中論 義疏通講

Canonical Grounding First: A Name Is a Gate, Not Something to Grasp

Seng Rui begins by saying that the Treatise on the Middle is named “Middle” to illuminate the true mark and called a “treatise” to explain the meaning of its words. The true mark is not the word “Middle” itself, yet words must be borrowed to lead people toward awakening. Discourse is not the true mark itself, yet exposition must be borrowed to make the path clear. This opening establishes a boundary for the whole treatise: names and words can guide, but they cannot in turn become substantial things to which one clings.

If the Extremes Are Existence and Nonexistence, How Can Practice Itself Also Become One-Sided?

The concern expressed in the preface is not limited to worldly wrong views. Inverted views immerse people in the three realms, but a narrow wisdom of aversion may also leave a practitioner aloof and isolated. If one’s illumination is not vast enough, existence and nonexistence, the Way and the conventional, cannot be comprehended on equal terms. If one’s understanding is not thorough enough, one cannot travel the middle course and dissolve the two limits. The Middle Way is therefore not a compromise chosen between existence and nonexistence. It is the seeing that both extremes stand only in mutual dependence, so that the Way and the conventional no longer bar one another.

If Reasoned Exposition Refutes Everything, How Can It Still Become a Path?

Seng Rui praises Nāgārjuna for “analyzing them through the Middle Way.” This makes clear that critical analysis is not meant to leave a wasteland behind. The treatise dismantles congealed views layer by layer, allowing those lost in fixed directions to glimpse its purport and turn around. It then draws understanding back into the teaching of the present moment, so that one need not seek an answer somewhere beyond concepts. What refutation removes is constricted clinging. What appears when it has been removed is a broad, level, practicable road.

Whose Voice, Exactly, Do We Hear in the Treatise on the Middle Before Us?

The preface draws a crucial textual boundary: Nāgārjuna composed the verses, Qingmu wrote the commentary, and Kumārajīva’s translation workshop edited places in Qingmu’s commentary that were defective or cumbersome. A reader of this work therefore cannot treat the verses alone as the entire Treatise on the Middle, nor quietly present Qingmu’s commentary as Nāgārjuna’s self-commentary. Verse, commentary, and translation workshop all inhabit the same transmitted text; each has its own identity, and together they make the book.

Because Qingmu commented and Kumārajīva edited what was defective or cumbersome, a reader hears more than one voice: the verses state the meaning established by Nāgārjuna, the commentary opens it through Qingmu’s explanation, and the translation workshop has made editorial interventions between them. Chinese transmission has preserved all these strata, but they cannot be collapsed into a single conclusion with no discernible author. The more clearly they are distinguished, the less likely a later explanation is to eclipse the wording of the root verses.

Carried Forward: Jizang’s Mirror

Jizang’s commentary on the preface carries forward the transmission account of “five hundred verses” and opens the meaning of the preface line by line. This mirror reminds us that ancient masters already treated the preface as part of the work as transmitted, not as a dispensable appendix. Yet a mirror serves to reveal the transmission and purport already present in the preface; it does not impose a different history upon it.

Seng Rui closes by placing the Hundred Treatise, the Treatise on the Middle, the Great Treatise on the Perfection of Wisdom, and the Twelve Gate Treatise side by side: one treats external error, one clears internal obstructions, one is vast, and one is profound and exact. This does not flatten four treatises into one. It offers a common path of reading. Entering the Three Treatises through this preface, the first discipline to learn is precisely this measure: let the treatises illuminate one another without causing any one of them to lose its own voice.

Open the Treatise on the Middle. The first thing you see is not a proposition, but a person bowing to the Buddha.

“Neither arising nor ceasing; neither permanence nor annihilation; neither identity nor difference; neither coming nor going forth. Able to teach these causes and conditions, he skillfully ends all conceptual proliferation. I bow my head to the Buddha, foremost among all who teach.”

Eight negations come first. Only the final two lines give the reason for the bow: because the Buddha can teach causes and conditions in this way, he can skillfully end all conceptual proliferation; because he can end all proliferation, what he teaches is foremost among everything expressed in words. Qingmu gives only one judgment on these two verses: “By praising the Buddha with these two verses, the ultimate meaning has already been stated in brief.”

Before the twenty-seven chapters have begun, the ultimate meaning has already been said in full. Every refutation that follows only says these two verses again. The Treatise on the Middle does not advance layer by layer toward its conclusion. It gives you the conclusion first, then lets you walk through one gate after another until you arrive there yourself.

For Whom Was This Treatise Composed?

To understand the eight negations, first ask whom this treatise addresses. The chapter begins with the question, “Why was this treatise composed?” Qingmu’s answer has a distant and an immediate part.

The distant occasion is the many non-Buddhist accounts of production. Qingmu lists eight in a single breath: “Some say all things are produced by Maheśvara; some by Viṣṇu; some by conjunction; some by time; some by primordial nature; some by transformation; some by spontaneity; and some by atoms.” The eight schools differ, but their illness is one: each places a source capable of producing behind the myriad things. Once such a source is posited, one cannot avoid “falling into wrong views of no cause, wrong cause, annihilation, and permanence, while speaking in many ways of self and what belongs to self.”

How did the Buddha treat this? Qingmu describes two steps. He “first taught the twelve links of dependent origination in the śrāvaka teaching,” then “for those already practiced, possessed of the great resolve, and capable of receiving the profound Dharma, taught through the Mahāyāna the mark of causes and conditions.” That mark is that “all dharmas neither arise nor cease, are neither identical nor different, and so forth; they are ultimately empty and have no possession of their own.” Qingmu grounds this in scripture, saying that “as stated in the Prajñā-pāramitā,” the Buddha told Subhūti: “When a bodhisattva sits at the place of awakening, he contemplates the twelve links of dependent origination as inexhaustible like space.” (The corresponding passage in the Smaller Prajñā-pāramitā Sūtra reads: “When a bodhisattva sits at the place of awakening and contemplates the twelve links of dependent origination in this way, he is free from the two extremes. This is a dharma unique to a bodhisattva.”) The twelve links are not canceled by emptiness. They are to be contemplated like space.

The immediate occasion identifies the treatise’s true audience. Qingmu says that during the Semblance Dharma five hundred years after the Buddha’s passing, “people’s faculties became dull and they clung deeply to dharmas.” They sought fixed marks, “did not know the Buddha’s intent, and clung only to words.” Such people heard of ultimate emptiness but did not understand the causes and conditions on account of which the Buddha taught emptiness. They wondered: If everything is ultimately empty, how can the karmic results of demerit and merit be distinguished?

Qingmu then writes the pivotal sentence of the entire passage: “Grasping this mark of emptiness, they become attached and produce many faults within ultimate emptiness.”

The illness is not emptiness. It is turning emptiness into something else to clutch. The eight schools place a source behind the myriad things. These people place outside dharmas a real object called “emptiness” and embrace it. The two illnesses appear opposed, but perform the same action: the hand reaches out to grasp. Nāgārjuna composed the treatise precisely “for such people.” Only here does the weight of the eight negations settle into place. They are not eight propositions. They are eight releases of the hand.

Why Are Eight Matters Enough? Non-arising and Non-ceasing Are the Root; the Other Six Establish Them

Someone immediately asks, “Dharmas are limitless; why refute them through only these eight matters?” Qingmu answers without hesitation: “Although dharmas are limitless, stating these eight matters in brief comprehensively refutes all dharmas.”

He begins with non-arising. The masters of various treatises describe the mark of arising in many ways: “cause and effect are identical; cause and effect are different; the effect already exists in the cause; the effect does not yet exist in the cause; a thing arises from itself; it arises from another; it arises from both; there is arising; or there is no arising.” Qingmu judges them all in a single sentence: “All such descriptions of arising are incorrect.” His conclusion is: “No fixed mark of arising can be found; therefore there is non-arising.”

Once non-arising is established, non-ceasing follows: “If there is no arising, how could there be ceasing?” Qingmu takes another step: “Because there is neither arising nor ceasing, the other six matters are likewise absent.”

Another question follows. If non-arising and non-ceasing already refute all dharmas comprehensively, why mention the other six?

Qingmu answers: “To establish the meaning of non-arising and non-ceasing.” The latter six are not six additional refutations. They say the same thing—non-arising and non-ceasing—again for people standing in different places.

Some reject non-arising and non-ceasing but accept neither permanence nor annihilation. Qingmu says: “If one investigates non-permanence and non-annihilation deeply, it is precisely non-arising and non-ceasing.” The reason lies in the statement that “if a dharma truly existed, it should not become nonexistent.” To have existed before and not exist now is annihilation; to have possessed a nature beforehand is permanence. To say neither permanence nor annihilation is therefore already to have entered the gate of neither arising nor ceasing.

Some hear four refutations but still try to establish dharmas through four gateways. Qingmu says, “If identical, there is no condition; if different, there is no continuity.” Identity leaves no producer and produced; difference leaves earlier and later unable to connect.

Some hear six refutations but still establish dharmas through coming and going forth. Coming means that “dharmas come from Maheśvara, primordial nature, atoms, and so forth”; going forth means “returning to their original place.” This returns directly to the eight accounts at the beginning of the chapter. Coming and going forth give the myriad things a home to come from and return to.

The eight negations are not eight knives. They are one knife turned toward eight ways of gripping. The hand refuses to open in eight different postures, and so the teaching is spoken eight times.

What the Worldly Eye Sees: One Grain Exhausts the Eight Matters

Qingmu suddenly leaves abstract reasoning and turns to what the eye can see. He repeats the same reason each time: “Because it is visibly evident in the world.” Throughout, he uses only one object—a grain.

Non-arising: “Apart from the grain at the beginning of the eon, present grain cannot be found.” This grain today did not originate from nowhere as an entirely new event. Non-ceasing: “If it had ceased, grain should not exist now.” If it had truly vanished without remainder, where did the grain in your hand come from?

Not permanent: “When a grain sprouts, the seed changes and breaks down.” How could it be permanent? Not annihilated: “A sprout comes from the grain,” followed by, “If it were annihilated, there should be no continuity.”

Not identical: “A grain does not make the sprout, nor does the sprout make the grain.” It is not one thing merely changing its name. Not different: “If they were different, why distinguish grain sprout, grain stalk, and grain leaf, rather than tree sprout, tree stalk, and tree leaf?” If wholly unrelated, why call this particular thing a grain sprout?

No coming: “The sprout within a grain has nowhere from which it comes.” If it came, “the sprout should arrive from elsewhere like a bird flying in to perch on a tree.” A sprout is not a bird; it does not fly into the grain from somewhere else. No going forth: if it went forth, “one should see the sprout emerge from the grain like a snake emerging from its hole.” Who has ever seen a sprout tunnel out like a snake leaving a cave?

The eight negations sound abstruse, yet Qingmu places them upon a single grain and repeats each time, “Because it is visibly evident in the world.” The eight negations are not conclusions produced by inference. They name what has always been so and what anyone can see. Between grain and sprout, no arising or ceasing, permanence or annihilation, identity or difference, coming or going forth can be found. It is not absent because the reasoning is unusually subtle. It was never there. Each negation denies a claim imposed by us, not something inherent in the grain.

The Treatise Author Takes Up the Question Himself: Arising Is Sought through Four Gateways and Found in None

After listening for some time, the questioner says something entirely honest: “Although you have explained the meaning of non-arising and non-ceasing, I want to hear what the treatise author himself has said.” The verse appears:

“Dharmas do not arise from themselves, nor from another; neither from both nor without a cause—thus one knows non-arising.”

Self, other, both, and no cause: these are the only four routes from which arising could come. When all four gateways are impassable, arising cannot be found. Qingmu explains each one.

Self-arising fails. “No thing arises from its own body; it must depend on many causes.” If it truly arose from itself, “one dharma would have two bodies: one called arising, the other that which arises.” The producer itself would also need to be produced, and thus “arising would have further arising, without end.”

Arising from another fails for a very light-footed reason: “Because there is no self, there is no other. Why? An other exists because of a self.” If self cannot stand, other has no ground on which to stand. Arising from both fails because it combines self and other and inherits the faults on both sides: “Arising from both has two faults.”

Arising without a cause fails. Qingmu says, “If all things existed without a cause, they would be permanent,” and gives the consequence: “Without a cause there is no effect. If an effect existed without a cause, giving and keeping precepts should lead to hell, while the ten evil acts and five heinous offenses should lead to heaven.” If effects truly required no causes, good and evil, suffering and happiness, could be mismatched at will. This directly answers the doubters at the opening of the chapter: refuting arising does not refute cause and effect; the assertion that truly destroys cause and effect is arising without a cause.

The verse digs further and exposes the root:

“The intrinsic nature of dharmas is not present in conditions; because intrinsic nature is absent, other-nature is likewise absent.”

Qingmu’s explanation has only two levels. First: “The intrinsic nature of dharmas is not present among the many conditions; a name is obtained only because the conditions come together.” Search through the conditions and no intrinsic nature can be found; when they assemble, what is obtained is only a name. Second: “Other-nature exists because of intrinsic nature; relative to the other, other-nature is itself intrinsic nature.” On its own side, what is called other-nature is still intrinsic nature. Thus, “refuting intrinsic nature refutes other-nature.”

Once intrinsic nature falls, all four gateways fall with it. Qingmu concludes: “Arising without a cause entails a great fault. If arising with a cause can be refuted, how much more arising without one! Arising cannot be found within the four alternatives; therefore there is non-arising.”

Only here has the first “not” in the homage been fully accounted for. Non-arising does not say that nothing happens. It says that in the event that happens, no self exists that can set it in motion.

Abhidharma Establishes Four Conditions: The Strongest Position in the Chapter

At this point a genuine opponent arrives. Abhidharma proponents say that “dharmas arise from four conditions.” Non-Buddhists establish Maheśvara or atoms, taking a substance as their ground. Abhidharma instead takes the Buddha’s own terminology: the myriad things arise through many conditions. This is almost exactly what Nāgārjuna himself wishes to say.

The verse on the four conditions reads:

“The causal condition, immediately preceding condition, object condition, and dominant condition: these four conditions produce dharmas; there is no fifth condition.”

The position is exacting: “Every condition is included within the four,” and each receives a definition. Causal condition “names all conditioned dharmas.” Immediately preceding condition comprises “all past and present mental events and mental factors apart from the final mental event and factors of an arhat in the past or present.” Object and dominant conditions include “all dharmas.” Even the fine point that an arhat’s final mind does not serve as an immediately preceding condition has been accounted for.

This is the position most worthy of respect in the chapter, and its refutation is the chapter’s central work. What is refuted below is not the fact that dharmas arise through many conditions. It is the concealed reality smuggled into the word “arise.”

Comprehensive Refutation: The Effect Gives “Condition” Its Name

The respondent does not yet dismantle the four one by one. He first asks a comprehensive question:

“Does the effect arise from conditions or from non-conditions? Do these conditions possess an effect, or do they lack one?”

Qingmu says, “Neither is so,” then gives the sharpest reason in the chapter:

“Because this dharma produces an effect, it is called a condition; before the effect has arisen, why is it not called a non-condition?”

Qingmu unfolds this: “Conditions have no fixed status.” “Before the effect has arisen, something is not called a condition; it is only because the eye sees an effect arise from it that it is called a condition.” Then comes the judgment: “A condition is established through the effect, because the effect is later and the condition earlier.”

We imagine that condition stands first and effect comes later, so the condition must have made the effect. Qingmu says that the effect turns back and gives the condition its name. His example is a jar. Water and clay come together and a jar arises; “seeing the jar, one knows that water, clay, and so forth are the jar’s conditions.” But “before the jar arises, why not call the water, clay, and so forth non-conditions?” Hence the conclusion: “Therefore the effect does not arise from conditions. If conditions do not produce it, how much less non-conditions.”

Another verse closes the remaining route:

“Whether the effect already exists or does not exist in the conditions, neither is possible. If it was absent before, for whom are they conditions? If present before, what use are conditions?”

If the effect already exists in the conditions, they are not called conditions—“because the effect already exists.” The thing is already there; what did the conditions do? If the effect did not exist in them, they likewise are not called conditions—“because they do not produce some other thing.” Qingmu concludes in one sentence: “The effect neither exists nor does not exist beforehand in conditions.”

A condition is a name within a relationship, not a thing that stands by itself. Once this is seen, refuting the four conditions one by one simply puts it into practice.

Refuting the Four Conditions One by One

The questioner is unsatisfied: “You have comprehensively refuted all causes and conditions. Now I wish to hear each of the four conditions refuted in turn.”

Causal condition.

“If an effect is produced neither as existent, nor as nonexistent, nor as both existent and nonexistent, how can there be a causal condition?”

There are only three possibilities for production: the effect exists beforehand, does not exist beforehand, or both exists and does not exist. What already exists need not be called produced, “because it already exists.” What is absent cannot be called produced either, “because it was absent beforehand”; moreover, this “would make conditions the same as non-conditions.” As for both existence and nonexistence, Qingmu says, “Both existent and nonexistent means half existent and half nonexistent, entailing the faults of both,” and adds, “Existence and nonexistence contradict one another, and nonexistence contradicts existence. How can one dharma have two marks?” None of the three routes works, so “how can there be a causal condition?”

Immediately preceding condition.

“Before an effect has arisen, there should be no ceasing. How can a ceased dharma condition anything? Thus there is no immediately preceding condition.”

An immediately preceding condition means that the prior mental moment ceases and yields to the next. Qingmu asks simply, “The future dharma has not arisen; for whom does it serve as an immediately preceding condition?” The one to receive it has not arrived. Conversely, “If the future dharma already exists, it has arisen; what need has it of an immediately preceding condition?” He does not spare the present side either. Present mental events and mental factors “have no time in which they abide.” If they do not abide, they cannot serve as conditions; if they do abide, they are not conditioned dharmas. As for the claim that “a ceased dharma still exists,” it would be permanent, and “if permanent there would be no demerit or merit.” Retreating to the claim that it conditions at the time of ceasing also fails: “At the time of ceasing it is half ceased and half not yet ceased; there is no third dharma called the time of ceasing.” This refutation proceeds from the opponent’s own doctrine.

Qingmu finally invokes the Buddha’s words as evidence: “All conditioned dharmas cease moment by moment, with no moment in which they abide.” If there is moment-by-moment ceasing and not one moment of abiding, the Abhidharma distinction between dharmas intending and not intending to cease has nowhere to be placed: “If you say this distinction does not exist within a single moment, you refute your own doctrine.”

Object condition.

“In the true and subtle Dharma taught by the buddhas, how could there be an object condition regarding dharmas that have no object?”

Qingmu says that the Mahāyāna dharmas taught by the Buddha—material and immaterial, formed and formless, contaminated and uncontaminated, conditioned and unconditioned—“when their marks enter dharma-nature, are all empty, signless, and objectless.” He immediately gives an image: “Like the many rivers entering the sea and becoming one flavor.” Once each river enters the sea, its water is no longer separately its own.

He then adds a criterion: “The true Dharma is worthy of trust; teachings adapted to capacities cannot be taken as ultimately true.” This sentence is a key to reading every category in the Buddha’s teaching. The same Buddha speaks some words ultimately and others according to what a person can receive at that time. To take the latter as the former is to cling.

Dominant condition.

“Dharmas have no intrinsic nature, and therefore no mark of existence; the claim that because this exists, that exists does not stand.”

The dominant condition rests on the familiar statement “because this exists, that exists.” Qingmu answers: “This does not stand here.” Why? “Because dharmas arise from many conditions, they have no fixed nature of their own; because they have no fixed nature, no mark of existence is present.” If no freestanding “exists” can be pointed to, what do the two occurrences of “exists” in that statement each denote?

He adds the most important final stroke: “The Buddha spoke in accommodation to ordinary people’s distinctions between existence and nonexistence.” The statement meets ordinary minds where they habitually divide what exists from what does not. Its level is guidance, not ultimacy.

Conclusion: With No Effect, Conditions and Non-conditions Fall Together

Once the four conditions have been refuted, two final sweeps remain. The first concerns collective and separate search:

“Sought among causes and conditions collectively or separately, an effect cannot be found; if none is present in causes and conditions, how can it emerge from conditions?”

Qingmu distinguishes them clearly: “‘Collectively’ means no effect in the conjunction of causes and conditions. ‘Separately’ means no effect in any individual condition.” We often suppose that combining things adds something extra. Here, nothing is present either in combination or in separation.

The second sweep concerns non-conditions:

“If an effect absent from conditions could emerge from conditions, why would that effect not emerge from non-conditions?”

Qingmu presses the question to its end: “If there is no jar in clay, why does it not emerge from milk?” There is no jar in clay, and equally no jar in milk. Why should a jar emerge from clay in particular? To say an effect emerges from conditions rests on habitual experience, not on an effect actually waiting inside the conditions to come out.

The final verse then gathers the entire chapter:

“If an effect arises from conditions, those conditions lack intrinsic nature. How can what arises from the natureless be said to arise from conditions? An effect arises neither from conditions nor from non-conditions; because no effect exists, neither conditions nor non-conditions exist.”

Qingmu’s explanation is clean. First the condition: “An effect arises from many conditions, so those conditions lack intrinsic nature. If they lack intrinsic nature, there is no dharma; how can no dharma produce?” Then the non-condition: “‘Non-condition’ is spoken of because condition is refuted. In reality there is no dharma called a non-condition.” The name non-condition is posited only in order to refute condition; it is not itself a thing. Finally, both are brought together: “If the effect arises from neither, there is no effect. Because there is no effect, neither conditions nor non-conditions exist.”

This is the chapter’s final hammer stroke: it is not that conditions exist first and are later discovered to be unable to produce an effect. Once the effect cannot be found, the name “condition” has nowhere to hang. Condition and non-condition are mutually supporting names. Remove the effect and both fall at once.

Return now to the first act of homage. Its meaning becomes clear: “Able to teach these causes and conditions, he skillfully ends all conceptual proliferation.” The praise is not that the Buddha gives a doctrine about causes and conditions. The very way he teaches them brings proliferation to rest. That is why the verse continues, “I bow my head to the Buddha, foremost among all who teach.” His primacy lies precisely here.

What Can a Reader Carry Away from This Chapter?

First, remember the two kinds of person for whom the treatise was composed: one “clings only to words,” the other “grasps the mark of emptiness and becomes attached.” The eight negations treat the same action in both.

Second, keep “a condition is established through the effect” close to heart. Familiarity with a name does not mean that something capable of producing stands there.

Third, know that refuting conditions is not refuting cause and effect. What truly destroys cause and effect is the assertion of no cause. This chapter refutes only the speck of intrinsic nature placed upon conditions.

Carried Forward: Jizang’s Mirror

The preceding is the meaning of the chapter itself. What follows lets the Sui–Tang Three Treatises tradition illuminate it from the side. The mirror is not the treatise; it offers another way to read the same verses.

Jizang’s Commentary on the Treatise on the Middle first reads the chapter title together with the title of the treatise. The true causes and conditions contemplated here are originally neither arising nor ceasing, through neither coming nor going, and so causes and conditions reveal the Middle Way. Contemplation discerns in the mind; treatise declares through the mouth. He also gives “causes and conditions” three levels: the four conditions refuted here; the general fact that saṃsāra and nirvāṇa, ordinary and noble, understanding and delusion, all stand through relative provisional designation; and the twelve links of dependent origination. All three readings coexist. He does not ask the reader to choose only one.

Jizang divides the chapter’s movement into three parts: first, two verses clarify non-arising through the four gateways; next, an outsider establishes the four conditions to prove arising; finally, from “Does the effect arise from conditions?” onward, the four conditions are refuted to reveal non-arising. His mirror language for non-arising through conditions is direct: precisely because something provisionally depends on conditions, it has no intrinsic nature; because it has none, it does not arise. He uses the moon in water: moon and water are both present, yet the reflected moon has no fixed arising. At the final verse, the commentary says again that conditions exist through the effect and therefore lack intrinsic nature, while the effect exists through conditions and therefore lacks intrinsic nature.

Jizang gives the most attention to the object condition. He treats cognitive engagement as the root of illness and says that seeking apprehender and apprehended and finding neither ends that illness through non-attainment. He also preserves five explanations: mind and object named together; the apprehending mind again apprehending the object; the object producing the apprehending mind; the myriad dharmas serving as conditions for the mind that apprehends them; and the general and specific senses of condition named together. He explicitly says that the object condition is precisely the absence of a fixed object condition, not that its meaning is swept away as well; he explains “signless” as no apprehended object and “objectless” as no apprehending mind.

This mirror connects the chapter to the Three Treatises tradition’s interpretive structure of refuting the false and revealing the correct. Its judgments are Jizang’s additions, not the original words of Nāgārjuna’s verses or Qingmu’s long commentary. The readings are recorded side by side without adjudication, each returned to its own place.

The person who opens this chapter with a question offers no abstruse doctrine. He invokes the most ordinary fact: “The worldly eye sees activity in the three times.” Anyone who opens their eyes can see activity in all three phases—the stretch already walked, the stretch not yet walked, and this step being walked now. If the activity is visible, then where there is activity, there must be a dharma.

This is a difficult position to refute. One may doubt an invisible soul; it is much harder to doubt the road one has just walked.

Nāgārjuna does not deny that road. What he does next is simply take the three phases apart, one cell at a time.

Already Gone, Not Yet Gone, and Presently Going

The opening verse lays out the three cells:

“There is no going in what has already gone, nor in what has not yet gone; apart from the already gone and not yet gone, there is no going at the time of going.”

Qingmu’s explanation takes only a few sentences, but it closes all three. Why is there no going in what has already gone? “Because it has already gone.” The activity of going has subsided there. To say that another going remains would mean that “an act of going exists apart from going”—separating the act and setting another copy down beside it. That cannot stand. Why is there no going in what has not yet gone? “Because no dharma of going has yet occurred.” Going has not happened there.

The mind naturally retreats into the middle: then let going reside in the instant presently going.

That retreat is exactly what this chapter asks us to see. Qingmu says, “The time of going names what is half gone and half not yet gone; it is not apart from the already gone and not yet gone.” The name “time of going” is assembled from what has gone and what has not: half already gone, half not yet gone. It is not a third independent patch of ground. To hide in the time of going is to hide again in the two cells just examined.

Put Another Going Inside the Time of Going, and the Count Increases by One

The questioner hardens the position: “Going occurs wherever there is motion.” Going is present where activity is occurring. Activity has ceased in what has gone and has not yet begun in what has not gone, so going can only be present at the time of going.

Nāgārjuna begins his answer with the plainest question: “How could a dharma of going be present at the time of going?”

The key is how the time of going comes to be. It is not a span of time waiting there first for going to enter it. It obtains its name through the presently operating dharma of going. Since it exists through going, it cannot then become a container for going. Qingmu gives an accessible comparison: if a time of going existed apart from the dharma of going, “going would have to be contained within the time of going as fruit is contained in a vessel.” The vessel exists first and fruit is placed into it later—but here, the fruit itself makes the vessel.

Moreover, once a time of going is said to exist apart from going, the two “no longer depend on one another.” The time of going can go by itself; the goer need have nothing further to do with it.

Insist on this, and the count fails. “If going exists at the time of going, there are two kinds of going”: one establishes the time of going, and the other is placed inside that time. The same event has been counted twice. As soon as the dharmas of going multiply, people multiply with them: “One person has two goings and two goers.” No one would accept that walking produces two walks and two people walking.

What If the Goer Is Established First?

The questioner tries another route: if the dharma of going cannot be separated from a goer, concede first that a real goer exists in the three times.

Nāgārjuna answers with several verses in succession, each closing one opening.

The first looks for a goer apart from the dharma of going. Yet the name “goer” comes from going in the first place. Where no dharma of going exists, no one can be called a goer.

The second establishes goer and going together, letting the goer employ the dharma of going. Once again, the same goer has two goings—one makes him a goer, the other is there for him to employ.

The third exposes a reversed sequence. Qingmu says directly: “To say the goer uses the dharma of going is to place the goer first and the dharma of going later.” A ready-made goer exists first and only then begins to go. But a person who is not going is not called a goer; they must go before they become one.

Another verse closes both human poles at once: “The goer does not go; the non-goer does not go,” and “there is no third goer.” The goer is already established through going; their going is not a separate task still to be performed. The non-goer has no going at all. Apart from these, no third kind of person—neither goer nor non-goer—can be found to carry out the act.

In Which Moment Does the First Step Fall?

Qingmu now pursues the inquiry one step further for the reader: “If going and a goer determinately exist, there should be an initial setting forth.” If going possessed intrinsic nature, there ought to be a first instant that can be pointed out.

The verse seeks this instant in the three times:

“There is no setting forth in what has gone, none in what has not gone, and none at the time of going. Where could setting forth occur?”

It cannot be found in what has already gone, for the walk is complete there. It cannot be found in what has not yet gone, for there is no walking there. Nor can it be found at the time of going, because that time must wait for the dharma of going before it can be established; it comes later than the first step.

A second verse illuminates the point from the other side: “Before setting forth, there is no time of going and nothing already gone.” Before the first step, the two fields called “time of going” and “already gone” do not yet exist. How can the first step be placed in a field that does not exist? The verse ends with the question: “If there is no setting forth anywhere, why make these distinctions?”

Qingmu dismantles the series to its end: “Because there is no setting forth, there is no going; because there is no going, there is no goer.” Once the goer is gone, the three names already gone, not yet gone, and time of going become empty as well.

When we speak of an event’s beginning, we ordinarily assume that time is already there and the event occurs at one of its points. This passage sees the reverse: the event itself draws the point.

The One Who Stops Is the Same

The questioner changes direction. If going cannot be established, then surely standing still can—the motionless thing, at least, must be real.

The verse answers in the same form used to refute going: “The goer does not stand, nor does the non-goer stand,” and apart from them there is no third standing. Qingmu explains both sides. The goer does not stand “because going has not ceased; what opposes going is called standing.” They are still going, while standing is precisely its opposite. The non-goer does not stand either, because “standing exists because the dharma of going ceases; without going, there is no standing.” The name standing comes from the cessation of going. If no going ceases, there is no standing to name.

Standing is not the thing left behind once motion has departed. Like going, it is a name that can be spoken only through its counterpart.

Qingmu then extends this gateway: “Every dharma of continuation and cessation follows the same reasoning as going.” His examples are not obscure. A grain continues into sprout, stalk, and leaf: this is continuation. The grain ceases, and therefore sprout, stalk, and leaf also cease: this is cessation—“succession is called continuation; interruption is called cessation.” He gives a still more intimate sequence: “Ignorance conditions formations, and so on through aging-and-death: this is called continuation.” When ignorance ceases and formations and the rest cease, that is cessation.

One seed and one cycle of birth and death are measured with the same rule.

Then Trust the Eye

The questioner’s final card is the first one again. Every gateway has been refuted, yet the eye plainly sees people walking and stopping.

Qingmu answers in six characters: “What the physical eye sees is not trustworthy.”

It is easy to make this sentence too heavy. He is not saying that you failed to see a body moving. He asks something else: can the movement you see prove a going and a goer with intrinsic nature? He immediately presents two routes from which the opponent must choose: “Are they established through one dharma or through two?” Identity or difference.

Identity means that the dharma of going is the goer. Then “agent and action would be one”: walking and the person walking collapse into the same thing. Qingmu says this “causes confusion and destroys causes and conditions,” because originally “the goer exists because of going and going because of the goer.” Merge them and there are no longer two ends that can condition one another. He also looks from another side: “The person is permanent, the dharma impermanent.” The goer is a person, going a dharma. If both are one, both must be permanent or both impermanent.

Difference causes a more obvious difficulty. Difference means each is established independently: “Before the dharma of going exists, the goer should exist; before the goer exists, the dharma of going should exist.” A person who walks could be present with no walking at all, and walking could walk by itself with no one walking. Since they “do not depend on one another,” “when one dharma ceases, the other should remain.” One disappears and the other continues unchanged.

They are neither identical nor different, and outside these two gateways there is no third way to establish them.

Each End Depends on the Other; Neither Exists from Its Own Side

The next two verses state mutual dependence directly.

“The goer known through going cannot employ this same going; before the dharma of going existed, no goer went.”

The goer cannot then employ the going through which they were recognized as a goer, because before that going there was no such goer. Qingmu gives a subtle contrast: “When a person and a city exist first, setting forth can occur.” Person and city can already be there, and only then can departure take place. A goer is not something like that: “The goer is established through the dharma of going, and the dharma of going through the goer.” A person exists first. A goer does not.

Substituting another going also fails. One goer has no room for a second going.

In the final passage Qingmu names both positions. “Fixed means truly existent from the beginning, not produced through the dharma of going.” A fixed goer is already present without depending on going. “Unfixed means not truly existent from the beginning, but obtaining the name goer through the dharma of going.” An unfixed goer has no share of its own and is merely named through going. If the former stood, he would remain a goer apart from going and could never stop. The latter has no independent part to begin with and therefore cannot command going in the three times. The dharma of going is examined along with him: “Dharma of going names the body’s motion.” It too cannot be established before and apart from a goer.

Qingmu then gives the chapter’s conclusion: “These dharmas all depend on one another.” Going, goer, and destination are all mutually dependent names. Immediately follows the crucial statement: “They cannot be called determinately existent or determinately nonexistent.”

Only then does he arrive at the final sentence: “The three dharmas are false, empty, and without possession of their own; they exist only as provisional designations, like an illusion or transformation.”

A provisional designation is not empty talk. A person still rises, walks to the doorway, and stops. Yet within the whole event no going can be found that stands there without depending on anything else, and no goer can be found who stands there without depending on going. “Like an illusion or transformation” means precisely this: the person conjured by an illusionist walks, and you see him walk, yet no one who walks can be extracted from him.

What Can a Reader Carry Away from This Chapter?

First, notice the word “presently.” When past and future both prove unreliable, the mind retreats to the present instant, imagining that something real must remain there. This chapter shows that the instant itself is a name assembled from before and after.

Second, see where a name comes from. The goer comes from going, standing from the cessation of going, and the destination from both. Many everyday identities arise in the same way—the person who acts, the person who speaks, the person who walks this road. When a name is established before the event, it usually has already lost its footing.

Third, do not stop walking because walking has been refuted. Qingmu himself says, “They cannot be called determinately existent or determinately nonexistent.” What is refuted is the “determinately,” not your two feet.

Carried Forward: Jizang’s Mirror

The preceding is the meaning of the chapter itself. What follows lets the Sui–Tang Three Treatises tradition illuminate it from the side. The mirror is not the treatise; it offers another way to read the same verses.

Jizang’s Commentary on the Treatise on the Middle first classifies the chapter. In a broad division it has seven gateways named after what they refute: a three-time gateway refuting going; a three-time gateway refuting the goer; then refutations of initial setting forth, standing and stander, going and goer through identity and difference, through causes and conditions, and through fixed and unfixed. In an intermediate division he condenses these into four gateways named after the means of refutation: the three times, identity and difference, causes and conditions, and existence and nonexistence. These are a commentator’s divisions, not chapter headings supplied by Nāgārjuna’s verses or Qingmu’s prose.

He also establishes the chapter’s terms through the condition of moving a foot. A person seated still is simply a person; once the foot moves, the person is called a goer and that motion a dharma of going. Before the step is “not yet gone”; the place crossed by the foot is “already gone”; its present motion is the “time of going”; its initial motion is “setting forth”; the motion of the current step is “this going”; other steps compared with it are “other going”; and the place on which the foot falls is “destination.” All these names are falsely received through the conjunction of many conditions; examined closely, not one can be found. From the fault of two goings he infers two times, two bodies, two form aggregates, and ultimately two persons. In refuting standing he gives two purposes: first refute motion and then stillness, so that amid the evident four postures of walking, standing, sitting, and lying down one falls into neither; then, through standing, refute going again, so that once standing is refuted the mind that seeks going also rests. In the fixed–unfixed gateway he connects the two extremes to permanence and annihilation, real and provisional: fixed existence is permanence and fixed nonexistence annihilation; a person existing independently of dharmas is permanent and real, while a person existing through dharmas is impermanent and provisional. The commentary ends with illusory going to prevent refutation of intrinsic nature from being read as annihilating absence: illusory going is plainly present and therefore not fixed nonexistence; illusory going is not real going and therefore not fixed existence.

This mirror connects the chapter to the Three Treatises tradition’s interpretive structure. Its divisions, terminology, and layers of challenge are Jizang’s additions, not the original words of Nāgārjuna’s verses or Qingmu’s long commentary. The readings are recorded side by side without adjudication, each returned to its own place.

The instant the eyes open, the world is there. It happens so seamlessly that no one stops to ask: where is the one who sees?

This chapter begins there. It does not ask what you have seen. It asks whether “seeing” itself can stand.

The questioner at the beginning of the chapter comes holding a sūtra. He says, “The sūtra teaches that six faculties exist,” then recites its verse:

“The eye and ear, together with nose, tongue, body, and mind, are the six faculties; these six, beginning with the eye, operate in relation to the six objects, beginning with form.”

He then unfolds its meaning. The eye is inside, form outside, and “the eye sees form.” The sequence continues until mind is inside, dharmas outside, and mind knows dharmas. Six faculties each face one of six fields, corresponding one by one.

The question is disciplined and forceful. It is not a theory of the questioner’s invention; the words are already present in the sūtra.

The respondent gives only two words.

“No. Why?”

A Lamp Illuminates Itself; the Eye Does Not

To see why the treatise author dares answer this way, first look at the measure he holds.

The verse says:

“This eye cannot see its own body; if it cannot see itself, how can it see other things?”

Qingmu turns the verse into an analogy. He says, “A lamp illuminates both itself and other things.” When a lamp shines, it stands in the light together with everything else. Illumination need not be divided into an inside and an outside; as soon as it operates, it reaches both.

Qingmu then draws the comparison: “If the eye bore the mark of seeing, it should see itself and other things.” If the eye itself were the mark of seeing, it should be like a lamp: it should illuminate itself as well as everything else.

“Yet that is not so.”

It is not. A pair of eyes sees mountains, people, and a point of fire in the distance, but never itself. What appears in a mirror is a reflected image, not the thing presently seeing.

Notice the treatise author’s method. He does not say “eyes are useless”—the chapter never says this. He takes what the name “seeing” would have to mean and uses it to measure an eye said to possess vision in itself. A lamp’s illumination is pervasive; the eye’s vision is partial. Since it is partial, seeing is not a ready-made capacity carried inside the eye.

Being able to see other things but not itself already tells us that seeing is not a thing. It is an event established through conditions.

Fire Burns Other Things but Not Itself—This Rescue Cannot Rescue the Claim

The questioner immediately supplies another analogy, and does so cleverly.

He says: “Although the eye cannot see itself, it can see other things, just as fire burns other things but cannot burn itself.”

What weight does this carry? He effectively concedes that the lamp analogy fails, then substitutes another ready example: things like this already exist in the world. Fire burns wood but not itself; therefore nothing is strange about the eye seeing form but not itself.

The respondent does not debate whether fire can burn itself. He says: “Although you offer the analogy of fire, it cannot establish a dharma of seeing in the eye.” However cleverly the analogy is exchanged, it still does not establish a seeing that belongs to the eye.

The verse then points out that this question has already been answered:

“The analogy of fire cannot establish a dharma of seeing in the eye; what has gone, not gone, and the time of going have already answered this matter.”

The words “already answered” point back to “Examination of Going and Coming.” That chapter seeks “going” in three periods. In the stretch already walked, going is over, so present going cannot be found. In the stretch not yet walked, going has not begun. What about the present moment of going? Spread the moment open and it is still only two sections, already gone and not yet gone. No independent going can be touched in any of the three.

Qingmu brings that measure directly to burning and seeing: “Just as no burning exists in what has burned, what has not yet burned, or the time of burning, no mark of seeing exists in what has been seen, what has not yet been seen, or the time of seeing.”

Changing the analogy does not solve the problem, because the problem is not the analogy but time. For the analogy to work, a real burning must first stand. Yet burning cannot be produced in any of the three times. One unfindable event cannot prove another; both ends remain suspended.

Seeing Is a Name Obtained Only When the Eye Faces Form

The treatise author now brings the words closer, to the most everyday part of experience.

The verse says:

“Before seeing has seen, it is not called seeing; the claim that seeing can see is therefore not so.”

Qingmu’s explanation is plain: “Before the eye faces form, it cannot see and is not called seeing. It is called seeing because it faces form.”

When your eyes are closed, are they still there? Yes. But you do not say that they are seeing. Only when the eye faces a form can the name seeing be placed upon the event.

The difficulty now appears. The mind habitually imagines that a ready-made seeing exists first and then goes out to see things. Qingmu says, however, “Before seeing, there is no seeing.” Before an act of seeing, the name seeing cannot be established at all. If it cannot be established, no preexisting seeing waits there, prepared to go out and see.

It is not a tool that exists first and is then used. It is an event that receives its name only when the encounter occurs.

This step matters. The preceding sections refuted a seeing-nature inherent in the eye. This section refutes extracting “seeing” from a particular relation and making it an independent item.

Neither Seeing nor Non-seeing Can House Another Seeing

At this point the treatise author closes the path with a pair of tongs.

The verse says:

“Seeing cannot possess seeing, nor does non-seeing see; once seeing is refuted, the seer is refuted with it.”

Anything in the world is either seeing or not seeing; there is no third kind. Examine both sides in turn.

First, seeing. Qingmu says, “Seeing cannot see, because this was explained before.” The preceding sections have already examined it: seeing does not see itself, and cannot be established before it faces form. To say that another seeing operates inside it merely repeats the same event without adding anything.

Next, non-seeing. Qingmu says, “Non-seeing does not see because it lacks the mark of seeing.” Something that is not seeing bears no mark of seeing. Without that mark, no seeing can occur.

Neither side has room for it, so the verse gives a linked conclusion: “Once seeing is refuted, the seer is refuted with it.”

Once seeing cannot be found, the one named through seeing—the seer—cannot be found either. Qingmu supplies two levels of reason.

First: “If a seer existed apart from seeing, a person without eyes should see through another faculty.” If the seer were someone existing separately from seeing, the seer would remain present in a blind person, who should then be able to see form with the ear or nose. This is not so.

Second: “If seeing is used to see, the mark of seeing resides in seeing.” If one instead says that the seer can see only by relying on seeing, the mark resides on the side of seeing, while “the seer lacks the mark of seeing.” The seer carries no such mark in itself.

One route has no support; on the other, what supports it does not belong to the seer. The position of seer cannot be reached by either path.

With No One Who Sees, There Is No Set of Things Seen

The verse drives the matter to its end:

“Whether apart from seeing or not apart from seeing, the seer cannot be found; since there is no seer, where can seeing and the visible exist?”

Qingmu explains it simply: “If seeing exists, the seer cannot be established; if seeing does not exist, the seer likewise cannot be established.”

When seeing exists, the event is already there and no separate owner need be established. When seeing does not exist, still less can anyone be said to see. Neither side succeeds; the seer has nowhere to stand.

Qingmu then asks a question close to the bone: “If there is no seer, who can use the dharma of seeing to distinguish external forms?”

This sounds like a challenge to the treatise author, but it actually points to the last shadow in the reader’s mind. Even if seeing depends on conditions, we feel, surely “I” still use it. Qingmu is saying that this position of “using” cannot be found either.

When that position is empty, the other end empties with it. The verse asks, “Where can seeing and the visible exist?” “The visible” is what stands on the seen side. If the person who sees cannot stand alone, the thing seen cannot stand alone either. Two ready-made things do not first exist and then meet; the paired names can be spoken only within their relation.

Release One Link and the Whole Chain Releases

The treatise author suddenly widens the view:

“Because seeing and the visible are absent, consciousness and the other three dharmas are absent; how, then, can the four kinds of clinging and the other conditions exist?”

Qingmu names them: “Because seeing and the visible are absent, consciousness, contact, feeling, and craving are all absent.”

The point becomes clear along the twelve links of dependent origination. Only when eye faces form is there eye consciousness. Only when faculty, object, and consciousness gather is there contact; only with contact is there feeling; only with feeling, craving. The links grow one after another. If the first section contains no seeing or visible that stands by itself, the later sections likewise contain no consciousness, contact, feeling, or craving that stands by itself.

Qingmu takes one more step: “Because craving and the rest are absent, the four kinds of clinging and the other links of dependent origination are absent.” Since craving has no intrinsic nature, the four kinds of clinging that arise through it—and existence, birth, and aging-and-death after clinging—have none either.

Notice what is said to be “absent.” The treatise author does not say that eyes never see form, that craving never arises, or that no one ever clings. What he bars is the “fixed.” The placement of this section also matters. It takes the chapter from what looks like a fine-grained problem of cognition all the way through the chain of saṃsāric continuation. What releases is not one link; from the beginning, no link in the chain was iron.

The Other Five Faculties Need Not Be Explained Again

The final verse closes:

“The ear, nose, tongue, body, and mind, together with sound, the hearer, and the rest— know that their meaning is wholly the same as explained above.”

Qingmu gives the reason: “Seeing and the visible are empty because they belong to many conditions and therefore are not fixed.” Each pair—ear and sound, nose and scent, tongue and flavor, body and touch, mind and dharmas—can be examined in the same way: the ear does not hear itself; hearing has no name before hearing; apart from hearing there is no hearer; hearing and the heard cannot be established separately.

Thus he says, “Because the meaning is the same, they are not explained separately.”

The sentence sounds like an economy of ink, but it is really an act of trust. The treatise author returns the method to the reader: this chapter does not give seven conclusions. It gives you a measure you can use again.

What Can a Reader Carry Away from This Chapter?

First, notice the place that “cannot see itself.” “A lamp illuminates itself and other things”: a lamp reaches both; the eye does not. Whenever some capacity seems ready-made within you and independent of anything else, ask: can it illuminate itself?

Second, do not establish an owner before an event. “If there is no seer, who can use the dharma of seeing to distinguish external forms?” The “who” beneath that question is the link most easily overlooked. When seeing occurs, there is simply seeing; no additional person need be installed behind it.

Third, remember that release of one link releases the whole chain. “Because craving and the rest are absent, the four kinds of clinging and the other links of dependent origination are absent.” This present act of seeing is not a small event unrelated to birth and death; it is the first section of the chain. See that this section depends on conditions, and the rest need no longer be treated as iron.

Carried Forward: Jizang’s Mirror

The preceding is the meaning of the chapter itself. What follows lets the Sui–Tang Three Treatises tradition illuminate it from the side. The mirror is not the treatise; it offers another way to read the same verses.

Jizang’s Commentary on the Treatise on the Middle first resolves the apparent conflict between sūtra and treatise. The Buddha spoke of six faculties so that people might understand the six as empty; the outsider hears this and clings to six fixed and real faculties, failing to recognize the Buddha’s intent. Jizang also says that the Buddha’s teaching of six faculties belongs to conventional truth, and that the Tathāgata teaches conventional truth precisely to lead people through it into the ultimate meaning. Lesser Vehicle practitioners instead take the six as real and have not entered the ultimate meaning. Thus, when the treatise author now says there are no six, his intent is to clarify the Buddha’s original purpose in teaching six, not to destroy the sūtra teaching.

Jizang also classifies the chapter. The respondent’s seven verses divide into two sections: the first six directly refute the eye faculty; the final verse contemplates the other five by analogy. The six verses are further marked in sequence as direct refutation, referral to the preceding refutation, renewed refutation, no person because no dharma, no dharma because no person, and no effect because no cause. He also gathers them into two levels: the first five refute seeing as cause and the last refutes its produced effect. Within the five, he distinguishes refuting the eye’s seeing, form’s seeing, and a person’s seeing. He reads the final consciousness and other three dharmas as no effect because no cause: eye and form conjoin to produce eye consciousness and then the four kinds of clinging; eye and form are causes, consciousness and the rest effects. Since the preceding search for three kinds of seeing found none, the causes are unestablished, and the effects are then shown not to arise.

This mirror supplies names and levels for the chapter’s divisions and connects the tension between sūtra and treatise to the Three Treatises tradition’s interpretation of conventional and ultimate truth. The classification and judgment are Jizang’s additions, not the original words of Nāgārjuna’s verses or Qingmu’s long commentary. The readings are recorded side by side without adjudication, each returned to its own place.

The questioner opens with a single sentence: “The sūtra teaches five aggregates. How is this to be understood?”

