks it into two verses, “the existence of death and the existence of birth, in between the five aggregates” and “not having reached what should be reached, what is not yet born is named intermediate existence,” with prose commentary between them; and the passage of the simile of grain Xuanzang writes as eight lines in one breath, while Paramārtha breaks it into several short verses each with its embedded explanation.

The divergences of rendering fall into three classes. The first is difference of designation: the Heaven of the Multitude of the Four Great Kings is rendered the Heaven of the Four Heavenly Kings, the Yāma Heaven the Heaven of Proclaimed Joy, the Tuṣita Heaven the Heaven of Good Contentment, and the Heaven of Delight in Working Transformations the Heaven of Transformed Delights; the Hell of Equal Reviving is rendered Renewed Life, Massed Crushing Massed Striking, Howling Crying Out, and Burning Heat Blazing; Jambudvīpa in the south is rendered the Yanfu Continent and Uttarakuru in the north the Northern Superior Birth; the Brahmapurohita Heaven is rendered the Heaven of Brahmā’s Forerunners, the Avṛha Heaven the Heaven of No Great Seeking, and the Akaniṣṭha Heaven the Lowest of None; and Kaśmīra is rendered Jibin. The second is systematic difference of rendering graphs: destiny is rendered way, sentient being living being, intermediate existence the intermediate skandha, image reflection, gandharva given in another transcription, non-returner anāgāmin, and attaining parinirvāṇa extinction. The third is difference of dividing and joining and of syntax: the order of the two similes is exchanged (Xuanzang has the pepper drink and the vajra ring, Paramārtha the vajra earring and the maricha drink); in the four stations of consciousness Paramārtha embeds the word “craving” in each line, giving “the station of consciousness that craves form” and the rest, and so displays more plainly than Xuanzang the sense that it abides by reason of craving; and in the four modes of birth Paramārtha adds the word “mixed” and sets himself the question “what is the sense of mixed?” to explain it, which Xuanzang’s version does not have.

There are two places where the divergence goes beyond the level of rendering, and they deserve particular note. First, the identity of the objector: Xuanzang has a non-Buddhist of wrong livelihood, Paramārtha the Nirgrantha’s son — the two do not point to the same non-Buddhist school. Second, the name of the sūtra adduced: Xuanzang has the Sūtra of the Horse-Keeping Clan, Paramārtha the Sūtra of Aśvalāyana, and the order in which the four castes are listed differs as well. Both places are recorded without adjudication.

A negative proposition is recorded besides: Xuanzang closes the refutation of “no attaining parinirvāṇa in the interval in the formless realm” with a stanza of the Udāna, while at the same place of refutation Paramārtha’s version states it outright in prose — “if so, in the formless realm too there should be said to be those who attain extinction in the interval, since there also they attain parinirvāṇa in the middle of the lifespan; and since it is not said, this grasping is mere private discrimination” — and passes directly on without quoting the stanza. This is a real difference of fullness between the two versions and not a failure of search, and is recorded as it stands. The relative merit of the two translations is not adjudicated.

4. Coordinates for Collateral Reading

What this fascicle establishes is the first register of the setting up of sentient beings, and also the first place where the name of intermediate existence is proved out in full. The three places below may be read against later treatises, and the words of those systems are given in section VII of this unit’s apparatus.

First, the criterion of the five destinies. This fascicle establishes the entity of a destiny as “unobscured-indeterminate alone,” and the whole ground for it lies in the one sentence “were it otherwise the destinies would be mixed together” — that is, mixture must be kept out. Later ages of the Yogācāra family prove the eighth consciousness with the same set of enumerated dharmas, and the criterion they give also runs, “only what is really existent and constant and pervasive and unmixed may be established as a true and real destiny and birth,” and, “dharmas that are not maturation would make destiny and birth mixed and disordered, since while dwelling in this destiny one arouses dharmas of the destiny and birth of another” — for whose words and page locators see section VII of this unit’s apparatus. ⇒ The only explicit text the two places have in common is the single item “unmixed”; the three items “really existent,” “constant,” and “pervasive” are added by that system and are not established in this fascicle. The entity to which each attaches differs as well: this fascicle attaches it to the unobscured-indeterminate, that system to the eighth consciousness. This is the most direct place of collateral reading between the two systems.

Second, the time intermediate existence abides. This fascicle establishes intermediate existence without touching the measure of the time it abides (the argument over the time it abides being in the next fascicle of this chapter). The Yogācārabhūmi on the same matter gives the measure of that time, saying that if intermediate existence has not obtained the condition of birth it abides at most seven days — one and the same dharma, and two systems differing in fullness, which may serve as a coordinate. For its words and page locators see section VII of this unit’s apparatus.

Third, the three conditions for entering the womb. This fascicle adduces the sūtra as three things present together, that the mother’s body be in fit condition, that father and mother join in mutual love, and that a gandharva be just present; the Yogācārabhūmi has three places present, its three being in order that the mother be in fit condition and the time also right, that father and mother join and arouse craving together, and that a gandharva be just present. The number of the three conditions and the name of the third are the same and the syntax slightly different — one and the same translation workshop and yet a difference of wording, which is only recorded, without adjudication. For its words and page locators see section VII of this unit’s apparatus.