We need to see the weight of this sentence clearly. He is not putting forward a non-Buddhist thesis; he is citing the sūtra. The sūtra unmistakably speaks of form, feeling, perception, formations, and consciousness. Since the Buddha himself spoke in this way, surely the five aggregates must be five real things?

The respondent does not debate the sūtra. From among the five aggregates he takes the coarsest, the one that looks most truly existent—form—and turns it over and dismantles it before you. Only after he is finished does he say: the other four aggregates are the same.

Yet what finally settles at the end of the chapter is not form, but a passage teaching people how to ask and how to answer.

The Opening Verse Closes Both Ends

“If one departs from the cause of form, form cannot be found; if one departs from form, the cause of form cannot be found.”

The two lines point in precisely opposite directions. The first says: pull away the cause and form cannot be found. The second says: pull away form and that so-called “cause” cannot be found either.

Most people nod when they hear the first line—without material, where could the thing come from? But the second does not sit so easily: the material is plainly right there; how could it disappear just because the finished object is gone?

That second line is exactly what matters. It does not say that this bundle of threads has vanished. It says that the identity “cause of cloth” has vanished.

Cloth and Threads: An Example Anyone Can Touch

Qingmu immediately gives an example:

“The cause of form is like the threads that cause cloth: remove the threads and there is no cloth; remove the cloth and there are no threads. Cloth is like form, and threads are like its cause.”

Pull away the threads and there is no cloth—that direction is easy to understand. Pull away the cloth and there are no threads—this direction requires close thought. That bundle of thread is still in one’s hand. How can it be said not to exist?

Because the “threads” spoken of here are not threads in general; they are the cause of cloth. They are called a cause because they are spoken of in relation to a bolt of cloth. If no cloth is made, they are simply threads, not anyone’s cause. “Cause” is never a label carried on the body of a thing itself; it is a name attached only in relation to a fruit.

Thus, the verse does not say that cloth and threads are the same thing, nor that there is no cloth or thread in the world. It asks whether you can find one thing that “is fruit in itself” and another that “is cause in itself,” each standing on its own before the two are joined.

The answer is no. Each side must await the other before receiving its name.

Form Apart from Its Cause Would Be a Thing without a Provenance

The questioner first seizes on the first half and asks in return: What fault would there be if form existed apart from the cause of form?

The answering verse says:

“If form existed apart from the cause of form, that form would be without a cause; that a dharma should exist without a cause does not make sense.”

Qingmu again uses cloth: “It would be like cloth existing apart from threads: the cloth would have no cause. The world contains no dharma that exists without a cause.”

A bolt of cloth that exists without depending on threads has appeared from nowhere. For every thing in the world, ask where it came from and a succession can always be traced; the kind for which nothing can be traced has never actually been seen.

But the Opponent Immediately Produces a Host of “Uncaused Dharmas”

Here begins the longest dispute in the chapter. The questioner is unconvinced and lists three classes: Buddhism has three unconditioned dharmas; the unconditioned is permanent, and the permanent is therefore uncaused. Non-Buddhist teachings have space, time, direction, spirit, atoms, and nirvāṇa; worldly teachings also speak of space, time, and direction. The reasoning is orderly—these three are “found everywhere and therefore called permanent; because they are permanent, they are uncaused.” Since Buddhism itself accepts uncaused dharmas, on what grounds can you say that the world contains none?

The respondent’s reply is short:

“These uncaused dharmas exist only as things said. Once examined and distinguished in thought, none exists.”

There is only something said. Everyone speaks of space, time, and direction, but set the words down and investigate within them, and no thing can be traced. Being able to say a name and finding a real thing standing beneath that name are two different matters.

He then adds a sentence of exacting precision: “If a dharma exists through causes and conditions, it should not be called uncaused.” If you say that it comes through causes and conditions, it is not uncaused; if you say that it truly has no causes and conditions, it falls precisely on the side just denied. Both routes are closed.

The questioner retreats another step and adopts a more refined formulation: there are two kinds of cause, an operative cause and a cause in speech. Although these permanent dharmas have no operative cause, they do have a cause in speech, “because it enables people to know them.”

This is a clever move: he no longer insists that permanent dharmas have a cause that produces them, only a cause that enables people to know them—a name. But the respondent still does not accept it:

“Even if there is a cause in speech, the claim does not stand.”

The reason has two levels. First, he does not overstep: space will be refuted later among the six elements, while the remaining items will each be addressed in what follows. He does not pretend to have finished refuting them here. Only then comes the crux:

“Moreover, even things visibly present before us can all be refuted; how much more atoms and other dharmas that cannot be seen?”

Things everyone can see before them still cannot stand when investigated; could what no one can see somehow stand more firmly? Being unseen is not a safer place. People often imagine that if a thing is hidden where experience cannot reach it, it no longer needs to account for where it came from.

Now Look at the Other End: A Cause Apart from Its Fruit Is Whose Cause?

Once the first half has been addressed, the argument returns to the second.

“If a cause existed apart from form, it would be a cause without a fruit; if one speaks of a cause without a fruit, there is no place for it.”

Qingmu is economical: “If form as fruit is removed and only the cause of form remains, it is a cause without a fruit.”

The questioner still asks: What is wrong with a cause existing without a fruit?

The respondent gives two answers. The first is the same principle as the earlier “cause of cloth,” now stated more directly: “It is in relation to a fruit that something is called a cause. If there is no fruit, how can it be called a cause?”

The second cuts more sharply: “If the cause contains no fruit, why should a thing not arise from what is not its cause?” If you say that the fruit was never present within the cause, then it was equally absent within what is not its cause. Both sides share the same “absence.” On what grounds does the fruit emerge from one rather than the other? Qingmu points down a path in passing: “This matter was explained in ‘Examination of Conditions.’”

The question asks for a difference. If you are to call one thing a cause, you must be able to say how it differs from all the things that are not its cause; yet “the fruit is absent within it” is precisely what it shares with what is not the cause.

Form in the Cause, Form Not in the Cause: Neither Side Has Room

This presses the argument into the next verse:

“If form already existed, it would have no need of a cause of form; if form did not exist, it likewise would have no need of a cause of form.”

Qingmu explains: “If form already exists within the cause, it is not called the cause of form; if form does not exist within the cause, it is likewise not called the cause of form.”

If the bolt of cloth were already present in the threads, why weave it? What already exists need not be produced, so this “cause” has nothing to do. If the cloth were utterly absent from the threads, the threads would be no different from a heap of sand, which also lacks cloth; nothing would yet entitle them to be called the cause of cloth.

Presence does not work, and absence does not work. Put the fruit into the position of cause and the cause becomes superfluous; take the fruit out and the cause hangs empty.

Then Call It Uncaused—That Path Does Not Go Through Either

The questioner now takes a final step: since neither form present nor form absent can stand, may I simply say that there is an uncaused form?

The answering verse:

“That form should exist without a cause ultimately cannot stand. Therefore, a person of wisdom should not discriminate form.”

Qingmu replies with extreme brevity: “Neither a fruit existing in its cause nor a fruit not existing in its cause can be found—how much less form without a cause.” Even what has a cause cannot stand, so what has none needs no further discussion.

The second half of the verse deserves a pause. The verse abruptly stops speaking of form and begins speaking of the person. Qingmu explains “discrimination” very concretely:

“‘Discrimination’ refers to ordinary people: through ignorance and the defilement of craving they covet and cling to form, then through wrong views produce discriminative proliferation, claiming that the fruit exists in the cause, that it does not exist in the cause, and so forth.”

The sequence has three sections: first ignorance—not understanding; next the defilement of craving—the mind has stuck to form; only last come the assertions born of wrong views: that the fruit exists in the cause, that it does not exist in the cause, and all the other arguments.

See this sequence clearly: fixed claims about cause and fruit do not grow from nowhere. First there is a mind that has stuck fast; only then does it devise a theory for the place where it is stuck. Thus, at the end of the refutation, the respondent turns the words toward the person conducting the debate.

Nor does “should not discriminate” mean that one should not think—this entire chapter is thinking at its most exacting. It says that since “form is sought here but cannot be found,” one should not install a fixed claim upon form.

Does the Fruit Resemble the Cause, or Not?

One gate remains to be traversed.

〔T30 0007a05〕 “If the fruit resembles the cause, that cannot stand; if the fruit does not resemble the cause, that cannot stand either.”

First consider resemblance. Qingmu gives three differences: “the cause is fine and the fruit coarse”—threads are fine, cloth thick; “cause and fruit differ in color, efficacy, and so forth”; and the most interesting, “if cloth resembled threads it would not be called cloth, because the threads are many and the cloth one.” If cloth truly looked just like threads, it would no longer be called cloth. Moreover, there are many threads but one whole bolt of cloth—how are many and one alike?

Then consider unlikeness. “Hemp threads cannot make silk, and coarse threads cannot produce fine cloth.” Thread made of hemp cannot be woven into silk; coarse thread cannot be woven into fine cloth. If cause and fruit were wholly unrelated, any thread should be able to make any cloth, but this is not so.

The two sides close in: call them alike and the difference is erased, so cloth can no longer be cloth; call them unlike and the relation is severed, so on what grounds was this cloth woven from this bundle of threads? Qingmu concludes: “Since both positions fail, there is neither form nor a cause of form.”

In daily life, there is nothing wrong with saying, “Cloth is woven from threads.” The mistake is to assign that event of “weaving” to a fixed thing that must bear it, then ask whether that thing is ultimately like or unlike another.

The Other Four Aggregates Are to Be Seen in the Same Way

Once the gates examining the form aggregate have been traversed, Nāgārjuna passes over the others in a single verse:

“The aggregate of feeling and the aggregate of perception, formations, consciousness, and the rest— all other dharmas are the same as the aggregate of form.”

Qingmu likewise adds only one sentence: “The other four aggregates and all dharmas should likewise be contemplated and refuted in this way.”

The sentence seems light, but in fact it returns the work to the reader’s own hands. It does not say, “Feeling, perception, formations, and consciousness are also empty; simply remember that.” It says they “should likewise be contemplated”: walk through the preceding gates yourself—apart from cause, apart from fruit, fruit present or absent, resembling the cause or not resembling it—and try each gate with feeling, then with perception.

A conclusion can be heard from someone else; no one else can traverse those gates for you.

The Final Two Verses: The Treatise Author Teaches How to Ask and Answer

At this point the treatise has effectively finished. But Qingmu says, “Now the treatise author, wishing to praise the meaning of emptiness, speaks further in verse,” and adds two verses:

“If someone asks a question and one tries to answer apart from emptiness, it does not constitute an answer but remains caught in the very same doubt. If someone poses a challenge and one states another’s fault apart from emptiness, it does not constitute a challenge but remains caught in the very same doubt.”

These two verses do not discuss the five aggregates. They discuss question and answer themselves. It is as though the treatise author, on completing his work, turns back and says: you have just watched me question and answer all the way through; there is a rule within it.

Qingmu’s example could not be more ordinary. Someone says, “A jar is impermanent.” Another asks, “Why is it impermanent?” He answers, “Because it arises from an impermanent cause.”

Qingmu says this does not count as an answer. Why? “Within the cause and condition there is still doubt: it is not known whether it is permanent or impermanent.” The “impermanent cause” offered as an answer has not itself been settled as permanent or impermanent. You have only moved the doubt back one step; you have not dissolved it.

The challenging side is the same. “You use impermanence to refute my permanence; I too can use permanence to refute your impermanence.” You strike my grasped permanence with your grasped impermanence; I strike back with permanence. The positions keep changing hands, and neither person moves the other. Moreover, if there truly were a real impermanence, Qingmu says it would incur three faults: “If impermanence were real, there would be no karmic retribution; the eyes, ears, and other dharmas would perish moment after moment; and there would be no discrimination.” Karma could not connect with its fruit, the eyes and other faculties would break off at every moment, and even discrimination would have nowhere to be established.

Then does refuting by reliance on emptiness not pose the same problem—have you not made “emptiness” your foothold?

Here the respondent speaks the chapter’s most crucial sentence:

“If one relies on emptiness to refute permanence, there is no fault. Why? Because this person does not grasp the mark of emptiness.”

The difference is not which weapon has been picked up, but whether the hand closes around it. Permanence and impermanence are both grasped things, so they can strike at each other forever. Emptiness can refute without being refuted in return because the person who speaks emptiness does not clench emptiness into yet another thing.

The final sentence carries the matter to its end:

“Thus, if one wishes to conduct questions and answers, one should rely on the dharma of emptiness—how much more one who wishes to seek the mark of freedom from suffering and quiescent cessation?”

Even an everyday debate must rely on this if it is not to become an exchange of stones. What, then, of someone truly seeking freedom from suffering and seeking cessation?

A chapter that begins by asking about the five aggregates ends by teaching one how to speak. This ending does not stray from the subject: the five aggregates are established as truly existent precisely through one question and answer after another.

What Can a Reader Carry Away from This Chapter?

First, remember the two directions. “Remove the threads and there is no cloth; remove the cloth and there are no threads.” We habitually trace downward toward causes but rarely turn around to see that what is called a cause receives its identity from the fruit. Cause and fruit do not first stand separately and then become connected; the two names arise together.

Second, notice whether you began arguing only after you had already stuck to something. Qingmu gives the sequence clearly: ignorance and the defilement of craving come first; discriminative proliferation follows. Beneath a fiercely contested argument there is often something the mind cannot bear to release.

Third, learn to distinguish an answer from a displacement. An answer such as “because it arises from an impermanent cause” sounds complete, but merely pushes the unsettled matter back one step. Ask yourself: has the thing I am using as my answer itself been settled?

Fourth, emptiness is for releasing the hand, not closing it. “Because this person does not grasp the mark of emptiness”—these words are the most important sentence at the chapter’s close.

Carried Forward: Jizang’s Mirror

The preceding is the meaning of the chapter itself. What follows lets the Sui–Tang Three Treatises tradition illuminate it from the side. The mirror is not the treatise; it offers another way to read the same verses.

Jizang’s Commentary on the Treatise on the Middle divides the chapter’s nine verses into two parts: the first seven seek the ultimate emptiness of the five aggregates; the final two praise ultimate emptiness. He further places the six verses examining form under three gates: first, cause and fruit cannot be separated; second, cause and no cause; and last, resemblance and non-resemblance. This classification returns the long sequence of questions and answers in the chapter’s middle to their places and makes its shape easier to see.

On the line “a person of wisdom should not discriminate form,” he connects what is discriminated back to the preceding claims—cause and fruit separate or not separate, caused or uncaused, and so forth—and rebukes them together. He explains the section on ignorance and the defilement of craving as craving arising from delusion, which is affliction; later, using wrong views to discriminate existence and nonexistence in the three times is a new affliction arising on top of the old delusion. This layered interpretation renders Qingmu’s threefold sequence in still finer detail.

As for whether relying on emptiness becomes another dependence, Jizang frames and answers the question: if one says that emptiness is relied upon to refute permanence, does this not posit an emptiness on which to rely, leaving one vulnerable to refutation by the non-Buddhists? His answer is that one does not grasp the mark of emptiness or any of the four alternatives; the mind has nothing on which it relies. He also says that even worldly questions and answers must rely on emptiness—how much more should someone seeking the highest path not retain an existent thing?

This mirror joins the chapter to the Three Treatises tradition’s framework of doctrinal classification and disclosing the right by releasing views. Its divisions and layered judgments are Jizang’s additions, not the words of Nāgārjuna’s verses or Qingmu’s long commentary. The readings are recorded side by side without adjudication, each returned to its own place.

Two other alternative commentaries also transmit this chapter. At the opening verse, the Bhāviveka-line Prajñā Lamp Commentary identifies the four great elements—earth and the rest—as the cause of form, saying that within the ultimate meaning form cannot be found apart from them, while conventionally form is provisionally established in reliance on the four elements as causes. At the chapter’s beginning, the Sthiramati-line Mahāyāna Commentary on the Middle Contemplation first distinguishes levels: another school establishes the twelve sense fields as truly existent in the ultimate because they are included within the aggregates; the commentator grants their inclusion in conventional truth but not in the ultimate. At the close, it says that because nothing departs from emptiness, everything is accomplished, and conventional acts and attainments do not fall into annihilation. The wording transmitted by the three lineages differs; on the mutual dependence of the names cause and fruit, their meaning converges.

We recognize things by their marks. What is hard is earth, what is wet water, what is warm fire, and what moves wind. After recognizing them this way for long enough, a principle naturally forms in the mind: since each has its own mark, each must be its own thing.

The questioner at the opening of this chapter says precisely this:

“The six elements each have a fixed mark; because they have fixed marks, the six elements exist.”

Earth, water, fire, wind, space, and consciousness each possess a determinate mark; since the marks are determinate, the six things must naturally be truly existent. This question does not come from scripture but from daily life. Its force lies exactly there—it flows so smoothly that we do not notice it is a thesis.

The respondent does not begin with earth, water, fire, or wind. He begins with space. And his first sentence does not ask whether space exists; it asks: When did its mark begin to exist?

Which Exists First, the Mark or the Dharma?

The verse says:

“Before the mark of space exists, no dharma of space exists; if space already existed beforehand, it would be without a mark.”

The two lines form a pair of tongs.

First consider the opening line. If you wish to use its mark to prove that space exists, then before the mark of space exists, the dharma called space cannot yet be spoken of.

Now consider the second. Yet you also say that space was there first—it must be there before the mark has something to which it can attach. Very well: during the interval before the mark arrives, space is a thing without a mark.

What was brought in as proof turns back and leaves what it was meant to prove suspended. If the mark comes first, the dharma does not yet exist; if the dharma comes first, there is an interval in which no mark points to it. The questioner wanted mark and dharma each to stand securely and then support one another. The verse shows that both ends cannot be there first.

How the Mark of Space Arises: It Too Depends on Something

Qingmu next explains the side of the “mark.” What is the mark of space? “A place without form is called the mark of space.”

This may look like a mere definition, but it pulls up the questioner’s foundation. Qingmu says: “Form is a fabricated and impermanent dharma. If form has not yet arisen, then, because it has not arisen, it has not ceased; at that time there is no mark of space.” Form is fabricated and impermanent. If form has not arisen, its passing away cannot yet be spoken of, and at that time there is no mark of space. Only “because form exists can there be a place without form”; only then is there a place devoid of form to which one can point, and this place is called the mark of space.

The mark of space is not an indicator that exists by itself apart from form. It can be spoken of only in relation to form.

This step is crucial. The questioner originally treated the mark as the firmest evidence—the mark is fixed, therefore the dharma is fixed. Now the mark itself depends on something else: it can be spoken of only in dependence upon form. A thing that stands only by depending on another cannot prove a thing that stands without depending on another.

Then Grant That It Has No Mark—This Path Does Not Go Far Either

The questioner retreats a step: “What fault is there if space exists without a mark?” Why not simply allow space to remain existent without a mark?

The respondent does not say, “Anything without a mark must not exist.” Instead, he returns the matter to recognition: how will you point it out?

The verse says:

“This markless dharma exists nowhere at all; within a markless dharma, a mark has nothing that it can mark.”

Qingmu fills this out concretely: “If one searches among permanent and impermanent dharmas for a markless dharma, it cannot be found.” The dharmas people discuss fall into the permanent and impermanent, and both have their marks—“arising, abiding, and ceasing are the marks of the conditioned; the absence of arising, abiding, and ceasing is the mark of the unconditioned.” The conditioned is marked by arising, abiding, and ceasing; the unconditioned by their absence. People recognize dharmas by their marks. Consequently, there is no route to a dharma wholly without a mark: “If space had no mark, there would be no space.”

Another rescue remains: perhaps it first lacks a mark and later a mark comes to indicate it. Qingmu closes this as well: “If it initially has no mark, there is no dharma that can be marked.” If a mark is to arrive, something must be there for it to mark; if no mark can initially distinguish anything, there is no ready-made object waiting there.

A name may be posited first; the thing supposedly identified cannot remain empty.

Where Can a Mark Abide? None of Three Places Works

At this point the respondent changes the point of entry. Instead of asking about the dharma, he asks about the mark itself: where does this mark abide?

“Within the marked and unmarked, a mark has nowhere to abide; apart from the marked and unmarked, it likewise abides nowhere else.”

A dharma either already has a mark or lacks one; there is no third class. Each place is examined in turn.

First, a markless dharma. Qingmu uses an ox as an example: “A fleshy hump, horns, hair at the end of the tail, and a dewlap hanging beneath the neck are called the marks of an ox.” These together are called the marks of an ox. Then he presses in both directions: “Apart from these marks, there is no ox.” “If there is no ox, these marks have nowhere to abide.” Apart from the marks, no ox can be indicated; without an ox, the marks have nowhere to abide. Ox and ox-marks do not first exist separately and then claim one another.

Second, a dharma that already has a mark. “Within the marked, a mark likewise does not abide, because a mark is already present.” Since it already has its mark, another cannot enter. Qingmu uses water and fire as an analogy: “The mark of fire does not abide within the mark of water, because water already has its own mark.” Water already bears the mark of water; the mark of fire cannot enter it.

Third, suppose a mark could enter a markless dharma. Qingmu says it would then “be uncaused; uncaused means that a mark exists although no dharma exists.” Without a cause, a mark appears although the dharma was originally absent. Then he speaks an important sentence: “Mark, marked, and permanent mark are established through dependence.” Mark, marked, and even what is called a permanent mark all stand through dependence.

Once the three places have been examined, Qingmu concludes: “Apart from dharmas with a mark and dharmas without a mark, there is no third place that can be marked.”

This section is not the same sentence repeated three times. It seals three different escape routes: the object is absent, the position is occupied, and the mark appears without a cause. Once all three are closed, the thought that “a mark carries its own power of indication and later attaches to some already established dharma” has nowhere to land.

Mark and Marked: They Stand Together and Fail Together

The verse then carries the point through:

“Because there is no dharma of marking, there is no dharma that is marked; because there is no dharma that is marked, there is likewise no dharma of marking.”

Qingmu explains with utmost brevity: “The marked exists because of a mark, and a mark exists because of the marked; the two depend upon one another.”

We need to see clearly what “depend upon one another” means here. It does not mean that two ready-made things help each other stand—in that case each would still have its own body and merely lean on the other. It says: this pair of terms can be spoken only as they face one another. What marks is called a mark because there is something marked; what is marked is so called because there is a mark. Remove one end, and even the other’s name cannot be established.

The final verse gathers the entire mechanism through which the six elements were inferred:

“Thus, there is now no mark, nor anything that is marked; apart from mark and marked, there is no further thing.”

Qingmu says: “When causes and conditions are examined from beginning to end, mark and marked are definitively unfindable.” Yet all dharmas discussed by people “are included within the two dharmas, mark and marked.” He adds the most loosening sentence: “Sometimes the mark functions as the marked, and sometimes the marked as the mark. Fire has smoke as its mark, while smoke in turn has fire as its mark.”

Between fire and smoke, which is the mark and which the marked? It depends on which direction you look. See smoke from afar, and smoke is the mark of fire; see fire close at hand, and fire can also be the mark of smoke. The roles within the same pair can reverse. Their reversibility shows precisely that the two identities are not intrinsic natures carried separately within each thing but roles assigned within a relation.

Refuting Existence Does Not Require Positing a Nonexistence

The questioner now takes a road many people would take: “If there is no existence, there should be nonexistence.” If existence cannot stand, surely there must at least be a “nonexistence.”

The verse answers:

“If existence itself does not exist, how could nonexistence exist? since existence and nonexistence are both absent, who knows existence and nonexistence?”

Qingmu is concrete: “When a thing destroys itself or is destroyed by another, it is called nonexistent.” When something breaks down of itself or is destroyed by another, people call it absent. Thus, “Nonexistence does not exist on its own but comes from existence.” The word nonexistence is borrowed from existence; once existence itself has not been established, the nonexistence named through it has nothing on which to depend. He adds a counterquestion: “Even what the eye sees and the ear hears cannot be found, how much less a nonexistent thing.” Even visibly seen and audibly heard things are unfindable under examination, much less a thing that is absent from the outset.

The questioner’s final move is to preserve beyond existence and nonexistence someone who “knows existence and nonexistence.” The respondent is direct: “If there were a knower, would the knower be within existence or within nonexistence? Since existence and nonexistence have been refuted, the knower is refuted together with them.” The knower you establish must fall on the existent side or the nonexistent side; since neither can be found, the knower too has nowhere to stand.

Thus, the summarizing verse:

“Therefore, know that space is neither existent nor nonexistent, neither marking mark nor marked; the other five are like space.”

Qingmu passes over the rest in one sentence: “Just as the mark of space cannot be found however it is sought, so too with the other five elements.”

Why Begin Precisely with Space?

Someone asks well: “Space is neither first nor last. Why refute it first?” Space comes neither first nor last in the list of six elements, so why begin there?

Qingmu gives two reasons.

The first concerns difficulty. “Earth, water, fire, and wind conjoin through many conditions and are therefore easy to refute.” That the four great elements are brought together through many conditions is easy to see. “Because consciousness takes suffering and pleasure as causes, it can be known to be impermanent and changing and is therefore easy to refute.” It is likewise easy to feel consciousness changing with suffering and pleasure. Space is different: “Space has no such mark; ordinary people merely hope that it exists, and so it is refuted first.” It shows neither an obvious mark of conjunction nor change with suffering and pleasure; it is said to exist through a hope in the ordinary mind. The one whose flaw is hardest to see is exactly the one to examine first.

The second concerns sequence. “Moreover, space can support the four great elements, and consciousness exists through the four elements as causes and conditions; therefore, refute the root first, and the others are refuted of themselves.” One account makes space the support of the four great elements and makes those four conditions for consciousness. Following that path, examine first what is taken as the root, and the rest naturally loosen with it.

Together, these reasons show that space is refuted first for argumentative convenience, not because any one of the six elements is exceptionally important or unimportant.

The Final Question: Does This Not Put You at Odds with the World?

The final question carries the greatest force: “Everyone in the world sees dharmas as existent or nonexistent. Why do you alone oppose the world and say that nothing is seen?” Everyone plainly sees dharmas as existent or nonexistent. Why do you alone contradict the world and claim that nothing is seen?

The answering verse says only one thing, but it is where the chapter comes to rest:

“Those of shallow wisdom see dharmas as marked by existence or nonexistence; they cannot see the secure dharma in which views are brought to an end.”

Notice that the verse does not say “everything worldly people see is false.” It says that those of shallow wisdom, on seeing dharmas, see only the two marks of existence or nonexistence; stopping upon these two marks, they cannot see the secure dharma in which all views are ended.

Qingmu speaks with care: “If a person has not attained the path and does not see the true mark of dharmas, craving and views condition all kinds of proliferation.” The root of the difficulty is not seeing but craving and views—what is seen must be grasped. Thus, “Seeing a dharma arise, one calls it existent, grasps the mark, and asserts existence; seeing a dharma cease, one calls it annihilated, grasps the mark, and asserts nonexistence.” Arising and ceasing are things before one’s eyes; the error lies in the instant of “grasping the mark.” See a thing appear and at once press it into existence; see it disappear and at once press it into annihilation.

The wise are different. Qingmu says: “Seeing dharmas arise, the wise at once end the view of nonexistence; seeing dharmas cease, they at once end the view of existence.” Notice the ingenuity of the two sentences: the wise use the very same arising and ceasing. On seeing a dharma arise, they immediately know that all cannot be said to be determinately nonexistent; on seeing it cease, they immediately know that all cannot be said to be truly existent. The very same event nourishes two views in shallow wisdom and dissolves two views in wisdom.

Thus, the respondent is not disputing with the world over whether anything is “seen.” What is seen need not be cancelled; the grasping of its mark can be released.

At the close, Qingmu turns the principle upon itself: “Thus, although all dharmas are seen, they are like illusions and dreams. Even the view of the uncontaminated path is to be ended, how much more other views.” Even the view of the uncontaminated path is to be ended, much more the rest. This is the chapter’s softest and most thorough point—the view used to refute views is not retained as a final attainment.

The concluding sentence says: “Thus, one who does not see the secure dharma in which views are ended sees existence and sees nonexistence.” Security is not another thing placed outside existence and nonexistence; it is where the views of existence and nonexistence have both ceased.

What Can a Reader Carry Away from This Chapter?

First, notice what the word “mark” depends upon. “A place without form is called the mark of space.” Even space’s mark can be spoken only in relation to form. Whenever something in daily life seems “just naturally this way,” ask: in relation to what is this mark being spoken?

Second, refuting one side does not require filling the gap with the other. “Nonexistence does not exist on its own but comes from existence.” After releasing existence, one need not turn around and seek reassurance in a nonexistence; that nonexistence still borrows its name from existence.

Third, use arising and ceasing without grasping either. “Seeing dharmas arise, the wise at once end the view of nonexistence; seeing dharmas cease, they at once end the view of existence.” Arising and ceasing before one’s eyes are seen exactly as before; one simply does not press them, at that instant, into existence or annihilation.

Carried Forward: Jizang’s Mirror

The preceding is the meaning of the chapter itself. What follows lets the Sui–Tang Three Treatises tradition illuminate it from the side. The mirror is not the treatise; it offers another way to read the same verses.

Jizang’s Commentary on the Treatise on the Middle says that the Buddha, adapting to deluded beings, gives various teachings, sometimes naming the five aggregates and sometimes the six elements, while intending to refute the self. People stuck to the teaching do not understand no-self and instead cling only to dharmas; the preceding chapter refuted their confusion over the five aggregates, and this chapter removes their delusion over the six elements. He also reads the chapter sequence as a transition of gates: the preceding chapter clarified the gate of emptiness, and this one next teaches the gate of the signless. If one awakens through the gate of emptiness, there is no need for the signless; but because someone may grasp the mark of emptiness and thereby turn it into no gate at all, this chapter next refutes the mark of emptiness and lets one enter through the gate of the signless.

He also divides the eight answering verses into two sections: the first seven refute marks and clarify the gate of the signless; the final verse is a gate of admonition and refutation. Among the first seven, the first six refute space and the next extends the examination to the other five elements. The refutation of space further divides into five and a half verses refuting the dharma of space and half a verse refuting the person who knows space. The five verses refuting existence divide into four: the first verse and a half use what marks to refute what is marked; the next verse and a half use what is marked to refute what marks; the third verse refutes through the interdependence of mark and marked; and the fourth concludes. For the passage on the wise seeing arising and ceasing, he pairs arising through causes and conditions with ceasing when conditions disperse: since dharmas are seen to arise through causes and conditions, how could they be annihilated nonexistence; since they cease when conditions part, how could they be determinately existent?

This mirror supplies a classification for the chapter, names its gates, and joins its sequence to the Three Treatises tradition’s interpretive transition from the gate of emptiness to the gate of the signless. The classification and names of gates are Jizang’s additions, not the words of Nāgārjuna’s verses or Qingmu’s long commentary. The readings are recorded side by side without adjudication, each returned to its own place.

By the time one reaches this chapter of the Treatise on the Middle, the questioner comes holding a sentence from the sūtra.

He speaks with confidence: the sūtra teaches that craving, anger, and delusion are the root of the world. Moreover, craving has more than one name—“first it is called love, next attachment, next defilement, next sexual desire, and next covetous desire.” As the names unfold one after another, the terminology is complete and its levels distinct. When the names are this numerous and precise, could what they indicate really be empty? He therefore installs two positions in passing: “Resting upon sentient beings, the being is called the defiled, and craving is called the defiling dharma.” A person stands there to be defiled, and a dharma stands there to defile. Once both ends are established, craving naturally exists. The inference then continues: the three poisons produce three kinds of action; those actions produce the three realms; “therefore, all dharmas exist.”

One sūtra sentence and a few inferential steps, and the entire world has been erected.

The respondent gives only one sentence, but it is the gate of the entire chapter:

“Although the sūtra teaches the names of the three poisons, their reality cannot be found.”

See clearly the distinction this sentence makes. He does not say the sūtra is wrong, nor that there is no such thing as craving. He first accepts the names—“although the sūtra teaches the names of the three poisons”—and only then asks: if we follow these names to seek the real thing, can it be found?

A name’s presence does not mean that something standing by itself exists beneath it. The ten verses that follow all do the same thing: taking the two names defiling dharma and defiled person, they pass through one gate after another seeking the reality those names indicate.

The Gate of Sequence: Who Is There First?

The first verse begins with before and after:

“If, apart from the defiling dharma, a defiled person already existed, because this person was already in desire, defilement would have to arise again.”

Suppose a defiled person stands there first and the defiling dharma arrives afterward. Qingmu topples the supposition in one sentence: “If it is already determined that a defiled person exists, no further defilement is needed, because that person has already been defiled.”

The step is utterly plain. Calling someone “the defiled” means the person is already defiled; what is left for a defiling dharma to do upon arrival? What can it add? If the person still lacks a measure of defilement before qualifying as defiled, then the earlier name “the defiled” cannot be attached.

What if the order is reversed? The verse continues: “If there were no defiled person, how could there be defilement?” Qingmu says: “If it is already determined that there is no defiled person, defilement likewise should not arise. A defiled person must first exist, and only then can defilement arise. If there is initially no defiled person, there is no one to receive defilement.”

Prior absence of the defiled does not work either—upon whom could defilement operate? With no one to receive it, defilement has nowhere to land.

At one end, a person already defiled has no need to be defiled again; at the other, no person exists to be defiled. Neither end of this path goes through.

The verse then turns the same measure upon the other side: “Whether defilement exists or does not exist, the defiled is the same.” Qingmu’s analogy is vivid: “If a defiling dharma determinately existed before and apart from a person, it would be without a cause. How could it arise? It would be like fire without fuel.”

A defiling dharma existing prior to and apart from a person is like a fire without fuel. What would the fire burn? Upon what would craving crave?

Then Say They Are Simultaneous—But Simultaneity Removes Dependence

The questioner changes his footing: since sequence fails, let defiling dharma and defiled person arise at the same time. He asks directly: “What fault is there if they arise simultaneously?”

The move looks clever. Arising together, neither owes the other priority.

But the verse points out what has been lost:

“That the defiled person and the defiling dharma are established together cannot stand; if the defiled and the defiling dharma exist together, there is no dependence.”

Two things established simultaneously no longer depend upon one another. Qingmu makes this explicit: “The defiling dharma does not exist because of the defiled. The defiled does not exist because of the defiling dharma. These two should be permanent, because each has been established without a cause.”

The questioner originally wanted the fact of craving’s existence; if he truly secures simultaneous establishment, what he secures instead are two things that exist without causes. What exists without a cause is permanent. Qingmu states the cost of permanence directly: “If permanent, many faults follow; there is no dharma of liberation.”

This sentence should be remembered. Examining the fixed reality of defilement and the defiled is not an effort to win an argument; it preserves the possibility of liberation. If defilement truly were a permanent thing, it could never be washed away.

Identity and Difference: How Can One Dharma Conjoin with Itself?

Having questioned sequence and simultaneity, the respondent establishes another gate: “Next, defilement and the defiled should be refuted through identity and difference.”

“If the defiled and the defiling dharma are one, how could one dharma conjoin? if the defiled and the defiling dharma are different, how could different dharmas conjoin?”

First consider identity. Qingmu gives an example in only a few words, yet anyone understands it: “How could one dharma conjoin with itself? A fingertip cannot touch itself.”

A fingertip cannot touch itself. Conjunction needs at least two ends; with only one end, conjunction cannot be spoken of. Since craving is said to occur when defiled person and defiling dharma conjoin, they cannot at the same time be said to be one.

Next consider difference. Difference fails too, not because the two cannot touch, but because no touch is needed: “Because they are established as different. If each is already fully established, no further conjunction is needed; even if conjoined, they remain different.”

Two things already fully established in themselves gain nothing by conjoining or not conjoining. Even placed together, they remain two.

The respondent presses a level further:

“If what is one can conjoin, conjunction should exist apart from a companion; if what is different can conjoin, conjunction should likewise exist apart from a companion.”

If one can still be called conjoined, it should conjoin apart from a companion—one end is enough, so why wait for another? If different things can conjoin, they should likewise conjoin apart from a companion—each is independently established and originally does not await the other.

On the side of identity, Qingmu adds a more immediate point: “Moreover, if they are one, there should not be the two names defilement and the defiled. Defilement is a dharma, while the defiled is a person. If person and dharma are one, there is utter confusion.”

Defilement is a dharma; the defiled is a person. If person and dharma were one, utter confusion would follow—when craving ceased, the person would cease with it; as long as the person remained, craving could never be removed. Everyday speech distinguishes the matter clearly: this person has craving. Making the distinction already concedes that the two are not one.

On the side of difference, Qingmu’s inference is equally uncompromising: “If what is different can conjoin, even the most distant should conjoin.” If two things truly stand independently yet can conjoin, distance cannot be a reason against it; even the most distant should conjoin.

The Eye Sees Different Dharmas Conjoin—That Conjunction Was Added Later

The questioner appeals to experience: “It may be granted that one thing does not conjoin, but the eye sees different dharmas conjoining.” I concede that one cannot conjoin, but the eye sees different things come together every day.

The respondent does not deny what the eye sees. He asks what the word “different” had already been made to mean:

“If what is different can conjoin, what does conjunction do for defilement and the defiled? the marks of these two are already different, and only afterward is a mark of conjunction asserted.”

You first call the two things independently possessed of fixed, different marks, then say that they conjoin. What has conjunction done for them? Qingmu says: “The marks of these two were already different, and only afterward is conjunction forcibly asserted.” The word “forcibly” strikes the crux—conjunction is added later, but the addition does not alter the independently established identities assigned to them beforehand.

The verse asks once again:

“If defilement and the defiled have each already established a different mark, once the different marks have been established, how can they be said to conjoin?”

Qingmu makes the words still plainer for the reader: “If defilement and the defiled have each already established a separate mark, why do you now forcibly assert a mark of conjunction?”

Two Unestablished Accounts Hold Each Other Up

Here the chapter’s most interesting move appears. The respondent stops refuting one claim at a time and instead diagrams the opponent’s changing footing:

“Because the mark of difference has not been established, you wish to speak of conjunction; when the mark of conjunction is ultimately unestablished, you turn back and speak of the mark of difference.”

Qingmu is more explicit: “Because the different marks of defilement and the defiled are unestablished, you next assert a mark of conjunction. But the mark of conjunction contains faults, and defilement and the defiled are not established. In order to establish the mark of conjunction, you then assert a mark of difference again.”

When difference does not work, conjunction is moved in to rescue it; when conjunction causes trouble, difference is moved back. Neither side yet stands, but they lean upon each other for support.

Qingmu then delivers the chapter’s sharpest sentence:

“You yourself believe the matter settled, but what you say is unsettled.”

You believe you have settled it, yet the words you speak never settle. This sentence is not addressed only to the ancient questioner. When the same matter is given one account from one angle, another after the angle shifts, and then the first again—the mind beneath them that imagines it has grasped something has never come to rest.

The final gate closes in one verse:

“Because the mark of difference is unestablished, the mark of conjunction cannot be established; within which mark of difference do you intend to assert a mark of conjunction?”

Once the mark of difference fails, conjunction has nothing on which to depend. Qingmu hands the words back unchanged: “Within which mark of difference do you intend to assert a mark of conjunction?”

Conclusion: Neither Conjunction nor Nonconjunction; All Dharmas Are the Same

“Thus, defilement and the defiled are established through neither conjunction nor nonconjunction; all dharmas are the same, established through neither conjunction nor nonconjunction.”

The second half of the verse carries the chapter outward from the single matter of craving. Qingmu follows it: “As with defilement, so too with anger and delusion. As with the three poisons, so too with all afflictions and all dharmas.”

Examine anger and delusion as defilement was examined; examine all afflictions and all dharmas as the three poisons were examined. What this chapter really tests is not merely craving but the thought that “an agent and an acted-upon thing each first stand on their own, then come together.”

The crucial point is the final sentence. Do not read the preceding ten verses as saying that nothing exists:

“Neither before nor after, neither conjoined nor dispersed—they are established through causes and conditions.”

Not before, not after, not conjoined, not dispersed—and then the words “established through causes and conditions.” The four negations dismantle fixed arrangements of sequence, conjunction, and dispersion; the final words say that these things still exist, only in the mode of existing through causes and conditions.

Craving does arise. It still burns people and still must be counteracted. What is denied here is only that, behind the arising of craving, a ready-made defiled person and a ready-made defiling dharma already wait to be brought together.

What Can a Reader Carry Away from This Chapter?

First, a complete set of names does not make a ready-made entity. “Although the sūtra teaches the names of the three poisons, their reality cannot be found.” The more finely love, attachment, defilement, sexual desire, and covetous desire are distinguished, the easier it is to imagine that a corresponding object must be sitting beneath them.

Second, a person already defiled does not need to be defiled again. This is a useful measure: whenever an identity is installed first and an event is then sought to accomplish that identity, the same fault appears.

Third, examining fixed reality preserves liberation. “If permanent, many faults follow; there is no dharma of liberation.” If defilement were a permanent thing, practice would have no point at which to begin.

Fourth, notice whether you are making two sides rescue one another. “You yourself believe the matter settled, but what you say is unsettled.” This sentence is best used as a mirror for oneself, not a weapon against someone else.

Carried Forward: Jizang’s Mirror

The preceding is the meaning of the chapter itself. What follows lets the Sui–Tang Three Treatises tradition illuminate it from the side. The mirror is not the treatise; it offers another way to read the same verses.

Jizang’s Commentary on the Treatise on the Middle reads the five names love, attachment, defilement, sexual desire, and covetous desire as a sequence from shallow to deep: the initial arising of thought is called love; the mind’s continuing fixation, attachment; deep and persistent entanglement, defilement; the stirring of a frantic mind, sexual desire; and the expedient effort to take something as one’s own, covetous desire. He says the first three are still light and extend through the three kinds of existence, while the latter two are heavy and belong only to the desire realm. He also divides the chapter’s ten verses into six sections: refutation through before and after; refutation through simultaneity; refutation through identity and difference; focused refutation through difference; admonition; and analogical refutation of other dharmas. He judges these six gates to be a contemplation that cuts off afflictions: seeking affliction through six gates and finding none is right contemplation, called the uninterrupted path; the contemplation arising in the next moment is called the path of liberation. For the final verse, he says the first half recapitulates the preceding vertical refutation of five claims, while the second extends the analogy horizontally to all dharmas. He especially warns that, once conjunction has been refuted, someone might inversely grasp nonconjunction as establishing person and dharma; therefore nonconjunction too is refuted. The first two verses concern person and dharma before and after and so belong to the gate of nonconjunction; from simultaneity through the chapter’s end, the gate of conjunction is refuted.

This mirror joins the chapter to the Three Treatises tradition’s interpretive framework of refuting the deviant and disclosing the right. The sequence assigned to the names and the sixfold classification are Jizang’s additions, not the original words of Nāgārjuna’s verses or Qingmu’s long commentary. The readings are recorded side by side without adjudication, each returned to its own place.

The person at the opening of this chapter says something to which everyone would nod.

He says: the sūtra plainly teaches that conditioned dharmas have three marks—arising, abiding, and ceasing. All things “arise through the dharma of arising, abide through the dharma of abiding, and cease through the dharma of ceasing”; therefore, the many dharmas truly exist. He has not imagined an entity from nothing. He holds the facts before his eyes: the flower opens, remains in bloom, and withers. Everyone can see the three stages. From them he infers backward: since the three operations of arising, abiding, and ceasing perform their work upon all things, those things exist.

The respondent does not deny that flowers bloom and wither. He replies with only one sentence: “The three marks have no fixed determination.”

He then asks a question from which the chapter’s more than thirty verses all grow: you say the three marks can mark conditioned dharmas—then to what class do the three marks themselves belong?

The First Cut: Is What Marks Conditioned or Unconditioned?

There are only two options.

If arising is a conditioned dharma, then by the questioner’s own rule every conditioned dharma must have three marks; arising itself must therefore possess arising, abiding, and ceasing. Qingmu does not begin here with infinite regress. He first addresses something closer: the three cannot crowd into the same place. “Contradiction means that the mark of arising should attend the dharma that is arising, the mark of abiding the dharma that is abiding, and the mark of ceasing the dharma that is ceasing.” While one dharma is arising, how could the contradictory marks of abiding and ceasing be present at the same time? His analogy is exceedingly spare: “As light and darkness do not coexist.”

If arising is called unconditioned, the path is even less open. “Because the unconditioned has no nature.” The name unconditioned is itself only “given because the conditioned has ceased”; it is attached after the conditioned comes to rest. Apart from not arising and not ceasing, “it has no further mark of its own.” How could something without a mark of its own mark something else? Qingmu uses two ready-made absences: “A rabbit’s horn and a turtle’s hair cannot mark a dharma.”

What is brought to prove another must first stand itself. After this first cut, the three marks have already lost the ground beneath them. Every rescue that follows attempts to restore this ground.

Neither Gathered nor Dispersed

The question retreats a step: do the three marks each mark separately, or do they gather and mark together?

If they mark separately, then “at the time of arising there is no abiding or ceasing; at the time of abiding no arising or ceasing; at the time of ceasing no arising or abiding.” At the same locus, marks are sometimes present and sometimes absent, so the three are not universal marks of the conditioned. What if they gather and mark together? “How could mutually contradictory dharmas all be present at one time?” Three operations that conflict with one another cannot crowd simultaneously into one dharma.

Someone then adds another layer for them: above the three marks, supply another set of three marks. The respondent merely follows this road to its end: “This becomes endless; without further marks, they are not conditioned.” Add one layer, and another must be added forever. Add none, and the original marks are not conditioned dharmas and are unqualified to mark the conditioned.

Set another brick beneath something that cannot stand, and the brick itself needs support.

Seven Dharmas Arising Together: Two Things Not Yet Standing Support Each Other

The questioner now offers an intricate mechanism. When one dharma arises, seven dharmas arise together, including its own body: the dharma, arising, abiding, ceasing, secondary arising, secondary abiding, and secondary ceasing. Primary arising, “apart from itself, can produce the other six dharmas,” while secondary arising in turn produces primary arising. In this way, each pair attends to the other and the chain closes at the second layer without becoming infinite.

This move is genuinely elegant. The respondent does not take a detour, but asks one question: where does secondary arising come from? From primary arising. If so—“Secondary arising comes from primary arising; how can it produce primary arising?”

Turn the question around toward primary arising. Primary arising comes from secondary arising: “Primary arising comes from that; how can it produce secondary arising?” Qingmu states the crux plainly: “Secondary arising does not yet possess its own body—how can it produce primary arising?”

The questioner finally relocates the action to the present instant: regardless of before or after, in the very moment secondary arising is arising, it can produce primary arising. The answer remains the same: “Secondary arising does not yet exist—how can it produce primary arising?”

Two people can support one another only if at least one is already standing. If both are still falling, mutual support is not support; it is falling together. This passage lays bare the chapter’s structure: the questioner believes he has found a closed circle, but has merely multiplied “not yet” by two.

The Lamp Analogy: A Lamp Cannot Illuminate Darkness on Its Own Body

The questioner changes to an image familiar to everyone: “A lamp entering a dark room illuminates all objects and can also illuminate itself.” A lamp illuminates another and itself; arising produces another and itself.

The respondent’s reply is the chapter’s finest move. He does not begin with the logic of “self” and “other,” but with the word “illuminate”:

“There is no darkness within the lamp, nor darkness where it abides; only dispelling darkness is called illumination; where there is no darkness, there is no illumination.”

What is illumination? Dispelling darkness is called illumination. Then look: is there darkness in the body of the lamp? No. Is there darkness where the light is present? No. Since neither place contains darkness to begin with, the lamp dispels nothing there—where there is no darkness to dispel, the word “illumination” has no application. The lamp only appears to illuminate itself because we have secretly treated “brightness” as “illumination.”

The questioner retreats once more: the lamp does not illuminate before it arises or after it has arisen, but while it is arising. The respondent says: “A lamp at the time of arising is called half-arisen and half-not-yet-arisen.” The lamp’s own body is not yet established; with what could it dispel darkness? Moreover, it never reaches darkness—“Only when a person catches a thief is it called defeating the thief.” To catch a thief, one must reach the person; shouting from another street does not defeat the thief.

If a lamp could dispel darkness without reaching it, then lighting one lamp here “would dispel all darkness.” It is equally unable to reach darkness everywhere in the world.

The final blow returns symmetry to the opponent: if a lamp illuminates itself and another because it opposes darkness, darkness also opposes the lamp and should therefore “darken itself and darken another.” In fact, no mass of darkness makes itself dark. Since that side fails, the other fails as well. “Therefore, the lamp analogy fails.”

Arising Sought in Three Places Cannot Be Found in Any

Once the analogy has been removed, the respondent examines arising itself. Does arising arise within what has arisen, or within what has not yet arisen?

What has not arisen is no dharma; with what could a non-dharma produce itself? What has arisen is already established: “Once it has arisen, what use is further arising?”—“What has already been done should not be done again.” The question is simple but closes both ends. Together with the gate of the time of arising, already answered in “Examination of Going and Coming,” there is no arising in any of the three places.

For the claim that what has not arisen can arise, Qingmu gives three piercing examples. If a not-yet-arisen dharma could arise, then “all ordinary people in whom bodhi has not arisen should now give rise to bodhi”; “arhats who abide in the indestructible dharma and have no afflictions should now produce afflictions”; and “rabbits and the rest, which have no horns, should now produce horns.” If “not yet present” were enough to produce “present,” practice and the severing of afflictions would become accidents, and horns should grow upon rabbits. If a not-yet-arisen dharma could produce itself, causes and conditions would have no meaning.

The questioner therefore places all the weight upon conditions: when condition, act, agent, time, and place are all complete, the not-yet-arisen dharma arises. The respondent points out that this still must fall into one of three possibilities—prior existence, prior nonexistence, or both—and all three have already been refuted. Moreover, saying arising occurs at the time of arising incurs two arisings: one because “time of arising” receives its name from arising, and another placed inside the time of arising. “There are not two dharmas; how could there be two arisings?”

Conditions Are Not a Refuge; Conditions Are Precisely the Reason for Emptiness

The questioner finally changes his words: I specify none of the three gates; I say only that arising exists because many conditions conjoin.

The respondent does not refute conjunction yet again. He turns the conclusion of conjunction around and places it before the questioner:

“If a dharma arises from many conditions, its nature is quiescent cessation; therefore, arising and the time of arising are both quiescent cessation.”

This is the chapter’s pivot. You bring in many conditions to rescue arising, but to arise from many conditions is precisely to lack intrinsic nature. Qingmu explains quiescent cessation with great clarity: “Quiescent cessation means neither this nor that, without marks; the path of speech is cut off and all proliferations cease.” He then uses woven things: “Cloth exists because of threads, and a mat because of rushes.” If threads had a fixed mark of their own, they should not come from hemp; if cloth had a fixed mark, it should not come from threads. Yet one layer actually depends on another, so none has a fixed nature. Fire and fuel are the same: when one end is absent, the other is absent with it. “Because they lack intrinsic nature, they are empty, like a mirage without reality.”

At this point one thing must be seen clearly: the respondent has never removed the arising and ceasing before one’s eyes, only the determinate body added to them. The flower truly blooms; what cannot be found is a thing called “arising” standing there to execute the blooming.

Abiding: Aging and Death Follow Every Dharma

Once arising has been refuted, the questioner turns to abiding: even if a dharma does not arise, surely it abides.

The respondent does not begin with inference but observation: “All dharmas whatsoever bear the marks of aging and death; one never sees any dharma abide apart from aging and death.” Qingmu adds that impermanence follows from the instant of arising, and “impermanence has two forms, called aging and death.” What we call abiding is only an interval in which change proceeds too slowly for us to see.

Next, abiding is examined through self and other. If abiding abides through its own mark, it becomes permanent and therefore is not a conditioned dharma arising through many conditions. Moreover, self-action upon self has already failed—“The eye cannot see itself; abiding is the same.” If abiding occurs through another mark, that other must rely on yet another, producing an infinite regress.

The eye sees everything but itself. This is the same point running from the lamp analogy through the chapter: what acts cannot act upon itself.

Ceasing: Milk Does Not Cease While It Is Milk; a Second Head Cannot Be Cut Off

Ceasing is likewise sought in three places: what has ceased does not cease again, what has not ceased is apart from the mark of ceasing, and apart from these two there is no distinct time of ceasing.

Two analogies deserve to be remembered. The first is milk: “Milk does not cease while it is milk.” While it remains milk, the mark of milk stands there intact and is not ceasing; once it is not milk, one cannot say that milk is ceasing. No exact instant of decay can be indicated. The second is a second head: if the dharma does not exist, no ceasing can be spoken of—“Like a second head: because it does not exist, it cannot be cut off.” What never existed cannot be severed.

That ceasing occurs neither from self nor from other has the same reason as arising: “A fingertip cannot touch itself.” As for another, that other likewise lacks a body of its own—“Because its own body is absent, the other is absent too.”

Once the Conditioned Fails, the Unconditioned Has No Support

When the three marks fail, conditioned dharmas fail. What, then, of the unconditioned? Qingmu speaks honestly about where the name comes from: “Because the marks of the conditioned cease, it is called the mark of the unconditioned. The unconditioned has no distinct mark of its own; it is through these three marks that the mark of the unconditioned exists.” Fire has heat, earth hardness, and water cold, each as its own mark; “the unconditioned is not so.” It is a name established only against the three marks on the other side.

Something established purely through contrast is removed when the contrast is removed. Thus, at the end of the chapter’s refutation, no secure unconditioned remains for one to inhabit.

Illusion, Dream, and Gandharva City: What, Then, Can Be Said?

The questioner finally asks an honest question: if arising, abiding, and ceasing ultimately do not exist, why does the treatise keep using their names?

The respondent closes with three analogies: “Like an illusion, like a dream, like a gandharva city.”

Qingmu’s explanation should be read slowly. The marks of arising, abiding, and ceasing have no fixed determination; ordinary people, through craving and attachment, call them fixed. Yet “the noble ones, moved by compassion and wishing to stop their inversion, teach them using the very names to which they cling. Although the words are the same, the minds are different.” The teaching uses the listener’s words, not the speaker’s view. With the same word “arising,” an ordinary person indicates a real event, while a noble one uses it to stop an inversion. The word has not changed; the mind has. To take these names as an objection therefore aims at the wrong place.

Under the illusion analogy, Qingmu adds the chapter’s most immediately practical sentence:

“For what is made by illusion, one should not demand its source. One should not form thoughts of sorrow or joy within it, but simply regard it as something seen by the eye.”

The first half concerns principle: one need not demand the origin of something conjured by illusion, because it never possessed an origin to trace. The latter half reaches the mind—one should not give rise to thoughts of sorrow or joy within it; one should take it simply as something the eye has seen.

This sentence is where the chapter turns from doctrine to cultivation of mind. After more than thirty verses have pressed arising, abiding, and ceasing into nonattainment, the reader can easily stop at a clever conclusion: so nothing exists after all. But that is not Qingmu’s point. His point is: since what is seen is like illusion, seeing is simply seeing. Do not add a portion of joy to it, and do not add a portion of sorrow. Sorrow and joy are the two faces of attachment. Once either arises, one has tacitly conceded that the event bears enough real weight to make the mind rise and fall.

Dream is the same: “What is seen in a dream should not be sought as real.” A gandharva city is the same: it appears at sunrise but has no body, “merely provisionally given a name and soon gone.” None of the three analogies tells one to close one’s eyes and stop seeing; they tell one to see without pursuing.

The final sentence gathers the chapter: “Ordinary people discriminate it as existent; the wise investigate it and cannot find it.” The same event appears existent when viewed through discrimination and unfindable when investigated. The object has not changed; the mode of seeing has.

What Can a Reader Carry Away from This Chapter?

First, what is refuted is fixed determination, not appearance. At the end of the search the treatise indeed says “therefore, no arising,” and “because arising, abiding, and ceasing cannot stand, there is no conditioned.” But what is absent is arising and ceasing with intrinsic nature; what is removed is their “fixed determination”—the three marks “have no fixed determination,” while “ordinary people, through craving and attachment, call them fixed.” The flower still blooms; there is simply no thing called arising behind it performing the act.

Second, what acts cannot act upon itself. There is no darkness within a lamp, the eye cannot see itself, and a fingertip cannot touch itself. Any attempt to use one thing to guarantee its own account breaks at this sentence.

Third, two things holding each other up cannot both remain ungrounded. The difficulty of primary and secondary arising appears throughout daily life: reason is used to certify confidence and confidence to certify reason; result certifies method and method result. Look to see whether either end stood first.

Fourth, seeing is simply seeing. “One should not form thoughts of sorrow or joy within it, but simply regard it as something seen by the eye.” This is the sentence the chapter leaves in the heart. There is no need to deny an object when it arrives, and no need to rise and fall with it. Only when no sorrow or joy is added have the preceding thirty-odd verses truly been read into the mind.

Carried Forward: Jizang’s Mirror

The preceding is the meaning of the chapter itself. What follows lets the Sui–Tang Three Treatises tradition illuminate it from the side. The mirror is not the treatise; it offers another way to read the same verses.

Jizang’s Commentary on the Treatise on the Middle places the chapter within the sequence of the whole treatise. From “Examination of Conditions” through “Examination of Defilement and the Defiled,” the treatise clarified that no dharma is marked; this chapter clarifies that there is nothing that marks. When both marking and marked are unfindable, the conditioned is empty; when the conditioned is empty, the unconditioned is empty with it. He also divides the chapter’s thirty-five verses in two: the first thirty-four refute the existence of three marks, and the final verse refutes their nonexistence. Together, the two sections disclose that the three marks can be determined neither as existent nor nonexistent, which he calls the Middle Way within conventional truth. As for the illusion and dream of the final verse, he reads them back across several earlier chapters: if the three marks are like illusion and dream, then the four conditions, going and coming, six faculties, five aggregates, and the rest likewise enter the ten analogies.

This mirror joins the chapter to the Three Treatises tradition’s interpretation of the treatise’s whole sequence and the Middle Way of the two truths. Its division, doctrinal labels, and retrospective application to previous chapters are Jizang’s additions, not the original words of Nāgārjuna’s verses or Qingmu’s long commentary. The readings are recorded side by side without adjudication, each returned to its own place.

A questioner begins by bringing forward the most solid-looking event before our eyes:

“Action, an agent, and the means used in acting now exist; because the three conjoin, karmic fruition exists. Therefore, agent and action should exist.”

The statement seems almost beyond dispute. Someone acts, something is acted upon, and tools and materials are at hand. When the three gather, a result appears—the carpenter, the house, the axe and saw: which is not plainly there?

The respondent’s first answer is cool: “In the preceding chapters, all dharmas have been refuted without remainder.” The chapters before this have already carried out the examination. The three marks fail, so the conditioned fails; when the conditioned fails, the unconditioned cannot be spoken of. If neither conditioned nor unconditioned stands, where can action and agent stand? Strictly speaking, there is no need to ask again.

But the respondent adds a sentence: “Because your clinging is deep, you ask again; I will now answer again.”

This deserves a pause. He does not say, “You do not understand,” but “your clinging is deep”—clinging has struck deep roots. The principle has already been explained, but an explanation’s end is not the heart’s release. Attachment to “I am doing something” does not loosen after a single refutation of marks. The answer is therefore given again. This chapter exists not because something was omitted before, but because the questioner’s hand has not opened.

First See Clearly: What Does “Determinate” Mean?

The first verse establishes the gate of examination:

“A determinately existent agent does not perform a determinate action; a determinately absent agent does not perform an indeterminate action.”

Here “determinate” does not mean making a firm decision in the everyday sense. It means that before action occurs, this end has already been established through itself and does not await the other end.

Once this is clear, the verse is not hard to understand. Qingmu says: “If agent and action are already determinately existent, there should be no acting.” If the agent is already a ready-made agent before acting and the action already a ready-made action before being performed, what use remains for the event of “acting”? Everything that needed establishment is established. Conversely, if both ends are determinately absent before acting, there is no one to act and nothing to act upon.

At one end everything is complete and action is superfluous; at the other everything is absent and action has nowhere to begin.

Fix One End First and the Other Becomes Idle

The second verse separates the matter more finely:

“A determinate action is not performed, and such an action has no agent; a determinate agent does not act, and such an agent has no action.”

Qingmu’s inference is direct: “Moreover, action should exist apart from the agent. But this is not so.”

The principle lies in the names. A person is called an “agent” because that person does something; something is called an “action” because someone performs it. The two names are originally a pair. If action is already determinately action before being done, it must be action without depending on an agent—and therefore action should exist apart from an agent. But on what grounds is an “action” never done by anyone and never awaiting anyone’s doing still called action?

The agent’s side is the same. If a person is already an agent without doing anything, “agent” becomes a designation unrelated to acting.

Qingmu then gathers in the indeterminate side as well: “Even existent agent and action cannot act, how much less absent agent and action.” The reason is only that “they are originally absent.”

When Both Ends Stand First, Conditions Lose Their Place

The third verse states the joint of the examination:

“If the agent determinately exists and the action also determinately exists, agent and action fall into causelessness.”

At first this seems surprising: the claim is that “both sides truly exist,” so how does it become causeless?

Qingmu reveals the point: “They do not exist through causes and conditions.” Because each side is complete by itself in advance, neither needs to await the other; if neither awaits the other, neither arises through causes and conditions. “Causeless” does not mean that nothing at all exists. It means that you have installed both too securely—so securely that no condition is needed.

This is the chapter’s crucial reversal: to make a thing too real does not secure cause and fruit; it sends cause and fruit away. A person looking for an absolutely reliable support for action ends by dismantling the ground on which support could stand.

After Causelessness, Everything Gives Way in Sequence

The questioner is unconvinced and asks what fault there is in causelessness. The respondent counts the consequences one by one:

“If they fall into causelessness, there is neither cause nor fruit, neither action nor agent, nor any means used in acting.”

The count continues:

“If the karmic fruits of demerit and merit are absent, nirvāṇa too is absent; every matter that could be acted upon is empty and bears no fruit.”

Qingmu first fixes the definitions: “A dharma capable of producing is called a cause, and a dharma produced is called a fruit.” Once these two names have no footing, everything after them empties in sequence: no action, no agent, no means used in acting, and therefore no demerit or merit; without demerit and merit, no corresponding karmic fruits; without karmic fruits, no path to nirvāṇa.

This sequence must be read as a whole, not by extracting one sentence. The entire chain is the consequence of the questioner’s thesis, and Qingmu concludes: “Therefore, things cannot arise without a cause.”

Notice what has given way: precisely the things that one who asserts a real agent most wishes to preserve—actions bear fruit, good and evil receive recompense, practice reaches a destination. An agent was made real in order to preserve these, yet that real agent carries them all away.

Leaving Both Ends Empty Produces Nothing Either

The questioner changes paths: then let agent and action both be indeterminate.

The respondent is brief: “If one thing is absent, action cannot arise; how much less if both are absent.” He then gives an analogy:

“Like a conjured person taking space as a house: there is only something said, with neither agent nor action.”

A conjured person lives in space as a house. The sentence can be produced, read smoothly, and understood—but from beginning to end there is neither a person nor a house. A sentence’s intelligibility does not make an act possible.

The analogy does not say that all worldly speech is useless. It addresses one specific thought: empty both ends and imagine this avoids the difficulty that prior establishment is superfluous. Once both are emptied, only a string of words remains.

Existence and Nonexistence Cannot Occupy One Place

The questioner infers in reverse: if dual absence cannot act, surely the agent and action plainly present now can act.

The respondent turns the examination toward their combinations:

“An agent, whether determinate or indeterminate, cannot perform the two kinds of action; because existence and nonexistence contradict one another, one place cannot contain both.”

Qingmu is utterly plain: “Existence is determinate, nonexistence is indeterminate; how can one person and one matter possess both existence and nonexistence?” Determinate means existent and indeterminate nonexistent. Placing both existence and nonexistence on the same person or matter forces two contradictions into one position.

Now inspect the cross-pairings:

“What exists cannot act upon what does not; what does not exist cannot act upon what exists.”

Qingmu asks: “If an agent exists but no action exists, how can there be acting?” A person exists with no thing to do; a thing exists with no person to do it. Each side lacks one end. On the side of existence he presses again: “If the action already exists, what remains for the agent to do? If the action does not exist first, how can it be acted upon?” If action exists beforehand, the agent has nothing to perform; if it does not exist beforehand, the agent has nothing to take up.

The verse then gathers all three kinds of action:

“An agent performs neither determinate action, indeterminate action, nor action both determinate and indeterminate; the faults are as previously explained.”

Qingmu says this verse “refutes all three at once.” Determinate, indeterminate, and both-determinate-and-indeterminate exhaust the three gates. Move the same division to the side of the agent, and the result is identical. Qingmu concludes: “In this way, agent and action are sought everywhere but cannot be found.”

Notice where the weight of this sentence falls. What cannot be found is an agent and action standing by themselves outside causes and conditions. It does not mean one cannot say that someone is doing something.

Then How Are Action and Agent Established?

The questioner finally asks the crucial question: “If one says there is no action and no agent, one falls again into causelessness.”

The respondent’s answer is the chapter’s affirmative side:

“This action arises from many conditions and is provisionally said to exist; it has no fixed determination.”

It arises from many conditions, so it is said to exist—an existence spoken through provisional naming. To call it existent does not claim a fixed and unchanging body. What is refuted is “determination,” not “existence.”

Then comes the chapter’s central verse:

“Because of action there is an agent; because of the agent there is action; this is how the meaning of action is established, and there is no further matter.”

The final words are precisely measured—“there is no further matter.” Apart from the names obtained through this mutual dependence, no third thing appears. There is no need to search for an underlying real agent to guarantee the pair of names.

Qingmu unfolds the verse: “Action is initially unfixed and arises because of a person.” When a person moves, action arises; once the action arises, the person who performs it is called the agent. Thus, “Through the conjunction of two matters, action and agent are established.”

What arises from conjunction has no nature of its own: “What arises from conjunction lacks intrinsic nature; because it lacks intrinsic nature, it is empty; because empty, it is unarisen.”

Finally, he clarifies two levels: “Only in conformity with ordinary people’s memories and discriminations are action and agent said to exist; within the ultimate meaning, there is neither action nor agent.”

At the conventional level, speech continues: this deed was done by this person, this action was committed by this person, and responsibility remains. At the ultimate level, no ready-made agent apart from conditions can be found, nor any ready-made action apart from the agent. The two levels do not interfere. Precisely because there is no intrinsic nature, conditions can move; if a truly unchanging agent existed, that agent could not perform anything new.

This Measure Does Not Apply Only to the Agent

The final verse extends the examination:

“As action and agent are refuted, experience and experiencer are the same; all dharmas whatsoever should likewise be refuted in this way.”

Qingmu first restates the reasoning: “Because they cannot be separated, they are not fixed; because unfixed, they lack intrinsic nature.” Since they cannot be separated, neither is independently determinate; since they are not determinate, they lack intrinsic nature.

He then tests another pair of person and dharma: “Experience means the body of the five aggregates; experiencer means the person.” Experience is this body and mind of five aggregates; experiencer is the person who undergoes it. The result is the same: “Apart from the person there are no five aggregates; apart from the five aggregates there is no person.”

This step reaches far. The preceding discussion concerned “I act”; here it concerns “I experience.” Someone may understand that agent is a provisional name and still tightly guard an experiencer—pain is experienced by me, pleasure by me; surely this “me” is real? The final verse says the same measure reaches this pair. The one who experiences and the body-mind experienced are likewise an inseparable and unfixed pair.

“All other dharmas should likewise be refuted in this way.” Wherever one can ask, “Which of these two ends was established first?” the same examination can be made.

What Can a Reader Carry Away from This Chapter?

First, the more real a matter is made, the less secure it becomes. “If the agent determinately exists and the action also determinately exists” concludes by “falling into causelessness.” To seek an absolutely reliable owner for action leaves conditions nowhere to operate.

Second, the identity agent is not received first and then enacted. “Because of action there is an agent; because of the agent there is action.” A person becomes that agent at the very time of acting. This can be applied directly to oneself: there is no need to decide first what kind of person one is and then behave according to the identity.

Third, refuting a determinate agent does not remove demerit, merit, cause, and fruit. What truly makes demerit, merit, and karmic fruition collapse is the path of causelessness. What is preserved here is “provisionally said to exist, without fixed determination”—the name can be spoken without making the body real.

Carried Forward: Jizang’s Mirror

The preceding is the meaning of the chapter itself. What follows lets the Sui–Tang Three Treatises tradition illuminate it from the side. The mirror is not the treatise; it offers another way to read the same verses.

Jizang’s Commentary on the Treatise on the Middle places this chapter and those following it within the gate of nonaction, after emptiness and the signless, as one stage in the sequence of the three gates of liberation. He also exhaustively tests acting through five combinations of person and dharma: person and dharma both existent; both nonexistent; half existent and half nonexistent; one existent and one nonexistent; and one person with three dharmas or three persons with one dharma. Each arrangement changes the fault but leaves no fifth road. He further classifies the chapter: the first eleven verses refute action and agent, meaning person and dharma within the cause; the final verse refutes experience and experiencer, meaning person and dharma within the fruit. Once person and dharma are absent in both cause and fruit, all is empty.

This interpretation—the chapter’s position among the three gates of liberation, the fivefold table of tests, and the division between person and dharma within cause and fruit—is Jizang’s addition, not the original wording of Nāgārjuna’s verses or Qingmu’s long commentary. The reader layer still follows Kumārajīva’s verse groups and sequence of questions and answers. The readings are recorded side by side without adjudication, each returned to its own place.

A person closes the eyes, and sound remains; covers the ears, and the light before the eyes remains. Each of the six faculties does its own work, yet all these acts seem to belong to the same person. The mind therefore naturally asks: to whom does all this belong?

The question may seem merely a habit of speech. Yet at the opening of this chapter, someone presses that habitual expression all the way down.

From the Word “Who,” a Preexistent Person Is Set Up

The questioner at the chapter’s opening first cites another’s claim:

“The faculties such as eye and ear, the dharmas such as pain and pleasure— who possesses such things? That one is called the prior abider.”

The faculties of eye and ear and the experiences of pain and pleasure must have someone who possesses them. That one is called the prior abider.

Qingmu first fixes the two terms: eye, ear, nose, tongue, body, vitality, and the rest are faculties; painful, pleasant, and neither-painful-nor-pleasant experience, together with mind and associated mental dharmas such as perception, volition, and recollection, are the dharmas such as pain and pleasure. He then states the questioner’s reason: “By virtue of this prior abider, the faculties such as the eye are able to grow.” Not only must there be an owner, but that owner must come first: he exists first, and only then can the faculties such as the eye grow.

The questioner therefore gathers the claim into one sentence: “If there were no prior abider, who would possess the dharmas such as the eye?” Unable to infer anything else, he concludes that “a prior abider already existed before them.”

We must see where the force of this question lies. It is not established from scripture but from the feel of language: if one says “my eye,” there ought to be an I; if one says “I am suffering,” there ought to be a person who suffers. From a possessive, it infers an owner; from the owner’s supposed power to nurture the faculties, it infers that he precedes them.

An ordinary phrase has thus been stretched into an entity.

The First Step Does Not Ask Whether It Exists, Only How It Is Known

The respondent’s first verse does not immediately say that the prior abider is absent. It asks a question that comes earlier:

“If a prior abider existed first, apart from the faculties such as the eye, and the dharmas such as pain and pleasure, by what could it be known?”

You say that this one exists first, apart from the faculties such as eye and ear and apart from the dharmas such as pain and pleasure. Then—from where do you know him?

Qingmu makes the question concrete: “External dharmas such as jars and robes are known through faculties such as the eye; internal dharmas are known through faculties such as pain and pleasure.” Jars and robes outside are known through eye, ear, nose, tongue, and body; inner events are known through experiences such as pain and pleasure. Nothing a person can speak of is known without some avenue. Qingmu also cites a sūtra: “Susceptibility to destruction is the mark of form; the capacity to experience is the mark of experience; the capacity to cognize is the mark of consciousness.” Form has susceptibility to destruction as its mark, experience the capacity to experience, and consciousness the capacity to cognize. Each has its own mark, and people recognize it by that mark.

Yet the prior abider posited by the questioner is declared to be apart from every faculty and every experience. If so, it has no mark to indicate; with no mark to indicate, no path can reach it.

This step matters. The respondent is not saying that whatever cannot be seen does not exist. He returns the burden to the one advancing the thesis: if you say it exists, explain how you know it. A name may be set down, but its evidence cannot be left empty.

Marks of a Self: Using Operations of Body and Mind to Prove a Master Apart from Body and Mind

The questioner immediately supplies evidence. He now says: “Inhalation and exhalation, looking and blinking, vitality, thinking, pain and pleasure, hate and love, initiating movement, and so forth are marks of a self.” Breathing, blinking, being alive, thinking, pain and pleasure, hate and love, raising a hand or taking a step—are these not all plainly present? “If there were no self, how could there be such marks as inhalation and exhalation?”

The rescue accords readily with ordinary feeling. Yet it has a difficulty: every example is precisely an operation of the faculties such as eye and ear or of the dharmas such as pain and pleasure. A moment ago the prior abider was said to precede and stand apart from these; now these very things are asked to prove it. The one to be proven stands outside the door, while all the witnesses brought forward stand inside.

The respondent therefore takes up the claim and searches for the “self” in relation to the body.

Pillar, Armor, Severed Head, Lamplight, and Madness: Five Searches

In this passage Qingmu’s extended commentary applies five successive tests. They do not repeat one sentence five times; each closes off a different route of rescue.

The first asks about location. If a self exists, it can only be inside or outside the body: inside “like a pillar within a wall,” or outside “like a person wearing armor.” If inside, trouble follows: “If it were inside the body, the body should be indestructible, because the self always remains within it.” If a permanent thing braces the body from inside, how could the body decay? Yet the body plainly decays. Therefore, saying the self is inside the body is “nothing more than a statement, false and without reality.” If outside and densely covering the body like armor, the body should be invisible and indestructible—yet “we plainly see the body destroyed.” Neither inside nor outside can be sustained.

The second closes the escape route. Someone says that when an arm is cut off, the self contracts inward, so it cannot be severed. Qingmu’s response is direct: “When the head is cut off it likewise should contract inward and the person should not die; yet death plainly occurs.” If it truly has the ability to contract, the moment of beheading is when it should do so most of all; nevertheless, the person dies.

The third asks about size. Another says that a large body has a large self and a small body a small self, “just as a large lamp gives a large light and a small lamp gives a small light.” The rescue sounds natural but surrenders permanence by itself: “The self would change with the body and should not be permanent.” If it follows the body, then it follows the body to the end: “When the body is absent the self is absent, just as when the lamp goes out its light goes out.” Qingmu’s next sentence cuts deepest: “If the self is impermanent, it is of the same kind as eye, ear, pain, pleasure, and the rest.” If impermanent, it belongs to the same class as eye, ear, pain, and pleasure. Why posit it at all? It was posited precisely to supply something that preceded and differed from them.

The fourth asks about mastery. “A person afflicted by madness is not in control and does what should not be done.” Such a person does what ought not be done and cannot govern himself. But the self was introduced as master: “If a self existed as the master of every action, how could the person be said not to be in control?” If the illness cannot afflict that self, then the body is still moving and something other than the self must be moving it; one would have to admit a separate agent.

Across these four tests, one rescue route after another closes: location fails, escape from severance fails, permanence fails, and mastery fails. Each fails at the same place—every power added to save the self must be borrowed from body and mind; once borrowed, the self is no longer a prior abider apart from body and mind.

If It Can Truly Be Separate, Separation Should Work at Both Ends

At this point the respondent changes the point of attack. The verse says:

“If a prior abider exists apart from eye, ear, and the rest, then eye, ear, and the rest should likewise exist apart from the prior abider.”

You say that the prior abider can exist first, apart from eye, ear, and the rest. Very well—then follow the reasoning honestly to its end: eye, ear, and the rest should also be able to exist apart from the prior abider.

This sentence is not arguing for the independence of eye and ear. It is testing the questioner’s partiality. Separation means separation; there is no reason to allow only one side to separate while forbidding the other. What the questioner really wants is one-way separation: the prior abider need not depend on the faculties and dharmas, but the faculties and dharmas must depend on the prior abider. He can offer no reason for that one-way privilege.

The questioner’s retreat is revealing: “The two things may indeed be separate from one another, so long as a prior abider exists.” He would rather surrender his reasoning than surrender that one.

The respondent therefore opens the last layer:

“Through dharmas one knows a person; through a person one knows dharmas. Apart from dharmas, where is the person? Apart from the person, where are the dharmas?”

Qingmu aligns the terms: “Dharmas means eye, ear, pain, pleasure, and the rest; person means the prior abider.” Your earlier argument had two inferences: dharmas prove a person, and a person proves dharmas. Then answer this: apart from dharmas, where is the person? Apart from the person, where are the dharmas?

Person and dharmas are a pair known in dependence upon one another. Such mutual indication establishes the names through which the two sides refer to each other; it cannot establish either side as a body that stands apart from the other. The questioner thought mutual indication could prove a preexistent person. In fact, it proves precisely that his person cannot move a single step apart from dharmas.

No Overseer Need Come First: The Six Faculties Distinguish for Themselves

Without a prior abider, then, would seeing and hearing dissolve into confusion? The next verse answers:

“All the faculties such as the eye in truth have no prior abider; the faculties such as eye and ear are discriminated through their different marks.”

Qingmu is concrete: “Eye-consciousness arises with the eye faculty as cause and visible form as condition. The faculties such as eye and ear are known through causes and conditions in conjunction, not through a prior abider.” When the eye faculty meets visible form, eye-consciousness arises. The eye faculty is known through this conjunction of causes and conditions, not through a prior abider. The faculties “can each discriminate”: each has its own mark and each meets its own object, so there is no confusion.

This answers a common worry. One imagines that without an owner the eye will hear and the ear will see. The truth is the reverse: order is not given by an owner but by each faculty’s own mark. The eye sees only form because the eye is just such a set of conditions; no one above it assigns the work.

Then Who Knows the Object—One Knower, or Six?

The questioner tries another path.〔T30 0014a04〕

“If the faculties such as the eye have no prior abider, how can each individual faculty such as the eye know an external object?”

Qingmu fills out the reason: “The faculties such as eye and ear do not think and should therefore have no knowing.” A faculty is an insensible thing and cannot think. How, then, could it know an external object? Since knowing plainly occurs, “one should know that apart from the faculties such as eye and ear, there is another knower of objects.”

This question is cleverer than the first. It no longer begins from “whose?” but from “who knows?”

The respondent first divides the road: “Is there a separate knower in each faculty, or is there one knower within all the faculties? Both have faults.” The knower you infer must be either one per faculty or one shared by all six. There is no third path. Each path then receives one verse.

First, the shared knower:

“The one who sees would be the one who hears, the one who hears the one who experiences; in that case, these faculties should have a prior abider.”

If the one who sees, hears, and experiences is one, that one is not bound to any particular faculty. If not bound to a faculty, when operating through the eye it should not be limited to seeing: perhaps “one could hear a sound with the eye,” or even act “as though a person had six directions in which to see or hear at will,” choosing any of the six. Qingmu cuts it off in one sentence: “But this is not so.” The eye simply cannot hear sound.

Now consider one knower per faculty:

“If seeing and hearing are distinct, and the experiencer too is distinct, then while seeing one should also hear; in that case, there would be many selves.”

If seer, hearer, and experiencer are distinct, each is independent and continually present. While the seer sees, the hearer has not ceased and should hear at the same time. The selves in nose, tongue, and body should likewise all operate together. The result is “one person but many selves, because all the faculties would know their various objects at the same time.” Qingmu again says simply: “Yet this is not so.”

With one knower, the six faculties should exchange functions; with six knowers, one person splits into six. Since present seeing and hearing neither exchange functions nor split apart, they are not produced by a knower standing apart from the faculties.

Nor Is It in the Four Great Elements

The search can go one layer deeper. Someone may think: if it is not found in the faculties, perhaps it hides in the material from which they are made. The verse blocks that retreat:

“The faculties such as eye and ear and the dharmas such as pain and pleasure arise from the great elements—and in those great elements too there is no self.”

Qingmu is concise: the four great elements on which eye, ear, and the rest depend contain no prior abider. It is not found in an operation, and it remains unfound when the search retreats to the operation’s cause. However deep the location changes, the same thing is still missing—the mark that could indicate it.

Once the Person Is Withdrawn, Dharmas Cannot Be Preserved

The questioner’s last move is to preserve half the claim: perhaps the prior abider can be abandoned, but surely eye, ear, pain, and pleasure exist?

The respondent does not allow it:

“If the faculties such as eye and ear and the dharmas such as pain and pleasure have no prior abider, then the dharmas such as the eye should likewise be absent.”

Qingmu presses two questions: “Who possesses this eye, ear, and the rest? Through what conditions do they exist?” You previously called eye, ear, pain, and pleasure a ready-made group only because they were placed under the name of one owner. Once the owner is withdrawn, on what basis does this group remain ready-made?

This does not say that seeing, hearing, and feeling do not occur. It says that they can be understood only through the conditions on which they exist; they cannot change their signboard and continue as self-standing bodies. Once the person is refuted, no dharma may remain behind as a substitute.

Unfindable in the Three Times, the Question and Answer Dissolve

The last verse gathers the chapter:

“Before the faculties such as the eye there is no prior abider; in the present and future there likewise is none. Because it is absent in all three times, that discrimination too is absent.”

Qingmu says: when the prior abider is considered and sought, it is absent before the faculties such as the eye, absent now, and absent afterward. “If it is absent in the three times, this is nonarising and quiescence.”

The final sentences deserve attention. Qingmu does not say, “Therefore the answer is that the prior abider does not exist.” He says: “Such questions and answers and conceptual proliferation then cease. Because conceptual proliferation ceases, dharmas are empty.”

It is not the answer that ceases, but the entire chain of question and answer. The chapter’s opening question—if there is no prior abider, who possesses the dharmas such as the eye?—already placed an owner inside the question. When the thing placed there cannot be found in any of the three times, the question has no way to arise. It has not been defeated; it has dissolved on its own.

This also continues the movement of the preceding chapters. First, the one installed behind seeing, hearing, pain, and pleasure is refuted. But the search does not stop there. The respondent next says: “If there is no prior abider, then the dharmas such as the eye should likewise be absent.” If the dharmas such as eye, ear, pain, and pleasure have no prior abider, “who possesses this eye, ear, and the rest? Through what conditions do they exist? Therefore, eye, ear, and the rest are likewise absent.” Once that one cannot be indicated, the dharmas supposedly owned by it undergo the same search and cannot be left standing as realities. The end says further: “Because conceptual proliferation ceases, dharmas are empty.” What ceases is the whole exchange, together with the entities presupposed at both its ends. What remains is not a hard set of seeing, hearing, pain, and pleasure, but only what exists through conditions and is distinguished by differing marks.

What a Reader Can Carry Away from This Chapter

First, replace “whose?” with “through what conditions?” When pain comes, one habitually searches first for the self who suffers. This chapter shows that such a one cannot be indicated in any of the three times. Only conditions can be indicated.

Second, notice every power added in an attempt to rescue him. The self must contract, change size with the body, and escape illness. Each addition is borrowed from body and mind. The more it borrows, the less the one apart from body and mind can stand.

Third, do not fear that without an owner there will be chaos. “Eye-consciousness arises with the eye faculty as cause and visible form as condition.” The eye sees form, the ear hears sound; each is discriminated through its different mark. Order never awaited an overseer.

Carried Forward: Jizang’s Mirror

The preceding is the meaning of the chapter itself. What follows uses the Sui–Tang Three Treatises tradition as a mirror held alongside it—a mirror is not the treatise, but another way to see the same verses.

In his Commentary on the Treatise on the Middle, Jizang divides the chapter into three parts: first refuting the prior abider to illuminate the emptiness of sentient beings; next refuting the faculties to illuminate the emptiness of dharmas; and finally rebuking the opponent’s imposed constructions of person and dharmas. He also divides the places searched for a prior abider into three levels: before the six faculties (also described as outside them), within the six faculties, and within the four great elements. This is to seek a self in every place and find none. As for why this chapter follows the refutation of action and agent, he says that the preceding chapter refuted the functioning of person and dharmas, while this chapter refutes their body: only after a self and faculties are posited can the function of acting be posited. The function has been refuted, but its root has not yet been removed, so this chapter is required.

He also gives a name to each of four passages in Qingmu’s responses: searching inside and outside the body is the refutation through location; contraction when a limb is severed is the refutation through drawing out the implication; change with bodily size is the refutation through impermanence; and lack of control in madness is the refutation through non-mastery. He says that these four refutations respectively disclose emptiness, suffering, impermanence, and non-mastery. As for the two verses on the knower, he divides them into the refutation of one self and the refutation of many selves. If one self occupies all six faculties, they incur the fault of exchanging functions: confusion among faculties means confusion among their objects. If six selves occupy six faculties, one person has six selves and incurs the further fault of simultaneously apprehending all six objects. In both verses the first half states the other’s position and the second half exposes its fault.

This mirror supplies the chapter with an outline, names the four refutations, and places the chapter sequence within the Three Treatises interpretive structure of refuting error and disclosing the correct. The outline and names are Jizang’s additions, not the original words of Nāgārjuna’s verses or Qingmu’s extended commentary. The readings are recorded side by side without adjudication, each returned to its own place.

By the time the Treatise on the Middle reaches this chapter, the questioner has been driven into a corner: the preceding chapters searched through agent and experiencer and found neither. So he changes tactics. Instead of speaking directly about a self, he turns to something everyone has seen with their own eyes.

He says: “There should be experience and an experiencer, just as there are fire and fuel. Fire is the experiencer, and fuel is experience, namely the five aggregates.”

Fire burns wood. Fire is the burning side; wood is the side being burned. The two are plainly distinct and plainly come together. If fire and wood can be so, why could the experiencer and the five aggregates not be the same?

The full weight of the question rests on the word “just as.” He is not discussing fire; he is using fire as proof. Proof by analogy has one vulnerable point: if the analogy cannot stand by itself, the conclusion it supports falls with it. So the respondent turns around to inspect the fire—“because neither fire nor fuel is established.”

All sixteen verses in the chapter inspect this one fire.

1. First Objection: If You Open Your Mouth to Refute It, Have You Not Already Admitted That It Exists?

Before the fire can be examined, the questioner blocks the way with a deft objection.

He says: “Because the horn of a hare and the hair of a tortoise do not exist, they cannot be refuted.” A hare has no horn and a tortoise no hair—precisely because they do not exist at all, there is nothing to refute. Since you use the marks of identity and difference to refute fire and fuel, fire and fuel must exist; otherwise, what are you refuting? He adds an even more concrete example: gold must first exist before one can fire and hammer it. Thus he says that “only after people in the world plainly see that a thing exists can it be reflected upon,” and concludes with a tightly drawn line: “If you allow its existence, it is thereby already existent.”

What makes this challenge difficult is not fire, but speech itself: whatever one speaks of has already been admitted.

The respondent answers by drawing just one distinction, but it governs the manner of speech throughout the Treatise on the Middle: “There should be no fault in speaking according to worldly convention.” He gives an example that works on the spot: “A dialectician who wishes to refute existence and nonexistence must speak of existence and nonexistence, but does not accept them merely by naming them.” Anyone who wishes to refute existence and nonexistence has no choice but to say the words “existence” and “nonexistence.” Yet saying them aloud does not mean receiving them into the mind as substances. Speaking and accepting are two different things.

The respondent then returns the challenge unchanged: “Your word ‘refute’ would refute you.” If you insist that every utterance is an admission, then the word “refute” that you utter is also an utterance and, by your own rule, refutes you first.

Thus, “although there is speech, there is likewise no acceptance.” The words are spoken; the things are not taken up. These eight Chinese characters stop a familiar misunderstanding—the idea that someone who speaks of emptiness should remain silent or else contradict himself. One must be able to point something out in order to refute it. Pointing lets it show itself; it does not invite it in.

Once this objection has been stopped, fire and fuel formally undergo examination.

2. Identity: If Fire and Fuel Are One, the Potter Is the Jar

The first verse opens two routes:

“If fire were fuel, act and agent would be one; 　if fire were different from fuel, there would be fire apart from fuel.”

Qingmu first fixes the names: “Fire is flame, and fuel is wood.” Then he exchanges them for another familiar pair of names: “The agent is a person, and the act is the deed.” Fire is what burns and wood what is burned, just as a person is what acts and a deed what is done.

To say that fire and fuel are one is therefore to say that act and agent are one. Qingmu immediately illuminates this through a scene of craft: potter and jar. “The potter is not the jar, and the jar is not the potter; how could they be one?”

No debate is needed to understand the sentence. The jar is made by the potter, but the potter is not the jar—the potter remains when the jar breaks. If maker and made were truly one, the person making the jar would have to shatter with it.

So the route of identity first collapses. Fire is not wood.

3. Difference: A Fire Apart from Wood Would Burn Forever Yet Burn Nothing

Once identity fails, one naturally slides to the other side: then they must be different.

The respondent follows the word “different” and watches where it leads. If fire and wood were really two things each able to stand on its own, then “one should be able to distinguish this as fuel and that as fire, and wherever there was no fuel there should be a fire.” You should be able to point to a place with no wood at all and say: there is fire. Qingmu closes the case with four Chinese characters: “But it is not in fact so.”

Then three consequences follow in succession:

“It should then burn forever, without arising through fuel; 　there would be no labor of kindling fire, and it would also be called a fire without an act.”

The first is perpetual burning. If fire does not rely on wood, it has neither rising nor falling and burns forever.

The second is that human effort becomes vain. Qingmu introduces an utterly domestic detail: “Human effort means tending and protecting the fire so that it burns.” This effort is nothing lofty; it means crouching before the stove, adding wood, raking ash, blocking wind, and keeping watch over the flame. Qingmu uses this concrete fact as evidence: “Such effort is plainly present; therefore one knows that fire is not different from fuel.” If fire truly had nothing to do with wood, for whom did your hand add that piece of fuel? This is the chapter’s homeliest and most forceful move: it does not argue from an abstraction, but from a pair of hands.

The third consequence is the most glaring: a fire without an act. “Apart from fuel, what could fire burn?” What does a fire burn once wood is gone? Can something that burns nothing still be called fire? Qingmu’s verdict is final: “There is no such thing as a fire without an act.”

The route of difference also fails. What it gives you is not an independent fire, but a fire that neither goes out nor burns anything—the name remains, while fire has vanished.

The respondent then restates these two layers in another verse: “If fire does not wait for fuel, it does not arise from conditions; if fire burns forever, human effort should be in vain.” Not waiting, it does not arise from conditions; not arising from conditions, no amount of tending can reach it.

4. It Is Called Fuel Only While Burning—Then What Was the Piece in Your Hand at That Time?

The questioner has one route of retreat left: I do not say that wood exists apart from fire; I say it is called fuel only while it is burning.

This step sounds very much as if dependent arising has already been understood. Yet a single verse exposes it:

“If you claim that it is called fuel only while it burns, 　at that time there is only wood; what is it that burns the fuel?”

By saying “it is called fuel only while burning,” you admit that the name “fuel” arises from a relation. Yet you previously used fuel as a prior object for fire. Before burning, the piece in your hand—by your own rule—was not yet called fuel: “At that time there is only wood.” It is only a stick of wood. Since fuel has not yet been established, what is fire burning?

Qingmu places the two ends side by side, and the contradiction shows itself: “If fuel exists separately apart from fire, how can one say it is called fuel only while burning?”

One name cannot serve both uses. Either it exists only within the relation, in which case it cannot be treated as a substance prior to the relation; or it is a substance, in which case one should not say that it is “called fuel only while burning.”

5. No Reaching Means No Burning: Two Independent Things Cannot Even Touch

The route of difference contains a deeper and especially elegant layer:

“If they are different, fire does not reach fuel; if it does not reach fuel, it does not burn; 　if it does not burn, it is not extinguished; if it is not extinguished, it abides forever.”

Ordinarily one assumes that two independent things can at least come into contact—fire comes over and burns the wood. But the respondent observes that things truly established without depending on one another cannot even accomplish the act of meeting. The reason is that they “are not established in mutual dependence.” Since each is already established in itself, “what use would fire have for fuel?” It has no reason to move toward the wood.

And without reaching it, fire cannot burn it—“because nothing can burn what it does not reach.” If it does not burn, there is no extinction; without extinction, it “abides forever in its own mark.” A fire that never goes out is no fire.

Each of these four links generates the next, until the route of difference is driven to the same place as the view of permanence.

The questioner objects with a very natural counterexample:

“Although fire and fuel are different, fire can reach fuel, 　just as this person goes to that person and that person comes to this person.”

Qingmu puts it more directly: “Just as a man goes to a woman and a woman goes to a man.” Are a man and a woman not two people? Can they not still come together?

The respondent accepts the analogy and distinguishes the decisive condition in one move:

“If fire and fuel could both be established separately apart from each other, 　only then could fire reach that fuel.”

To use man and woman as an analogy, the two sides must first meet one condition: each must remain established apart from the other. Man and woman do—“a woman exists apart from a man, and a man apart from a woman.” A woman need not rely on a particular man in order to be a woman; each is a complete person.

Fire and fuel are not like this: “Apart from fire there is no fuel, and apart from fuel there is no fire.” Apart from fire, the piece of wood is only wood and is not called fuel; apart from wood, the heat has nothing on which to depend and is not called fire. The two names arise together.

So Qingmu judges: “Therefore your analogy is wrong.” If the analogy fails, the conclusion fails with it.

The questioner thinks he has found an example of the same kind, but in fact he has found a counterexample: man and woman can approach each other precisely because each is established without depending on the other—fire and fuel are exactly otherwise. The more the analogy is pressed, the more clearly their difference appears.

6. Mutual Dependence: Then Let Us Say the Two Wait upon Each Other

With both identity and difference shattered, the questioner advances what most resembles the correct answer: “Fire and fuel exist in mutual dependence”—fire exists because of wood, wood is fuel because of fire, and each establishes the other.

This is nearly dependent arising. Why does the respondent still inspect it? Because one must see what “mutual dependence” is being made to do. If it merely means that neither side has substance of its own, there is no problem. But the questioner uses mutual dependence to prove two real things. That requires a question:

“If fire exists because of fuel, and fuel exists because of fire, 　which dharma is first determinately present, thereby allowing fire and fuel to exist?”

Which one first stands firm?

If fuel is determinate first, then “fuel comes first and fire afterward”—the wood was already there on its own, so one need not say that it waits for fire. Yet how can wood that has never been burned be called fuel? “If fire does not burn the fuel, then the fuel is not established.” Conversely, if fire is determinate first, exactly the same fault returns unchanged.

The next verse reveals the self-contradiction on fire’s side:

“If fire exists because of fuel, then fire is established after being established; 　consequently, there is no fire within fuel.”

If fire is said to be real, then fire has already “abided by itself within fire”—it is established in its own position. Yet you also say that it must depend on wood to be established. It is therefore established a second time after already being established, accomplishing the same thing twice. The other side is equally awkward: “Fuel abides in its own substance apart from fire”—wood is independently established, so there is no fire within it. You try to bind two entities together through mutual dependence, but both remain where they were; neither actually depends on the other.

The respondent therefore turns mutual dependence over and examines it:

“If a dharma is established because of what it depends on, this dharma in turn establishes that on which it depends; 　there is then no cause that serves as the ground of dependence, nor any dharma being established.”

Fire depends on wood, and wood in turn depends on fire. Once the circle begins to turn, there is nothing first established: “Neither is determinate. Because they are not determinate, they cannot be found.” Both wait for the other to stand firm first, so neither stands firm. One thing must be seen clearly here: the problem is not mutual dependence; the problem is using mutual dependence to prove substance. Mutual dependence says precisely that neither side has feet of its own. Yet you would use it to prove two pairs of feet.

The final stroke is the cleanest:

“If a dharma is established through dependence, how can it depend while unestablished? 　If after being established it has dependence, once established, what use has it for dependence?”

While unestablished, it is not yet a thing—“the unestablished is nonexistent; how could the nonexistent depend on anything?” A nonexistent thing cannot wait for someone. Once established, it is already a thing—“once established, what use has it for dependence?” Someone standing firmly need not be held up. The two ends close from front and back, leaving no moment between them in which a real thing could be “established through mutual dependence.”

A final verse seals both routes together: “There is no fire because of fuel, and none without fuel; there is no fuel because of fire, and none without fire.”

7. The Three Times of Going and Coming, Used Here Once More

At this point, the respondent adds another approach:

“Fire does not come from elsewhere, nor is there fire in the place where it burns; 　fuel is likewise so, and the rest is as explained for going and coming.”

Fire does not fly from somewhere else and burrow into the wood; nor is fire hidden in advance inside the wood. Qingmu offers a hands-on method: “Because when one splits the wood and seeks fire, it cannot be found.” Split the wood and look. What is split is wood; split it again and it is still wood. Fire is not inside.

He then carries the examination through the three times in a single line: “What has burned does not burn, what has not burned does not burn, and while burning it does not burn.” What has already burned burns no more; what has not burned has yet to burn; and in the very moment of burning no independent “present burning” can be separated out. The chapter on going and coming has already performed this examination once. Here it is not repeated, but only named: this meaning is as explained there.

This stroke deserves notice. The chapters of the Treatise on the Middle are not unrelated short essays. Once a mode of examination has been used here, it remains effective elsewhere. Going, coming, and burning change the subject, not the question: after removing the relation, can you still point to the thing?

8. Five Statements Gather Everything Together

After identity and difference, mutual dependence, and coming and going have all been examined, Nāgārjuna concludes in five statements:

“Fuel is not itself fire; apart from fuel there is no fire. 　Fire does not possess fuel; there is no fuel within fire, and there is no fire within fuel.”

Not itself, not apart, not possessing, not within it, and it not within me. Qingmu identifies the fault supporting each statement: the first rests on “the fault that act and agent would be one”; the second on “faults such as perpetual burning”; the last three all on the fault of difference, so “none of the three is established.”

All five routes have been exhausted. If one still wishes to install an independent fire between flame and wood, there is nowhere left to put it.

9. For Whom Was This Fire Explained?

At this point the questioner himself comes to his senses and asks the chapter’s most important question: “Why speak of fire and fuel?”

You have spent all this time examining fire. Why examine it at all?

Only then does the respondent return the matter: “Just as there is fire because of fuel, so there is an experiencer because of experience. Experience means the five aggregates, and experiencer means the person.”

At the beginning, the questioner used fire and wood to prove experience and experiencer, and the respondent simply followed his analogy to the end. Since fire and wood are established neither through identity nor through difference, the experience and experiencer supported by that analogy are likewise unestablished.

Nor does the matter stop there:

“Through the dharmas of fire and fuel, the dharmas of experience and experiencer are explained, 　as are jars and robes and all such dharmas.”

The same examination applies not only to the person and the five aggregates, but also to jars, robes, and “all such dharmas.” Qingmu says it plainly: “All external dharmas such as jars and robes are just as explained above: unborn and ultimately empty.”

This move raises the weight of the whole chapter. Fire and fuel are not an isolated example but a sample. Wherever something can be divided into an “agent” and an “object”—what burns and what is burned, what experiences and what is experienced, what acts and what is acted upon, what contains and what is contained—the same questions can be asked: Are the two identical or different? Which one first stands firm? Apart from the other, does either remain?

Pressed to the end, agent and object are not two antecedent things that later come together. First there is the event, and only within the event do the two names divide.

10. The Sound at the Chapter’s End

The final verse suddenly grows severe:

“If someone asserts a self and says that dharmas each have different marks, 　know that such a person cannot taste the Dharma of the Buddha.”

Qingmu first gives the reason: “Dharmas have never arisen from the beginning and bear the mark of ultimate quiescence; therefore this verse is spoken at the chapter’s end.”

He then gives two examples from his own time. One is the Vātsīputrīyas: form cannot be said to be the self, and the self cannot be said to exist apart from form; “the self resides in a fifth, inexpressible repository.” This school carefully avoids both identity and difference, yet opens a separate inexpressible repository in which to place the self. Avoiding identity and difference is not the same as releasing the self; it merely hides the self more deeply.

The other is the Sarvāstivādins: “Dharmas each have their own marks: wholesome, unwholesome, or indeterminate,” tainted or untainted, conditioned or unconditioned—each bearing a determinate mark. This school does not establish a self, but establishes fixed marks for dharmas one square at a time.

The two schools suffer the same illness, though it settles in different places: one locates substance in the person, the other in dharmas. Qingmu’s verdict is therefore the same for both: such people “do not attain the quiescent mark of dharmas, but use the Buddha’s words to construct all kinds of conceptual proliferation.”

The words “the Buddha’s words” weigh most heavily. They are not using non-Buddhist language, but precisely the names spoken by the Buddha. The more finely they divide them into separate squares, the farther they move from the mark of quiescence.

The phrase “cannot taste the Dharma of the Buddha” is also perfectly chosen. It does not say that they fail to understand the reasoning. It says that they cannot taste it—like memorizing every dish on a menu without eating a single bite.

11. What Can a Reader Carry Away from This Chapter?

First, an analogy must itself be examined before it can serve as proof. The things about which the mind is least guarded are often precisely those that “everyone has seen.” The questioner uses this lack of guard to smuggle in experience and experiencer.

Second, speaking and accepting are two different things. “Although there is speech, there is likewise no acceptance”—one speaks in order to point something out, not in order to take it up. If this distinction does not stand, one either dares not speak or establishes another entity with every word.

Third, the stroke concerning human effort. “Tending and protecting a fire so that it burns—such effort is plainly present.” Every piece of wood added while crouching before the stove is immediate evidence that fire is not apart from fuel. The principle need not be sought far away.

Fourth, mutual dependence does not prove two entities. It says exactly the opposite: neither side has feet of its own and so they lean on one another. The instant one uses it to prove that they have their own feet, the unestablished cannot depend and the established need not depend; at neither end is there a place to begin.

Fifth, this fire is addressed to me. The examination of fire and fuel is not for the sake of fire and fuel, but to see clearly the “one who experiences” and “what is experienced” between the person and the aggregates. Extend it outward, and jars, robes, and all dharmas are no different.

Carried Forward: Jizang’s Mirror

The foregoing is the meaning of the chapter itself. What follows uses the eye of the Sui-Tang Three Treatise lineage as a sidewise illumination—a mirror is not the treatise, but another way of reading the same verses.

In his Commentary on the Treatise on the Middle, Jizang divides the chapter’s refutation of the analogy into six approaches: identity and difference, mutual dependence, cause and noncause, inside and outside, the three times, and the fivefold search. He treats the final verse as a rebuke after both thesis and analogy have been exhausted: the first two approaches investigate nonexistence through correct contemplation, while the last rebukes the deluded mind that supposes existence.

On the questioner’s linguistic objection, Jizang answers: although there is speech in accord with convention, there is no acceptance in accord with convention—the language is the same as that of the world, but what is accepted in the mind is different.

As for the fivefold search, Jizang directly maps it onto the self and the five aggregates: form is the self, the self is apart from form, form is in the self, the self is in form, and the self possesses form. Five statements for each of five aggregates make twenty-five selves. He also resolves a small question for the reader: Why say only that the self possesses form, rather than that form possesses the self? Because the self is the governing subject and controls form, so form belongs to the self. One cannot reverse this by saying that form governs the self and the self belongs to form; therefore no such statement appears.

This mirror connects the chapter to the Three Treatise lineage’s interpretive structure of refuting error and disclosing correctness. Its sixfold classification is Jizang’s addition, not the original speech of Nāgārjuna’s verses and Qingmu’s prose commentary. The readings are recorded side by side without adjudication, each returned to its own place.

When questioning reaches its end, it always arrives at the same sentence: How did it all begin?

The person asking has already installed one thing in the mind—belief that there is a beginning somewhere, though it has not yet been found. It cannot be found because it is too distant, too ancient, or incompletely recorded. The task is merely to locate that point.

This chapter is called “Examination of the Prior Limit.” The prior limit is the earliest boundary being sought. From beginning to end, the chapter says only one thing: that point is not hidden; it cannot stand in the first place.

One Question: The Sūtra Speaks of Existence, and the Same Sūtra Says It Cannot Be Found

The chapter opens with a challenge from the questioner, and what he holds in his hand is a sūtra.

He says that the Sūtra on the Beginningless Limit teaches that “beings come and go in saṃsāra, while the prior limit cannot be found,” yet the same sūtra also says, “Here it speaks of beings and saṃsāra.” Since it says there are beings who come and go and a saṃsāra through which they pass, how can it also say that the prior limit cannot be found?

The question is forceful because it does not come from outside. It takes the tradition’s own sūtra and points to what looks like an internal contradiction: the same text speaks of existence and says that the limit cannot be found.

Where is the skeleton of this question? In an unspoken assumption: to say that something exists is to say that it has a beginning. If beings exist, there must be a moment when beings began; if saṃsāra exists, there must be a point when saṃsāra began. The entire chapter’s answer consists in dismantling this assumption.

First Establish the Speaker: Who Is the Great Sage?

The respondent does not enter the argument immediately, but first establishes the speaker.

The verse says, “What the Great Sage taught is that the prior limit cannot be found.” Qingmu pauses over this sentence to explain what “Great Sage” means: “There are three kinds of sage.” First are non-Buddhists who have attained the five supernormal powers; second are arhats and pratyekabuddhas; third are great bodhisattvas who have attained supernormal powers. The Buddha is supreme among these three and is therefore called the Great Sage.

Readers may easily skip this passage, yet it does something important: it acknowledges that “sage” is not a designation monopolized by Buddhists. Non-Buddhists who attain the five supernormal powers also count as sages; so do arhats, pratyekabuddhas, and great bodhisattvas. Qingmu does not expel others from sagehood, but says only that the Buddha is supreme among them. It is a very candid way of speaking—first order the titles honestly, then proceed.

Only then comes the decisive sentence: “Nothing the Buddha says is other than a true statement.” Since the questioner brings a sūtra, the respondent first sets the sūtra’s standing firmly in place. If the speaker makes true statements, then “the prior limit cannot be found” cannot be read as expedient language, exaggeration, or approximation. It is a true statement and must be taken seriously word by word—and what follows in the chapter does exactly that, all the way to the end.

Beginningless Does Not Mean “A Very, Very Long Time Ago”

The verse continues: “Saṃsāra has no beginning, and likewise has no end.” Qingmu offers only one sentence of explanation, but it sets the direction of the whole chapter: “The first and last limits of saṃsāra cannot be found; therefore it is said to have no beginning.”

Notice the direction of inference. It is not that saṃsāra is first known to be immensely long, so long that its beginning cannot be seen, and is therefore called beginningless. Rather, the two positions of first and last cannot stand at all, and therefore it is called beginningless.

These two kinds of “beginningless” are entirely different. The first still concedes a starting point but says that people do not know where it is; the second says that the position called “starting point” is itself unestablished. The first leaves a mystery, while the second dissolves the puzzle.

When people hear that saṃsāra is beginningless, most first imagine the former—an expanse of time so long that it has no visible end. Yet not one word in Qingmu’s reason refers to length. He says only that the first and last limits cannot be found.

The questioner immediately hears the difference and retreats one step: Fine, even if there is no beginning and no end, there should still be a middle. Qingmu directly judges: “This too is not so.”

Without Two Ends, No Midpoint Can Be Measured

The verse follows quickly: “If there is neither beginning nor end, how could there be a middle?” Its conclusion is: “Therefore, within this, neither before, after, nor simultaneous occurrence exists.”

Qingmu gives an utterly plain reason: “A beginning exists because of middle and end, and an end exists because of beginning and middle.”

A beginning is called a beginning because a middle and end follow it; an end is called an end because a beginning and middle precede it. The three positions are not three things each able to stand by itself, but three names that delimit one another. Remove two, and the remaining one cannot even acquire its name.

Since beginning and end cannot be found, a middle has no boundary on which to stand. It is not negated by anyone; nothing is there to draw its boundary. Qingmu therefore concludes: “In saṃsāra there is no beginning, middle, or end.”

On a line without two endpoints, no midpoint can be measured. This is not inaccurate measurement; there is nothing to measure.

Birth and Aging-and-Death: Each of the Three Positions Tried in Turn

Once beginning, middle, and end prove incoherent, the argument changes to a question closer to the body: take birth and aging-and-death—which comes first?

There are only three possibilities: birth first, aging-and-death first, or both at once. The verse and commentary try each in turn.

First route: birth first, aging-and-death afterward. The verse proposes, “If birth existed first and aging-and-death afterward,” then immediately lists two consequences: “There would be birth without aging-and-death, and aging-and-death without birth.”

If birth were first established on its own, aging-and-death would not yet have arrived, so this would be a birth containing no aging-and-death. Yet Qingmu says plainly: “By principle, birth should include aging-and-death, and aging-and-death should include birth.” Birth and aging-and-death stand as a mutually dependent pair; remove one, and the other no longer becomes that side of the pair. A birth that never ages or dies is not the birth of which we speak. The aging-and-death that later appears separately would likewise become aging-and-death that did not arise from birth: “Nor does aging-and-death exist without being caused by birth.”

Second route: aging-and-death first, birth afterward. Reverse the order. The verse proposes, “If aging-and-death existed first and birth afterward,” with the consequence: “It would be without a cause, and there would be aging-and-death without birth.”

Birth still comes later, so from where did the aging-and-death that appears first come? Qingmu says, “Aging-and-death would be without a cause,” and gives the reason directly: “Because birth comes later.” Then he presses with another question: “Without birth, how could there be aging-and-death?” How could something not yet born already have its aging and its death?

Reversing the sequence saves nothing; it merely moves the fault to the other side.

Third route: both at once. Once neither sequence succeeds, the questioner changes the claim and says that the two occur simultaneously. Qingmu first marks this as a third proposal forced by the previous failures, then immediately judges: “This too incurs a fault.”

The verse says: “Birth and aging-and-death cannot coexist at the same time”; “death would then exist at the time of birth, and both would be without a cause.”

The fault has two layers, and Qingmu distinguishes them clearly. The first is contradiction of function: “By principle there should be existence at birth and nonexistence at death.” At birth there is existence; at death there is nonexistence—these are opposed phases. If death were already present at birth, existence and nonexistence would occur at the same time. Qingmu judges: “This is not so.”

The second layer is harder to notice: mutual causation fails. “If they arise at the same time, they do not mutually serve as causes.” If two things truly appear simultaneously, one cannot say which produces which or which waits upon which. Qingmu uses a homely analogy: “Just as a cow’s two horns emerge at the same time and are not causes of one another.” A cow’s two horns grow out together; the left horn does not cause the right, nor the right the left.

The exact scope of this analogy must be seen. It does not say that a cow’s horns do not exist, but that simultaneous appearance by itself contains no relation of mutual causation. People often mistake events happening together for events being causally related; this cuts off that step.

Once the Three Routes Are Exhausted, What Is Left to Say?

After all three positions have been tried, the verse concludes: “If beginning, end, and simultaneous occurrence are all impossible,” then “why continue conceptual proliferation and assert that birth, aging, and death exist?”

Qingmu presents this conclusion in two parts. The first is the result: because reflection on birth, aging, and death reveals a fault in all three, they are “unborn and ultimately empty.” The second is a rebuke, and its target is exact—he does not rebuke speaking of birth, aging, and death, but “asserting determinate marks.”

The distinction here must be very fine. The chapter does not say that no one is born or grows old and dies in the world. The sūtra speaks of beings and saṃsāra, and Qingmu likewise speaks of beings in saṃsāra. What is rebuked is the determinate mark: the view that takes birth as one thing standing on its own and aging-and-death as another thing standing on its own, then asks which of these two comes first. The failure of all three positions shows precisely that the two were never that kind of thing.

The words “conceptual proliferation” fall heavily. Once every position has been tried, to continue discussing their priority is no longer to say anything; it merely spins.

Not Only Saṃsāra: All Dharmas Are Like This

If one imagined at this point that the chapter treated only saṃsāra, its final two verses open the claim outward.

The verse gives a list: “All causes and fruits, marks and what bears a mark,” then “experience and experiencer and every dharma whatsoever.” It says that “not only in saṃsāra can the prior limit not be found,” and concludes: “The prior limit of all such dharmas likewise does not exist.”

Cause and fruit—which comes first? Mark and what bears a mark, the marking and marked—which comes first? Experience and experiencer, the experienced dharma and one who undergoes it—which comes first? Every pair can be subjected to the three routes in this chapter, and every pair yields the same result: before fails, after fails, and simultaneity fails.

Qingmu makes this layer explicit at the end: these “all lack a prior limit; it is not only saṃsāra that lacks one.” The discussion spoke only of saṃsāra up to this point “in order to disclose the teaching briefly.”

Saṃsāra is only an example—one that everyone experiences personally and can enter most easily. The example is given so that one learns the method there and then applies it elsewhere. This also answers the opening question. The sūtra speaks of beings and saṃsāra at the worldly level; when it says that the prior limit cannot be found, it says that the thing spoken of as existing has no beginning able to stand. The two statements do not contradict each other: one speaks of the event, the other of what lies beneath the event.

Comparative Readings: Alternative Meanings in the Transmitted Lines

The Bhāviveka-line Prajñāpradīpa Commentary adds another layer: before the noble path’s counteraction arises, birth, aging, and death continue without interruption and in turn serve as causes, while their first arising is indeterminate. Therefore they are said to have neither limit nor boundary—yet this is spoken in conventional truth, not to establish a new limitless entity in ultimate truth. This addition is useful because it blocks one reading: turning “beginningless” around and grasping it as a new thing.

The Sthiramati-line Mahāyāna Madhyamaka Commentary applies each of the three routes to cause and fruit and to mark and what bears a mark. If fruit exists first and cause afterward, fruit is causeless; if cause exists first while fruit does not yet exist, cause and fruit do not cohere; if both dharmas exist simultaneously, the so-called nature of cause and fruit lacks substance. The same explanation applies to mark and what bears a mark. It does exactly what the final two verses invite—move the same examination to another pair and run it again.

Each line adds within its proper scope, and neither alters the text of Kumārajīva’s verses or Qingmu’s extended commentary.

Carried Forward: Jizang’s Mirror

The foregoing is the meaning of the chapter itself. What follows uses the eye of the Sui-Tang Three Treatise lineage as a sidewise illumination—a mirror is not the treatise, but another way of reading the same verses.

In his Commentary on the Treatise on the Middle, Jizang opens four intentions in the word “beginningless.” First, it does not mean that saṃsāra is so long that its beginning is difficult to know; the beginning of saṃsāra cannot be found at all, so saṃsāra has no root. Second, beginninglessness relinquishes both sides: to say “beginningless” merely distinguishes the absence of a beginning and does not establish another thing called beginninglessness. Thus none of the statements involving beginning and beginninglessness functions, enabling awakening to and entry into true reality. Third, no beginning means no end and no middle, just as a tree without roots also has no branches and leaves. This reveals that the six destinies were never born and are not now extinguished, that there is neither saṃsāra nor nirvāṇa. Fourth, for pedagogical guidance one may still say that saṃsāra is long, prompting followers of both vehicles to become disenchanted, diligently cultivate contemplation, and sever the afflictions.

He also maps the same term onto the refutation of four views. Saying “beginningless” refutes the view of a beginning; since it is called beginningless because no beginning exists, not because a beginningless entity exists, it refutes the view of beginninglessness; since beginning is absent and beginninglessness likewise absent, there is no end, refuting the view of an end; since both divisions are absent, there is no passage through a middle, refuting the view of a middle. Once the four views are refuted, one is enabled to awaken to and enter true reality.

As for the simultaneous route, he divides causelessness into two layers. The first is rigorous exclusion: if death exists at birth, birth is absent at death, so there are fundamentally no two dharmas able to cause one another. The second is provisional allowance: even if birth and death are conceded to coexist simultaneously, they remain like a cow’s two horns and do not cause one another.

This mirror connects the chapter to the Three Treatise lineage’s interpretive structure of refuting error and disclosing correctness. The levels of four intentions, four views, and two forms of causelessness are Jizang’s pedagogical expansions, not the original words of Nāgārjuna’s verses or Qingmu’s extended commentary. The readings are recorded side by side without adjudication, each returned to its own place.

When people suffer, a sentence always rises in the mind: Who did this?

There are only a few possible answers. Perhaps I brought it on myself—it is all my fault. Perhaps another gave it to me—if not for that person, how could I have come to this? Perhaps both sides had a hand in it—the other began it, but I too was wrong. Or perhaps there is simply no reason: it happened; it was only bad luck.

At the very opening of the chapter, the questioner arranges these four explanations in a neat row:

“Self-made and made by another, jointly made and made without a cause— 　when suffering is explained in these ways, none applies to the fruit.”

Four explanations are swept away by the single phrase “none applies to the fruit.”

First see clearly what is being swept away.

Four Answers Are Swept Away, Not Suffering

When explaining this sentence, Qingmu adds a crucial passage: “Beings incur suffering through many conditions; weary of suffering, they wish to seek its cessation. Not knowing that suffering actually exists through causes and conditions, they form four errors.”

This passage has three layers, and none can be omitted. First, suffering exists—beings truly suffer. Second, suffering has a way it comes about: it is “incurred through many conditions.” Third, suffering can be brought to cessation: being weary of it and wishing it to cease is not itself wrong.

Where, then, is the error? In “not knowing that suffering actually exists through causes and conditions.” Not knowing that suffering depends on causes and conditions, one searches outside them for someone who made it. Failing to find such a one, the mind guesses in turn at self-making, making by another, joint making, and causeless making. The four answers are four branches of the same ignorance.

Thus all the refutations that follow never refute the claim that suffering has causes, nor the claim that suffering can cease. What they refute is the maker of suffering installed behind causes and conditions.

Self-Making: If It Truly Produced Itself, Conditions Would Have No Work to Do

The first answer undergoes examination. The verse says:

“If suffering were self-made, it would not arise from conditions; 　because these aggregates exist, those aggregates arise.”

The force of the verse lies entirely in the five words “it would not arise from conditions.”

Qingmu exposes the weight of “self”: “Self means arising from intrinsic nature.” Self-making does not mean that my present suffering is related to my earlier conduct. It means that suffering produces itself through its own substance. If itself is enough, then no other condition is needed.

But that is not what happens. Qingmu continues: “The later five aggregates arise through the preceding five aggregates.” The body and mind of this life depend upon the body and mind of the preceding life. Since preceding aggregates must be awaited, the later ones do not arise through themselves. Since they do not arise through themselves, the words “self-made” fall away.

Notice that nothing here denies continuity between before and after. Precisely because the continuity is real, “self-arising” is unestablished. What is refuted is not cause and fruit, but the “self” that does not wait for cause and fruit.

Making by Another: Apart from Thread, Is There Still Cloth?

The questioner immediately seizes on this continuity and objects: since “these five aggregates make those five aggregates, it is made by another,” is this not precisely making by another?

The respondent’s answer is conditional:

“If one says that these five aggregates are different from those five aggregates, 　only then should one say that suffering is made by another.”

Before saying that suffering is made by another, one must prove that these aggregates and those are truly two things wholly unrelated to one another.

So Qingmu offers an analogy from weaving: “If thread and cloth were different, cloth should exist apart from thread. If cloth does not exist apart from thread, then cloth is not different from thread.”

This analogy must be viewed slowly. Thread is yarn; cloth is a woven bolt. Cloth is of course not the same as thread: threads are individual strands, cloth is a whole expanse, and they differ to touch and in use. But can all the thread be removed while a bolt of cloth remains? No. Pull out the thread and the cloth comes apart.

Thus cloth and thread are neither one thing nor two things each able to stand by itself. Their difference is a difference that cannot be separated.

The preceding and later five aggregates are just so. Today’s body differs in appearance, circumstances, and mental activity from the body of the past; it is not wrong to call them “different.” Yet if the preceding aggregates are entirely removed, the later ones have nowhere to land. A difference that cannot be removed cannot support making by another. Other-making requires a maker standing outside, wholly unrelated to me; no such outside can be found here.

Then Does a Person Make and Experience His Own Suffering?

The questioner changes levels. Instead of discussing the dharma of the aggregates, he discusses a person: “Each person makes and experiences his own suffering.”

This question lies closest to our ordinary feeling. The verse responds with a question:

“If a person makes suffering himself, apart from suffering, where is the person?”

Apart from suffering, where is that person?

Qingmu makes the line still tighter: “Apart from the suffering of the five aggregates, where is there a separate person who could make suffering himself? One should state who this person is, but he cannot be stated.” If you say a person makes suffering himself, point the person out. Yet the person cannot be pointed out.

He cannot be pointed out because apart from the five aggregates, there is nowhere to point to a person. What is called “I” is posited precisely upon these five aggregates. If a ready-made “I” first existed outside the aggregates and outside suffering, and then this “I” set to work making suffering for itself, on what was that “I” established before it made suffering?

Giver and Receiver: Both Ends Vanish

The questioner retreats one more step: I did not make it; someone else made it and gave it to me.

The respondent follows the sentence and divides it into two ends.

First ask about the receiver:

“If suffering is made by another and given to this person, 　apart from suffering, where is this person to receive it?”

Qingmu says directly: “Apart from the five aggregates there is no such person to receive it.” To whom is suffering to be handed? To the person who remains apart from the five aggregates. But apart from the five aggregates, the person who would take it disappears.

Then ask about the giver:

“If suffering is made by that person and brought to be given to this one, 　apart from suffering, where is the person who could give it to this one?”

Qingmu again presses to the end: “If such a person exists, one should state his mark.” If you say there truly is such a “that person” who makes suffering, carries it over, and hands it across, describe his mark.

One hands it over and another receives it; each end first requires a person standing apart from suffering. Neither can be pointed out, so the transfer has no way to occur.

Change the Pronoun, and the Error Remains the Same

At this point, the verse connects self-making and making by another:

“If self-making is not established, how can that person make suffering? 　If that person makes suffering, it is likewise called self-making.”

Qingmu identifies the hinge: “This and that exist in mutual dependence.”

This and that are not two kinds of thing, but two positions. Whoever stands on the speaking side is “this”; the one on the other side is “that.” Change where you stand, and this and that exchange places. So when you say that person makes suffering, viewed from that person’s side it is precisely his making his own suffering—“for that person it is likewise called self-made suffering.”

But the route of self-made suffering has already failed. Qingmu therefore concludes: “Self-made suffering has already been refuted. Since the self-made suffering you accept is not established, making by another is likewise not established.”

The same difficulty does not vanish when the pronoun changes. This sentence is worth remembering: blaming oneself and blaming another look like opposed paths, but beneath them lies the same assumption—a ready-made maker of suffering. When the assumption falls, both paths fall together.

A Knife Cannot Cut Itself

The final verse returns from the approach of the person to the approach of dharmas:

“Suffering is not called self-made; a dharma does not make itself. 　Since that has no substance of its own, where is that which makes suffering?”

Qingmu offers the chapter’s most memorable analogy: “Just as a knife cannot cut itself.”

A knife can cut things; that is what a knife does. But however sharp it is, it cannot cut itself. Cutting requires two ends, cutter and cut; one thing serving as both at once cannot be established. A dharma is likewise so: it cannot use itself as material and make that very dharma itself.

As for making by another, Qingmu’s reason is: “Apart from suffering, that has no intrinsic nature.” Apart from suffering, “that” has no substance of its own. If one forcibly establishes an intrinsic nature of that apart from suffering, another question follows: “That itself is suffering; how could suffering make suffering itself?” The “that” you establish is ultimately suffering. The circle returns to the start: suffering is still making suffering.

Joint Making and Causeless Making Gathered Together

The questioner has two final moves. If neither self-making nor making by another works, surely it is jointly made.

The verse answers cleanly:

“If suffering made by this and that were established, jointly made suffering should exist; 　neither this nor that makes anything—how much less could there be making without a cause!”

Joint making combines two things. But to combine them, each must first exist. Making by this is unestablished and making by that is also unestablished; with nothing on either side to combine, the combination is likewise absent.

As for causeless making, Qingmu says: “Self-making and making by another already incur faults; how much more so making without a cause. Causelessness incurs many faults, as explained in the chapter that refutes act and agent.” Self and other can at least present a line of reasoning, yet still fail. Removing all dependence on causes and saying that suffering simply appears from nowhere only multiplies the faults. Moreover, to call suffering causeless simultaneously cancels its cessation: what arises without a cause has no basis on which it can be made not to arise. This directly conflicts with the opening statement that beings, “weary of suffering, wish to seek its cessation.”

Not Only Suffering

The final verse extends this measuring line outward:

“It is not only in speaking of suffering that the four theses are unestablished; 　for all external things, the four theses likewise are unestablished.”

Qingmu first distinguishes one matter: “Although the Buddhadharma says that the five aggregates subject to appropriation are suffering, some non-Buddhists take painful experience as suffering.” The suffering spoken of in the Buddhadharma is the whole of the five aggregates subject to appropriation; non-Buddhists refer only to the painful kind of experience. Whether suffering is taken broadly or narrowly, none of the four modes of making can stand.

Then he extends the scope to the end: “For all external things—all dharmas such as earth, water, mountains, and trees—they are likewise unestablished.” Earth, water, mountains, trees—whatever one can ask “How did this come about?” of likewise cannot be said to make itself, be made by another, be jointly made, or be made without a cause.

This chapter is therefore more than a debate about suffering. Applied to any dharma, the question “Who made it?” yields no answer. Only the question “Through what conditions does it exist?” yields an answer.

What Can a Reader Carry Away from This Chapter?

First, when suffering comes, one can ask “Who did this?” one time less. The four answers have been tried one by one, and none can stand. To keep pursuing them merely changes pronouns.

Second, blaming oneself and blaming another are not two opposed states of mind. “This and that exist in mutual dependence”—they share the same assumption. Release the assumption and both ends loosen together.

Third, refuting the four modes of making does not declare suffering irrelevant or cessation impossible. The opening sentence remains standing throughout: “Beings incur suffering through many conditions; weary of suffering, they wish to seek its cessation.” Suffering exists through many conditions; precisely because it depends on conditions, when the conditions disperse it does not arise. The place to begin is with conditions, not with the unfindable maker of suffering.

Carried Forward: Jizang’s Mirror

The foregoing is the meaning of the chapter itself. What follows uses the eye of the Sui-Tang Three Treatise lineage as a sidewise illumination—a mirror is not the treatise, but another way of reading the same verses.

When explaining the chapter title in his Commentary on the Treatise on the Middle, Jizang says that no affliction of the world exceeds suffering. Every teaching urgently seeks release, yet failure to understand suffering’s cause gives rise to four erroneous interpretations. Clinging to suffering as having determinate nature prevents the fruit of suffering from ceasing and instead produces more causes of suffering. The examination of suffering therefore shows that conditioned suffering has no determinate nature, enabling its fruit to cease and its causes to arise no more.

He also constructs a doctrinal outline for the whole chapter. The eight verses of refutation divide into two: the first seven explain neither self-making nor making by another; the last explains neither joint nor causeless making. The first seven divide into three layers: two verses first treat neither self nor other at the level of dharmas; the next three treat neither self nor other at the level of persons; the final two conclude person and dharma together. This outline precisely follows the chapter’s movement from the five aggregates as dharmas to the person and then to their joint conclusion.

For the two concluding verses, Jizang distinguishes the approaches of mutual dependence and reciprocal identity. Mutual dependence says that other exists by depending on self; since self is unestablished, other is also unestablished. Reciprocal identity says that what is other for another is precisely self. He also uses the two examples of a knife that does not cut itself and an eye that does not see itself to explain that a dharma does not make itself, and uses the lack of substance in “that” to explain why another cannot make suffering.

The knife analogy is already present in Qingmu, while the eye not seeing itself is added by Jizang. The layered doctrinal outline and the interpretive structure of determinate nature and cessation likewise belong to the Three Treatise lineage’s reading, not to the original words of Nāgārjuna’s verses and Qingmu’s prose commentary. The readings are recorded side by side without adjudication, each returned to its own place.

Across the Treatise on the Middle, chapters refute going and coming, the six faculties, fuel and fire, and the prior abider. By this point, most readers have learned one thing: whatever is established as “just what it is in itself” cannot stand.

The subject here is “Examination of Formations.” What are formations? Qingmu speaks clearly: “Formations name the five aggregates, because they arise from formative activity.” He also says: “What are called formations are activities of body, speech, and mind.” On one side stands this current of body and mind—form, feeling, perception, formations, and consciousness—flowing thought-moment by thought-moment. On the other are the bodily, verbal, and mental actions people perform every day. Both meanings inhabit the word “formations.” This chapter examines this thing most intimately bound to us and most readily taken as real.

Yet the sentence with which the chapter finally lands does not refute formations. It refutes emptiness.

I. The Questioner’s Step: Since the Buddha Calls Them False, There Must Be Something False

The questioner opens by bringing forward the Buddhist scriptures:

“As the Buddhist scriptures teach, falsity is the mark of delusive grasping; because formations are delusively grasped, they are called false.”

Qingmu states the intended meaning plainly: “The first truth, namely nirvāṇa, is not the mark of delusive grasping.” Nirvāṇa is the first truth and is not delusively grasped; the formations opposed to it are therefore false and delusively grasped. Since the Buddha himself says so, it is a matter certified by the Buddha that “formations possess the false mark of delusive grasping.”

Look closely at this move. The questioner has not misquoted the Buddha. He has merely assigned a real position to each term in a scriptural contrast—the true and the false, nirvāṇa and formations. If something true stands over there, something false must stand here; and if the false really is false, “falsity” itself becomes something real.

This is among the hardest forms of grasping to detect: it is not grasping existence, but grasping the thing judged false as something that nonetheless stands there solidly and securely.

II. The Respondent Asks Only One Thing: What Exactly Has Been Grasped There?

The answering verse does not debate the scripture. It merely asks:

“Of what is false and delusively grasped, what within it is there to grasp? The Buddha taught such a matter because he wished to reveal the meaning of emptiness.”

Qingmu follows the question with extreme brevity: “If a dharma bearing the mark of delusive grasping is itself false, what is there within these formations to be grasped?”

Delusive grasping must grasp something. You say that formations are what is delusively grasped, so bring out that “object grasped.” Yet once you have acknowledged that formations are false, what real object remains to be grasped within something false?

Thus “false” is not a new identity fastened onto formations; it removes the identities already fastened to them. The Buddha speaks of false and delusive grasping not to establish a false dharma in opposition to a true one, but to let us see that no graspable thing was ever here. Qingmu delivers the conclusion: “Since the Buddha spoke in this way, one should know that he taught the meaning of emptiness.”

With this one question, the questioner’s framework of opposites collapses. The long passage that follows explains the same matter one aggregate at a time.

III. The Five Aggregates: From Infancy to Old Age, Where Is That “It”?

The questioner presses: how do we know that all formations are empty? The respondent’s general answer is two clauses: “Formations arise, cease, and do not abide; because they lack intrinsic nature, they are empty.”

He then begins with form. This is the chapter’s easiest passage to follow—and the one that cuts closest to the heart.

The form of infancy is not the form of crawling, the form of crawling is not the form of walking, and the sequence continues through adulthood and old age. Qingmu says: “Because form does not abide from thought-moment to thought-moment, a determinate nature cannot be distinguished and found.”

Then comes the dilemma: are the infant’s form and the old person’s form one, or two?

If they are one, it must remain infancy forever. Qingmu adds an analogy: “It would also be like a lump of clay that always remained a lump of clay and never became a jar.” If clay really were determinate clay, it could not become a jar; if form really were determinate form, the infant could not grow old. Whatever is determinately itself cannot move.

If they are two, the infant cannot become the crawling child and the crawling child cannot have been the infant; segment by segment, none has anything to do with another. Qingmu suddenly offers a thoroughly human line: “The dharma of kinship would be lost: there would be neither father nor child.” If the earlier and later were truly two things, only the infant-segment would have acquired a father; none of the later persons could count as his child. Parent and child, teacher and student, you and yesterday’s you—these relations all hold through “neither one nor different.” Once they fall into determinate identity or determinate difference, human relations are the first things to scatter.

The questioner retreats one step: there is no need to say one and no need to say different; surely “continuity” will do—the infant’s form ceases and the next form then arises. The respondent presses again: does it continue once ceased, or before ceasing? “If the infant’s form has ceased, how could continuity exist? There would be no cause.” Once it is wholly extinguished, what can serve as cause? Qingmu uses fire as his analogy: “Although firewood capable of burning may remain, once the fire is extinguished there is no continuity.” The wood remains, but once the fire has gone out there is no next portion of fire. Conversely, if it continues without ceasing, the infant’s form permanently abides in its original mark, and there is no continuation to speak of.

The questioner retreats again: I do not speak of cessation or noncessation, only of “similar arising without abiding.” The respondent says that if what “arises similarly” remains a determinate form, there will merely be tens of millions of determinate forms arrayed one after another; that is still not continuity.

At this point the respondent stops pressing questions and gathers the passage with three analogies. “It is like a plantain tree: when one seeks its solid heart, none can be found; there are only sheath and leaf.” Peel away one layer and there is another; peel to the end and no heart appears. Or it is like a lamp flame—“no determinate form can be distinguished and found within a lamp flame.” It looks like a single unmoving flame, yet no instant holds the same one.

What, then, is the “body of this person” spoken of in daily life? Qingmu gives the line that runs through the whole chapter: “It is said to exist only by worldly convention.” This is not nonexistence. It is existence only at the level of speech.

For feeling, perception, and consciousness, the respondent gives one passage and one analogy each. Feeling is “like the continuity of a stream”: it looks like one river, though not a drop abides. It is only because there is felt awareness that painful, pleasant, and neither-painful-nor-pleasant feeling are said to be present in the body. As for perception, “perception arises dependent on names and marks; apart from names and marks, it does not arise.” Qingmu compares it to a shadow: “Because a form is present, a shadow is present; without the form, there is no shadow. A shadow has no determinate nature.” Seek a shadow apart from its form and it cannot be found; perception apart from external names and marks is established no more successfully.

Consciousness is hardest to dismantle, so Qingmu questions it most closely. Is this consciousness located in form, in the eye, or between them? Nothing is determined. Is the consciousness recognizing this person one with or different from that recognizing another person? This too cannot be distinguished. He therefore uses a street performer as his analogy: “It is like an entertainer who holds a pearl in his mouth, takes it out, and shows it again: people wonder whether it is the original pearl or another one.” A conjurer puts a pearl in his mouth and then spits one out to show you; you wonder, is it the same pearl as before? In the same way, thought arises after thought and one wonders: is it the same “I” thinking as a moment ago? The conclusion is: “Because consciousness does not abide, it lacks intrinsic nature and is false like an illusion.”

IV. Karmic Formations Are the Same: What Increases and Decreases Cannot Be Determined

After the “formations” of the five aggregates, Qingmu turns to the “formations” of action: “What are called formations are activities of body, speech, and mind.” They divide into pure and impure. Harming sentient beings and craving attachment are impure; not harming sentient beings, speaking truthfully, and not craving attachment are pure.

What matters is his next observation. For those cultivating pure conduct, birth among humans and devas brings karmic fruition; once that fruition is received, the activity “decreases,” and when performed again, it “increases.” For those performing impure conduct and falling into the three evil destinies, the same applies: after fruition, it decreases; performed again, it increases. Therefore, “formations do not abide because they increase and decrease.”

He adds an analogy: “It is like a person who is ill: with suitable care the illness is cured, but without such care it gathers again.” With care it heals; without care it builds up again. Karmic formation is not a fixed stockpile but a process waxing and waning according to conditions. What can increase and decrease is not determinately existent.

Then comes the pivot of the passage: “Through conventional truth one is able to see the truth of the ultimate meaning.”

This sentence deserves a pause. After so much refutation, one can easily imagine that the worldly level must be thrown away. It must not. Qingmu immediately counts the twelve links one by one—ignorance conditions formations and so on through birth, aging, death, sorrow, lamentation, pain, and distress. All these are taught at the level of conventional truth. It is precisely through this sequence of cause and effect spoken by conventional truth that one reaches the cessation of ignorance and the nonaccumulation of formations; only then are identity-view, doubt, and attachment to precepts, abandoned by seeing, and craving, anger, and the rest, abandoned by cultivation, cut off. Conventional truth is the ladder; the truth of the ultimate meaning is where it climbs. Without the ladder, there is no ascent.

What follows complete cessation? Qingmu says: “The body of the five aggregates is ultimately exhausted without remainder, and emptiness alone remains.” Then he immediately returns to the opening question: “Therefore, because the Buddha wished to reveal the meaning of emptiness, he said that formations are false.”

The circle is complete. When the Buddha says “false,” his intent is precisely emptiness.

V. Non-Nature Too May Not Be Established as a Thing

With the prose explanation complete, Nāgārjuna gives the verse:

“Because dharmas undergo change, we know that all lack intrinsic nature; no dharma of non-nature exists either, because all dharmas are empty.”

The first half gathers the entire preceding argument. Precisely because the infant can become old, one sees that it never abided in intrinsic nature—Qingmu says, “Although dharmas arise, they do not abide in their own nature; therefore they are said to lack intrinsic nature.” Change itself is evidence of the absence of intrinsic nature.

What truly demands attention is the latter half: “No dharma of non-nature exists either.”

The questioner seizes on this immediately: since dharmas lack intrinsic nature, does that not mean there is a “dharma of non-nature”? What fault would that entail?

Qingmu’s answer has one sentence of greatest weight: “Non-nature is spoken of only to refute intrinsic nature.”

Non-nature is not a newly discovered thing; it is a remedial medicine. When someone grasps intrinsic nature, non-nature is taught. Once intrinsic nature is refuted, non-nature too has nothing to oppose. If non-nature is gathered back into a real dharma, one can no longer say that all dharmas are empty—“If this dharma of non-nature existed, one could not call all dharmas empty. If all dharmas are empty, how could there be a dharma of non-nature?”

Still unsatisfied, the questioner answers with a verse:

“If dharmas have no intrinsic nature, how can one speak of infancy and so on through old age as undergoing many changes?”

If you say that they lack intrinsic nature, how can you still speak of all the differences from infancy to old age? The answering verse cuts in both directions:

“If dharmas possessed intrinsic nature, how could they change? If dharmas lacked intrinsic nature, again how could change belong to them?”

The first cut meets the questioner: what has intrinsic nature cannot change. Qingmu says: “Intrinsic nature means determinate existence and cannot change.” Then he offers analogies: “Genuine gold cannot be changed; likewise, the nature of darkness does not become light, and the nature of light does not become darkness.” The second cut returns toward his own side: non-nature too cannot be used as a thing that undergoes change. “Change” stands on neither side. This is precisely to say that change is the appearance of dependent arising, not a property owned by either side.

Then the pressure goes one layer deeper:

“This very dharma does not change, nor does another dharma change; just as adulthood does not become old age, and old age does not become adulthood.”

If this very dharma changes, then old age must become old age—where could an identical thing change to? If another dharma changes, adulthood and old age are wholly separate things, so adulthood cannot become old age.

The questioner produces the counterexample anyone might offer: “We now see with our own eyes that a young person grows old with the passing days, months, and years.” I personally see the young become old; does that not count as change?

The answering verse replies with something from the kitchen:

“If this very dharma itself changed, milk should itself already be curd; apart from milk, what other dharma could become curd?”

If the same dharma could change by itself, milk should instantly be curd, without awaiting causes and conditions. Yet “milk and curd have many different marks”; milk is indeed not curd. But if curd is a wholly different thing, where apart from milk will you find it?

That we see a youth become old is true; the error is assigning “change” to some determinate thing as its bearer. Qingmu closes with care: “Therefore one should not cling one-sidedly.” Refusing either extreme is not vagueness; it is refusing to force an occurrence of dependent arising into the drawer of either “identical” or “wholly different.”

VI. The Last Move: Emptiness Is No Place to Settle

At this point, the questioner makes the chapter’s final and most subtle retreat: “You have refuted identity and refuted difference, yet emptiness remains. Emptiness is itself a dharma.”

You have refuted both identity and difference, but surely one thing remains: “emptiness.” That must be the dharma of emptiness.

The answering verse says:

“If a nonempty dharma existed, an empty dharma should exist; but no nonempty dharma truly exists, so how could an empty dharma be found?”

Qingmu’s explanation is spare: “Since a nonempty dharma does not exist, there is no mutual dependence; without mutual dependence, how could there be an empty dharma?” Emptiness is spoken in contrast with the nonempty. Once the nonempty is thoroughly refuted, emptiness has nowhere to stand. Emptiness was never the one thing left behind; it is a formulation established through mutual dependence.

The questioner makes a final turn that sounds almost as though he now understands: if there is no mutual dependence, there is no mark; if there is no mark, there is no grasping—is that not precisely the teaching of emptiness?

Yet the respondent’s final verse does not nod along:

“The Great Sage taught the dharma of emptiness in order to abandon all views; if someone once more sees that emptiness exists, such a person cannot be transformed by any buddha.”

Qingmu explains the function of the Buddha’s teaching of emptiness: “The Great Sage taught emptiness in order to refute the sixty-two views and the many afflictions of ignorance, craving, and the rest.” Emptiness is used to abandon views. But once you take “I have abandoned views” as something attained, emptiness turns from medicine into illness.

He gives two analogies in succession. First, medicine: “It is like an illness that must be treated with medicine; if the medicine itself becomes an illness, the person cannot be treated.” Illness can be treated with medicine; when the medicine itself becomes the illness, there is no second medicine. Second, fire: “It is like a fire arising from firewood, which can be extinguished with water; if fire arose from water, with what could it be extinguished?” Qingmu states the point: “Emptiness is like water and can extinguish the fires of the many afflictions.” The water is good water; it simply cannot extinguish the fire that arises from itself.

What does a person grasping emptiness look like? Qingmu describes the person concretely: weighed down by karmic offenses, deeply attached, and dull in wisdom, he gives rise to a view about emptiness, saying either that emptiness exists or that it does not. Once he falls into existence and nonexistence, afflictions arise again. The most difficult part follows: “He will say, ‘I have long known this emptiness already; apart from this emptiness there is no path to nirvāṇa.’” You try to teach him, and his first answer is, “I already knew that.” Emptiness has become one of his possessions, and so becomes a wall through which no one can pass him anything.

The last line is coldest of all: “To claim that one can attain liberation apart from the gateways of emptiness, signlessness, and wishlessness is nothing but words.” One may speak lucidly and at length about the three gateways to liberation—emptiness, signlessness, and wishlessness—but talk of liberation apart from entering them remains only talk.

VII. What Readers Can Carry Away from This Chapter

First, return to this very body and look. Infancy, crawling, childhood, adulthood, old age—this sequence is each person’s own life. If “I” were determinate, this moment could not age; because it ages, it was never fixed for a single moment. This is not merely a proposition to memorize but a mirror available at any time.

Second, recognize the measure of “said to exist only by worldly convention.” Qingmu never says that body and mind do not exist; he says they are established only at the level of speech. Thus one still cares for the body and still cultivates pure conduct—“It is like a person who is ill: with suitable care the illness is cured” says exactly this. What is refuted is determination, not daily functioning.

Third, notice whether you have collected emptiness as a possession. Whenever the first thought to arise on hearing the Dharma is “I already know this,” Qingmu has written us into the text. Emptiness is water and medicine. Water should not give birth to fire; medicine should not become illness.

Carried Forward: Jizang’s Mirror

The meaning above belongs to the chapter itself. What follows is a side illumination from the Sui–Tang Sanlun lineage—a mirror is not the treatise, but another way of reading the same verses.

Jizang’s Commentary on the Treatise on the Middle first distinguishes the word “formations.” Asked whether calling the five aggregates formations means causal or resultant formations, he says it carries both meanings: the five aggregates are produced from karmic formations and so are called formations; these same five aggregates themselves arise, cease, and act and so are called formations. This classification orders the two identities of “formations” more systematically than Qingmu does.

For “no dharma of non-nature exists either,” he labels the move the provisional use of intrinsic nature to refute non-nature. Only if intrinsic nature originally existed could non-nature be spoken of in opposition; when intrinsic nature itself does not exist, whence non-nature? He adds that the verse contains multiple implications.

At the final verse, he lays out the stages of the Buddha’s teaching of emptiness more finely. Simple emptiness is taught to free beings from the view of existence; doubled emptiness is taught to refute the view of emptiness; the eighteen emptinesses refute the view of existence; and solitary emptiness refutes mutual dependence and thus also refutes a view of emptiness. To say that such a person cannot be transformed by the buddhas means that one attached to simple emptiness cannot again be transformed by simple emptiness, one attached to doubled emptiness cannot be transformed by doubled emptiness, and the same applies to attachment to the eighteen emptinesses or solitary emptiness. Every level of emptiness can be grasped as a new foothold, at which point the same medicine can no longer be used.

This mirror joins the chapter to the Sanlun lineage’s exegetical architecture of abandoning views and disclosing the right. Its layered classification is Jizang’s addition, not the wording of Nāgārjuna’s verses or Qingmu’s prose. The readings are recorded side by side without adjudication, each returned to its own place.

Two other exegetical traditions also transmit this chapter. At the verse “Because dharmas undergo change,” the Bhāviveka-lineage Prajñā Lamp Commentary says that seeing dharmas change and fail to abide permanently reveals their absence of essence, attesting the same point as this chapter. At the final verse it compares grasping emptiness to taking a purgative that instead becomes a grave illness; those grasping emptiness cannot be cured, and emptiness was expressly taught to abandon wrong views, so abandoning emptiness entails no fault. At “no dharma of non-nature exists either,” the Sthiramati-lineage Mahāyāna Commentary on the Middle Contemplation says that although all dharmas are taught as empty, there is no separate dharma of non-nature to obtain. The three transmissions differ in wording but agree that remedial language must not be reified. The Chinese translation of the Sthiramati lineage ends with this chapter; no corresponding alternative reading remains after it.

Perhaps no word in everyday speech invites less suspicion than “conjunction.” The eye meets color and seeing occurs; two people come together and something happens. Conjunction appears to be only a connective, not a claim—it says nothing exists, only that this and that have met.

This chapter takes that connective itself as its object of examination.

As the examination proceeds, one discovers that conjunction is not neutral at all. Whenever one speaks of conjunction, one has already silently admitted two things: there are two things able to stand on their own, and they were originally separate. Both admissions hide beneath the word, unnoticed by the speaker. This chapter turns them over and questions them one by one.

Arising from the Preceding Chapter: When Difference Is Already Absent, Where Can Conjunction Be?

The chapter does not begin with a verse but with a statement from Qingmu: “The commentator says: the preceding ‘Chapter Refuting the Faculties’ explained that seeing, the seen, and the seer are all unestablished.”

The preceding chapter has already examined seeing, the seen, and the seer one by one; none can stand. Qingmu then gives the reason for the present chapter: “Since these three matters possess no mutually differing dharmas, there is no conjunction.”

This sentence is the skeleton of the entire chapter. For conjunction to be established, there must first be two different things. Since the three matters do not even possess differences of their own, nothing remains to conjoin. Thus he says: “The meaning of nonconjunction will now be explained.”

The questioner immediately follows: “Why is there no conjunction among the three matters of the eye and so forth?”

The question is concrete. The verse names three matters: “seeing, the visible, and the seer.” Qingmu identifies them at once: “Seeing is the eye-faculty, the visible is the form-object, and the seer is the self.” Eye-faculty, form-object, and seer—the self—plainly look like three different things, standing there distinctly. How could one say they possess no difference? All the questions and answers that follow respond to this one challenge.

Each Occupies a Different Place: What Is Refuted Is Not Distance

The first answering verse says:

“Seeing, the visible, and the seer each occupy a different place; since these three dharmas are thus different, there is never a time when they conjoin.”

Qingmu descends to specific locations: “Seeing is the eye-faculty, the visible is the form-object, and the seer is the self,” and then maps them—“the eye is inside the body, and form is outside it.” As for the self, he does not decide the opponent’s case but preserves two positions: “Some say it is inside the body, while others say it pervades every place.”

At this point a thought can easily arise: does this merely mean that the three are too far apart to touch? If so, the chapter would be an essay about space, and saying that sight reaches a distance would refute it.

Look closely: the verse says, “These three dharmas are thus different.” What is refuted is not nearness or distance, but what the arrangement of “each occupying a different place” presupposes—three independently established essences, each occupying its own place and unrelated to the others. Distance is only one exposed corner of that arrangement. The real difficulty is this: if each is independently established, on what basis can they still conjoin into a single event of seeing?

Qingmu then presses this very difficulty into view. He asks: “Does one see through conjunction or without conjunction? Neither is so.”

First consider seeing through conjunction. If seeing occurs only after conjunction, then “wherever an object exists, a faculty and a self should be present.” Wherever color exists, an eye and a self would have to be there at the same time. The world is plainly not like this.

Then consider seeing without conjunction. If seeing can occur without conjunction, then the faculty, self, and object should see without error while each occupies a different place. Qingmu offers a fact anyone can test: “The eye-faculty is here and does not see a distant jar.”

Both gates collapse. Neither conjunction nor nonconjunction succeeds. This is precisely the chapter’s method: it does not try to prove that “seeing does not exist”; it simply walks down the two roads supplied by the claimant and lets him discover that neither leads through.

Add “Mind,” and the Same Difficulty Remains

The questioner changes his approach. He no longer speaks only of three matters but adds mind: “Because the four matters—self, mind, faculty, and object—conjoin, cognition arises.” His reason is ready at hand: we genuinely know a jar and know cloth; since cognition exists, the four things that produce it must exist.

This reasoning moves backward from the result and appears firm.

The respondent first refers back to the preceding chapter: “This has already been refuted in the ‘Chapter on the Faculties,’ but it should now be explained again.” He does not overturn the previous account; he simply traverses it again from another angle.

This time, the route is time. You say that the four matters conjoin and produce cognition. When does cognition arise?

If it arises only after seeing, “if it arises after seeing, cognition is useless.” The object has already been seen; what purpose does that cognition serve? Cognition arrives too late and becomes a superfluous procedure.

If it arises before seeing, “if it arises before seeing, there has been no conjunction.” The four things have not yet gathered; whence would cognition emerge?

Then say they arise simultaneously. The respondent refuses this too: “If they arise simultaneously, there is no mutual dependence.” What does mutual dependence mean? He immediately explains: “The jar exists first, seeing comes next, and cognition then arises.” The claim that four matters produce cognition depends precisely on a sequence—the object first, then seeing, then cognition. Compress all of them into one time, and the sequence disappears; once sequence disappears, “therefore it arises” becomes empty—what produces what?

None of the three times succeeds. Qingmu therefore gathers the argument: “Since cognition does not exist, seeing, the visible, and the seer do not exist either.”

He then gives the landing point of the passage: “All dharmas in this way are like illusions and dreams and have no determinate marks; how could conjunction exist? Since there is no conjunction, they are empty.”

This sentence must be read carefully. “Like illusion and dream” does not mean that nothing exists; it means “without determinate marks”—no fixed essence can be found. Precisely because nothing possesses a determinate mark, the claim that “two fixed things meet” has no starting point. Conjunction fails not because there are too few things, but because no thing is hard enough to be taken up and conjoined.

The Same Measure Extends to Defilement and Every Subject-Object Division

The next verse extends the point:

“Defilement, what is defiled, and the defiled one follow the same principle; the other entrances and the other afflictions are all likewise so.”

Qingmu fills in the parallels carefully instead of dismissing them with “the rest are the same.” He says, “hearing, the audible, the hearer, and the other entrances,” and again, “anger, what is hated, the angry one, and the other afflictions.”

Why list each group? Because what is refuted is not the eye but a structure—the threefold division into activity, object, and subject. In seeing: seeing, the visible, and the seer. In hearing: hearing, the audible, and the hearer. In defilement: defilement, what is defiled, and the defiled one. In anger: anger, what is hated, and the angry one. Almost every event people experience is divided into three parts in this fashion, and these three are then said to “conjoin” to form the event.

The chapter refutes this division itself. It therefore cannot refute only one group: if it refuted only the eye, the reader could take vision as an exception. Once three or four groups are set out, the measure reveals itself as universal. Defilement and anger matter especially. If the threefold subject-object division were only one epistemological account, perhaps little would be at stake; but the afflictions are established precisely upon these three parts—a defiled one encounters something defilable, and defilement occurs. When the three parts fail, affliction has no seat on which to sit.

The Turn: Conjunction Depends on Difference—What Does Difference Depend On?

The chapter now turns. Its first half examines conjunction; its second half suddenly speaks entirely about difference.

The turning point is this verse:

“Only differing dharmas should conjoin, yet seeing and the rest are not different; since a mark of difference cannot be established, how could seeing and the rest conjoin?”

Qingmu gathers it in one sentence: “Things are said to conjoin only because they differ, but the mark of difference in seeing and the rest cannot be found; therefore there is no conjunction.”

This is the chapter’s pivot. Conjunction has no feet of its own; it stands on “difference.” To say that two things conjoin, one must first say that they are two things—for if they are one, why would conjunction be needed? Thus, once a mark of difference fails to stand, conjunction collapses by itself; there is no need to refute conjunction item by item.

The edge of the treatise immediately widens:

“It is not only the mark of difference in seeing and the rest that cannot be found; all dharmas whatsoever likewise lack a mark of difference.”

Qingmu repeats it exactly: “It is not only the mark of difference among the three matters—seeing, the visible, the seer, and the rest—that cannot be found. All dharmas lack a mark of difference.”

The questioner therefore asks all the way down: “Why is there no mark of difference?”

The four rounds of question and answer below respond entirely to that question.

Difference Exists Dependent upon Difference: House, Beams, and Rafters

The first round says:

“Difference is different because of another differing dharma; apart from that differing dharma, there is no difference. If a dharma emerges from a cause, that dharma does not differ from its cause.”

Qingmu explains the first half: “What you call difference is named difference only because another differing dharma exists. Apart from a differing dharma, it is not called difference.”

The meaning is simple but firm: the word “different” is relational. To call this different always means different from that. Remove that, and this is no longer called different. Difference was never a mark carried on one thing; it is an event only between two ends. Remove one end and difference disappears.

The latter half expresses the same matter from another angle: “If a dharma arises from many conditions, it does not differ from its cause, because when the cause is destroyed, the fruit is destroyed along with it.”

The analogy is ready-made: “It is like a house that exists because of beams, rafters, and the rest. The house does not differ from its beams and rafters, because when the beams, rafters, and the rest are destroyed, the house is destroyed too.”

A house is gathered from beams, rafters, and the rest. If the house truly were something independent of the beams and rafters, it should remain when they were taken apart—but dismantle the beams and rafters, and the house is gone. Thus the house has no essence different from beams and rafters. Whatever emerges from causes and conditions is like this: no true boundary can be found between it and its causes.

Yet conjunction requires exactly that boundary.

The Fist and Five Fingers

The questioner retreats one step: “What fault is there if there is a fixed differing dharma?” Can one not simply admit a fixed difference?

The answer says:

“If apart from the difference upon which it depends there were still another difference, that difference should exist apart from all other differences as well. Yet apart from the difference upon which it depends, there is no difference; therefore no difference exists.”

Qingmu gives a demonstration one can make with one’s own hand: “If a difference belonging to a fist existed apart from the difference of five fingers, that difference of the fist should also distinguish it from jars and other things.”

It makes sense to say that a fist differs from five fingers. But if this “difference of the fist” were a real thing carried by the fist itself, it should hold in relation to anything—the fist should manifest that same difference in relation to a jar. Yet where is the fist apart from the five fingers? “Apart from the difference of the five fingers, a difference of the fist cannot be found.”

This analogy is powerful because it is right in the hand. Close the five fingers and call it a fist; open them and the fist is gone. No thing called a fist subsists apart from the five fingers. The so-called difference of the fist has meaning only when spoken in relation to the fingers; apart from that relation, it disappears.

Yet people habitually take the difference that exists only within a relation as something carried by the thing itself. See another person and feel he is innately separate from oneself; encounter a matter and feel it is simply that way in itself. The analogy of the fist says this: what is “just itself” cannot withstand opening the hand.

Can the Mark of Difference Stand by Itself?

The questioner retreats again, this time bringing forward scripture: “My scripture teaches that a mark of difference does not arise from many conditions. A mark of difference exists because a general mark is discriminated, and because a mark of difference exists, differing dharmas exist.”

His meaning is that a mark of difference does not arise from conditions but is obtained by discriminating a general mark; the mark of difference comes first, and differing dharmas then exist.

The respondent’s verse says:

“Within what already differs there is no mark of difference; within what does not differ there is none either. Since there is no mark of difference, there is no difference between this and that.”

Qingmu first turns the opponent’s words back on him. If you say that a mark of difference exists only through discriminating a general mark, does that not mean the mark of difference arises from many conditions?—“If so, the mark of difference arises from many conditions; what you describe is therefore precisely a conditioned dharma.” You intended to establish a mark of difference that did not arise from conditions; as soon as you state the case, you instead describe it as dependently arisen. One conclusion follows immediately: “A mark of difference exists dependent on differing dharmas and cannot be established alone.”

He then gathers the argument through two gates.

Within a dharma that already differs, there is no mark of difference: “Since a differing dharma already exists, what use is a further mark of difference?” It is already a differing dharma; what work remains for another mark of difference?

Within a dharma that does not differ, there is no mark of difference either: “If a mark of difference existed within a dharma that does not differ, it could not be called a dharma that does not differ.” If a mark of difference truly entered it, it would no longer be nondifferent.

The mark fits in neither place and therefore has nowhere to abide. Qingmu concludes: “Since there is no mark of difference, there are no dharmas of this and that.” Not only does the mark of difference disappear; even the division into this dharma and that dharma disappears with it. This and that were drawn apart by the mark of difference from the beginning.

Thus the chapter returns to its subject: “Since differing dharmas do not exist, there is no conjunction.”

Final Verse: Conjoiner, Time of Conjunction, and What Conjoins Are All Absent

“This dharma does not conjoin with itself, and a differing dharma does not conjoin either; the conjoiner, the time of conjunction, and the dharma that conjoins are all absent.”

Qingmu explains it from both sides.

An identical dharma does not conjoin with itself because it is one: “Just as one finger does not conjoin with itself.” Hold one finger out and ask whether it has conjoined with itself—the word “conjunction” loses all use here.

Differing dharmas do not conjoin either, because “the matter of difference is already established and no conjunction is needed.” Once they are already two different matters, difference stands there complete. What further work can conjunction do? If conjunction changes something, they were not originally different in that way; if conjunction changes nothing, conjunction is an empty assertion.

One gate treats identity, the other difference. Once both roads are exhausted, “when the matter is examined in this way, what conjoins cannot be found.”

The final sentence sweeps all three places away at once: “Therefore the conjoiner, the time of conjunction, and the dharma that conjoins cannot be found.”

Notice that all three fall together. When refuting conjunction, people often refute only the dharma conjoined while secretly retaining a conjoiner, or they retain “the moment of conjunction.” Qingmu retains none: the one that conjoins, the time of conjunction, and the dharma that conjoins are unfindable in all three places. Whenever refutation secretly leaves a position intact, sooner or later that position grows back into the original thing.

What Readers Can Carry Away from This Chapter

First, notice when you say “conjunction.” An eye meets an object, a person meets a person, conditions meet conditions—each time one says conjunction, one has also said, “There are two independently established things.” One does not notice while speaking; looking back makes it visible.

Second, remember that difference is not located in the thing. The fist’s difference exists only in relation to five fingers. Whenever someone or something seems “just that way by nature,” open the hand and look.

Third, do not secretly preserve a position while refuting. “The conjoiner, the time of conjunction, and the dharma that conjoins are all absent”—subject, object, and time must be released together. To retain one place is to retain a seed.

Fourth, refute to the end and speak to the end. Qingmu explicitly says the three are also absent: “Since cognition does not exist, seeing, the visible, and the seer do not exist either.” This sentence cannot be evaded. He immediately states the exact measure of what remains: “All dharmas in this way are like illusions and dreams and have no determinate marks; how could conjunction exist? Since there is no conjunction, they are empty.” The two sentences must be read together. What is absent is the determinate mark and self-essence; what remains is the manifest appearance, like illusion and dream. A fixed seer, a fixed visible, and a fixed conjunction cannot be found, but events are not therefore annihilated.

Carried Forward: Jizang’s Mirror

The meaning above belongs to the chapter itself. What follows is a side illumination from the Sui–Tang Sanlun lineage—a mirror is not the treatise, but another way of reading the same verses.

Jizang’s Commentary on the Treatise on the Middle first places this chapter within the whole treatise. The “Chapter on the Five Aggregates” examined the emptiness of the five aggregates but did not exhaust the meaning; the “Chapter on the Six Faculties” examined the unfindability of the six faculties but likewise did not exhaust it. The present chapter therefore examines faculty and object again through the gate of conjunction. He further classifies the two refutations: the earlier chapter refuted the five aggregates lacking intrinsic nature, while the present one refutes the conjunction of faculty and object through conditions. Thus the preceding chapters refuted real existence, and this chapter refutes provisional existence.

He also gives the chapter a path of practice. Deluded beings take faculty and object as two; therefore they say faculty and object conjoin. From conjunction arise the three poisonous afflictions; from affliction comes karma, and from karma comes the fruit of suffering. When faculty and object are examined as originally without conjunction, affliction does not arise, and therefore there is no karma and suffering. He then explains why a chapter titled the refutation of conjunction spends its entire course refuting difference: conjunction arises from difference; difference is the root of conjunction; where there is no difference, there is no conjunction. Thus difference must be refuted. The preceding chapter refuted difference only briefly; the present chapter extensively refutes the meaning of difference among all dharmas. As for why sentient beings circle through saṃsāra, he attributes it to conscious beings taking the myriad phenomena as different, thereby giving rise to delusion and failing to attain liberation. When this chapter seeks the meaning of difference in all dharmas and finds no path to it, affliction does not arise.

This mirror joins the chapter to the Sanlun lineage’s architecture for ordering earlier and later refutations and distinguishing real from provisional existence. It also aligns the text with the sequence of affliction, karma, and suffering. These classifications and alignments are Jizang’s additions, not the wording of Nāgārjuna’s verses or Qingmu’s prose. The readings are recorded side by side without adjudication, each returned to its own place.

The dispute in this chapter circles one word from beginning to end: intrinsic nature.

The questioner opens by saying: “Dharmas each possess intrinsic nature because each has a distinctive efficacy. A jar possesses the intrinsic nature of a jar, and cloth the intrinsic nature of cloth. This intrinsic nature appears when many conditions conjoin.” The claim sounds entirely reasonable. A jar can hold water and cloth can cover the body; each has its own efficacy. Since their efficacies differ, some distinct thing underneath must support each one—a jar is a jar because it possesses jar-nature. This intrinsic nature cannot ordinarily be seen; it “appears” only when clay, wheel, cord, water, potter, and all the other conditions gather.

One sentence conceals two things: intrinsic nature already exists, and intrinsic nature emerges only when many conditions gather.

The questioner sees no conflict between them. Perhaps he imagines intrinsic nature as something buried in the ground and conditions as the spade that digs it out; the thing was always there, and the spade merely makes it visible.

What Nāgārjuna does throughout the chapter is bring both the spade and that buried thing into the light.

What Emerges from Conditions Is Fabricated by Them

The answering verse has only two clauses: “That intrinsic nature exists among many conditions cannot be established. If intrinsic nature emerges from many conditions, it is called a fabricated dharma.”

Qingmu closes the matter: “If it emerged from many conditions, it would be a fabricated dharma and would possess no determinate intrinsic nature.”

Everything that exists only by relying on conditions is fabricated by those conditions; what is fabricated has no determinate intrinsic nature. The “appearance” imagined by the questioner is not appearance but production. Only when clay, wheel, cord, water, and potter gather is there a jar; if one is absent, the jar does not come about. Since it is assembled, what is assembled must obey the conditions assembling it.

Whatever must wait for conditions must change with conditions; whatever changes with conditions is not the unchanging thing.

Thus the questioner’s own words—“this intrinsic nature appears when many conditions conjoin”—complete the argument against themselves. He hoped that “appears” would preserve the priority of intrinsic nature, but as soon as the word is spoken, intrinsic nature enters the hands of conditions.

“Intrinsic Nature” Means Precisely What Does Not Depend on Another

The questioner retreats one step: “What fault is there if the intrinsic nature of dharmas is fabricated from many conditions?” Even if intrinsic nature is fabricated by conditions, so what?

The answering verse is severe: “If intrinsic nature were something fabricated, how could the meaning of intrinsic nature still hold?” It adds a definition: “The name ‘intrinsic nature’ means what is unfabricated and does not depend upon another dharma for establishment.”

This sentence is the pivot of the whole chapter. Nāgārjuna does not set up a new theory to oppose the questioner; he merely lays out the meaning of the word that the questioner himself employs: in the mouths of both sides, intrinsic nature means that which is unfabricated and does not depend on another dharma for establishment. If it is fabricated, it is no longer what this word indicates. Two positions are not contending; one position contends with itself.

Qingmu supplies two analogies, each governing a different point.

The first is gold: “Gold mixed with copper is not genuine gold.” Once gold is mixed with something else, it is no longer pure gold. This analogy governs “not dependent on conditions”: genuine intrinsic nature admits no admixture from outside—“if intrinsic nature truly existed, it would not require many conditions”; conversely, “if it emerged from many conditions, one should know that no genuine intrinsic nature exists.”

The second is long and short, this and that: “If intrinsic nature were determinate, it should not await something else before appearing. It is unlike long and short or this and that, which exist through mutual dependence because they lack determinate intrinsic nature.” This analogy governs “not dependent on another.” Is a bamboo pole long or short? Without another beside it, the question cannot be answered. Names such as long and short require the opposite term precisely because they lack determinate intrinsic nature of their own. Qingmu’s point is that if intrinsic nature were determinate, it should not fall into the situation of long and short, which depend on the opposite term.

Read together, the two analogies block both ends of “unfabricated and not dependent on another dharma”: independent of conditions and independent of another.

Retreat to Other-Nature, and One Turns in Place

The questioner changes direction: “If dharmas lack intrinsic nature, they should possess other-nature.” If nothing is available on this side, surely one can still depend on something else.

The answering verse says: “If a dharma lacks intrinsic nature, how could other-nature exist? Intrinsic nature, when spoken of in relation to another nature, is itself called other-nature.”

This requires slow reading. “Other” is spoken from this side; stand on the other side and that “other” is its own “self.” The name changes, but the situation does not. Qingmu speaks more plainly: “Other-nature for its own part is itself intrinsic nature,” and that thing likewise “arises from many conditions and exists through mutual dependence,” so it too is absent. The final sentence is decisive: “Other-nature is itself intrinsic nature.”

In other words, the questioner thinks he has found a new foothold but has merely moved the same question next door. That house too is rented.

Apart from Self and Other, Can a Third Thing Be Established?

Before saying, “If existence is refuted through intrinsic nature and other-nature,” the questioner first asks: what fault is there if dharmas exist apart from intrinsic nature and other-nature?

The answering verse says: “Apart from intrinsic nature and other-nature, how could another dharma exist? Only if intrinsic nature and other-nature existed could dharmas be established.”

The latter half requires attention. Nāgārjuna is not endorsing intrinsic nature and other-nature; he is pointing out that the “establishment of a dharma” sought by the opponent has only these two available routes. Qingmu gives an example of the opponent’s classification: “A jar’s essence is intrinsic nature, while the thing upon which it depends is other-nature.” The jar’s own essence counts as self; what it depends upon counts as other. The ways one can imagine a “dharma” being established do not exceed these two gates. When neither gate succeeds, no third thing has a place to stand; placing one there from nowhere merely starts again under a new name.

Refute Existence, and the “Nonexistence” Left in the Hand

The questioner now makes a move many people make: “If existence is refuted through intrinsic nature and other-nature, then nonexistence should now exist.” You have dismantled existence; what remains must be nonexistence.

The answering verse says: “If existence cannot be established, how could nonexistence be established? Only because an existent dharma exists is its destruction called nonexistence.” Qingmu repeats: “It is the destruction of an existent dharma that is called nonexistence; this nonexistence exists only because existence was destroyed.”

This move is crucial. Nonexistence is not an independent thing; it is the name left after existence is destroyed. When a jar breaks, we say, “The jar is gone.” The “gone” in this sentence rests entirely on the former jar. If that jar was never established through determinate intrinsic nature in the first place, “the jar is gone” likewise has no determinate footing.

Thus whoever accepts that existence is unestablished must accept that nonexistence is also unestablished. People often imagine that refuting existence automatically lands them on the side of nonexistence, like a balance where one side rises when the other sinks. But here, the beam itself rests on existence. When existence is gone, the beam is gone too.

A Downward-Sliding Slope

Nāgārjuna next turns from argument toward the person: “If someone sees existence and nonexistence and sees intrinsic nature and other-nature, that person does not see the true meaning of the Buddha-Dharma.”

Qingmu’s long explanation is the passage closest to the human heart in this chapter. He first names the root illness: “If someone deeply grasps dharmas, that person will insist upon finding a view of existence.” Deeply attached people must grasp something. He then describes the slide: “When intrinsic nature is refuted, the person sees other-nature; when other-nature is refuted, the person sees existence; when existence is refuted, the person sees nonexistence; when nonexistence is refuted, the person is bewildered.”

Intrinsic nature, other-nature, existence, nonexistence—these are exactly the four gates examined in sequence above. Qingmu arranges them into a slope: dismantle one gate and the hand instantly reaches for the next; when all four have been dismantled, the person is stunned.

The earlier reasoning looks horizontally across four gates; this passage looks vertically at one person. The same four things are four propositions in the argument and four changes of grip in a person. The act of grasping has not changed; only its object changes. Thus after all four gates are refuted, that person has not been resolved; he has merely run out of things to grasp. This is what “bewildered” means.

Where, then, is the way out? Qingmu says: “Those with sharp faculties and slight attachment know that extinguishing all views is peaceful and secure, so they no longer give rise to these four kinds of conceptual proliferation.”

Notice that the quality he selects is not cleverness but “slight attachment.” A person whose attachment is light does not rush to replace whatever has been dismantled. Through the dismantling itself, he tastes the security of opening the hand, and the four kinds of conceptual proliferation no longer arise. This sentence pulls the chapter from debate back into practice: the four gates are not four questions but four movements of the hand. When the hand does not reach, the questions naturally disappear.

This Is Not Nāgārjuna’s New Teaching

After so much refutation, a reader can hardly avoid asking: is forbidding a stand on either existence or nonexistence not too severe?

Nāgārjuna’s next verse answers precisely that doubt: “The Buddha can extinguish existence and nonexistence, just as when he transformed Kātyāyana; as stated in the scripture, he is free from both existence and nonexistence.”

Qingmu identifies the source: “In the Śāṇḍa Kātyāyana Sūtra, the Buddha taught him the meaning of right view, free from existence and nonexistence.”

This citation carries weight because freedom from existence and nonexistence is not a new rule established by this treatise; it is the meaning of right view as the Buddha already taught it. In moving from the first verse to this point, the chapter has not stepped beyond the Buddha’s teaching. Nāgārjuna has only retraced, in the manner of a treatise, what the Buddha taught Kātyāyana.

Qingmu then resolves a real doubt for the reader. He says: “If there were the slightest determinate existence among dharmas, the Buddha should not have refuted existence.” If the Buddha saw even a trace of determinate existence remaining, he would not refute it. Yet people have a habit: “As for nonexistence, once existence is refuted, people take the refutation to mean nonexistence.” Refute existence and they assume nonexistence is being established. This is precisely the third step down the preceding slope. Thus Qingmu says: “Because the Buddha thoroughly comprehends the marks of dharmas, he teaches that neither exists.” Neither exists in order to cut off every path by which people might land at either extreme. The final sentence is a direct admonition: “Therefore you should abandon views of existence and nonexistence.”

What Is Before Our Eyes Speaks for Itself

The preceding passages reason through the matter. Here Nāgārjuna points to what stands before the eyes.

The verse says: “If a dharma truly possessed intrinsic nature, it should not later change,” and again, “That intrinsic nature might bear a mark of change can never be established.” Qingmu’s explanation is spare: “If determinate intrinsic nature existed, there should be no mark of change, as in the preceding analogy of genuine gold.” He returns to the analogy: if genuine gold really is genuine gold, it cannot be gold today and something else tomorrow.

Then comes the statement of direct perception: “Because we now directly see that dharmas possess marks of change, we should know that they have no determinate marks.”

This sentence comes not from reasoning but from the eyes. One need not understand a formal argument; one need only look. Jars break, cloth decays, people age. Change is something directly seen before us. The fact of change itself bears witness to “no determinate mark.”

Perhaps the questioner still wishes to defend the side of nonexistence. Nāgārjuna therefore adds another verse, questioning both ends at once: “If a dharma truly possessed intrinsic nature, how could it change? If a dharma truly lacked all intrinsic nature, again how could it change?” Qingmu follows both questions: what determinately possesses intrinsic nature cannot move, while that which “has no intrinsic nature and therefore no essence of its own” has nothing to change. Change can be placed within neither determinate existence nor determinate nonexistence—yet it is happening before our eyes.

Determinate Existence Falls into Permanence; Determinate Nonexistence into Annihilation

In the final section, Nāgārjuna states the practical destination of the two views: “Determinate existence means attachment to permanence; determinate nonexistence means attachment to annihilation. Therefore the wise should not cling to existence or nonexistence.”

Qingmu gives examples of each.

For permanence, he cites two accounts. First, the three times: “A dharma’s mark exists in the future; the dharma then arrives in the present and passes into the past without abandoning its original mark. This is permanence.” The dharma travels through the three times and changes location while losing nothing of its original mark; it is therefore permanent. Second, the fruit preexists within the cause: “To say that the fruit already exists within the cause is likewise permanence.” The fruit lies ready in the cause and merely awaits the moment to appear; that too is permanence.

On the side of annihilation, to assert determinate nonexistence means that something “must have existed before but not exist now.” It existed before and is now gone; this is annihilation. Qingmu defines it: “Annihilation means the absence of a cause for continuity.” Annihilation means that the cause of continuity is gone, so the earlier and later segments cannot connect.

Qingmu gathers the outcome of the two views in one sentence: “Through these two views one departs from the Buddha-Dharma.”

The questioner asks a good question here: “Why does existence produce the view of permanence and nonexistence the view of annihilation?”

The answering verse says: “If a dharma has determinate intrinsic nature and is not nonexistent, that is permanence. If it existed before but does not exist now, that is annihilation.” Qingmu’s long explanation follows both roads to their ends. On the side of permanence: if intrinsic nature is determinately existent, it bears a mark of existence rather than nonexistence and “should never become nonexistent.” If one truly says it is nonexistent, it is no longer existent but becomes a nonexistent dharma—a fault already examined, “as previously explained.” Thus one can only preserve existence all the way through, falling into the view of permanence. On the side of annihilation, “if a dharma existed before and, after destruction, became nonexistent, this is called annihilation.” He also identifies the premise of this route: “This follows from your claim that existence and nonexistence each possess a determinate mark.” It is precisely because you first assign a determinate mark to existence and another to nonexistence that the interval between them breaks.

The Final Landing Is Demerit and Merit

Qingmu’s closing sentence brings the whole chapter from doctrine into ordinary human life: “If one holds views of annihilation or permanence, demerit, merit, and other such matters are absent and worldly affairs are destroyed. Therefore these views should be abandoned.”

This deserves to be read slowly. If all dharmas are permanent and never abandon their original marks, one who does evil is forever that way, doing good cannot change anything, and practice has no place to begin. If all dharmas are annihilated and the earlier and later do not continue, what one does now cannot connect with what one experiences later; demerit and merit again become empty.

Permanence leaves a person unable to change; annihilation leaves a person without responsibility. Both overthrow worldly affairs. Thus abandoning views of existence and nonexistence is not merely about tidiness in argument; it lets demerit, merit, and practice continue to stand.

The chapter’s movement is now complete. It begins with the harmless-looking common sense of jar-nature and cloth-nature, passes through intrinsic nature and fabricated dharmas, self and other, existence and nonexistence, permanence and annihilation, and finally returns to whether a person will open the hand. “Therefore the wise should not cling to existence or nonexistence” does not mean understanding a theory; it means no longer extending that hand.

What Readers Can Carry Away from This Chapter

First, one word cannot be used in two ways. Whenever someone says, “It already exists, but appears only when the conditions arrive,” first ask: does it appear, or is it fabricated?

Second, refuting one side does not mean landing on the other. If existence cannot be established, nonexistence cannot be established either. An empty hand is not the problem; reaching downward to grasp again is.

Third, change stands before the eyes. One need not wait to understand the formal argument before trusting that things have no determinate intrinsic nature. To see a jar break and a person age is already to see what this chapter teaches.

Fourth, views of existence and nonexistence are abandoned so that demerit and merit can stand. This chapter does not empty out the world. On the contrary, permanence and annihilation are the two views that empty out the world.

Carried Forward: Jizang’s Mirror

The meaning above belongs to the chapter itself. What follows is a side illumination from the Sui–Tang Sanlun lineage—a mirror is not the treatise, but another way of reading the same verses.

In his Commentary on the Treatise on the Middle, Jizang reads the two views of existence and nonexistence through the two truths. The Tathāgata always teaches based on the two truths, which are the conditionally empty and existent. The opponent hears “existent” and interprets it as existence through intrinsic nature; he hears “nonexistent” and interprets it as nonexistence through intrinsic nature. The two truths are therefore obstructed and the two wisdoms do not arise. Jizang also identifies the joint in the chapter’s opening passage: many conditions and intrinsic nature have contradictory meanings. If intrinsic nature exists, it does not rely on conditions; if it relies on conditions, it loses its own essence. To claim both that intrinsic nature exists and that it relies on conditions is self-contradictory. Concerning change in the later passage, he says that if arising is treated as prior nonexistence becoming present existence and cessation as previous existence returning to nonexistence, change is reified as nonexistence turning into existence and existence turning into nonexistence. When change is refuted, both existence and nonexistence are refuted with it.

This mirror joins the chapter to the Sanlun lineage’s architecture of refuting the wrong and revealing the right. Its classification is Jizang’s addition, not the wording of Nāgārjuna’s verses or Qingmu’s prose. The readings are recorded side by side without adjudication, each returned to its own place.

The preceding chapters refuted coming and going, arising and ceasing, agent and action. A reader might still feel that these belong to the arena of debate. In this chapter the question suddenly comes close to the body: bondage and release—being tied and being untied. What practitioners seek throughout their lives is nothing other than this passage from bondage to release.

Yet this chapter says: bondage is unobtainable, and release is unobtainable too.

We must first see clearly the standpoint from which this sentence is spoken. Otherwise, it sounds as if cultivation itself were being canceled.

What the Questioner Wants Is Someone Who Comes and Goes in Saṃsāra

The opening question begins this way: “Saṃsāra cannot be wholly without a basis; within it, surely sentient beings come and go, or conditioned activities come and go.”

The questioner is not being unreasonable. He sees what appears to be an obvious fact: people have former and later lives, and karma continues from before to after. If all this is altogether empty, then who passes from this life to the next? So he presses: “On what grounds do you say that sentient beings and conditioned activities are all empty and that there is no coming and going?”

He wants one who comes and goes. This question hands us the thread running through the whole chapter: for coming and going to be established, there must first be something that passes from here to there. What the respondent does next is follow the questioner’s own words and see whether that thing can stand.

Neither the Permanent nor the Impermanent Comes and Goes

The verse answers first:

“If conditioned activities that come and go are permanent, they should not come and go; if impermanent, they likewise should not; the same is true of sentient beings.”

It offers only two paths, and neither is passable.

If conditioned activities are permanent, Qingmu speaks plainly: “because they are determinate and because their intrinsic nature abides.” What is already determinate and abides in intrinsic nature never changes. What does not change has no difference between before and after, and therefore “there can be no continuity of birth and death.” Saṃsāra is saṃsāra precisely because it is continuous. Something permanently abiding and unmoving has not even continuity—what could come and go?

If conditioned activities are impermanent, Qingmu says: “because they are indeterminate and lack intrinsic nature.” Since they lack intrinsic nature, no self-subsisting thing exists there that could be moved from here to there. Impermanence is not one thing changing rapidly; it means that the thing supposedly moving all the way through cannot be found. Thus “there likewise can be no coming and going or continuity of birth and death.”

What is permanent does not move; what is impermanent has no master. Coming and going would require something that both remains itself and yet can move. Those two requirements never meet.

The same holds for sentient beings: “If it is said that sentient beings come and go, the same faults follow.” Replace conditioned activities with sentient beings, and the fault remains unchanged.

Seeking the One Who Comes and Goes among Aggregates, Realms, and Entrances

The second verse narrows the field to the sentient being:

“If sentient beings come and go among aggregates, realms, and entrances, exhaustive search by the five modes finds nothing—who is the one that comes and goes?”

Qingmu first fixes the field: “Saṃsāra and the aggregates, realms, and entrances have the same meaning.” Saṃsāra is not another place outside aggregates, realms, and entrances. What people call a lifetime is just this collection of five aggregates, eighteen realms, and entrances. If we seek the sentient being that comes and goes, we can seek it only here.

He then refers back: “That sentient being was sought by the five modes in the ‘Examination of Fire and Fuel’ and found unobtainable.” The fivefold search was already conducted in that chapter: whether sentient being and aggregates are identical, different, contained in one another, or possessed by one another, every examination fails. Since no sentient being can be indicated among aggregates, realms, and entrances, Qingmu returns the questioner’s own question: “Who, then, comes and goes among aggregates, realms, and entrances?”

No new principle is added here. The questioner’s “surely a sentient being comes and goes” is simply brought back to the field he himself accepts, and he is asked to point. When he cannot point, the statement stops.

From Body to Body: Who Is There in That Moment?

The third verse changes the angle and directly takes up what “coming and going” would actually look like:

“If one goes from body to body, the one who comes and goes becomes bodiless; if that one has no body, there is no coming and going.”

Coming and going, plainly stated, means relinquishing this body and taking another. Ask, then: at the very moment of passage, does that one have a body or not?

Qingmu examines both paths: “Does it come and go with a body or without a body? Neither is tenable.”

If one says it has a body, the difficulty is this: “Passing from one body to another, the one who comes and goes becomes bodiless.” If it carries the former body along, then upon reaching the later body it has two bodies. If it does not carry that body, then during the moment between the two it is precisely bodiless. Qingmu adds another cut: “If it already had a body beforehand, it should not pass again from body to body.” If it already has a body, why must it take another?

If one says it is bodiless, the answer is swifter: “If beforehand it had no body, then it does not exist.” Without a body, that “one” has no basis for establishment. “If nonexistent, how could there be coming and going in birth and death?” If no person can be indicated, whose coming and going is it?

All three faces of the questioner’s traveler are now closed.

Nirvāṇa Extinguishes Neither Conditioned Activities nor Sentient Beings

The questioner changes his point of entry: “The sūtras say that nirvāṇa exists and extinguishes all suffering. What this extinction extinguishes must be either conditioned activities or sentient beings.”

The question is candid: since the Buddha says nirvāṇa can extinguish all suffering, surely there must be something extinguished. It must be either conditioned activities or sentient beings.

The answer is: “Neither is extinguished.”

The reason is not new. Qingmu directly says, “This has already been answered”: “Conditioned activities have no intrinsic nature, while sentient beings are unobtainable when sought in every way through the coming and going of birth and death. Therefore, conditioned activities are not extinguished, nor are sentient beings extinguished.”

We must see clearly what this “not extinguished” means. It does not mean that conditioned activities endure forever or that sentient beings never perish. It means that extinction requires something extinguished. Conditioned activities lack intrinsic nature: no intrinsically existent thing stands there waiting to be extinguished. Sentient beings are unobtainable when sought: no real person is erased by nirvāṇa. What nirvāṇa does has never been the destruction of an intrinsically real thing.

“If So, Then There Is Neither Bondage nor Release”

Here the questioner becomes urgent and voices the chapter’s most important challenge: “If so, then there is neither bondage nor release, because the basis is unobtainable.”

This sentence asks on the reader’s behalf. If both person and dharmas are unobtainable, who is tied and who is untied? Does cultivation not come to nothing?

The respondent does not evade the question:

“Conditioned activities bear the marks of arising and ceasing; they are neither bound nor released. Sentient beings, as explained before, are likewise neither bound nor released.”

He directly affirms: there is neither bondage nor release.

But the reason must be heard clearly. Why are conditioned activities neither bound nor released? “Because conditioned activities arise and cease thought by thought, they should not undergo bondage and release.” For something arising and ceasing thought by thought, the former thought has vanished and the later thought has not arrived; between them, no enduring thing remains that could be tied, and no enduring thing remains that could be untied. Why are sentient beings neither bound nor released? “Sentient beings, as already explained, are unobtainable when sought by the five modes; how could they undergo bondage and release?”

Thus neither bondage nor release means that there is no intrinsic bondage binding an intrinsic person. It does not mean that cultivation is useless, nor does it mean that there is no nirvāṇa.

If the Body Is Bondage, There Must Be Two Bodies

The next three verses examine, one by one, what bondage could be established upon.

First they test the most natural statement: the body of five aggregates is itself bondage.

“If the body is called bondage, one with a body is not bound; one without a body is not bound either—upon what could bondage occur?”

Qingmu makes the calculation: “If a sentient being already had the five aggregates, it should not be bound. Why? Because one person would have two bodies.” If the sentient being already has a body of five aggregates and the five aggregates are then said to bind it, the bound one has one body while the binding aggregates form another—“one person would have two bodies.” That is absurd.

What if one says the sentient being has no body? “Without a body, there are no five aggregates; without the five aggregates, there is emptiness—how could that be bound?” With not even the five aggregates present, what is there to bind?

Having a body fails; being bodiless fails. Qingmu seals the road with the next line: “Beyond these two, no third thing that is bound exists.” Outside existence and nonexistence, there is no third thing that could serve as the bound.

Bondage and What Can Be Bound: Which Comes First?

The seventh verse examines temporal order:

“If what can be bound is already bound, bondage should bind what can be bound; yet before it there is in fact no bondage—the rest is as answered for coming and going.”

Suppose what can be bound must already be in bondage before bondage can bind it. Since it is already bound, why bind it again? But if it has no bondage beforehand, where does that “bondage” come from? Qingmu answers directly: “Apart from what can be bound, no bondage exists beforehand; therefore one cannot say that sentient beings are bound.”

He then names two common allocations: “Some say that the sentient being is what can be bound and the five aggregates are bondage. Others say that the various afflictions within the five aggregates are bondage and that the remaining aggregates are what can be bound.”

One account has the five aggregates bind the sentient being; another has afflictions bind the five aggregates. Each appears to distinguish binder from bound with perfect clarity. Qingmu answers both with the same point: two things cannot be separated. “If a sentient being existed before and apart from the five aggregates, the five aggregates should bind the sentient being. Yet no distinct sentient being exists apart from the five aggregates.” If no sentient being can be indicated apart from the five aggregates, then for the aggregates to bind the sentient being is the aggregates binding themselves. “If afflictions existed apart from the five aggregates, afflictions should bind the five aggregates. Yet no distinct afflictions exist apart from the five aggregates.” Afflictions too are not outside the aggregates, so for afflictions to bind the aggregates is again something binding itself.

Binder and bound are an event that requires two things. Once two things cannot be indicated, the event cannot be established.

Finally, Qingmu brings out a blade already used in an earlier chapter—the three times of coming and going: “As said in the ‘Examination of Coming and Going’: what has gone does not go, what has not gone does not go, and during going there is no going. Likewise, what is not bound is not bound, what has been bound is not bound, and during binding there is no binding.”

For one already bound, the event of binding has passed and no further binding is needed. For one not yet bound, binding has not occurred. As for “the time of being bound,” apart from already bound and not yet bound, no third interval can be indicated as a time of binding. The same cut made in the “Examination of Coming and Going” falls here unchanged.

Release Is the Same

“One who is bound has no release, and one who is unbound likewise has no release; if release exists during binding, bondage and release occur at the same time.”

Qingmu’s commentary is brief and sharp: “One who is bound has no release. Why? Because that one is already bound. One who is unbound likewise has no release. Why? Because there is no bondage.”

One already bound does not obtain release: that person is in bondage, so where can release enter? One never bound does not obtain release either: without bondage, release has no work to do. Suppose one compromises and says that release exists in the very act of binding. Then “bondage and release occur at the same time. This is not so, all the more because bondage and release oppose one another.” Bondage and release are contrary events; at the same time and place, something cannot be both tied and untied.

Since bondage has nowhere to be established, release has nowhere to be established. These are not two conclusions but two faces of one conclusion.

The Thought “I Shall Attain Nirvāṇa”

At last the questioner points to an evident fact: “Some people cultivate the path, enter nirvāṇa here and now, and attain release. How can you say there is none?”

People plainly cultivate and plainly attain release—how can you simply say there is none?

The responding verse is the chapter’s closest line:

“If I do not appropriate any dharmas, I shall attain nirvāṇa; one who is like this is instead bound by appropriation.”

Qingmu explains it in one sentence: “If someone forms this thought—‘I depart from appropriation and attain nirvāṇa’—that person is at once bound by appropriation.”

Notice that this does not tell anyone not to cultivate the path, nor does it say nirvāṇa is false. What it exposes is the shape of the thought: an “I” that departs from appropriation and will attain nirvāṇa. Within that thought stands an “I,” and there also stands a thing that I will acquire. “Not appropriating any dharmas” is meant as an open hand, but the thought “I shall attain nirvāṇa” is itself a reaching hand.

Appropriation means grasping. A person relinquishes worldly grasping, yet grasps with exceptional force at the project “I will attain release.” The subject-object configuration refuted above returns in its most refined form: an “I” that will be released and a nirvāṇa sought by me.

Thus, if the person invoked by the questioner—“someone cultivates the path, enters nirvāṇa here and now, and attains release”—carries this structure in mind, what that person obtains is not release. A rope remains in the hand; it merely seems to be an excellent rope.

No Separate Nirvāṇa Apart from Saṃsāra

The final verse gathers the whole chapter:

“There is no separate nirvāṇa apart from saṃsāra; such is the meaning of true character—what distinction could remain?”

Qingmu’s commentary is utterly steady: “In the ultimate meaning of the true character of dharmas, one does not say that a separate nirvāṇa exists apart from saṃsāra.”

The chapter has examined coming and going and examined bondage and release. Throughout, it examines the same illness: dividing one event in two and then installing a transitional subject between the halves. Saṃsāra is one side, nirvāṇa another, and a person is required to walk from one to the other. That is coming and going once again; it is passage from body to body. Since coming and going cannot stand, neither can this manner of crossing.

Qingmu cites the sūtra as witness: “Nirvāṇa is precisely saṃsāra, and saṃsāra is precisely nirvāṇa.”

He then closes with a question: “In the true character of dharmas, how could one say, ‘This is saṃsāra; that is nirvāṇa’?”

To point within true character and say, “This is saṃsāra; that is nirvāṇa,” one must first have two fixed things to point at. The ten verses above remove those fixed things one by one: the traveler is removed, what is extinguished is removed, binder and bound are removed, and the “I” seeking nirvāṇa is removed. Here even this final pair of distinctions has nowhere to rest.

What a Reader Can Carry Away from This Chapter

First, there is no need to find a traveler for saṃsāra. “Exhaustive search by the five modes finds nothing—who is the one that comes and goes?” Not finding one is precisely the chapter’s point.

Second, understand what neither bondage nor release means. It does not mean that cultivation comes to nothing. It means that no intrinsic bondage binds an intrinsic person. It does not cancel the path; it cancels the two entities supposedly standing at its ends.

Third, watch the thought that looks most like cultivation. “I shall attain nirvāṇa”—while this thought remains, one is “instead bound by appropriation.” The posture of letting go very easily becomes a new fist.

Carried Forward: Jizang’s Mirror

What precedes is the meaning of the chapter itself. What follows is illuminated from the perspective of the Sui-Tang Three Treatise lineage—a mirror is not the treatise, but another way of reading the same group of verses.

Explaining why the Commentary on the Treatise on the Middle refutes bondage and release, Jizang says that seeing bondage and release as existent is itself called bondage; only when examination finds no source for either is there release. He further says that followers of the inner and outer teachings, Great and Lesser Vehicles, remove bondage and are no longer bound by bondage, yet have not removed release and remain bound by release. It is like escaping iron fetters while still wearing golden ones. The treatise-master refutes bondage and release so that both sets of fetters can be shed. Jizang also says these practitioners regard bondage and release as two things and concentrate on severing bondage and cultivating release; this passage instead lets one realize that bondage itself is release and know bondage and release as nondual. That is why it refutes both.

Regarding the claim that the five aggregates bind the sentient being, Jizang offers another analysis. Only the conjunction of the five aggregates is called a sentient being; before the sentient being is formed, no other five aggregates can already stand there. Moreover, “sentient being” is the collective name for the five aggregates. Once one has taken up the sentient being, no separate five aggregates remain to bind it. He further uses the image of a knotted rope: it is this same rope that is knotted and this same rope that is untied; there is no other thing. Therefore one cannot say that a separate name-and-form binds the sentient being, nor that a separate sentient being receives the bondage of name-and-form.

On the verse about not appropriating any dharmas, Jizang distinguishes two senses of “no appropriation”: first, the five aggregates are called appropriation; second, the mind of grasping is called appropriation. At entry into nirvāṇa without remainder, both kinds of appropriation are absent; hence one speaks of not appropriating any dharmas. Yet the one who entertains this thought does not appropriate appropriation but does appropriate nonappropriation, so nonappropriation becomes appropriation again. Jizang then concludes: this person claims that the mind appropriates nothing but ultimately still has something attained. Wherever something is attained, appropriation ultimately remains, so that person is still bound by appropriation. Thus one leaves the ordinary to attain sagehood, only for the sage to become ordinary again.

This mirror connects the chapter to the interpretive structure of refuting error and revealing rightness in the Three Treatise lineage. Its articulation is Jizang’s addition, not the words of Nāgārjuna’s verses or Qingmu’s extended commentary. The readings are recorded side by side without adjudication, each returned to its own place.

The preceding chapters have refuted going, coming, arising, ceasing, and the agent. By this point someone is bound to stand up and speak—and what he says is the hardest sentence in the entire treatise to withstand.

He says: everything else can be discussed, but not karma. “Karma determinately exists and can make every sentient being undergo its fruition.” His reason is not metaphysical theory but scripture: “All sentient beings are born according to karma”—those who commit evil fall into the hells, those who cultivate merit are born in the heavens, and those who practice the path attain nirvāṇa. If these three fail, the entire Buddhist path collapses. He therefore concludes: “All dharmas should not be empty.”

We must see the weight of this question. He is not defending an abstract real existence; he is defending cause and fruit. If the price of emptiness is that demerit and merit become indistinguishable, then emptiness is not worth having. This chapter establishes and refutes again and again precisely because the respondent too recognizes what is at stake.

He Begins by Speaking of Goodness

The questioner’s first example is not evil karma but wholesome karma.

“A person can subdue the mind and benefit sentient beings; this is called loving goodness, the seed of fruition in two lives.”

Qingmu explains this verse very concretely. People have the three poisons and give rise to various activities because they wish to trouble others. One who cultivates goodness “first extinguishes one’s own evil,” and therefore subdues the mind and benefits others. Benefiting others is not empty rhetoric either: “practicing generosity, keeping precepts, patience, and the rest without troubling sentient beings.”

Notice the order here: only after subduing one’s own mind can one speak of benefiting others. Activities that trouble others are rooted in the three poisons. Unless those poisons are first extinguished, what is called benefiting others is most often merely another way of troubling them. Placed at the very beginning, this line establishes the chapter’s ground: no matter what is refuted below, this does not move.

What Is Karma? Volition and What Arises from Volition

The questioner next establishes karma’s substance in two levels. First the general statement: “The Great Sage taught two karmas: volition and what arises from volition.” Then the concrete statement: “The volition spoken of by the Buddha is mental karma,” and “what arises from volition is bodily and verbal karma.”

Why volition? Qingmu answers with precision: volition is an associated mental dharma and among such dharmas “can initiate and produce action, and therefore is called karma.” Action needs a point of initiation; that is volition. He adds an important sentence: other mind-dharmas and associated mental dharmas can also act, but “volition is the basis of what is done.” Volition alone is therefore singled out and called karma.

Karma’s root lies neither in the hand nor the mouth, but in the thought at the instant it moves.

Seven Dharmas: The Moment of Acting and What Follows the Act

The questioner further divides karma’s character into seven items: “Bodily karma and verbal karma, performed and nonperformed karma—among these four things some are wholesome and some unwholesome.” Add the two items “merit arising from use; demerit arising likewise,” and “together with volition these are seven dharmas that disclose the character of all karma.”

What most deserves a pause is performed and nonperformed karma. Qingmu defines them: “What is done while acting is called performed karma; what constantly follows and arises after the act is called nonperformed karma.”

While the deed is being done, it is performed karma. Once done, the event does not simply scatter; it “constantly follows.” This is the skeleton of the questioner’s thesis: only by following can it connect with fruition in a later life. Qingmu also draws the wholesome-unwholesome line cleanly: “Unwholesome means not stopping evil; wholesome means stopping evil.”

Now consider the two that arise from use. A donor gives; if the recipient uses the gift, the donor obtains two merits: “one arising from giving, the second arising from use.” Conversely, if someone shoots another with an arrow and the arrow kills, there are two demerits: “one arising from shooting, the second arising from killing.” Yet “if it strikes but does not kill, the archer obtains only the demerit of shooting and not the demerit of killing.”

This pair of analogies is the questioner’s most forceful move: a karma’s weight lies not only in the moment it is performed but also in the effect it has upon another. Therefore he says that karma has fruition in this and later lives; karma and fruition determinately exist, so “dharmas should not be empty.”

The Dilemma: If It Remains until Fruition, It Is Permanent; If It Ceases, It Cannot Produce Fruit

〔T30 0022a05〕The respondent takes no detour and directly poses a dilemma:

“If karma remains until fruition is undergone, this karma becomes permanent; if it ceases, there is no karma—how could it produce fruition?”

For karma to yield fruition, it must wait there until fruition becomes present. If it can wait, it is permanent. But Qingmu says: “Karma bears the marks of arising and ceasing and cannot remain for even a single thought, much less until fruition!”

Take the other path: karma ceases. Cessation is absence. How could what is absent “produce fruition”?

For karma to connect with fruit, it must remain; once it remains, it becomes permanent. If karma is not permanent, it must cease; once it ceases, it cannot connect with fruit. Both paths are blocked. This is no word game. It asks: just what sort of thing is the “karma” you establish, that it can both arise and cease and yet cross time?

First Rescue: Continuity Like a Sprout

The questioner’s first rescue brings out the seed:

“A continuum such as sprout and the rest all arises from a seed; from this continuum fruit is produced, and apart from the seed there is no continuum.”

Qingmu explains smoothly: from grain comes sprout, from sprout stalk and leaves, and through this continuum fruit finally appears. Apart from the seed there is no such succession; thus “the seed comes first and fruit afterward, neither annihilated nor permanent.” Move this to mind: the initial thought gives rise to demerit or merit like a grain seed; because of this thought, the remaining mind-dharmas and associated mental dharmas arise in continuity until fruition—“karma comes first and fruit afterward, neither annihilated nor permanent.”

It is an elegant rescue. Karma itself need not remain until fruition; it hands on the baton. Each former link ceases after the later has connected, so there is no annihilation; no link permanently abides, so there is no permanence.

The questioner then lists the fruit of wholesome karma: “What accomplishes merit is the ten white paths of karma”—not killing, not stealing, no sexual misconduct, no false speech, no divisive speech, no harsh speech, no useless speech, no jealousy, no anger, and no wrong view. Their fruit is reputation and benefit in this life and honored stations among humans and devas in later lives. He says that the many merits of generosity and respect are, in brief, all contained within the ten wholesome paths.

Refuting the Sprout Analogy: An Analogy Is Not Proof

The respondent gives only one verse: “If one analyzes as you do, the faults are extremely numerous. Therefore, what you say cannot be established in principle.”

Qingmu explains it through two paths.

The first is that the terms of the analogy are unequal. “A grain can be touched, has form, and its continuum can be seen.” Grain can be touched and seen; its continuum can be tested before one’s eyes. Yet Qingmu says that even with visible grain, “after considering this, I still do not accept the account.” How much less mind and karma, “intangible, formless, and invisible,” which arise and cease without remaining, when continuity is used to explain them.

Even a visible continuum is insufficient to prove that one thing runs through it; how much less when it is used to analogize what cannot be seen. An analogy can aid understanding but cannot substitute for establishment.

The second path presses the grain itself: “Does the continuum occur after cessation, or without cessation?”

If “the continuum occurs after the grain has ceased, it becomes causeless”—once the former link has wholly ceased, where does the later arise? If it continues without cessation, this grain remains there producing forever, so “one grain seed would produce every grain in the world.”

This pressure has the same skeleton as the earlier dilemma: after cessation it is causeless; without cessation it becomes permanent. Replace karma with a seed, and the dilemma remains inside it.

Second Rescue: The Dharma of Nonloss Is Like a Bond

The questioner tries again with a more refined doctrine, first hanging up a warrant: “Another doctrine congruent with karma and fruition should now be explained, one praised by the Buddhas, pratyekabuddhas, and worthy sages.”

The doctrine is: “The dharma of nonloss is like a bond, and karma like property taken on debt.”

This is an ingenious design. He concedes that karma ceases once performed. Yet as the karma is performed, it gives rise to another thing called the dharma of nonloss. Qingmu explains it clearly: “The dharma of nonloss should be understood as like a bond, and karma like taking property.” The property has long been spent, but the bond remains, and therefore the debt remains. Thus “because this dharma of nonloss exists, all karmas have fruition.”

He gives this bond a detailed specification. “Its nature is indeterminate, and analysis distinguishes four kinds”: bound to the desire realm, form realm, or formless realm, and unbound. Among wholesome, unwholesome, and indeterminate, it is only indeterminate. It is “not severed by seeing the truths but only by cultivation.” If it were severed by seeing the truths while karma extended to a similar stage, “faults such as the destruction of karma would follow.” Further, “when one first receives a body in a realm, fruition independently arises then”; two kinds of karma undergo fruition in the present life, and “some say that after fruition has been undergone, karma remains as before,” because “it does not cease thought by thought.” When does it cease? “Either after passing beyond a fruit-stage or after death”: stream-enterers and the rest after passing beyond a fruit-stage, ordinary persons and arhats after death. Outflowing and nonoutflowing must also be distinguished within it.

This now resembles a complete mechanism. But the respondent gives one sentence:

“On both sides, this doctrine does not escape the faults of annihilation and permanence; therefore it too should not be accepted.”

The bond itself cannot escape the same dilemma. If it “does not cease thought by thought” and remains until it yields fruit, it is permanent. If it too ceases thought by thought, it shares karma’s illness and likewise cannot connect. The questioner has moved permanence onto the bond; permanence has not been refuted, only deposited under another name. If a rescue merely relocates the difficulty, the difficulty remains.

Here the questioner is forced to say what he really means: “If so, then there is no karma and no fruition.”

Right Meaning: Though Empty, Not Annihilated; Though Existent, Not Permanent

The pivot of the entire chapter is this verse:

“Though empty, it is not annihilated; though existent, it is not permanent; karma and fruition are not lost—this is what the Buddha taught.”

The respondent does not seek a compromise between annihilation and permanence; he does not enter from there. Qingmu’s explanation must be read sentence by sentence.

First, why not annihilated? “Karma is ultimately empty, bears the character of quiescent extinction, and is separate from intrinsic nature. What dharma is there to annihilate? What dharma is there to lose?”

Annihilation means annihilating something existent; loss means losing something existent. If karma has no intrinsic nature obtainable from the beginning, there is no thing placed there waiting to be annihilated or lost. Annihilationist view is established only by first tacitly admitting existence. Here emptiness cuts at the root. It does not say, “It existed and has now been annihilated”; it says that this existence by intrinsic nature was never established.

Next, why not permanent? “Because of inverted causes and conditions, there is coming and going in saṃsāra; this too is not permanent.” Why? “If a dharma arises from inversion, it is false and unreal. Because unreal, it is not permanent.”

Coming and going in saṃsāra is not nonexistent; it exists. But it arises from inverted causes and conditions. Whatever arises from inversion is false and unreal; being unreal, it cannot be permanent. The reason it is not permanent is not that it will break, but that it was never real.

At what level, then, is it said that “karma and fruition are not lost”? Qingmu answers: “Because one greedily clings to inversion and does not know true character, it is said that karma is not lost.” Nonloss is spoken truthfully with reference to sentient beings who have not seen true character; it does not establish another imperishable karmic substance. The one who says this is the Buddha: “This is what the Buddha taught.”

Karmas Originally Do Not Arise: What Strange Consequences Follow from Determinate Karma?

The respondent presses the point to its end:

“Karmas originally do not arise because they have no determinate intrinsic nature; 〔T30 0023a01〕nor do karmas cease, because they do not arise.”

Qingmu identifies the level: “In the ultimate meaning, karmas do not arise.” Why do they not arise? Because they have no intrinsic nature. The reason for noncessation must be seen carefully: “Because they do not arise, they do not cease—not because they are permanent and therefore do not cease.” They are not unceasing because they permanently abide, but because they never arose. What never arose cannot be said to cease. This is wholly different from the questioner’s “remaining and not ceasing.”

Now the reverse: what happens if karma truly has intrinsic nature? The respondent lists five strange consequences.

First, it is permanent and cannot be performed: “If karma had intrinsic nature, it would be called permanent. What is not performed would still be called karma, while the permanent cannot be performed.” Qingmu adds: “because a permanent dharma cannot be performed.” What intrinsically exists is already so without awaiting performance, and cultivation becomes superfluous.

Second, another acts and I undergo: “If karma could exist without being performed, there would be demerit without commission.” Qingmu states it directly: “another person commits the wrong and this person undergoes its fruition.”

Third, pure conduct loses its basis: “Pure conduct would not have been broken, yet the fault of impurity would still be incurred”—“another person breaks pure conduct and this person incurs the offense, destroying worldly convention.”

Fourth, all worldly language is thrown into disorder: “This would destroy every convention of worldly language.” Qingmu gives an example. If everything already existed beforehand, “in winter one should not plan for spring, nor in spring plan for summer.” People plan for what is to come precisely because it does not yet exist. If it existed beforehand, planning could not be spoken of. Once intrinsic existence is established, time dies.

Fifth, demerit and merit become indistinguishable: “Committing wrong and cultivating merit would have no difference.” Generosity and keeping precepts are called cultivating merit; killing and stealing are called committing wrong. If karma exists without being performed, no distinction remains.

Add one more: “If karma were determinate and possessed intrinsic nature, after undergoing fruition one should undergo it once again.” After it has been undergone, it must be undergone forever. Qingmu charges the account back to the questioner: “Therefore, your claim that karmic fruition exists because of the dharma of nonloss entails these many faults.”

The final move traces one level higher: “If worldly karma arises from afflictions and these afflictions are unreal, how could karma be real?” Qingmu says afflictions “have no determinate existence and exist only through conceptual recollection and discrimination,” and concludes: “Because the cause has no intrinsic nature, karma too has no intrinsic nature.”

What about This Present Body?

The questioner changes angles. Even if afflictions and karma are unreal, the body of fruition plainly stands here. The respondent merely brings out the causal order: “Afflictions and karmas are said to be the causes and conditions of the body. Afflictions and karmas are empty—how much more the bodies?”

Qingmu makes the causes concrete. Worthy sages say afflictions and karma are causes and conditions for the body: “craving moistens birth,” while karma produces higher, middle, and lower, attractive and ugly, noble and base fruitions. Since these two causes, when sought in many ways, “have no determinate existence,” how could the body formed from them have determinate fruit? It exists “according to causes and conditions.”

Since what produces has no intrinsic nature, what is produced has still less intrinsic nature. Touching a body does not prove a body with intrinsic nature; it proves only that causes and conditions remain.

What about the Agent?

The questioner’s final ground is the person. He cites scripture: “Obscured by ignorance and bound by the fetter of craving, the present undergoer is neither identical with nor different from the original agent.” Qingmu says this is taught in the Beginningless Sūtra: sentient beings covered by ignorance and bound by craving’s fetter come and go through beginningless saṃsāra and undergo suffering and pleasure.

The four words “neither identical nor different” show that the questioner has already considered the dilemma. “If identical, a person who commits wrong and receives the form of a cow would mean that the person did not become the cow and the cow did not become the person.” If the present undergoer simply is the prior agent, then when a human receives a cow’s body, human and cow collapse into one. “If different, karma and fruition are lost and fall into causelessness; causelessness is annihilation.” If entirely two, the line of karma and fruit is broken. He therefore takes the middle: neither identical nor different.

The respondent does not entangle himself in identity and difference but removes their ground:

“Karma does not arise from conditions, nor from nonconditions; therefore there is no one who can originate karma. With neither karma nor agent, how could there be fruit arising from karma? If there is no fruit, how could there be one who undergoes fruit?”

A chain follows: karma does not arise with intrinsic nature, therefore there is no intrinsically existent originator of karma; without karma or agent, there is no intrinsically existent fruit; without fruit, there is no intrinsically existent undergoer. Qingmu simultaneously preserves the conventional side: “The person provisionally named among the five aggregates is the agent.” The agent is provisionally established, while karma producing fruit in good or bad destinations is called fruition. Since even an originator of karma cannot be obtained, “how much less karma, fruition, and one who undergoes fruition?”

“Neither identical nor different” is established precisely because neither side is a thing with intrinsic nature. The questioner assumes that an actual agent must first exist before one can speak of neither identity nor difference. The respondent shows that once the agent is made real, neither identity nor difference can stand.

A Conjured Person Conjures Another Person

The questioner retreats to the final and most honest step: “We now plainly see sentient beings perform karma and undergo fruition. How is this to be explained?”

The respondent answers with the chapter’s conclusion:

“It is like a person whom the World-Honored One conjures through supernormal power; this conjured person in turn conjures another person. The first conjured person is called the agent; what the conjured person does is called karma.”

Through supernormal power, the Buddha conjures a person who then conjures another person. The former is the agent; what the latter does is karma.

Qingmu makes the pivot very clear. A conjured person “has no real affair and can merely be seen.” Yet the conjured person also uses verbal karma to teach Dharma and bodily karma to give alms: “Although this karma is unreal, it can be seen.”

This directly answers “plainly see.” Visibility does not equal intrinsic nature; lacking intrinsic nature does not prevent visibility. The questioner takes “plainly seen” as evidence of real existence. The analogy separates them. Not one act or fruition before the eyes must be denied; beneath it, however, no unmoving thing provides support.

The layer in which the conjured person “in turn conjures another” is especially important. The agent itself is also conjured. Karma has no intrinsic nature, and the agent has no intrinsic nature; empty karma does not hang upon a real person. Thus Qingmu says: “In this way, the body in saṃsāra, the agent, and karma should also be understood.”

At the end Qingmu lists four things one by one. Afflictions are the three poisons, further divided into ninety-eight tendencies, nine fetters, ten enwrapments, six stains, and immeasurably many more. Karma is bodily, verbal, and mental karma, divided into this and later lives; wholesome, unwholesome, and indeterminate; painful, pleasant, and neither-painful-nor-pleasant fruition; present-life, next-life, and later-life fruition, and immeasurably more. Agent means one who can give rise to afflictions and karma and undergo fruition. Fruition means the indeterminate five aggregates born from wholesome and unwholesome karma. After opening them out so finely, one sentence gathers all: “all are empty and without intrinsic nature.”

The verse gives four analogies: “All are like illusion and dream, like a mirage and an echo.” They share one point: each truly appears, and each truly lacks an intrinsic body. The echo truly sounds, but no body of sound resides in the mountain. Karma is truly performed, but no intrinsic body of karma resides in the continuum.

What a Reader Can Carry Away from This Chapter

First, refutation and preservation must be read together. The respondent undeniably has a line of refutation and speaks without reserve: “With neither karma nor agent, how could there be fruit arising from karma?” “If even the one who originates karma does not exist, how much less karma, fruition, and one who undergoes fruition?” Karma, agent, and fruition with intrinsic nature are each examined away without remainder. Yet the chapter also preserves another line: “Karma and fruition are not lost—this is what the Buddha taught,” and “Although this karma is unreal, it can be seen.” What is removed is the determinate intrinsic nature added to karma; what is preserved is the causality spoken at the level of conventional truth. Omit either line and the reading tilts.

Second, watch the thought at the instant it moves. “Volition is the basis of what is done.” The karma of hand and mouth comes after that thought has already moved.

Third, there is no need to find a custodian for cause and fruit. Bond or continuum, each seeks something to keep custody of the karma one has performed. “Karma is ultimately empty, bears the character of quiescent extinction, and is separate from intrinsic nature. What dharma is there to annihilate, what dharma to lose?” Since there was never a thing that could be lost, no custodian need be established.

Fourth, what is plainly seen is not evidence of real existence. “Although this karma is unreal, it can be seen.” What can be seen may still lack intrinsic nature.

Carried Forward: Jizang’s Mirror

What precedes is the meaning of the chapter itself. What follows is illuminated from the perspective of the Sui-Tang Three Treatise lineage—a mirror is not the treatise, but another way of reading the same group of verses.

Jizang’s Commentary on the Treatise on the Middle divides this chapter into seven sections, saying that each first establishes and then refutes, for seven establishments and seven refutations; the final section refutes what is plainly seen. He further says the chapter addresses two targets. First are wrong-view non-Buddhists such as the six teachers, who say there is no black karma and no black karmic fruition, and likewise no white karma. Second are proponents of a “vast emptiness” who say everything is empty and there is neither demerit nor merit. He therefore states the chapter’s thesis: though ultimately empty, wholesome and unwholesome karma plainly are not lost. As for the concluding analogy of the conjured person, he distinguishes the two truths. That the illusory person is not a real person is spoken from ultimate truth. Yet there is not no illusory person, revealing the meaning of not nonexistent; not being a real person reveals the meaning of not existent. Together, this clarifies that conventionally the six destinies are plain while the four alternatives are always cut off, and while the four alternatives are cut off, the six destinies plainly are not lost.

This mirror connects the chapter to the Three Treatise framework of refuting error, revealing rightness, and distinguishing the two truths. Its articulation is Jizang’s addition, not the words of Nāgārjuna’s verses or Qingmu’s extended commentary. The readings are recorded side by side without adjudication, each returned to its own place.

Across the preceding seventeen chapters, readers have grown accustomed to one kind of ending: this or that cannot be established; this or that is unobtainable. Yet this chapter’s first sentence is not a refutation but a request for directions.

The questioner says: if all dharmas are ultimately empty, without arising or ceasing, and this is called the true character of dharmas, then—“How does one enter?”

It is an excellent question. Emptiness has been understood, but after understanding it, one remains standing in place. Where does the first step fall? If emptiness is only a conclusion, one can do nothing with it.

The respondent gives only one place to begin: “By extinguishing attachment to self and what belongs to self, one obtains the emptiness of all dharmas. The wisdom of no-self is called entry.”

Notice the order. It is not that one first realizes the emptiness of all dharmas and then, incidentally, self disappears. Only after attachment to self and what belongs to self is extinguished does the emptiness of all dharmas come into hand. Moreover, what one enters is not a place; entry is that wisdom—“The wisdom of no-self is called entry.” This sentence sets the direction of the chapter: it is not discussing emptiness but explaining how to walk.

Self and the Five Aggregates: Both Gates of Identity and Difference Are Closed

The questioner next asks: how do we know that dharmas are without self? The verses begin:

“If self is the five aggregates, self undergoes arising and ceasing; if self differs from the five aggregates, it does not have the aggregates’ character.”

Qingmu says the “spiritual self” posited by non-Buddhists can stand in only two ways: either the five aggregates are the spiritual self, or a separate spiritual self exists apart from them.

The first position collapses by itself. The five aggregates arise and cease; Qingmu states it with utter brevity: “because once arisen they deteriorate.” What arises is already deteriorating. If self is the five aggregates, it must arise and cease with them. But something that arises and ceases cannot bear the work non-Buddhists need it to bear—the self that passes from this life to the next and remains the same throughout.

The second position cannot even be stated coherently. If the self differs from the aggregates, it has none of their character; but “apart from the five aggregates there is no further dharma.” Qingmu therefore asks: by “what character and as what dharma does it exist?” If the self is denied the aggregates’ character yet is said to exist, another character must be produced. If none can be produced, “self” is only a name.

Someone tries to rescue it through space: self exists apart from the five aggregates just as space does. Qingmu points back to an earlier chapter. The “Examination of the Six Elements” already said that “no dharma exists that can be called space.” Something unable to stand in its own right cannot provide support.

When both gates of identity and difference are closed, a thing has nowhere to stand. The Treatise on the Middle always tests whether something can be obtained in this way: it does not search for the thing but examines, one by one, every place where it could remain.

No Spiritual Self Can Be Found among the Four Kinds of Warrant

Someone retreats a step. Do not argue from character, but from warrant: because there is trustworthy warrant, the spiritual self exists.

Qingmu lists the four kinds of warrant accepted in his setting: present fact, inferential knowledge, analogy, and the teaching of worthy sages, then searches within each. It is absent from present fact—no one directly perceives a spiritual self.

It is absent from inferential knowledge too. Qingmu states the rule of this gate clearly: “Inferential knowledge means first having seen and then knowing by analogy, as someone first sees fire with smoke and later, merely seeing smoke, knows there is fire.” Inference has a precondition: you must first have directly seen the pair present together. First see fire and smoke together; later, smoke alone can yield the inference of fire. What of the spiritual self? “Who first saw the spiritual self joined with the five aggregates and then, seeing the aggregates, knew the spiritual self existed?” Who, and when, saw the spiritual self and the five aggregates present together? Not once. Since the precondition has never been met, this path has had no starting point from the beginning.

The opponent further subdivides inference into three kinds—as with the original, as with the remainder, and commonly seen—and gives the sun as example. No one sees the sun walking, but one sees it emerge in the east and disappear in the west. Since people have the appearance of going, the sun too is known to go. In the same way, suffering, pleasure, hate, love, and awareness should depend upon something—“as when seeing subjects, one knows they must depend upon a king.”

Qingmu answers with the same measure. What is commonly seen works only because “one first saw a person joined with the dharma of going and reaching another place.” Person and going were first seen together; the aggregates and self have never been so seen. The example changes; the precondition remains the same.

Last comes the teaching of worthy sages. This gate seems strongest. People have not seen hells, heavens, or Uttarakuru, but know them by trusting sacred testimony. Qingmu does not deny the gate. He only identifies the origin of sacred testimony’s trustworthiness: “What sages teach, they first saw with their own eyes and then taught.” Sacred speech is trustworthy not because the speaker has high status but because that speaker first saw. Concerning the spiritual self, however, “there is no one who first saw the spiritual self and later spoke of it.” No one speaks after having first seen it. This gate too therefore cannot produce a self.

After walking all four gates, Qingmu concludes: “Therefore, when the spiritual self is sought among the four warrants and every warrant, it is unobtainable.” He adds another cut. In the “Examination of the Faculties,” seeing, seer, and visible object have already been refuted, so the spiritual self is refuted likewise. “How much less could a spiritual self exist among false conceptual recollections when even coarse dharmas visible to the eye are unobtainable?”

The weight of this entire passage lies not in its conclusion but in its method. It does not announce no-self as dogma; it invites every support needed by “self exists” to appear one by one, then tests whether each can stand.

From Self to What Belongs to Self: The Path Empties All the Way Down

Once self is unobtainable, the verse immediately takes the next step:

“If no self exists, how could there be what belongs to self? Because self and what belongs to self are extinguished, this is called obtaining the wisdom of no-self.”

What belongs to self is “mine” only because that self is first present. Withdraw self and what belongs to self has nowhere to hang. The broad circle of things a person labors to protect are not each guarded separately; all are fastened to that one center. When the center disperses, the whole circle loosens.

The path of cultivation is not mysterious. Qingmu says plainly: “By cultivating the factors of the noble eightfold path and extinguishing the causes and conditions of self and what belongs to self, one obtains determinate wisdom of neither self nor what belongs to self.” No separate contemplation is established; it is the noble eightfold path. Nor is self what is extinguished—self was originally unobtainable and cannot be extinguished. What is extinguished are “the causes and conditions of self and what belongs to self,” the conditions that repeatedly establish self.

The verse next gives this wisdom two names:

“One who obtains the wisdom of no-self is said to contemplate reality; a person who obtains the wisdom of no-self is exceedingly rare.”

One name is “contemplation of reality”—an affirmative judgment rarely seen in the entire treatise: this way of seeing accords with the facts. The other is “rare.” This is not praise but fact. One may understand the sentence “no-self,” yet it is extraordinarily difficult truly to have no center amid events.

One more point must be watched. Qingmu says: “What is called neither self nor what belongs to self is itself unobtainable in the ultimate meaning.” Even neither self nor what belongs to self is unobtainable ultimately. This sentence plants a stake for what follows, preventing one from turning no-self into a new attainment.

Internal and External Utterly Cease: When Appropriation Ceases, the Body Ceases

The verse turns toward fruition:

“Because internal and external self and what belongs to self are utterly extinguished and absent, the various appropriations cease; when appropriation ceases, the body ceases.”

Qingmu’s explanation forms a causal chain. Self and what belongs to self obstruct the wisdom-eye; when that eye is obstructed, reality cannot be seen: “Ordinary people cannot see reality because self and what belongs to self obstruct the wisdom-eye.” Read this slowly. Ordinary people are not eyeless; the eye is covered. What covers it is precisely self and what belongs to self. When these cease, afflictions cease; when afflictions cease, true character appears. Then, “because internal and external self and what belongs to self cease, the various appropriations cease; because appropriations cease, immeasurable later bodies all cease. This is called nirvāṇa without remainder.”

Why does cessation of appropriation mean cessation of the body? Because appropriation is the link fastening one to later existence. For a person without a center, reception no longer congeals into “my suffering and pleasure.” Without congealing, there is no grasping; without grasping, a later body has nowhere to arise. Thus cessation here does not mean destroying the body, but the later segment no longer connects.

Karma and Affliction Arise from Conceptual Recollection and Discrimination

The questioner asks: what of nirvāṇa with remainder?

Qingmu answers: “Because the various afflictions and karmas cease, the mind is said to obtain release.” From where do afflictions and karma come? He traces back two levels: “These afflictions and karmas all arise from conceptual recollection and discrimination and have no reality; the various conceptual recollections and discriminations all arise from conceptual proliferation.”

See the line: conceptual proliferation → conceptual recollection and discrimination → karma and affliction. A person first tells an event into a certain shape within the mind; once told, it is taken as true, and once taken as true, karma and affliction arise. The verse concludes:

“Because karma and afflictions cease, this is called release; karma and afflictions are unreal—upon entering emptiness, conceptual proliferation ceases.”

“Karma and afflictions are unreal”—these four words matter. If karma and affliction really existed, they could only be counteracted one by one, an endless task. Precisely because they arise from that layer of telling, when that layer quiets, they have no support. Qingmu gives exactly this: “When one obtains the ultimately empty true character of dharmas, the various conceptual proliferations cease. This is called nirvāṇa with remainder.”

Nirvāṇa without remainder and with remainder here speak of one event from two ends: one says no later body arises; the other says the mind has obtained release.

Sometimes Self Is Taught; Sometimes No-Self

Here the whole chapter pivots:

“Buddhas sometimes teach self, and sometimes teach no-self; in the true character of dharmas, there is neither self nor nonself.”

How can the Buddha teach both? Qingmu sets out three listeners. Some “do not know to fear wrongdoing” and their minds are immature. To them, self is first taught so they know their actions have an owner and first stand firm in cause and fruit. Others have attained the path and know dharmas are empty; for them, self is merely a provisional name and causes no fault. Still others cultivate such merits as generosity and keeping precepts, yet “weary of saṃsāra’s suffering and fear nirvāṇa as eternal extinction.” For them, no-self is taught precisely to dismantle the self that “will disappear.”

Qingmu states the reason again: “Dharmas are merely conjoined through causes and conditions; at arising, they arise emptily, and at ceasing, they cease emptily.” Between these, there is no one who is born or destroyed.

Teaching self and teaching no-self therefore do not report the structure of the world; each dismantles a particular attachment in a particular person. The third and fourth lines of the verse give the ground: within true character “there is neither self nor nonself.” Neither side is established.

No-Self Is Not Allowed to Become a Determinate Teaching Either

The questioner immediately seizes this point. If no-self is true and self is taught only conventionally, what fault is there?

Qingmu’s answer displays the chapter’s sharpest discernment: “No-self exists because it refutes the dharma of self. Since self is determinately unobtainable, how could there be a determinate no-self?”

The statement of no-self is spoken in response to self. If self determinately existed, refuting it could yield a determinate absence. But since self is determinately unobtainable, the absence opposed to it loses its opponent and cannot stand as a determinate thing. Once words become fixed doctrine, people grasp them. Qingmu directly states the consequence: “If no-self determinately existed, that would be annihilation and would give rise to craving attachment.” To clutch no-self is not release; it is craving another thing. He cites the Prajñā as witness: for a bodhisattva, self is not practice, and no-self is not practice either.

This passage guards the chapter’s gate. The first half expends such effort refuting self; if the reader then enthrones “no-self,” all the preceding work is wasted.

Activities of Mind and Paths of Language Are Cut Off

The questioner retreats to the final step. If neither self nor nonself, neither empty nor nonempty, is taught, what does the Buddha’s Dharma teach?

The answer: “The Buddha teaches the true character of dharmas. In true character, there is no path of language, and the various activities of mind cease.”

Why must the activities of mind cease? Qingmu does not disparage mind; he explains its origin: “Mind arises through conditions by grasping marks and exists because of fruition from past karma; it cannot truly see dharmas.” Mind itself is dependently arisen, and arises by grasping marks—it has already taken a mark the moment it moves. Using something that must grasp marks to see no-character mismatches tool and object.

The questioner objects. Ordinary minds may not see reality, but surely worthy sages’ minds can. Why say all activities of mind cease?

Qingmu gives three pieces of evidence. If mind were real, why use gateways to release such as emptiness? Why, among all meditative absorptions, is the attainment of cessation ranked first? And why does one ultimately return to nirvāṇa without remainder? All three point to one fact, and he concludes: “All activities of mind are false; because false, they should cease.” True character “transcends the associated mental dharmas, has neither arising nor ceasing, bears the character of quiescence, and is like nirvāṇa.”

Why Say Only “Like Nirvāṇa”?

The questioner presses one word. The sūtras say dharmas have always borne the character of quiescence and this itself is nirvāṇa. Why say only “like nirvāṇa” here?

Qingmu answers that this is said for those attached to dharmas. They divide dharmas in two: “They say nirvāṇa is quiescent and do not say the world is quiescent.” They grant the name quiescence only to nirvāṇa. The treatise teaches instead that all dharmas are empty of intrinsic nature and bear the character of quiescence. “Because those attached to dharmas do not understand, nirvāṇa is used as an analogy.”

Thus the word “like” is not a retreat but a bridge built from what the listener already accepts. Since you accept that nirvāṇa is empty, signless, quiescent, and without conceptual proliferation, see the world in that same way: “All worldly dharmas are likewise.”

Here analogy is not rhetoric; it is one rung of a ladder lowered to the listener.

Four Alternatives: Four Teachings within One Sentence

The questioner asks again: if paths of language are cut off, how can people be made to know true character?

The respondent then speaks the famous four alternatives:

“All is real; all is unreal; all is both real and unreal; all is neither real nor unreal—this is called the Dharma of the Buddhas.”

Qingmu assigns each its use. All is real because when the real nature of dharmas is sought, “everything enters the one equal character of ultimate meaning, namely, no-character.” His analogy is “many streams of different colors and flavors entering the great ocean and becoming one color and one flavor.” All is unreal because, before entering true character, every dharma examined separately lacks real substance and “exists only through the conjunction of conditions.” All is both real and unreal because of three grades of sentient beings: “those of the higher grade contemplate dharmas as neither real nor unreal”; the middle grade sees all as real and all as unreal; the lower grade’s wisdom is shallow and can see only nirvāṇa, an unconditioned dharma, as indestructible and real, while seeing saṃsāra, a conditioned dharma, as false and unreal. Neither real nor unreal is taught to refute the two extremes of real and unreal.

Someone asks more sharply: elsewhere the Buddha clearly teaches departure from “neither existence nor nonexistence.” Why is neither existence nor nonexistence here called the Buddha’s Dharma? Qingmu distinguishes them clearly. Elsewhere it is spoken “to refute four kinds of craving attachment.” There the four alternatives are four nests people clutch, and so must be left. “Here there is no conceptual proliferation within the four alternatives, and upon hearing the Buddha’s teaching one attains the path.” Here the alternatives are not clutched; each is a gate leading out.

The same four alternatives, when grasped, are four craving attachments; when not grasped, they are four skillful means. The difference lies not in the sentence but in the listener.

The Character of True Character: Known for Oneself, Not Followed from Another

At last the questioner reaches the root. By what character can someone who has obtained true character be recognized? What is true character like?

The verse answers:

“Known for oneself and not followed from another, quiescent and without conceptual proliferation, without difference or discrimination: this is called true character.”

Explaining “not following another,” Qingmu gives the most difficult circumstance. Non-Buddhists display supernormal power and say this is the path or this is not the path; the practitioner “trusts one’s own mind and does not follow.” Even if non-Buddhists transform their bodies and the practitioner does not know they are not the Buddha, because one “well understands true character, one’s mind cannot be turned.”

We must see that “known for oneself” is not stubborn insistence on one’s own view. It means one has seen for oneself. One who has seen cannot be turned by another’s supernormal display. This is the visible character of true character in a person.

Next comes quiescence. Qingmu says: “Because no dharma can be taken or rejected here, it is called the character of quiescence.” Quiescence is not a blank mind; it means nothing is held that must be taken, and nothing held that must be thrown away.

Next comes absence of conceptual proliferation. Qingmu divides proliferation in two: “one is the discourse of craving, the other the discourse of views.” One clings to what is loved; the other clings to what is seen. The first half of the chapter refutes the discourse of views; this passage guards against the discourse of craving. Even “no-self” can be loved, how much more everything else. When both proliferations are absent, conceptual recollection, discrimination, and the character of difference naturally disappear. This is called true character.

Neither Identical with nor Different from the Cause: Neither Annihilated nor Permanent

The questioner’s last doubt: if all dharmas are empty, is this not annihilation? If they neither arise nor cease, does this not fall into permanence?

Qingmu first identifies the source of the doubt: “You now greedily cling to grasping marks and see the faults of annihilation and permanence in true character.” Annihilation and permanence are not faults within true character; they are what the eye that grasps marks sees.

He then gives an affirmative answer:

“If a dharma arises from conditions, it is neither identical with nor different from its cause; therefore it is called true character, neither annihilated nor permanent.”

Qingmu states the reasoning very plainly: “If fruit differed from cause, that would be annihilation; if fruit did not differ from cause, that would be permanence.” If fruit were wholly different from cause, the cause would be severed with one cut: annihilation. If fruit did not differ at all, the cause would simply be carried unchanged into fruit: permanence. Dependently arisen reality is neither—because not identical, it is not permanent; because not different, it is not annihilated.

Annihilation and permanence entangle people because they first see cause and fruit as two things, then ask whether the two are one or two. Dependently arisen dharmas are not two things of that kind, so the question itself disperses.

The Flavor of Ambrosia and Three Ways of Attaining the Path

The final question asks about benefit. What good comes from understanding this way?

“Neither one nor different, neither permanent nor annihilated; this is called the flavor of ambrosia through which the World-Honored Ones teach.”

Qingmu says that if practitioners penetrate this meaning, then regarding all dharmas they are neither one nor different, neither annihilated nor permanent. They “at once extinguish afflictions and conceptual proliferation and attain the constant bliss of nirvāṇa.” He explains the ambrosia analogy: “As people in the world say, one who obtains the heavenly draught of ambrosia is without age, illness, death, or any decline and distress.” This dharma of true character is the genuine flavor of ambrosia.

The same flavor of ambrosia yields three fruits in different people. “If one obtains the true character of dharmas and extinguishes the afflictions, this is called the Śrāvaka teaching.” “If one gives rise to great compassion and awakens the unsurpassed mind, this is called the Mahāyāna.” If a Buddha does not appear and the Buddha’s Dharma is absent, the pratyekabuddha gives rise to wisdom through seclusion. The final verse addresses this third mode:

“If a Buddha does not appear in the world and the Buddha’s Dharma has vanished, the wisdom of pratyekabuddhas arises from seclusion.”

Qingmu completes the account. After a Buddha enters nirvāṇa without remainder and the teaching left behind has vanished, if someone from a past life is due to attain the path, that person briefly contemplates the conditions of disenchantment and “enters the mountain forest alone, withdraws from clamor, and attains the path. Such a person is called a pratyekabuddha.”

Three fruits, one flavor. The distinction lies not in several dharmas realized but in compassionate vow and time.

What a Reader Can Carry Away from This Chapter

First, know where to enter. Emptiness is not an esoteric principle to penetrate. There is only one entry: “By extinguishing attachment to self and what belongs to self, one obtains the emptiness of all dharmas.”

Second, know that even no-self cannot be clutched. “Self is determinately unobtainable—how could there be a determinate no-self?” The words that refute self are not a new conclusion to enthrone.

Third, know why teachings change. Sometimes self, sometimes no-self, sometimes real, sometimes unreal—the medicine changes, not the pathology. Within true character there is “neither self nor nonself,” and this has never changed.

Fourth, know the character by which to recognize a person. It is not how much one can say, but “known for oneself and not followed from another,” with nothing in the hand that must be taken or rejected.

Carried Forward: Jizang’s Mirror

What precedes is the meaning of the chapter itself. What follows is illuminated from the perspective of the Sui-Tang Three Treatise lineage—a mirror is not the treatise, but another way of reading the same group of verses.

Regarding the opening question “How does one enter?” Jizang’s Commentary on the Treatise on the Middle says that if one establishes the practitioner as what enters and true character as what is entered, these have already become two objects, agent and object. The person who enters and the dharma entered must both be removed; only when ultimately neither remains is it called entry. He further uses this to distinguish the depth of Great and Lesser Vehicle contemplation: Śrāvaka contemplation treats self as skillful means and no-self as reality, while this chapter clarifies that both self and no-self are skillful means and that neither self nor no-self is true. Therefore, bodhisattva contemplation is deeper. As for the four alternatives, he judges them to be skillful gates of true character. By letting the mind range through the four gates, one enters true character: the four gates are function, while not-four is substance.

This mirror connects the chapter to the Three Treatise framework of ranking depth and distinguishing substance and function. Its articulation is Jizang’s addition, not the words of Nāgārjuna’s verses or Qingmu’s extended commentary. The readings are recorded side by side without adjudication, each returned to its own place.

Time is the hardest thing to doubt. Money can be absent, the body can decay, thoughts can stop; yet the words “yesterday,” “now,” and “tomorrow” seem to remain whatever happens. Even if not one thing in the world were left, time would seem to keep flowing.

This chapter asks about that “keep.”

Nāgārjuna does not begin by trying to prove that time does not exist. He lets the one who asserts time give his own reason first, then follows that reason all the way to its end. At the end, the speaker discovers that the very reason he offered is precisely why time cannot stand by itself.

This Question Comes Prepared

〔0025c02〕The questioner opens by saying: “Time should exist, because it is established through dependence.”

We should see the weight of this sentence. He is not naively saying, “Of course time exists.” He offers an argument: time is established because the three times depend upon one another. With the past, future and present can be spoken; with the present, past and future can be spoken; with the future, past and present can be spoken. Remove any one, and the other two cannot be established.

Nor does he speak only of time. He continues: “Dharmas such as upper, middle, and lower and one and different likewise exist through mutual dependence.” The triad of upper, middle, and lower and the pair one and different are likewise established only through mutual dependence. He places an entire class of things on the same principle: whatever is established in dependence exists.

This is an intelligent thesis, because dependence is among our most familiar experiences: without upper, lower cannot be spoken; without long, short cannot be spoken. He turns that familiarity into evidence and asks: how could a relation this secure be absent?

What Nāgārjuna does next is turn the word “dependence” over and look at its other side.

If They Truly Exist Because of the Past, the Other Two Times Must Reside within the Past

Nāgārjuna’s first verse says: “If future and present exist because of past time,” then “future and present should be within past time.”

The key to this step is a general rule Qingmu states for the reader: “Whenever a dharma is established in dependence upon a place, that dharma should exist in that place.” If a dharma is established only by depending on a certain place, it should be present in that place.

Qingmu immediately gives a visible example: “Because light is established through a lamp, wherever the lamp exists, light should exist.” Light exists because of a lamp, so light is present where the lamp is. One does not say that light depends upon the lamp while placing the lamp here and the light somewhere else. If cause and fruit truly stand in dependence, the fruit is located where its cause is.

Apply the rule to the three times: because future and present are established through the past, they must be within past time.

Then trouble begins. Qingmu says that if so, “all three times would be called past time.” If future is within the past and present is also within the past, all three occupy this one place called past and must all be called past. Three names collapse into one.

Once collapsed into one, the trouble deepens. If everything is past, future and present are absent. Qingmu then presses: “If future and present time are absent, past time should likewise be absent.” Why can even the past not be preserved? Because “past time is called past time only in dependence upon future and present time.” The name past can be spoken only relative to future and present. When two of the three die, the one that remains can no longer be itself.

Here it is worth pausing. The questioner originally used dependence to establish time. Nāgārjuna does not deny dependence; he only follows it one step backward. If dependence is bidirectional, “just as future and present time are established because of past time, past time should likewise be established because of future and present time.” When one end falls, the other cannot stand.

Qingmu’s judgment is brief: “This cannot be so.”

Then Say They Are Not within the Past—but Dependence Disappears with That

The questioner still has a retreat: one could say that future and present are not within past time, but nevertheless are established because of the past. Would that not avoid collapsing the three times into one?

Nāgārjuna’s second verse blocks that point: “If future and present are absent within past time,” then “how can future and present time depend upon the past?”

Qingmu explains this step with another analogy. He says: “If the three times each have a different mark, they should not be established through mutual dependence,” but should be “like such objects as a jar and a robe, each established separately and not dependent upon the other.” A jar is a jar and a robe a robe; each has its own appearance and is established separately, with neither relying on the other. If the three times are not present within one another, they become unrelated things like jar and robe. Unrelated things have no relation of dependence.

The questioner is therefore trapped at both ends. Say they are present within one another, and the three times collapse into one and the entire system fails. Say they are not present within one another, and one admits that the three times stand independently, so the original reason—“established through dependence”—invalidates itself.

Qingmu adds an important passage that unfolds this bind through all three rotations: “Without dependence upon past time, future and present time are not established”; without the present, past and future are not established; without the future, past and present are not established. Each of the three times takes a turn as cause and fruit. None can stand outside the circle and guarantee the others.

Could They Simply Not Depend?

〔0026a04〕The questioner therefore asks: “What fault is there if future and present time are not established because of past time?” What happens if dependence is simply abandoned?

Nāgārjuna’s answer is extremely short: “Without dependence upon past time, there is no future time,” and “there is likewise no present time; therefore, both times are absent.”

Qingmu brings the sentence down to its most concrete point. He does not ask abstractly whether establishment succeeds, but asks about location: “Where does present time exist?” If present is not spoken relative to past, exactly where does this thing called present point? The same holds for future: without a relative reference, it cannot even be indicated.

Thus, by this passage both paths have been exhausted: dependence does not establish it, and nondependence does not establish it. Qingmu gathers the result into one sentence: “Because time exists only through such mutual dependence, time in truth is absent.”

This must be read precisely. It does not say that time has never appeared in our days. It says that time exists only in dependence; what exists in dependence has no body of time that stands by itself. Years still pass, but the “thing” said to pass cannot be found.

This Refutation Does Not Stop with Time

The fourth verse widens the field: “Because of this meaning, one knows the other two times,” and even “upper, middle, lower, one, and different—dharmas such as these are all absent.”

The questioner had casually staked upper, middle, lower, one, and different on the same reasoning. Nāgārjuna accepts the wager exactly as offered: since you say they are likewise established in dependence, they likewise cannot withstand this question.

Qingmu’s example belongs to everyone: “Because there is an upper, there are a middle and a lower; apart from upper there is no middle or lower.” Remove upper and middle and lower immediately lose their place. Conversely, “if middle and lower existed apart from upper, they should not depend upon one another.” If middle and lower can establish themselves apart from upper, they never depended upon it in the first place, and the thesis withdraws itself.

The pair one and different is tighter still. Qingmu says: “Because of one, there is different; because of different, there is one.” Different is said relative to one, and one relative to different. He then seals both sides: “If one truly existed, it should not exist because of different”; “if different truly existed, it should not exist because of one.” To exist truly means to be established by oneself. What is established by itself does not need the other in order to exist. Once one admits dependence on the other, one admits that it does not truly exist by itself.

Qingmu finally extends the point to the entire class in one sentence: “Dharmas of this kind should likewise be refuted in this way.”

Anything named only through comparison—long and short, large and small, high and low, before and after, this and that—is included in the sentence. Time is merely the most stubborn example.

Do Years, Months, Days, and Moments Not Count as Evidence?

The questioner finally invokes common sense: “Since distinctions such as years, months, days, and moments exist, one knows that time exists.”

This is the most powerful question because it appeals to experience anyone can check. Days truly pass one by one; moments truly move minute by minute. If measures can be distinguished, how could time be absent?

Nāgārjuna’s fifth verse does not deny these measures. It turns back to ask in what state this supposed time itself exists: “Time at rest cannot be found, and time gone cannot be found”; therefore, “if time cannot be found, how can the mark of time be spoken of?”

Qingmu opens the two gates: “If time does not remain at rest, it should not be findable; yet time at rest likewise is absent.” If time flows without remaining, at the moment one reaches to grasp it, it is no longer that moment, so it cannot be found. If time remains, what remains unmoving is no longer time: past, present, and future crowd into one place, and years and months lose their meaning. Moving, it cannot be grasped; unmoving, it cannot be grasped.

If it cannot be grasped, its appearance cannot be stated. Qingmu continues: “If there is no mark of time, there is no time.”

The measures can be distinguished—that is true. Yet the thing measured by those measures is never produced.

Time Exists Because of Things

The sixth verse is where the chapter comes to rest: “Time exists because of things; apart from things, where is time?” The next sentence completes the point: “Things themselves have no possession—how much less could time exist!”

Qingmu’s explanation has only two sentences, yet they are the root of the entire chapter: “Time is named because things arise; apart from things there is no time.”

The name time is established on the arising of things. Flowers bloom and fall, and so there are spring and autumn; a child grows taller, and so there are years. What is called a span of time is really a certain group of things arising, changing, and fading. Remove the things, and one does not leave behind an empty container of time—even the place on which the name time could be established is absent.

Nāgārjuna’s final question therefore advances by layers: things themselves have no possession; how much less time, which receives its name only in dependence upon things.

At the end, Qingmu gathers the preceding chapters of the entire treatise as the basis: “In the preceding chapters, things have already been refuted through many kinds of causes and conditions. Since things are absent, where could time exist?”

This sentence closes the lock on the chapter. The chapter can end so lightly not because it alone wields enormous force, but because the preceding chapters have already refuted things one by one. Time is a name hung on things; when the hook is absent, what was hung upon it naturally falls.

What a Reader Can Carry Away from This Chapter

First, see both sides of dependence. People habitually take dependence as sturdy evidence: when two ends fit so tightly, how could they be false? This chapter answers that the tighter they fit, the more clearly neither end can produce a share that belongs to itself. Dependence establishes names, not self-existing bodies.

Second, notice how you speak of time. When people discuss time, they often quietly treat it as an independent thing: time passes quickly, there is no time, time softens everything—as though something were flowing there. Qingmu will say only, “Time is named because things arise.” Apart from these very things that arise and cease, there is nothing else called time that flows.

Third, this question can be carried elsewhere. Nāgārjuna gathers upper, middle, lower, one, and different into the same refutation. In daily life, every judgment that stands only by comparison—good and bad, success and failure, fast and slow, winning and losing—can be asked: if the contrasting end is removed, what remains of this end?

Carried Forward: Jizang’s Mirror

The preceding is the meaning of the chapter itself. What follows uses the Sui–Tang Three Treatises tradition as a mirror held alongside it—a mirror is not the treatise, but another way to see the same verses.

In his Commentary on the Treatise on the Middle, Jizang first cites an older Vibhāṣā interpretation to set the chapter’s tone: dharmas themselves are time; time is provisionally named because of dharmas, and apart from dharmas there is no separate time. Although the time of the three periods has no separate body, the dharmas within time are certainly not absent. This distinction separates refuting a body of time from annihilating dharmas.

For the chapter’s first refutation, he lays out five successive challenges: first, mutual dependence means mutual presence; next, mutual presence makes all three share the name past; next, sharing the name past removes the other two times; next, removing the other two also removes the past; and finally, without the future as its cause of designation, the past itself is absent. He raises an objection: a lamp and its light exist simultaneously, so one can say light is present where the lamp is, but the three times are earlier and later—how can they be compared? His answer is that although earlier-and-later differs from simultaneity, the relation of mutual causation does not differ, so the challenge still applies. He extends the same set of challenges to the mutual dependence of saṃsāra and nirvāṇa.

He presses the two gates of remaining and going still further. If the body of time remains, it is constantly one single moment; within one moment there are no years or months, and without years or months that moment itself cannot count as a moment. If time does not remain, each moment ceases separately with no continuity, so how could years accumulate? He follows the worldly calculation—moments accumulate into days, days into months, months into years—and asks: if one day ceases, there is a later day without the earlier, so how can days accumulate into a month? If one day does not cease, there is only that one day, an earlier without a later, and again no month can accumulate. Neither gate succeeds.

In the final verse, he judges the first half to say that time exists because of things and thus that things are the body of time, while the second half closes by saying that since things are absent, time is absent. He gives flowers and fruits by which spring and autumn are known as his example: flowers, fruits, and other things have already been refuted in many ways; because things are absent, time is absent.

This mirror places the chapter within the interpretive structure of the Three Treatises lineage. Its divisions and levels of challenge are Jizang’s additions, not the original words of Nāgārjuna’s verses or Qingmu’s extended commentary. The readings are recorded side by side without adjudication, each returned to its own place.

Seed, soil, water, sunlight, and season come together one by one, and a sprout appears. Milk is left awhile and becomes curds. A lump of clay turns a few times in a potter’s hands and a jar is formed. Everyone sees these things, and they are where the questioner in this chapter begins:

“Question: Because a fruit is visibly seen to arise when many causes and conditions conjoin, one should know that this fruit exists from the conjunction of many conditions.”

What he says is not exactly wrong—he has merely taken one extra step. “A fruit is visibly seen to arise” is what is seen; “this fruit exists from the conjunction of many conditions” is what is inferred. From beginning to end, the chapter’s twenty-four verses pursue this one step: exactly where can the word “exists” be placed?

The First Threshold: Is the Fruit within Conjunction, or Not?

The respondent takes no detour. The first verse divides the questioner’s statement in two:

“If a fruit arises when many conditions conjoin, it already exists within conjunction; why would conjunction need to produce it?”

The meaning is direct: you say that the fruit exists from conjunction, so the fruit is already within conjunction. Since it is already there, what remains for conjunction to do? Qingmu fixes the sentence as a criterion—

“If the fruit already had a determinate body, it should not arise from conjunction.”

The four words “already had a determinate body” are the pivot of the entire chapter. If it has a determinate body, it is not something waiting for another to produce it; if it needs another to produce it, it has no determinate body. The two cannot coexist.

The questioner therefore retreats one step: then let the fruit originally be absent within conjunction, while arising from the many conditions. The respondent returns a single sentence:

“If a thing lacks intrinsic nature, that thing ultimately does not arise.”

The next two verses nail down one end apiece. Whoever says the fruit is present must produce it:

“It should be obtainable within conjunction, yet in fact it cannot be found.”

Qingmu supplies the means of retrieval: “If it is form, it should be visible to the eye; if not form, it should be knowable to the mind.” Both paths have been tried, “yet in fact the fruit cannot be found within conjunction.”

Whoever says the fruit is absent faces greater trouble:

“If the fruit is absent within the conjunction of many conditions, then those causes and conditions would be the same as noncauses and nonconditions.”

Here Qingmu gives the chapter’s first analogy and its most important one:

“Milk is a cause and condition for curds. If curds are absent within milk, they are likewise absent within water. If curds are absent within milk, milk is the same as water, and one should not say that curds come only from milk.”

Milk can become curds; water cannot. No one can erase this difference. But if “no fruit” means determinate and total absence, the total absence within milk is no different from the total absence within water. If they are the same, on what ground do curds come only from milk? This passage does not prove that curds are hidden inside milk. It shows that when “absence” is read as determinate absence, cause and noncause can no longer be distinguished. Neither end can hold. This is the questioner’s actual situation.

It is worth pausing to see how this pair bites together. Saying the fruit is present fails because it “cannot be found”: to say that something one cannot produce is already there is only a statement. Saying the fruit is absent fails through “sameness”: complete absence is the same everywhere, leaving no reason why milk produces curds but water does not. One side requires it to be producible; the other requires it to be distinguishable. Each time the questioner evades one side, he strikes the other. The many turns of the following verses—temporal sequence, transformation, pervasive possession, the three times, conjunction and nonconjunction—still reduce to this pair.

Does the Cause Cease Before, or After?

The questioner changes direction: instead of asking whether the fruit is within conjunction, he turns to time. The cause finishes serving as cause for the fruit, then ceases, and the fruit arises. Does this not avoid the difficulty?

“If a cause serves as cause for the fruit and ceases after serving as cause, that cause has two bodies: one gives, and one ceases.”

Qingmu is explicit: “One is the cause that gives; the second is the cause that ceases. This cannot be so, because one dharma would have two bodies.” If the cause that gives the fruit and the cause that ceases are the same, how could one half be giving while the other half ceases?

The questioner retreats again: then say that the cause ceases without ever serving as cause for the fruit, while the fruit still arises. The respondent follows through—

“The cause ceases and the fruit arises; this fruit is causeless.”

That end is one the questioner himself will not accept: “Among all fruits visibly seen, none arises without a cause.”

The questioner therefore turns to simultaneity: at the very moment the conditions conjoin, the fruit arises. The verse answers:

“Producer and what is to be produced would both exist at the same time.”

Qingmu adds an analogy: “A father and son cannot be born at the same time.” If what produces and what is produced are both present simultaneously, the words father and son have no place to stand.

What if the fruit is placed first? The fruit arises first, and the conditions conjoin later—

“It is apart from causes and conditions and is called a causeless fruit.”

If the fruit already exists without waiting for the conditions, the conditions that gather later are mere spectators. This branch returns by itself to “causeless.”

Transformation: The Measure of Clay-to-Jar and Milk-to-Curds

The questioner’s last device is “transformation”: the cause does not cease but transforms into the fruit. The verse first exposes what this conceals:

“If the cause transforms into the fruit, the cause itself reaches the fruit; then the previously arisen cause arises once more after having arisen.”

Qingmu distinguishes two kinds of cause—“one arises before; the second arises together.” If the previously arisen cause transforms into the fruit, it must arise a second time, yet “what has already arisen should not arise again.” He then gives a remarkably clean judgment:

“If it remains identical, it is not called transformation; if it has transformed, it cannot be called identical.”

The questioner objects with a good example: a lump of clay transforms into a jar. Only the name ceases; the body of the cause remains. Qingmu’s answer has three levels. First, “a lump of clay first ceases and then a jar arises; this is not called transformation”: clay ceases before the jar arises, which is prior cessation and later arising, not one thing migrating across. Second, “the body of the lump of clay does not produce only one thing; jars, basins, urns, and the rest all come from clay.” If a particular jar were truly hidden in the clay, where would basin and urn emerge from? Third, if lump of clay is merely a name, a name cannot transform into a jar.

Then what does count as transformation? Qingmu supplies the measure: “What is called transformation is like milk transforming into curds.”

Milk and curds appear twice in the chapter and serve different purposes. The first occurrence distinguishes cause from noncause; this one establishes the measure of “transformation.” The same analogy speaks from its proper place each time and the two uses must not be confused.

A Cause That Has Ceased, and a Cause That Already Possesses the Fruit

The questioner says that although the cause has ceased, it can still produce a fruit. The verse divides the claim into two questions:

“How can a cause that has ceased still produce a fruit? And if the cause remains within the fruit, how can the cause produce the fruit?”

What is no longer present has nothing with which to produce. What remains within the fruit confronts a fruit that already exists—what remains to be produced? Cessation and noncessation exhaust the branches, and each has its own difficulty. Whichever side the questioner takes, he must explain the same thing: in whom, at this very moment, does the power to produce reside? Qingmu gathers it in one sentence: “If the cause does not cease but conjoins with the fruit, how can it produce the fruit again?”

The questioner next says that the cause pervasively possesses its fruit. The verse answers:

“If the cause already pervasively possesses the fruit, what further fruit would it produce? Whether the cause sees the fruit or does not see it, neither can produce the fruit.”

Qingmu states the branches “seeing” and “not seeing” in turn. A cause that does not see the fruit has a fruit that “does not follow the cause”; moreover, “since the fruit does not yet exist, how could it produce the fruit?” A cause that sees the fruit “should not produce it again, because the fruit already exists.” Seeing is too late; not seeing cannot connect. “Pervasively possessing” was meant to make the cause contain the fruit in advance, but if it contains the fruit no further production is needed, and if it does not contain it, nothing determines which fruit it produces.

Nine Cells of the Three Times, Conjunction and Nonconjunction

Next come three verses arranged in a neat parallel test: a past cause against fruits of the three times, a future cause against fruits of the three times, and a present cause against fruits of the three times. Cell by cell, the answer is the same—“ultimately does not conjoin.” Qingmu recites all nine and concludes: “Thus, the three kinds of fruit ultimately do not conjoin with past, future, or present causes.”

Once the route through time is exhausted, only two options remain: conjunction or nonconjunction. A verse asks both away:

“If they are not conjoined, how can the cause produce the fruit? If they are conjoined, how can the cause produce the fruit?”

The question leaves no third path: conjunction and nonconjunction exhaust the branches. Without conjunction, there is no fruit to speak of; with conjunction, “if the fruit is within the cause, the fruit already exists within the cause—how could it arise again?” Qingmu provides a ready analogy: “It is like a person who has not conceived—how could that person bear a child?” What is absent cannot be produced; what already exists need not be produced.

Turn Around and Look at the Fruit

Everything above has asked about the cause. Qingmu says, “It is now time to speak of the fruit,” and the verses turn toward the fruit:

“If the fruit is not empty, it does not arise; if the fruit is not empty, it does not cease; because the fruit is not empty, it neither arises nor ceases.”

“Because the fruit is empty, it does not arise; because the fruit is empty, it does not cease; because the fruit is empty, it neither arises nor ceases.”

Qingmu’s explanation is brief. If the fruit is not empty, it “already exists determinately within the cause and need not arise again; because arising is absent, ceasing is absent.” If the fruit is empty, “emptiness means having nothing whatsoever; how could arising or ceasing occur?”

We should notice that Qingmu here uses “empty” in continuation of the questioner’s usage—the emptiness treated as “having nothing whatsoever.” Determinate existence cannot arise; determinate absence cannot cease. Neither end establishes arising and ceasing, showing precisely that arising and ceasing were never established through these two ends.

This turn matters. Every preceding question and answer was hung upon the cause: whether the cause first contains the fruit, when the cause ceases, how the cause transforms. Qingmu suddenly says, “It is now time to speak of the fruit,” and applies the same measure to the fruit itself. The result is the same. Whether you call the fruit nonempty or empty, if that “nonempty” is determinate existence and that “empty” is determinate nothingness, arising and ceasing have nowhere to stand. What cannot be obtained from the side of the cause cannot be obtained by moving to the side of the fruit.

One and Different: The Final Question

The last six verses gather the whole discussion. Qingmu says: “Cause and fruit will now be refuted through one and different.”

“To say that cause and fruit are one ultimately cannot be established; to say that cause and fruit are different likewise cannot be established.”

“If cause and fruit are one, producer and what is produced become one; if cause and fruit are different, the cause becomes the same as a noncause.”

If one, producer and produced collapse into one—what production remains? If different, cause and noncause no longer differ. This sentence returns exactly to the opening question concerning milk and water and clasps the end to the beginning.

Further:

“If the fruit determinately has intrinsic nature, what would the cause produce? If the fruit determinately lacks intrinsic nature, what again would the cause produce?”

“If the cause does not produce the fruit, it has no mark of being a cause; if it has no mark of being a cause, who could possess this fruit?”

Thus, the final move lands upon “conjunction” itself. The questioner has hung the fruit on conjunction throughout the chapter. But what of conjunction itself?

“If conjunction arises from many causes and conditions, conjunction itself does not arise; how could it produce the fruit?”

Qingmu says: “This dharma of the conjunction of many conditions cannot produce its own body. Since its own body is absent, how could it produce the fruit?” Conjunction too is made from many conditions and has no body capable of standing by itself. What itself depends upon conditions cannot support the statement that “the fruit truly exists from it.”

“Therefore, the fruit arises neither from conjoined nor unconjoined conditions; if there is no fruit, where could a dharma of conjunction exist?”

The last sentence turns the question around very lightly, and very heavily. If there is no truly existing fruit to speak of, where can “conjunction” itself be placed? Cause and fruit, conjunction and nonconjunction, are two ends spoken and established together. When one loosens, the other has nowhere to hang.

Look back at the path of these twenty-four verses. They first ask whether the fruit is within conjunction, then when the cause ceases, then how the cause transforms into the fruit, then how the three times could conjoin, and finally whether conjunction itself can stand. Layer after layer asks the same thing. Each time the questioner tries to find somewhere for the word “exists” to rest, placing it there either makes further production unnecessary or makes cause indistinguishable from noncause.

Seeds still sprout, milk still becomes curds, and clay still becomes jars. The chapter never once blocks these things. It blocks only the one extra statement—the “exists” in “one should know that this fruit exists from the conjunction of many conditions.” Dependent arising is plainly so, yet nowhere can one press down and say, “It is right here.”

Carried Forward: Jizang’s Mirror

The preceding is the meaning of the chapter itself. What follows uses the Sui–Tang Three Treatises tradition as a mirror held alongside it—a mirror is not the treatise, but another way to see the same verses.

In his Commentary on the Treatise on the Middle, Jizang first asks whether this chapter refutes cause and fruit or expounds cause and fruit. His answer preserves both meanings. What is refuted is the intrinsically real cause and fruit posited inside and outside, in Greater and Lesser Vehicles; only when intrinsically real cause and fruit are removed can cause and fruit through causes and conditions be discerned. Precisely because they are called causes and conditions, cause and fruit are vividly present and constantly quiescent. In the passage on simultaneity, he judges that the producer exists and so can produce another, while what is produced is absent and so is produced by another. If simultaneity is established, either both exist or both are absent: thus, to obtain simultaneity is to lose producer and produced, and to obtain producer and produced is to lose simultaneity. For the final gate of one and different, he lists four faults under one and four challenges under different. If one, there may be only cause with no fruit, only fruit with no cause, one thing that cannot be spoken as cause and fruit, or agent and action becoming one—the potter one with the jar. If different, cause and noncause both become noncauses, or both become causes, or cause becomes noncause and noncause becomes cause, or both different and nondifferent must be established.

This mirror places the chapter within the Three Treatises interpretive structure of refuting error and disclosing the correct. Its interpretation is Jizang’s addition, not the original words of Nāgārjuna’s verses or Qingmu’s extended commentary. The readings are recorded side by side without adjudication, each returned to its own place.

Everything is being destroyed. This is perhaps what people are least willing to doubt. Flowers fade, vessels break, bodies age, houses collapse. No treatise need be read; the eyes have already furnished all the proof.

The questioner in this chapter therefore opens with exactly that confidence: “All worldly things presently display the mark of ruin; therefore, destruction exists.” He is not speculating; he points to what lies before him.

What Nāgārjuna asks next is not whether flowers really fade. He asks something else: on what, exactly, does the “destruction” you speak of stand?

Formation and Destruction: Neither Can Stand Alone

The respondent’s first verse opens the pattern of the whole chapter in one sentence:

“Apart from formation or together with formation, no destruction exists there; apart from destruction or together with destruction, no formation exists there.”

Four positions: apart from formation, together with formation, apart from destruction, together with destruction. Destruction has only two possible ways to be established—either apart from formation or simultaneously with it. The same applies to formation. The four gates exhaust the field; there is no fifth. Qingmu restates it more simply: “Whether formation exists or is absent, destruction is absent in both cases. Whether destruction exists or is absent, formation is absent in both cases.”

First consider separation from formation. The verse says: “If separated from formation, how could destruction exist?” It follows with an analogy: “As death apart from birth—this cannot be so.”

The analogy should be read slowly. People in the world speak of birth and death as two events, one earlier and one later. But if they were truly two independent events, one should be able to have only the latter—death without birth. Can anyone die without ever having been born? No. Death is not an event that comes by itself; it is spoken in relation to birth. Remove birth, and the word death no longer has anything to indicate.

The same holds for destruction. Qingmu sets out the definitions of formation and destruction: “Formation means the conjunction of many conditions; destruction means the dispersal of many conditions.” Once formation is removed, what is dispersing? His analogy is utterly plain: “Without a jar, one cannot say that a jar is destroyed.”

Without a jar, you cannot break a jar. The fragments on the ground are called “fragments of a jar” only because there was a jar before. Destruction has never been an independent fact; it is the reverse face of a fact.

Together Does Not Work Either: Having Hair and Having No Hair

Try another formulation, then: can formation and destruction coexist simultaneously?

The verse says: “If formation and destruction exist together, how could formation and destruction exist?” The analogy remains birth and death: “Like birth and death in the world, they cannot both occur at one time.”

Qingmu presses more finely here. If destruction and formation are together, then before being together each must first be something in its own right: “Dharmas are first separately established and afterward conjoin; a conjoined dharma is not apart from difference.” For conjunction, there must first be two things that can conjoin. After they conjoin, their difference has not vanished. But if destruction truly occurs simultaneously and in the same body as formation, the difference disappears; once the difference disappears, destruction loses the cause on which it depends.

Conversely, can formation exist apart from destruction? Qingmu says: “If formation existed apart from destruction, formation would be permanent; permanence is the mark of indestructibility.” Does such a thing exist in the world? No. Thus, apart from destruction there is no formation.

Can formation exist together with destruction? Qingmu changes the analogy: “A person cannot have hair and have no hair at the same time.” Formation and destruction contradict one another. Two contradictory marks cannot occupy the same position simultaneously.

Someone devises a retreat: then say that formation constantly contains destruction. Would that not make them simultaneous? Qingmu’s answer is crisp: “If destruction were constantly present within formation, no dharma of remaining should exist; yet remaining does in fact occur.” If formation were being destroyed every moment, no interval could be called “remaining.” Yet remaining can be spoken, and we really do say that a thing “is there.” This route therefore fails as well.

After traversing the four gates, Nāgārjuna gathers the point: “Together with destruction there is no formation, and apart from it there is no formation,” then asks: “Since both are impossible, how could formation exist?” Qingmu adds a final sentence: “If it exists, explain it.” If you still assert the claim, please give the reason.

At this point the treatise observes every courtesy: it does not forbid you to speak; it asks you to speak.

Exhaustion and Nonexhaustion: Both Doors Close Together

The questioner changes direction. He says: “Dharmas displaying the mark of exhaustion and extinction are presently found.” Such dharmas can be spoken of as exhausted or not exhausted. Since both exhaustion and nonexhaustion can be spoken, formation and destruction should likewise be speakable.

This is an intelligent move. He no longer directly asserts that destruction truly exists, but begins from the fact that we undeniably use these expressions.

The respondent’s verse closes both doors at once:

“If exhausted, there is no formation; if not exhausted, there is likewise no formation. If exhausted, there is no destruction; if not exhausted, there is likewise no destruction.”

First, exhaustion. Qingmu describes it beautifully: “Dharmas constantly cease and are exhausted thought by thought, day and night; what is past does not remain, like flowing water. This is called exhaustion.” Moment after moment ceases throughout day and night, and what has passed refuses to stop even for an instant, like flowing water. Reach for such a thing, and it cannot be grasped; try to speak of it, and speech cannot catch it—“like a mirage, with no determinate nature obtainable.” It is like shimmering heat over a wilderness; approach it and there is nothing there. Since no determinate intrinsic nature can be obtained, from what moment could one cut and say, “This is the time of formation”? If formation cannot be indicated, destruction has nowhere to be placed.

Next, nonexhaustion. “Further, because arising and ceasing continue thought by thought without interruption, it is called nonexhaustion.” Arising and ceasing continue without interruption, and this is called nonexhaustion. But if the continuity is determined to be permanently abiding and unbroken, the difficulty deepens. Along a line that never breaks, “how can one discriminate and say, ‘Now is the time of formation’?” Where does it begin? Without a beginning, there is no formation; without formation, there is no destruction.

One is too fluid to indicate; the other too continuous to divide. Qingmu gathers them: “When sought in this way, a real thing cannot be found; therefore, there is no formation and no destruction.”

Trying to Keep the Dharma and Let Formation and Destruction Go

The questioner retreats a step—the same step we ourselves would likely take: “Set formation and destruction aside. What fault is there if only a dharma exists?” Let the dispute over formation and destruction rest; surely we can still say that “the thing exists.”

The verse blocks both directions:

“Apart from formation and destruction, no dharma exists either; apart from a dharma, formation and destruction likewise do not exist.”

What follows if a dharma exists apart from formation and destruction? Qingmu says: “That dharma should either be absent or permanent.” A thing that neither forms nor is destroyed must either not exist at all or permanently abide without change. No such permanent dharma exists in the world.

Conversely, what if the dharma is set aside and only the two marks formation and destruction are retained? Qingmu ends it with one question: “Apart from the dharma, who forms and who is destroyed?”

The question deserves a long pause. In daily speech, we call “change” something palpably real. Yet change must be the change of something. Remove that something, and “change” leaves only a word. A dharma is not separate from formation and destruction, and formation and destruction are not separate from a dharma. They are two ways of speaking of the same event, and neither can be extracted to stand as master.

The verse next asks again from the side of intrinsic nature: “If a dharma’s intrinsic nature is empty, who could have formation and destruction?” It also asks: “If intrinsic nature is not empty, formation and destruction likewise do not exist.” If dharmas fundamentally have no body of their own, on whom do formation and destruction fall? If dharmas have a determinate body—Qingmu specifies that “nonempty means determinately existent”—how could a determinately existent thing contain formation and destruction?

It asks again from the side of one and different: “If formation and destruction are one, this cannot be so; if formation and destruction are different, this likewise cannot be so.” To say they are one fails, because “their marks differ and they receive many discriminations.” The appearances of formation and destruction can plainly be distinguished. To say they are different also fails, because “there is no separate thing, and there is likewise no cause.” Apart from formation, no self-standing destruction can be indicated.

Can “Seen with the Eye” Serve as Proof?

After this circuit, the questioner’s original confidence is finally met directly. The verse says:

“If one says that because arising and ceasing are seen with the eye, then arising and ceasing exist, one sees arising and ceasing through delusion.”

Qingmu first restates the other’s claim: the eye plainly sees it, so how can you refute it with words?

He does not then deny “seeing.” He explains how this “seeing” arose: “Ordinary people obtained these eyes through inverted causes and conditions in prior lives; through the causes and conditions of recollection, conception, and discrimination in this life, they say that the eye sees arising and ceasing.” These eyes were obtained through inverted causes and conditions in prior lives, and what is called “seeing arising and ceasing” includes another operation of recollection, conception, and discrimination in this life.

In other words, visual seeing has never been pure reception. What a person sees is a result formed from the eye together with habitual discrimination. To use that result to prove arising and ceasing with intrinsic nature is to use the verdict as proof that the verdict is correct. Qingmu therefore says directly: “Within the ultimate meaning, arising and ceasing in truth do not exist.” This does not mean nothing happens before the eyes. It means that among everything that happens, not one event is arising or ceasing with intrinsic nature.

Neither from a Dharma nor from a Nondharma Can Anything Arise

If the questioner still wishes to rescue the thesis, only one path remains: arising. At least the later thing arises from the earlier thing, does it not?

The verse exhausts the sources of arising:

“From a dharma no dharma arises, nor does a nondharma arise; from a nondharma neither a dharma nor a nondharma arises.”

Dharma means existence; nondharma means having nothing whatsoever. Four combinations follow: existence producing existence, existence producing nonexistence, nonexistence producing existence, and nonexistence producing nonexistence.

Existence producing existence—does the causal dharma preserve its original form or lose it? If it preserves its form yet is still called production, it is permanent and what has already arisen must arise again. If it loses its form, the producer is gone and the fruit becomes causeless. Qingmu is clear: causelessness “falls into annihilation and permanence.”

Existence producing nonexistence and nonexistence producing existence both leap from one thing to its contrary, with no bridge between them. For nonexistence producing nonexistence, Qingmu uses the chapter’s lightest analogy: “As a hare’s horn does not produce a turtle’s hair.” Of two things that do not exist, the former cannot produce the latter.

The verse therefore also blocks arising from self, other, or both: “A dharma does not arise from itself, nor does it arise from another”; “it does not arise from self and other; how could arising exist?” Before a dharma arises, there is nothing. Since there is no “self” that could produce itself, there is no “other” to depend upon either—without self, where could other come from?

Continuity: Walking a Tightrope between Annihilation and Permanence

The most important passage lies in the latter half. A verse first gives a general judgment: “If one appropriates a dharma, one falls into annihilation or permanence.” Anyone who appropriates some dharma as determinately existent cannot escape the two extremes. The reason is almost cold in its simplicity: the appropriated dharma can only be permanent or impermanent. “If permanent, one falls into the extreme of permanence; if impermanent, one falls into the extreme of annihilation.”

The questioner now proposes an exceptionally elegant solution:

“One who appropriates dharmas does not fall into annihilation or permanence; because cause and fruit are continuous, there is neither annihilation nor permanence.”

Qingmu states the proposal in its strongest form—notice that this passage renders the opposing position as well as possible: “As a sūtra says, the five aggregates are impermanent, suffering, empty, and not-self, yet they are not annihilated; although demerit and merit are said not to be lost through immeasurable kalpas, they are not permanent.” The five aggregates are impermanent, so they are not permanent; demerit and merit are not lost through immeasurable kalpas, so they are not annihilated. Because of arising and ceasing, there is no permanence; because of continuity, there is no annihilation. Both extremes are avoided.

This is precisely where many people’s minds find rest: things are not eternal, yet death is not simply the end of everything. An unseen line joins what comes before to what comes after.

The respondent’s question cuts exactly at that line:

“If cause and fruit arise and cease, continuous and unbroken, because what ceases does not arise again, the cause is annihilated.”

You say continuity. Yet continuity is strung from causes that cease one by one. Once a particular cause has ceased, does it arise again? If it does not, it has been cut off. Something broken at every joint does not become unbroken merely because there are many joints. Qingmu follows with the question: “If the cause is cut off, how can continuity exist?”

The force of the blow lies here: the questioner’s proposal is not wrong in attitude, but in treating “continuity” as a real thing capable of sustaining itself. It tries to save both extremes with a real continuity, yet continuity itself still has to fall under either annihilation or nonannihilation.

Would Nirvāṇa Then Be Annihilation?

Following that line, the verse presses out a more serious consequence:

“A dharma abiding in its intrinsic nature should have neither existence nor nonexistence; if nirvāṇa extinguishes continuity, it falls into annihilation.”

The first half gives the principle. Qingmu again uses the jar: “If a jar is determinately fixed in the mark of jar, at that time it has no mark of loss or destruction.” If the jar is determinately jar, no mark of destruction can then be spoken. When the jar is absent, “if there is no jar, there is nothing to destroy”—even the object that could be destroyed is gone. Thus, when the jar exists there is no destruction; when it is absent there is no destruction. Extinction cannot be placed at either end. Once extinction cannot be placed, arising cannot be placed either, because “arising and ceasing depend upon one another.” Arising and ceasing are a pair that exist only through one another, not two self-standing things.

Only then does the latter half disclose the heavy consequence. If the three modes of existence are established through continuity and extinguishing that continuity is called nirvāṇa, does nirvāṇa not become cutting off something truly existent? Qingmu makes it explicit: if so, “at the time of nirvāṇa one should fall into annihilation.”

Every practitioner should attend to this step. If saṃsāra is regarded as a truly existent line, nirvāṇa can only be the act of cutting it; cutting off a truly existent thing is annihilation. Clinging to a real continuity does not rescue one from eternalism; it first destroys nirvāṇa.

The Segment That Cannot Be Joined

The treatise therefore pursues the matter one level further, to the most concrete point: how exactly does existence in this life join existence in the next?

“If the first existence ceases, there is no later existence; if the first existence does not cease, there is likewise no later existence.”

The first existence is existence in this life; later existence is existence in the next. If the first has ceased, from where does the later come? It is causeless. If the first has not yet ceased, later existence already exists—“then there are two existences at one time.” Two existences stand side by side simultaneously.

The questioner finds an exceptionally clever position: not after cessation and not before cessation, but arising at the very instant of cessation.

The verse answers: “At the time of ceasing there is one existence, and at the time of arising there is one existence.” The supposed single instant still contains two existences: one ceasing and one arising.

The questioner retreats again: no, there are not two existences. It is just as presently seen: the former ceases and the latter arises.

The verse unfolds the consequence of that statement:

“If one says that arising and ceasing occur at one time, then one dies in these very aggregates and is born in these very aggregates.”

If ceasing and arising truly occur at the same time and belong to the same existence, a person should be reborn within the very same five aggregates in which death occurs, not in another set of aggregates. For Qingmu says: “The one who dies is precisely the one who is born.”

The sentence sounds like a tongue twister, yet its edge is keen. Death and birth are contradictory events; contradictory events cannot occupy the same time and place. To join death and birth through “the same instant” requires the dead person to be the person born on the spot.

Seeking Existence through the Three Times

The final verse gathers the chapter:

“Seeking existence through the three times, continuity cannot be found; if it is absent within the three times, how could continuity of existence exist?”

The three modes of existence are existence in the desire realm, form realm, and formless realm. Qingmu says that because true wisdom has not been obtained throughout beginningless saṃsāra, continuity of the three modes of existence constantly occurs. Yet when it is carefully examined within past, present, and future, it cannot be obtained anywhere. If existence is absent throughout the three times, where could the continuity of existence be placed?

His final sentence is where the chapter lands: “One should know that the continuity of existence exists through ignorance and inversion; in reality, it is absent.”

We must read its measure precisely. It does not say “there is no saṃsāra.” It says that the continuity of saṃsāra appears through ignorance and inversion; in reality, no continuity with intrinsic nature revolves there.

What This Chapter Removes

After reading the chapter, one may easily misunderstand: is Nāgārjuna saying that flowers do not really fade and people do not really die?

No. From beginning to end, the treatise does not quarrel with what is before the eyes. Qingmu himself says “yet remaining does in fact occur,” and likewise says that “dharmas constantly cease and are exhausted thought by thought, day and night.” He does not deny worldly speech.

What the treatise removes is formation and destruction treated as able to stand by themselves. We fear destruction because we take it as a real event; we grasp formation because we take it as a real event. Yet through the four gates, formation exists in dependence upon destruction, destruction in dependence upon formation, dharmas in dependence upon formation and destruction, and formation and destruction in dependence upon dharmas. Exhaustion cannot indicate formation, and nonexhaustion cannot indicate formation. One is not established, and different is likewise not established. Arising from self, other, or both all fail to arise. Pursued to the end, nothing can be found that stands there by itself without depending upon another.

This is why the chapter closes by turning to annihilation and permanence. Views of existence and nonexistence do not arise without a root; they grow from the claim that “a truly existent thing is forming and being destroyed.” Determine that it has formed, and one falls into permanence; determine that it is utterly destroyed, and one falls into annihilation. Remove the root and neither extreme has a place to arise.

The plain jar remains throughout the chapter. When the jar exists, there is no mark of destruction; when the jar is absent, there is likewise no mark of destruction. Much of the destruction feared in a human life is the same: while it is feared, the event has not occurred; when it truly arrives, the one capable of fearing is no longer there.

Carried Forward: Jizang’s Mirror

The preceding is the meaning of the chapter itself. What follows uses the Sui–Tang Three Treatises tradition as a mirror held alongside it—a mirror is not the treatise, but another way to see the same verses.

In explaining the chapter title in his Commentary on the Treatise on the Middle, Jizang likewise takes the conjunction of many conditions as formation and their dispersal as destruction, and gives four levels of meaning: the conjunction and dispersal of things presently seen; the marks of arising and remaining as formation and the marks of change and ceasing as destruction; kalpas of formation and destruction in the vessel-world and sentient world; and even formation and destruction in the rise and decline of doctrinal meanings. He divides the whole chapter into eight gates—together and apart, exhausted and unexhausted, body and mark, empty and nonempty, one and different, arising and ceasing, annihilation and permanence, and the three times. To seek formation and destruction in all these ways and find them unobtainable shows that all dharmas were never formed from the beginning and are not now destroyed; this is the Middle Way.

Concerning the passage on visibly seeing arising and ceasing, Jizang says that this question runs through all sūtras and treatises and reveals both the fault in ordinary perception and the sages’ intention in breaking delusion. In principle there is no arising or ceasing; determinate arising and ceasing come from deluded feeling. As for the chapter’s final refutation of the continuity of the three modes of existence, his judgment is that practitioners of Greater and Lesser Vehicles see the arising of the three existences and seek to extinguish them. Yet if the arising and ceasing of the three existences are taken as real, the three existences cannot be extinguished. Only by awakening that the three existences were never born from the beginning and are not now extinguished can one leave them. The intention in refuting worldly dharmas is precisely to let sentient beings know that the world is ultimately empty.

This sequence of analytic gates and the designation “Middle Way” are Jizang’s additions, not the original words of Nāgārjuna’s verses or Qingmu’s extended commentary. The readings are recorded side by side without adjudication, each returned to its own place.

As the Treatise on the Middle refutes going, coming, arising, and ceasing, readers can usually keep their footing—those are dharmas. In this chapter, what is refuted is the Buddha.

Nor does the questioner come to provoke. He opens with the most ordinary statement imaginable: “In all the world, only the Tathāgata is most honored. He is the Perfectly and Completely Awakened One, called the Dharma King and the omniscient person. Such a one should therefore exist.” The Tathāgata is most honored in the world; he is perfectly awakened, the Dharma King, and omniscient. Surely such a one exists?

The turn concealed in this question must be seen clearly. Its first half is praise; its second half is inference. The praise is faultless. The problem lies in “should exist”: the questioner thinks that once the titles of honor are established, a thing bearing them is established as well.

The respondent does not reject the praise. He takes up only the word “exist” and tests it on the spot: “Now contemplate in accordance with reality. If he exists, he should be apprehensible; if he does not exist, what is there to apprehend?” If you say he exists, point out how he can be taken. If you say he does not exist, what is there to take?

The whole chapter begins here.

1. The Fivefold Search: Not It, Not Apart from It, Not Inside, Not Outside, Not Possessing It

Nāgārjuna’s verse has only four lines:

“Not the aggregates, not apart from the aggregates; neither resides within the other. 　The Tathāgata does not possess the aggregates—where is there a Tathāgata?”

These four lines form a net. To install a Tathāgata outside the five aggregates leaves only five routes: the Tathāgata is the five aggregates; another Tathāgata exists apart from them; the aggregates are within the Tathāgata; the Tathāgata is within the aggregates; or the Tathāgata possesses the aggregates. Qingmu walks each route for the reader.

The first does not work, “because they bear the mark of arising and ceasing.” The five aggregates arise and cease moment by moment. If the Tathāgata were the aggregates, the Tathāgata would also arise and cease moment by moment, and “would incur the faults of impermanence, annihilation, and the rest.” Qingmu adds another cut: “Experiencer and experienced dharma would be one.” The one appropriating the five aggregates and the aggregates appropriated would become the same thing. To merge what bears and what is borne into one is already incoherent.

The second route likewise fails. To establish a Tathāgata apart from the five aggregates would remove the mark of arising and ceasing from him. Unarisen and unceasing, he falls to the side of permanent abiding and “incurs faults such as permanence.” Qingmu also reminds us that such a Tathāgata apart from the aggregates “could not be seen or known by the faculties such as the eye.” Eyes could not see him, ears could not hear him, and none of the six faculties could meet him. Yet when the Buddha lived, people plainly saw him.

For the third and fourth routes, Qingmu uses two pairs of analogies. Five aggregates within the Tathāgata would be “like fruit in a vessel or fish in water”; the Tathāgata within the aggregates would be “like a person on a bed or milk in a vessel.” All these analogies point to the same thing: whenever A is said to be in B, A and B must be two things. Fruit is not the vessel, fish not water, person not bed. Once the two are separate, the Tathāgata has nothing to do with the aggregates and returns to the second route’s fault of permanence.

The fifth route is “like a person having a son.” To say the Tathāgata possesses the aggregates is like a person having a son—father and son are two persons. Once divided into two, all the preceding faults return unchanged.

Qingmu therefore concludes: “Sought in these five ways and not found, what is the Tathāgata?”

At this point, the reader’s heart will likely tighten. Notice what the fivefold search examines. It never asks whether the Buddha appeared in the world, taught Dharma, or liberated beings. It examines only this: can an independent, self-standing “Tathāgata-substance” that waits on no causes or conditions be identified outside the five aggregates? Not finding that thing does not mean there is no Buddha.

2. Then Let Us Say That “the Tathāgata Exists When the Five Aggregates Combine”

The questioner retreats one step, and honestly: “When the five aggregates combine, there is a Tathāgata.” No single aggregate viewed alone is the Tathāgata, but when the five come together, the Tathāgata is there.

This is almost correct. Yet Nāgārjuna points out what the claim itself has already said:

“If the Tathāgata exists through the combination of the aggregates, he has no intrinsic nature; 　without intrinsic nature, how could he exist by virtue of another?”

To say “combination” is to admit dependence on conditions; to admit dependence on conditions is to admit the absence of intrinsic nature. Qingmu is extremely concise: “If the Tathāgata exists because the five aggregates combine, he has no intrinsic nature. Why? Because he exists by virtue of the combination of the five aggregates.”

The questioner retreats again: then let us say that the Tathāgata does not exist through intrinsic nature but “exists by virtue of other-nature.”

The respondent gives two sentences. First: “Other-nature itself has no intrinsic nature.” The “other” on which you rely is itself produced by combination and has no substance of its own. Something unable to stand cannot support another. The second sentence cuts still deeper: “Because there is no ground of mutual dependence, other-nature cannot be found.” The name “other” can be established only in relation to “self.” Once self is absent, there is no basis for speaking of other: “Since it cannot be found, it is not called other.”

Read this slowly. People often think that once intrinsic nature is refuted, other-nature remains as a refuge—as though intrinsic nature were the front door and other-nature the back. Nāgārjuna says they are not two doors, but two sides of one door.

3. What Arises through Conditions Is Not a “Self”

Next comes the chapter’s most important inference:

“If a dharma arises through another, it is not a self; 　if a dharma is not a self, how could it be the Tathāgata?”

Qingmu gives it an analogy: “Just as a fist exists through five fingers, this fist has no substance of its own.”

Pause over the analogy. Close the fingers and it is called a fist; open them and no fist remains. One cannot call the fist false—it can strike someone. Nor can one call it a thing, because it exists nowhere outside the five fingers and disappears when they open. It functions but lacks substance of its own.

“So the self is named by virtue of the five aggregates and has no substance of its own.” The self is likewise so. Where the five aggregates come together, something functions under the name “I”; apart from the five aggregates, no one who bears them can be found.

Qingmu’s next sentence is the hinge of the whole chapter: “The self has many names: being, person, deva, Tathāgata, and so forth.”

Being, person, deva, and Tathāgata: all these names are hung in the same place. When an ordinary person clings to an “I,” he installs a master upon the five aggregates. If a Buddhist installs a “Tathāgata-substance” outside the five aggregates, the gesture is identical; only the name has become more magnificent. Thus Nāgārjuna says: “Without a self, how can he be named Tathāgata?” If even the self has no substance of its own, where would Tathāgata-substance come from?

The next verse finally gathers intrinsic and other-nature together: “Without intrinsic nature, how could there be other-nature? Apart from intrinsic nature and other-nature, what can be named Tathāgata?”

4. Before Appropriating the Aggregates, Was There an Antecedent Tathāgata?

One shadow remains in the questioner’s mind: perhaps a Tathāgata was originally there, then later appropriated this body of five aggregates, whereupon we called him Tathāgata.

Nāgārjuna dispels the shadow through a group of verses. To say this, one must first have a “Tathāgata before appropriating the five aggregates.” Yet “in reality there is no one who does not appropriate the aggregates, nor is there another dharma of the Tathāgata.” The antecedent one cannot be found; beyond him, no second dharma called Tathāgata exists. Since the appropriator was never there, who appropriates now? Qingmu says directly: “In reality, no Tathāgata first exists before appropriating the five aggregates. Who now could appropriate them?”

There is another layer: “If appropriation has not yet existed, what is appropriated is not called appropriation.” The name “what is appropriated” exists only in relation to an appropriator. If the appropriator is not established, the five aggregates are not “appropriated”; they are simply the five aggregates. “There is no dharma without appropriation that is named Tathāgata.” There is nothing that can avoid appropriating the aggregates yet still be called Tathāgata.

By now identity and difference have been sought, the five searches performed, and before and after examined. Nor does Nāgārjuna spare the appropriated five aggregates themselves: “The appropriated five aggregates do not exist through intrinsic nature.” Both appropriator and appropriated are empty.

5. Appropriation Is Empty and Appropriator Is Empty—Can “Emptiness” Then Be Used to Speak of the Tathāgata?

Thus the verse says:

“According to this reasoning, appropriation is empty and the appropriator is empty. 　How could emptiness be used to speak of an empty Tathāgata?”

This sentence is easy to miss. It does not say, “empty appropriation plus empty appropriator equals empty Tathāgata.” It blocks a subtler attachment: once both ends are empty, you still wish to use “emptiness” as material and mold an “empty Tathāgata” in which to settle.

The questioner indeed asks exactly that: “Does emptiness then definitely exist?” According to your explanation, is there not definitely an emptiness?

It is an excellent question, because it forces the final layer of attachment into the open. The respondent denies all four propositions together:

“Empty cannot be spoken; nonempty cannot be spoken. 　Both and neither cannot be spoken; they are spoken only through provisional names.”

Qingmu unfolds each line: one should not say dharmas are empty; one should not say they are nonempty; one should not say they are both empty and nonempty; nor neither empty nor nonempty. Why, then, is the word “empty” still spoken? Qingmu gives one reason only: “It is spoken as a provisional name only to dismantle contradiction.”

Emptiness is used to refute, not to establish. When someone clings to real existence, emptiness is spoken to that person. Its work is to untie the knot; once the knot is untied, the medicine is removed. To hold emptiness itself as a conclusion is to eat the medicine as food.

Qingmu therefore continues: “In such correct contemplation, within the true character of dharmas, the various challenges should not be taken as challenges.” At the level of true character, mutually contentious questions no longer apply.

6. None of the Eight Views Enters the Mark of Quiescence

Two verses have exactly the same form:

“Within the mark of quiescence there is no tetralemma of permanent and impermanent; 　within the mark of quiescence there is no tetralemma of bounded and unbounded.”

Four views arise regarding the past: the world is permanent, impermanent, both permanent and impermanent, or neither permanent nor impermanent. The same pattern generates four regarding the future: bounded, unbounded, both bounded and unbounded, or neither. Together they make eight.

Qingmu offers three reasons why they cannot enter, each deeper than the last.

First, the true character is “ultimately pure and cannot be appropriated.” There is nowhere to reach and grasp. Qingmu adds: “Even emptiness is not accepted, how much less the four views.” Even emptiness is not established, much less these eight.

Second, “the four views all arise through appropriation.” These views do not grow from nowhere; appropriation grows them. Only after taking something up does one judge it permanent or impermanent. Yet the true character “has nothing through which it appropriates.” Without an appropriated thing, views have no birthplace.

Third, and most piercing: “The four views all honor one’s own view and disparage another’s.” Wherever there is a view, there is hierarchy; wherever hierarchy appears, there are this and that. Yet “the true character of dharmas has neither this nor that.” This sentence exposes the foundation beneath the whole dispute. Arguing permanence and annihilation ultimately does not contend over principle, but over my view being higher than yours. No such seat exists within true character.

7. To Say “There Is No Tathāgata” Is the Deepest Wrong View

At last the questioner asks what he fears most: “If the Tathāgata is refuted in this way, does that mean there is no Tathāgata?”

Nāgārjuna’s answer lets neither side stand:

“Those whose wrong views are deeply rooted say there is no Tathāgata; 　yet within the Tathāgata’s mark of quiescence, to discriminate and say he exists is likewise wrong.”

Consider the first half. Qingmu divides wrong views into two: those that “destroy worldly well-being” and those that “destroy the path to nirvāṇa.”

The kind that destroys worldly well-being is coarse: saying there is no demerit, no merit, and no Tathāgata or other noble one. Once this view arises, one “abandons good and does evil,” losing even the well-being of the world. Qingmu’s verdict is severe: “Saying there is no Tathāgata is a deeply rooted wrong view: one loses even worldly well-being, how much more nirvāṇa!”

This must be stated clearly: throughout the chapter, neither failure of the fivefold search nor refutation of self and other ever points toward “there is no Buddha.” To assert no Buddha is not to reach the chapter’s depth, but to fall at its shallowest edge.

Now consider the second half. The view that destroys the path to nirvāṇa is subtle: “craving and clinging to a self and discriminating existence and nonexistence.” Such people cultivate good and end evil, so they attain worldly well-being. But they “do not attain nirvāṇa because of discriminating existence and nonexistence.” Their minds retain an object that must be judged existent or nonexistent, and the gesture of judgment itself keeps nirvāṇa outside.

Qingmu therefore speaks the sentence that requires the chapter’s closest reading: “To say the Tathāgata exists is likewise a wrong view.” Why? “Because one makes various discriminations about the Tathāgata’s mark of quiescence.”

The words “there is a Tathāgata” are not themselves wrong. What is wrong is adding discrimination to the mark of quiescence. Quiescence is not an object. Once placed opposite and judged “he is there,” what has been judged is already not it.

One side is annihilation, the other permanence. Deeply rooted wrong view stands at one end, subtle discrimination at the other. This chapter asks us to stand at neither.

8. After Cessation, Does He Exist or Not?

Following this line, Nāgārjuna takes the claim to its farthest point:

“Within such emptiness of nature, conception is likewise impossible: 　after the Tathāgata’s cessation, discriminating existence and nonexistence.”

Does the Tathāgata remain after cessation? This has been one of the most entangling questions since the Buddha’s lifetime. The respondent does not choose either side. He says that even the act of conceiving is impossible.

Qingmu’s reason is utterly concise and thorough: “The Tathāgata has always been ultimately empty—how much more after cessation!”

The question already assumes that a real one exists before cessation, and so asks where he goes afterward. If no intrinsically existent one has ever existed, the question “Does he exist after cessation?” hangs in the air with nowhere to land.

This leads to the chapter’s most forceful verse:

“The Tathāgata has passed beyond conceptual proliferation, yet people generate conceptual proliferation.” The Tathāgata has long passed beyond it, yet people insist on producing it; “conceptual proliferation destroys the eye of wisdom, and such people do not see the Buddha.”

Qingmu defines conceptual proliferation: “Recollection, apprehending marks, and discrimination between this and that.”

Unfolded, these are four movements. Recollection: the mind reaches toward an image already past. Apprehending marks: it grasps a mark on that image. Discrimination: it judges the mark this way or that. This and that: there now appears my side and the other side. “Those who say that the Buddha has ceased, has not ceased, and so forth engage in conceptual proliferation”—the question activates all four movements at once.

Its result is: “Because it covers the eye of wisdom, they cannot see the Tathāgata’s Dharma-body.”

See the direction of causality. The Buddha did not hide; the eye of wisdom covered itself. The sun did not leave; the eye became veiled. Thus Nāgārjuna says “such people do not see the Buddha.” Not seeing does not mean the Buddha is absent.

Qingmu finally gathers the chapter: “In this ‘Examination of the Tathāgata,’ contemplating from beginning through middle to end, the Tathāgata’s determinate nature cannot be found.” Notice the words “determinate nature.” What cannot be found is an unchanging substance, not the Buddha.

9. Whatever Nature the Tathāgata Has Is the Nature of the World

The final verse contains two of the quietest lines in the entire Treatise on the Middle:

“Whatever nature the Tathāgata has is the nature of the world; 　the Tathāgata has no intrinsic nature, and the world likewise has no intrinsic nature.”

Qingmu adds one sentence: “The contemplation and inquiry in this chapter reveal that the nature of the Tathāgata is the nature of the entire world.”

Still unsatisfied, the questioner presses to the end: “What is the nature of the Tathāgata?”

Qingmu’s answer has only nine Chinese characters: “The Tathāgata has no nature, just as the world has no nature.”

This exchange is where the chapter lands, and it deserves slow reception. After reading the whole way, one most easily expects that once the Tathāgata within the five aggregates has been refuted, a true Tathāgata-nature will finally be pointed out. Yet nothing is pointed out. Qingmu only says: the Tathāgata has no intrinsic nature, and neither does the world. Precisely in this “absence,” the two are the same.

How are they the same? It does not mean that Buddha and ordinary beings are alike, nor that purity and defilement do not differ. What is shared is the mode of existence: no worldly dharma exists apart from conditions, and the Tathāgata is no exception. No substance independent of conditions can be found among worldly dharmas, nor in the Tathāgata. Since both lack intrinsic nature, no wall of intrinsic nature separates the Tathāgata outside the world.

Why place this sentence last? Because the preceding eight sections all refute, leaving the reader’s hand emptier and emptier, and easily producing the thought that the Buddha must therefore be very far away. The verse turns back at exactly this point: the Tathāgata’s nature is the nature of the world. This very world in which you now stand is not another ground.

This is also a first sounding of what the later “Examination of Nirvāṇa” will say. For now, one sentence is established: the Tathāgata and the world meet in the absence of intrinsic nature.

10. What Can a Reader Carry Away from This Chapter?

First, do not install the Buddha outside the five aggregates. The fivefold search examines the gesture of installation. The more magnificent the installation, the more it resembles a self.

Second, emptiness unties knots; it is not placed upon an altar. “It is spoken as a provisional name only to dismantle contradiction”—medicine is not food.

Third, do not judge existence or nonexistence upon the mark of quiescence. To say there is no Tathāgata is a deeply rooted wrong view; to discriminate that the Tathāgata exists is likewise wrong. What is to be released is not the Buddha, but the gesture of judgment.

Fourth, conceptual proliferation is not merely saying the wrong thing; it is recollection, apprehending marks, and discriminating this and that. When the eye of wisdom is covered, one often feels most diligent.

Fifth, the Tathāgata’s nature is the nature of the world. This does not pull the Buddha down; it releases the thought that the Buddha is elsewhere.

Carried Forward: Jizang’s Mirror

The foregoing is the meaning of the chapter itself. What follows uses the eye of the Sui-Tang Three Treatise lineage as a sidewise illumination—a mirror is not the treatise, but another way of reading the same verses.

In his Commentary on the Treatise on the Middle, Jizang gathers the preceding search into three approaches: neither identical nor different, neither self nor other, neither before nor after. He also says that all three kinds of person and dharma sought above—non-Buddhist, Lesser Vehicle, and Great Vehicle—cannot be found. He reads the passage on empty appropriation and empty appropriator as a rebuke of clinging to existence: how can emptiness be taken as appropriation and emptiness as the Tathāgata?

On the four propositions being ineffable while the sūtras still say there is a Buddha, Jizang explains that provisional names and marks are spoken within what lacks name and mark. He cites the six ineffable statements beginning with arising-arising in the Nirvāṇa Sūtra, and the seventh statement that can be spoken because it has causes and conditions, to show that the Buddha is provisionally named in order to break the four contradictions. Since emptiness is provisionally established only to oppose existence, to cling to that emptiness is to fall again into every difficulty.

For the final verse, Jizang offers three readings. First, the questioner at the chapter’s opening raises the supramundane to prove the mundane, while the treatise master now raises the mundane to illustrate the supramundane. Second, against the hearer who supposes that the Buddha stands beyond the four propositions and outside the hundred negations, he explains that the Tathāgata’s nature is the nature of the world. Third, against the view dividing mundane and supramundane generated by the preceding words “the Tathāgata has passed beyond conceptual proliferation,” he explains that the fact these four propositions have never been four is itself the Tathāgata. Finally, he says that Tathāgata and world equally transcend the four propositions, so their two natures have no distinct nature.

This mirror connects the chapter to the Three Treatise lineage’s interpretive structure of refuting error and disclosing correctness. Its judgments are Jizang’s additions, not the original words of Nāgārjuna’s verses or Qingmu’s extended commentary. The readings are recorded side by side without adjudication, each returned to its own place.

People tend to place a strange trust in their afflictions. When anger rises, we do not doubt whether it is real. When greed rises, we do not doubt whether greed is real. It comes with such force—how could it be false?

This chapter asks precisely that question. Its method is unusual: rather than begin with the affliction itself, it begins with where the affliction came from.

The Questioner’s Single Step of Inference

The questioner does not open by inventing a thesis, but by citing a sūtra. The verse says:

“From recollection, conception, and discrimination arise greed, anger, and delusion; 　the inversions of pure and impure all arise from many conditions.”

Qingmu makes the questioner’s intent explicit: “The sūtra says that through inversions of pure and impure, recollection, conception, and discrimination produce greed, anger, and delusion. Therefore, one should know that greed, anger, and delusion exist.”

The hinge of the inference lies in its final sentence. The first two lines describe a sequence of arising: inversion and recollection, conception, and discrimination come first; greed, anger, and delusion follow. From this sequence, the questioner draws a conclusion—if we can explain how something arises, it must truly exist.

This is the step people take most easily. Whatever has an identifiable origin, we treat as real.

The Respondent Takes the Same Sentence Only One Step Farther

“If the three poisons arise through inversions of pure and impure, 　the three poisons have no intrinsic nature; therefore, the afflictions are unreal.”

The respondent does not deny the sūtra cited by the questioner, nor change its sequence. He accepts it, then points out that this very sequence explains the opposite. Qingmu states it more plainly: “If the afflictions arise through inversions of pure and impure and through recollection, conception, and discrimination, they have no intrinsic nature. Therefore, the afflictions are unreal.”

What arises through conditions is not there by itself. What is not there by itself has no nature that can be called its own. The reason the questioner uses to prove existence, the respondent uses to explain absence of intrinsic nature. Both use the same sūtra passage; they differ only in whether they are willing to take it one step farther.

Upon Whom Are the Afflictions Hung?

The respondent next changes angle. If afflictions exist, someone must possess them.

“Whether the dharma of a self exists or does not exist, it can never be established; 　without a self, the existence or nonexistence of the afflictions likewise is unestablished.”

“Who possesses these afflictions cannot be established; 　if they exist apart from that, the afflictions belong to no one.”

Qingmu first explains the term “affliction”: “Affliction means that which afflicts another.” If it afflicts another, someone must be afflicted, namely a being. Yet “this being cannot be found when sought anywhere.”

Someone then retreats: even if there is no self, afflictions surely belong to mind. Qingmu blocks this route too: “If one says that though there is no self, afflictions belong to mind, this too is not so.”

The Fivefold Search: Neither Side Can Be Found

“Just as the view of a body cannot be found through the fivefold search, 　afflictions likewise cannot be found through the fivefold search within a defiled mind.”

The view of a body cannot be found through the fivefold search within the five aggregates; the preceding chapters have made this method familiar. Here the same measure is moved between affliction and defiled mind and applied twice: affliction cannot be found within a defiled mind; Qingmu adds the reverse—“A defiled mind likewise cannot be found through the fivefold search within the afflictions.”

Neither side can be found. The affliction is not deeply hidden; between affliction and mind there was never a container relation that could be pointed out.

What Is False Cannot Produce What Is Real

“The inversions of pure and impure have no intrinsic nature; 　how could afflictions arise through these two?”

Qingmu gives the chapter’s briefest definition: “Inversion means false and delusive.” Then he infers: “What is false has no intrinsic nature; without intrinsic nature, there is no inversion.”

The questioner wishes to use inversion as a producing cause, but the cause itself is false. What is false has no intrinsic nature; without intrinsic nature, it cannot serve as an intrinsically existent producing cause. The point of support is removed.

The Six Objects: Like Mirage, Dream, and a City of Gandharvas

The questioner therefore changes to a harder place to stand. He no longer speaks of inversion, but of objects:

“Form, sound, smell, taste, touch, and dharmas make six; 　these six are the root of the three poisons.”

Qingmu explains his point: “These six entrances are the root of the three poisons.” Objects come first, inversion follows, and the three poisons come last.

The respondent’s verse is beautiful:

“Form, sound, smell, taste, touch, and dharmas are six in kind; 　all are empty, like a mirage, a dream, or a city of gandharvas.”

“Within these six, where are pure and impure? 　They are like an illusory person and an image in a mirror.”

Qingmu supplies a moment in time: before the substances of the six objects “combine with mind, they are empty and possess nothing,” then gives a verdict: “They merely deceive the mind and have no determinate mark.”

Each analogy deserves individual attention. A mirage is water seen from afar, but no water remains on approach. Dream events disappear on waking. A city of gandharvas looks like a city from afar and vanishes up close. An illusory person has form but is not a person. A mirror image can be seen but not extracted. All five say one thing: visible, yet impossible to seize.

Pure and impure are appropriated upon objects. When the object itself cannot be seized, pure and impure have nowhere to be placed.

Pure and Impure Are Names That Stand Only by Leaning on Each Other

“Without depending on the mark of pure, there is no impure; 　because impure exists through pure, there is no impure.”

The reverse is the same: without depending on impure, no pure can be spoken. The two names stand only by leaning on one another; neither is first established on its own.

If so:

“Without pure, through what could greed exist? 　without impure, through what could anger exist?”

Qingmu concludes in one sentence: “Because neither pure nor impure exists, neither greed nor anger arises.”

The Four Inversions: Refuting Permanence Does Not Leave Impermanence

The questioner now brings forward the four inversions beginning with permanence from the sūtra: seeing permanence within impermanence is inversion, while seeing impermanence within impermanence is not. Since inversion exists, one who is inverted must exist.

The respondent first answers the former half:

“Clinging to permanence in what is impermanent is called inversion; 　within emptiness there is no permanence—where is there an inversion of permanence?”

Qingmu says the other three inversions of self, pleasant, and pure are the same: “The remaining three are likewise so.”

Then comes the turn in this chapter that is easiest to overlook:

“If within impermanence one clings to impermanence and calls it noninversion, 　within emptiness there is no impermanence—where is there one not inverted?”

Most will accept the first verse’s refutation of permanence; this one refuses even the clinging to impermanence. The reason is the same: permanence is absent within emptiness of nature, but so is impermanence. The impermanence grasped is likewise not an existing thing available for grasping. Refuting the inversion of permanence is not establishing impermanence as a place to settle. It asks one to release even the measure used in refutation.

Even “Clinging” Itself Bears the Mark of Quiescence

“The object of clinging, the one who clings, the act of clinging, and the dharma used in clinging 　all bear the mark of quiescence. How could clinging exist?”

Qingmu explains the four separately: “The object of clinging means the thing; the one who clings means the agent; clinging means the act; and the dharma used means the matter employed.” He points back to a familiar place: “As explained in the ‘Examination of the Tathāgata.’”

He next defines clinging: “Clinging means recollection, conception, and discrimination regarding this and that, existence and nonexistence.” These are precisely the words cited by the questioner at the chapter’s opening. The chapter has gone full circle and returned to its beginning: clinging is discrimination and calculation among this, that, existence, and nonexistence.

Search through the Three Times: Who Generates Inversion?

The verse searches inversion through the three times: “One with inversion does not generate inversion; one without inversion does not generate inversion,” followed by “one inverted does not generate inversion, nor does one not inverted.”

“At the time of inversion, inversion likewise does not arise. 　Observe well for yourself: who generates inversion?”

Qingmu explains each: one already inverted does not generate inversion because he is already inverted; one not inverted does not generate it because there is no inversion; at the time of inversion, inversion likewise does not arise because two faults follow.

The final line is not a conclusion, but an admonition: “Now remove the mind of arrogance, observe well for yourself: who is the inverted one?” The question does not seek an answer; it asks the reader to return and look.

Even “Nonarising” Cannot Be Established as True Character

At this point, a new attachment easily arises: if inversion does not arise, then “nonarising” must be inversion’s true character.

Qingmu reveals the thought exactly as it is: “That person craves and clings to nonarising, saying nonarising is the true character of inversion.” Then he blocks it: “Even uncontaminated dharmas cannot be named the mark of nonarising, how much less can inversion be such a mark?”

Even uncontaminated dharmas cannot be fixed under the name “mark of nonarising.” When refutation reaches its end, the instrument of refutation is not left behind.

When One Side Is Absent, the Other Cannot Stand

“If permanent, self, pleasant, and pure truly existed, 　then permanent, self, pleasant, and pure would not be inversions.”

If the four truly possessed real nature, seeing them as they are could not be called inversion. Conversely, if one determines these four as truly nonexistent and allows impermanent, suffering, and impure on the other side to be real—

“If permanent, self, pleasant, and pure were truly nonexistent, 　then impermanent, suffering, and impure should likewise be nonexistent.”

Qingmu’s reason has only four Chinese characters: “There is no mutual causation or dependence.” When one side falls, the other loses what it stood against. Pure and impure are so; permanent and impermanent are likewise so.

Inversion Ceases, Ignorance Ceases, and the Twelve Links Cease in Sequence

“When inversion ceases in this way, ignorance likewise ceases; 　because ignorance ceases, formations and the rest likewise cease.”

Qingmu connects this sentence to the twelve links: “Because all inversions cease, ignorance, the root of the twelve links, likewise ceases; because ignorance ceases, the three kinds of karmic formation through aging-and-death all cease.”

This is the practical consequence toward which the whole chapter points. The first link of the twelve is ignorance, and what allows ignorance to stand is precisely inversion. When inversion, sought, does not arise, ignorance has nothing on which to depend. The first link does not connect, and the entire chain behind it ceases.

The long inquiry into the unfindability of inversion may appear merely to dismantle a term; followed to the end, it dismantles the first link of saṃsāra.

Then Can the Afflictions Ultimately Be Severed or Not?

The final two verses place even the word “sever” under the same measure.

“If the nature of the afflictions were real and belonged to someone, 　how could they be severed? Who could sever their intrinsic nature?”

“If the afflictions are false, without intrinsic nature and belonging, 　how could they be severed? Who could sever what has no intrinsic nature?”

What truly exists cannot be severed—if intrinsic nature were truly intrinsic nature, it would not disappear through anyone. What is false likewise cannot be severed—a dharma without nature is not a thing that can be lifted or put down. Qingmu’s final sentence remains a question: “Who could sever a dharma without intrinsic nature?”

The chapter ends on an unanswered question. This is not an unresolved matter. To speak of severing first requires a real thing, somewhere it belongs, and someone who severs; none of the three can be found. Severing is therefore not removing something, but the absence of anything that can be taken. This is the same matter as the preceding cessation of inversion and ignorance: cessation is not removing something by hand; it is nonarising.

What Can a Reader Carry Away from This Chapter?

First, being able to trace an origin does not make something real. The questioner’s step is utterly natural; we take it every day.

Second, after refuting one side, do not rush to stand on the other. Refuting permanence is not dwelling in impermanence, and refuting pure is not dwelling in impure. “Within emptiness there is no impermanence” closes the route of retreat.

Third, when affliction arises, first ask upon whom it is hung. The verse and commentary search all the way: the one afflicted cannot be found, belonging to mind likewise fails, and neither affliction within defiled mind nor defiled mind within affliction is found by the fivefold search. Following this question changes where effort is applied—not wrestling with affliction, but seeing that it originally belongs nowhere.

Carried Forward: Jizang’s Mirror

The foregoing is the meaning of the chapter itself. What follows uses the eye of the Sui-Tang Three Treatise lineage as a sidewise illumination—a mirror is not the treatise, but another way of reading the same verses.

When explaining the chapter title in his Commentary on the Treatise on the Middle, Jizang says that Buddhist and non-Buddhist, Great and Lesser Vehicle, all acknowledge inversion but do not know that its original nature is empty and quiescent. To seek inversion and find it unobtainable is precisely true character; therefore, the chapter is called the refutation of inversion. He further says that all suppose inversion arises first and is later extinguished through cultivation; if so, inversion cannot cease. When inversion is understood as originally unborn and presently unceasing, it truly ceases. He also treats the two marks of inverted and noninverted: conceiving them as two makes both inversion; understanding that they originally lack two marks names both noninverted. He divides the chapter into two sections: the first refutes the arising of inversion and affliction, and the second refutes the cessation of inversion and affliction. At the chapter’s end, he concludes that one who performs the chapter’s correct contemplation allows affliction and inversion to cease ultimately and forever, while contrary contemplation is wrong contemplation, leaving old delusions and generating new inversions.

This mirror connects the chapter to the Three Treatise lineage’s interpretive structure of refuting error and disclosing correctness. Its sectional division and judgments are Jizang’s additions, not the original words of Nāgārjuna’s verses or Qingmu’s extended commentary. The readings are recorded side by side without adjudication, each returned to its own place.

When the Treatise on the Middle reaches this chapter, its master receives the gravest blow yet.

Most earlier objections brought forward one thing—coming and going, arising and cessation, agent, karma, time, the Tathāgata—and waited for it to be dismantled. This chapter is different. What the objector holds is not something suspect but the Buddhadharma itself: the four truths, four progressions and four fruits, eight noble persons, and Jewels of Dharma, Saṅgha, and Buddha. He says: if emptiness means what you teach, all of these disappear.

His opening sentence first lays out the sequence of cultivation and realization: “Eliminate the four inversions, penetrate the four truths, attain the four fruits of a renunciant.” This is the path he himself follows, each link fastened to the next. Precisely because the links are tight, removing the lowest one brings the entire building down. The link he means to remove is arising and cessation.

1. The Objector’s Inference: From No Arising and Cessation All the Way to No Buddha

The objector’s verse drives through six steps in a single breath:

“If everything is empty, there is neither arising nor cessation; consequently, the four noble truths do not exist.”

First: if everything is empty, there is no arising or cessation; without arising and cessation, there are no four truths. Qingmu supplies the reason: the four truths have cause and result—origin is the cause and suffering the result; the path is the cause and cessation the result. Cause and result are arising and cessation. Without arising and cessation, the structure of the four truths itself disintegrates.

The next five steps follow in order. Without the four truths, the four activities of seeing, severing, realizing, and cultivating do not exist. Without those activities, the four fruits of a renunciant do not exist. Without the four fruits, no one attains or progresses toward them. Without these eight kinds of person, there is no Jewel of the Saṅgha; and without the four truths, there is no Jewel of the Dharma. Then comes the final stroke:

“One who teaches emptiness in this way destroys the Three Jewels.”

Qingmu restates the stroke even more plainly: “You say that dharmas are all empty; this destroys the Three Jewels.”

The objector adds another point: emptiness destroys not only the world-transcending but also the world:

“Emptiness destroys cause and result and also demerit and merit; moreover, it utterly destroys every conventional dharma.”

The indictment is now complete. One who teaches emptiness destroys the Three Jewels above and demerit and merit below. This is the weightiest and also the most sympathetic question in the entire treatise. The questioner does not lack faith in the Buddha. He believes so strongly that he fears emptiness will dismantle everything in which he believes.

2. The Reply’s First Sentence: Emptiness Is Not at Fault; You Do Not Know Emptiness

The treatise master’s first sentence does not defend emptiness. It identifies the misunderstanding:

“You do not in fact know emptiness or the reasons for emptiness, nor do you know the meaning of emptiness; therefore you distress yourself.”

“Distress yourself” is exact. The fault lies not in emptiness but in the distress you generate. Qingmu divides the failure to know into three: “You do not understand what the characteristic of emptiness is, the reasons for which emptiness is taught, or the meaning of emptiness.” You do not know what emptiness is like, why the Buddha teaches it, or what it means. Only by failing on all three counts can one derive that long chain of faults.

Notice the form of the answer. The treatise master does not say, “Your reasoning goes wrong at the third step.” He says: you have misidentified the very word “emptiness” in your premise. Treat emptiness as sheer nonexistence, substitute that nonexistence into the four truths, and of course everything comes up empty. The entire rest of the chapter therefore explains that word “emptiness” anew.

3. The Buddhas Teach by Relying on the Two Truths

The first step in that explanation is the two truths:

“The Buddhas rely on two truths when teaching the Dharma to beings: first, the conventional truth; second, the ultimate truth. If someone cannot know and distinguish these two truths, that person does not know the true meaning of the profound Buddhadharma.”

Qingmu’s commentary gives each truth its place. On the side of conventional truth: “The nature of all dharmas is empty, yet because the world is inverted, false dharmas arise and appear real to the world.” The nature of dharmas is originally empty, but because the world is inverted, various false dharmas arise, and at the level of the world they are real. On the side of ultimate truth: “Noble ones truly know the nature of inversion and therefore know all dharmas to be empty and unarisen; for noble ones, this is the ultimate truth and is called reality.”

Both passages use the same word “real,” but at different levels: one says “real to the world,” the other “for noble ones, this is the ultimate truth and is called reality.” The same dharmas, seen from within inversion, are a world of real existence; when the nature of inversion is seen as it really is, they are empty and unarisen. These are not two sets of dharmas but two truths concerning one set of dharmas.

The objector’s illness is recognizing only one truth. He places the four truths, four fruits, and Three Jewels on a level where they possess intrinsic natures and stand on their own. The moment he hears “empty,” he assumes that level has been removed and everything has fallen. The treatise master’s point is that they never stood on that level in the first place.

But could one reverse course and retain only the ultimate, discarding the conventional truth? Qingmu blocks this question in advance and introduces the next verse.

4. Without Relying on the Conventional, the Ultimate Cannot Be Spoken

〔T30 0033a02〕

“Without relying on the conventional truth, the ultimate cannot be attained; without attaining the ultimate, nirvāṇa cannot be attained.”

The reason appears in Qingmu’s sentence: “The ultimate is entirely disclosed through words, and words are conventional.”

The ultimate itself is beyond words, but whenever it can be spoken, it is disclosed through words, and words fall on the side of convention. Abolish convention and the ultimate has no door through which it can even be expressed. With no door, there is no way in, and nirvāṇa cannot be attained.

Thus the conventional truth is not the ultimate’s rival; it is the entrance to the ultimate. Far from destroying convention, emptiness must rely on convention even to be spoken. The objector fears that emptiness will destroy convention, but the actual situation is the reverse.

5. Venomous Snake and Incantation: Emptiness Is Useful Only When One Can Hold It

The treatise master next acknowledges something: emptiness really is dangerous.

“Unable to contemplate emptiness correctly, the dull-witted harm themselves, like one unskilled in incantations or unskilled in catching a venomous snake.”

Qingmu says it bluntly: “It is like catching a venomous snake for profit: failing to grasp it well, one is instead harmed.” If one catches a snake for gain but uses the wrong grip, one is bitten; if one cannot execute an incantation properly, it also turns back and causes harm.

This simile is not an insult. Treating emptiness as sheer nonexistence really does destroy demerit and merit and cause and result, exactly as the objector fears. And that really does happen—to the person who does not contemplate emptiness skillfully. Thus the next verse continues:

“The World-Honored One knew this Dharma to have profound and subtle characteristics beyond the reach of the dull-witted, and therefore did not wish to teach it.”

He did not withhold the Dharma out of stinginess. He knew it was too deep and would be received wrongly. The treatise master in effect concedes that the objector’s concern has a real basis—but the concern applies to people who misconstrue emptiness, not to emptiness itself.

6. Emptiness Itself Is Empty

What, then, is the emptiness taught by the treatise master?

“You say that I cling to emptiness and therefore assign me faults; yet the faults you now allege do not exist in emptiness.”

Qingmu supplies the chapter’s most crucial gloss: “I teach the emptiness of nature, and emptiness itself is empty; there is no such fault.”

The objector assigns faults by treating emptiness as an entity called “emptiness” that the treatise master embraces and refuses to release, thereby acquiring every ailment that comes with it. The master replies: I teach the emptiness of nature, and this emptiness is itself empty. Since emptiness is not a thing one can hold, faults caused by grasping emptiness have nowhere to lodge.

Once emptiness itself is empty, the question “Will emptiness destroy everything?” changes its shape. Emptiness is not a blade that cuts down the myriad dharmas. Emptiness says that the myriad dharmas never stood by themselves.

7. The Pivot: Because Emptiness Has Meaning, All Dharmas Can Be Established

Now the pivot of the entire chapter appears:

“Because emptiness has meaning, all dharmas can be established; without the meaning of emptiness, nothing could be established.”

Qingmu’s explanation adds not one unnecessary word: “Because emptiness has meaning, all worldly and world-transcending dharmas are accomplished.”

This is where the whole Treatise on the Middle turns. The preceding twenty-three chapters have all been refuting, so a reader can easily assume that the conclusion is “nothing can be established.” This verse reverses the direction: precisely because of emptiness, all dharmas can be established; if they were not empty, not one could be established.

We must see clearly what “established” means. For something to be established, it must be able to arise, change, and pass from absence into presence. It can arise and change precisely because it lacks a fixed and unchanging intrinsic nature. If it possessed such a nature, it would forever remain exactly as it is—neither arising nor ceasing, neither increasing nor diminishing—and nothing could ever be established. Emptiness is therefore not establishment’s opponent; it is establishment’s condition. The master immediately applies this reasoning to the objector:

“You now possess a fault yourself, yet turn it back upon me, like a rider who forgets what he is riding.”

The fault lies plainly in the determinate nature you assert, yet you turn it around and attach it to emptiness. Qingmu exposes the point: the objector “possesses a fault within existent dharmas but cannot notice it himself, and instead sees a fault in emptiness.”

8. Determinate Nature Means No Causes or Conditions

Why does the fault lie in determinate nature? The verse continues:

“If you see dharmas as determinately possessing natures, you thereby see dharmas as having neither causes nor conditions.”

Qingmu gives the inference: “If a dharma determinately possessed a nature, it should neither arise nor cease. What use would such a dharma have for causes and conditions?” A thing with a determinate intrinsic nature already is what it is. It need not be produced and cannot be extinguished; causes and conditions have no use for it. Conversely, whatever arises through causes and conditions has no intrinsic nature. One cannot have both.

Therefore:

“You thereby destroy cause and result, action, agent, and what is acted upon, and also destroy the arising and cessation of all things.”

The objector accuses emptiness of destroying cause and result. The treatise master shows that determinate nature is what truly destroys them. From this point, attack and defense exchange places.

9. Dharmas Born from Many Conditions: Empty, Provisional Names, and the Middle Way

At the turn, the master speaks the most widely recited verse in the entire treatise:

“Dharmas born from many conditions, I say, are precisely absence; they are also provisional names, and this is the meaning of the Middle Way. Never has there been a single dharma that did not arise from conditions; therefore no dharma is not empty.”

Qingmu begins by receiving the verse’s word “absence” as “empty”: “Dharmas born from many conditions, I say, are precisely empty.” He then unfolds it layer by layer:

First, why is it empty? “Because the thing belongs to many causes and conditions, it has no intrinsic nature; because it has no intrinsic nature, it is empty; emptiness too is empty.” Three layers descend: because it belongs to many conditions it has no intrinsic nature; because it has no intrinsic nature it is empty; and this emptiness itself is empty. Qingmu does not stop at “empty.” He takes one more step and relinquishes emptiness itself. Without this step, emptiness would become the last truly existent thing, and the objector’s accusation would actually stand.

Second, why is it a provisional name? “It is only to guide beings that one speaks by means of provisional names.” Since it has no intrinsic nature, it cannot finally be named. Names are still established and words still spoken in order to guide beings. A provisional name is not a deception; it is an establishment made for guidance.

Third, why is it the Middle Way? “Because this is free from the two extremes of existence and nonexistence, it is called the Middle Way.” Qingmu speaks more precisely: “Because this dharma has no nature, one cannot say it exists; and because even emptiness is absent, one cannot say it does not exist.” Because it lacks intrinsic nature, it cannot be called existent; because even emptiness cannot be grasped, it cannot be called nonexistent. When neither extreme can be asserted, this is called the Middle Way.

These three layers are three faces of the same event, not three stages. To call a dependently arisen dharma devoid of intrinsic nature is emptiness; to speak of its conditioned establishment is provisional naming; to say it falls into neither extreme is the Middle Way. There is no provisional name apart from emptiness, nor some separate middle beyond emptiness and provisional naming.

Qingmu returns the point to the dispute: “This fault now returns to you.” The next dozen or more verses return the objector’s chain of inference to him link by link.

10. Returning the Charge: If Everything Were Nonempty, the Four Truths Would Truly Not Exist

“If everything were nonempty, there would be neither arising nor cessation; consequently, the four noble truths would not exist.”

This precisely answers the objector’s opening verse. The objector says, “If everything is empty,” there are no four truths. The master says, “If everything were nonempty,” there would be no four truths—the same conclusion from the opposite premise. He then tests each truth in turn.

The truth of suffering: “If suffering did not arise from conditions, how could suffering exist? Impermanence is the meaning of suffering, yet determinate nature is not impermanent.”

If suffering possessed determinate nature, it would not await conditions; if it did not await conditions, there would be no suffering. Qingmu again adduces scripture: “Scripture says that impermanence is the meaning of suffering.” Suffering is suffering precisely in its impermanence. A thing of determinate nature never relinquishes itself, and so lacks impermanence and therefore suffering.

The truth of origin: “If suffering possessed determinate nature, why would it arise from origin?” If suffering already determinately existed, it would not need to arise from origin. The truth of origin would lose its function.

The truth of cessation: “If suffering possessed determinate nature, it should not cease.” A nature does not cease. If suffering were a nature, it could not be extinguished.

The truth of the path: “If suffering possessed determinate nature, there would be no cultivation of the path; if the path can be cultivated, it has no determinate nature.”

Qingmu’s reason is clean: “If a dharma were real, it would be permanent; what is permanent cannot be increased.” Cultivating the path is an increase, and a permanent dharma cannot increase. Either the path can be cultivated and lacks determinate nature, or it possesses determinate nature and cannot be cultivated. The objector can choose only one.

Having tested each of the four truths, the master gathers them and asks: “Where could the path that extinguishes suffering possibly lead?” If suffering, origin, and cessation cannot be established, where exactly could this path for extinguishing suffering go?

Now test the transformation involved in cultivation and realization:

“If suffering determinately possessed a nature, how could what was not seen before be seen now? Its nature would not have changed.”

An ordinary person did not see suffering before and sees suffering now; a transformation has occurred between the two. If nature does not change, how could what was not seen before now become visible?

Once this sentence stands, the four activities follow together:

“Just as seeing suffering cannot be established, severing origin, realizing cessation, cultivating the path, and the four fruits likewise cannot be established.”

Qingmu completes each item: origin’s nature cannot be severed, cessation was never realized before, and the path was never cultivated before—the four activities all fail. If the four activities fail, so do the four fruits: “The natures of these four fruits of the path have always been unattainable.”

The objector’s chain now repeats in its original form. Without the four fruits, there are no persons attaining or progressing toward them; without the eight noble persons, there is no Jewel of the Saṅgha. Qingmu pointedly records the scriptural basis here: “Yet scripture calls the eight noble persons the Jewel of the Saṅgha.” Then the chain moves upward:

“Because the four noble truths do not exist, the Jewel of the Dharma does not exist; without the Jewels of Dharma and Saṅgha, how could there be the Jewel of the Buddha?”

The objector challenged emptiness with the Three Jewels; the master returns the Three Jewels as a challenge to determinate nature. The very outcome the objector fears is produced by his own position.

11. Buddha and Awakening: Iron Lacks the Nature of Gold

The objector retreats one step. Dharmas may be refuted, but surely the ultimate path—unsurpassed complete awakening—must exist? It is because of this path that one is called a Buddha. The master replies:

“According to your thesis, there could be no Buddha by reason of awakening, nor could there be awakening by reason of a Buddha.”

If each possesses determinate nature, each is permanently fixed in itself and neither arises because of the other. The objector’s intended safeguard—“one is called a Buddha because of this path”—is dismantled by his own determinate nature.

The master goes further:

“Even if one strives diligently and cultivates the path to awakening, if one did not previously possess Buddha nature, one could not become a Buddha.”

Qingmu draws the treatise’s simile: “Just as iron lacks the nature of gold and never becomes gold despite every kind of refinement.” If one was determinately not a Buddha before, no amount of striving could ever make one a Buddha. The final thing canceled by the theory of determinate nature is the very meaning of practice.

12. Demerit and Merit and Their Results: Nonemptiness Leaves No One Able to Act

The objector began by accusing emptiness of destroying demerit and merit. The master returns each item:

“If dharmas were nonempty, there would be no one who performs demerit or merit; since they are nonempty, what could be done? Their natures would be determinate.”

If the natures of demerit and merit were already determined, no one could be producing either one. Producing means changing, and determinate nature cannot change. Qingmu adds: “There would be neither doing nor an agent.”

Then comes the question of karmic results:

“You maintain that karmic results do not arise within demerit and merit; this would mean that various karmic results exist apart from demerit and merit.”

If karmic results do not arise among the causes and conditions of demerit and merit, results exist apart from causes. Qingmu is explicit: “A karmic result would appear without awaiting its cause.”

The objector retreats again: apart from demerit and merit there need be no wholesome or unwholesome results, but such results do arise from demerit and merit. The master closes with one sentence:

“If karmic results arise from demerit and merit, then since results arise from demerit and merit, how could they be called nonempty?”

Once it is conceded that a result arises from demerit and merit, it is dependently arisen. How can something dependently arisen still be called nonempty? Every retreat takes the objector further into dependent arising, and with each step into dependent arising, emptiness establishes itself.

13. To Refute Emptiness Is to Refute Convention

The master now returns verbatim the objector’s charge that emptiness “utterly destroys every conventional dharma”:

“If you refute all dharmas and the meaning of emptiness in conditioned arising, you thereby refute convention and every other dharma.”

“If you refute the meaning of emptiness, nothing could be done; yet there would be doing without any act, and one who does nothing would be called an agent.”

Qingmu lists the absurdities that follow from refuting emptiness: results are unmade and causeless; there is doing without doing; agents do nothing; and karma, karmic results, and recipients exist apart from agents. Convention can function only because dharmas can arise and change. Refute emptiness, and all of this becomes disordered.

Consider the world itself:

“If determinate nature existed, the various features of the world would neither arise nor cease, but abide permanently and never perish.”

Qingmu appeals to what is presently seen. If gods, humans, animals, and all things possessed real natures, they should abide permanently and never perish. “Yet all things presently seen each display transformation, arising, cessation, and alteration. Therefore they should not possess determinate natures.” The impermanence before one’s eyes is the counterproof to determinate nature.

The world-transcending is no different:

“If emptiness did not exist, what had never been attained should not be attained; there would be no severing of afflictions and no exhaustion of suffering.”

Without emptiness, what has not been attained can never be attained, afflictions cannot be severed, and suffering cannot be exhausted. Practice is possible precisely because dharmas are not fixed. You can still change because you were never a fixed thing in the first place.

14. Conclusion: To See Dharmas of Causes and Conditions Is to See the Buddha

At the chapter’s end, the master concludes by citing scripture:

“Therefore a scripture says: one who sees dharmas of causes and conditions is able to see the Buddha and sees suffering, origin, cessation, and the path.”

Qingmu says that one who sees all dharmas arise from many conditions “can see the Buddha’s Dharma body and develop wisdom,” can see the four noble truths, and through seeing them attains the four fruits and extinguishes every distress.

The chapter returns to its starting point. The path named by the objector at the opening—eliminate the four inversions, penetrate the four truths, attain the four fruits of a renunciant—was what he feared emptiness would destroy. The master ends by saying that to see dharmas of causes and conditions is to see the four truths and see the Buddha. The same path has not been destroyed by emptiness; it can be traversed only within emptiness.

The final sentence strikes the chapter’s gavel: “To refute emptiness is to refute dharmas of causes and conditions; to refute dharmas of causes and conditions is to refute the Three Jewels; and to refute the Three Jewels is to refute oneself.”

The one who refutes emptiness refutes dharmas of causes and conditions; once those are refuted, the Three Jewels are refuted with them. Intending to protect the Three Jewels, he shatters what he protects in his own hands. The words “refute oneself” answer the opening challenge.

What Can a Reader Carry Away from This Chapter?

First, recognize the word “emptiness” in your own mind. If it means sheer nonexistence, every charge the objector makes is correct. If it is the emptiness in which “emptiness itself is empty,” not one charge can attach.

Second, know that the two truths are not two stories of a building. Convention is the mouth through which the ultimate can be spoken.

Third, hold fast to the pivotal sentence: “Because emptiness has meaning, all dharmas can be established.” What is refuted is intrinsic nature, not dharmas. What was not attained can be attained, afflictions can be severed, and suffering can be exhausted—all because this word “emptiness” bears their weight.

Carried Forward: Jizang’s Mirror

Everything above belongs to this chapter’s own meaning. What follows looks alongside it through the eye of the Sui–Tang Three Treatises tradition. A mirror is not the treatise; it is simply another way of reading the same verses.

In commenting on this chapter in the Commentary on the Treatise on the Middle, Jizang first locates the objector’s illness in rigidly grasping the characteristic of emptiness: precisely because he grasps emptiness’s characteristic, he does not know emptiness’s meaning. Explaining the verse “Because emptiness has meaning, all dharmas can be established,” Jizang says the upper half shows the treatise master free of fault and the lower half shows the outsider at fault. Everything can be established because of emptiness because the previous verse says emptiness itself is empty, which is neither empty nor existent; being neither empty nor existent, both emptiness and existence can be established. He also frames the point through the two truths: because of ultimate emptiness there is conventional truth; when the two truths are established, everything is established. Without emptiness, neither the ultimate nor the conventional truth is established.

As for the verse “Dharmas born from many conditions,” Jizang reads it through the three “are”s. A dependently arisen dharma is ultimately empty: if it possessed intrinsic nature, it would not arise through conditions; since it does arise through conditions, it lacks intrinsic nature and is therefore empty. It is also a provisional name: causes and conditions were never existent and now are not empty either; being neither empty nor existent, they cannot finally be named, so both existence and emptiness are spoken of provisionally. It is also the meaning of the Middle Way: lacking intrinsic nature, it is not existent; being not existent, neither is it empty. Neither existent nor empty is called the Middle Way.

This mirror joins the chapter to the Three Treatises tradition’s hermeneutical architecture of refuting error and disclosing the correct. That interpretation is Jizang’s addition, not the original words of Nāgārjuna’s verse or Qingmu’s extended commentary. The readings are recorded side by side without adjudication, each returned to its own place.

The Treatise on the Middle has refuted all the way to this point: going and coming, arising and ceasing, the five aggregates, the six faculties, the agent, and the Tathāgata—nothing can stand. At Chapter Twenty-Five, the challenger finally asks at the most painful place. He is not asking about reasoning but about his own destination.

The opening question takes the form of a verse:

“If all dharmas are empty, neither arisen nor ceased, what is cut off and what extinguished that it may be called nirvāṇa?”

Qingmu completes the thought for him. If there is neither arising nor ceasing, there are no afflictions to cut off and no five aggregates to extinguish; therefore, “all dharmas should not be empty.” His conclusion is that precisely because dharmas are not empty, things can be cut off and extinguished—“the many afflictions are cut off and the five aggregates extinguished; this is called nirvāṇa.”

See the assumption beneath the question. In the questioner’s mind, nirvāṇa is something produced: a heap of real afflictions lies there, and I cut them off one by one; real aggregates weigh upon me, and I extinguish them; the clean state left when cutting and extinction are complete is nirvāṇa. The thought is entirely natural. Almost everyone aspiring to release begins in this way. Precisely because it is natural, it must be treated separately near the treatise’s end.

The Question Returned: Those Who Say Dharmas Are Not Empty Can Explain Nirvāṇa Still Less

Nāgārjuna does not defend emptiness. He takes the questioner’s sentence exactly as given and changes a single word:

“If dharmas were not empty, there would be neither arising nor ceasing. What would be cut off and what extinguished that it might be called nirvāṇa?”

The same challenge is returned. What is the reasoning? If a dharma is not empty, it possesses intrinsic nature; what possesses intrinsic nature does not depend on conditions and cannot change. If affliction possessed intrinsic nature, it would forever be affliction and could not be cut; if the five aggregates possessed intrinsic nature, they would forever be the five aggregates and could not be extinguished. What truly cancels the “cuttable and extinguishable” is not emptiness but precisely nonemptiness.

Qingmu gathers this exchange into one sentence: “Thus neither the gate of existence nor the gate of nonexistence leads to nirvāṇa.”

This sentence is the threshold of the entire chapter. It does not say “existence is wrong and nonexistence right,” nor “nonexistence is wrong and existence right.” Neither gate opens. The next twenty-odd verses walk through both gates and their combinations one by one.

What, Then, Is Nirvāṇa? Nāgārjuna First Gives a Mark

Before dismantling the four gates, Nāgārjuna gives a positive verse:

“Nothing is attained and nowhere is reached; it is neither annihilation nor permanence, neither arising nor ceasing. What is spoken of in this way is called nirvāṇa.”

Qingmu explains word by word. “Nothing is attained means that nothing is acquired in either practice or fruition.” No object is acquired within causal practice or resultant realization. “Nowhere is reached means that no place exists at which to arrive.” Nirvāṇa is not a location, and no road delivers you there.

What does “not annihilated” mean? Qingmu’s reason must be noticed: “because the five aggregates have from the beginning been ultimately empty,” therefore “when one attains the path and enters nirvāṇa without remainder, nothing is cut off.” He does not say, “It is called nonannihilation because the cutting is so complete.” He says that the aggregates were empty from the start; no real object was ever there, so no real object is taken away upon attaining the path. This directly answers the questioner. Emptiness does not cancel cutting; it explains the truth of cutting.

What does “not permanent” mean? “Because nirvāṇa is quiescent extinction and no dharma can be discriminated, it is not called permanent.” The name permanent can attach only where something can be identified and discriminated. Nothing like that exists in nirvāṇa, so even “permanent” cannot be placed upon it.

Qingmu next cites scripture, establishing the general outline for all the reasoning that follows: “Nirvāṇa is neither existent nor nonexistent, neither both existent and nonexistent nor neither existent nor nonexistent. It appropriates no dharma; inwardly quiescent, it is called nirvāṇa.”

The four gates are complete. Each is now examined.

First Gate: Nirvāṇa Is Not Existent

“Nirvāṇa cannot be called existent, because existence bears the marks of aging and death; no existent dharma ever exists apart from the marks of aging and death.”

Qingmu’s explanation begins before our eyes: “Everything seen by the eye arises and ceases; these are the marks of aging and death.” Nothing we can point to and count as “existent” stands outside arising and ceasing. Arising and ceasing are aging and death. If nirvāṇa were another such existent thing, it would follow aging and death; yet what one seeks in nirvāṇa is precisely freedom from aging and death. To seek an object bearing aging and death cancels itself.

The second reason is tighter:

“If nirvāṇa were existent, nirvāṇa would be conditioned; yet no single dharma that exists is unconditioned.”

Qingmu says: “All things arise from many conditions and are conditioned.” Whatever exists is dependently arisen, and whatever is dependently arisen is conditioned. Thus “existent” and “unconditioned” cannot attach to one another. He also blocks a retreat: someone may say that a permanent dharma is provisionally called unconditioned. Qingmu answers that not even an impermanent dharma can be found when sought, much less that permanent dharma that can neither be seen nor attained.

The third reason comes from scripture:

“If nirvāṇa were existent, why would it be called nonappropriation? No existent dharma is ever named without appropriation.”

The scripture calls nirvāṇa nonappropriation. Whatever earns the name “existent” is established only within a relation of receiving and grasping; apart from appropriation, a thing cannot even obtain the name existent. Since the scripture says nonappropriation, nirvāṇa does not occupy the cell of existence.

Second Gate: Nirvāṇa Is Not Nonexistent Either

The questioner immediately retreats to the second cell: if existence is not nirvāṇa, surely nonexistence must be.

“Existence is not nirvāṇa; how much less nonexistence! Nirvāṇa contains no existence—where could nonexistence be?”

Qingmu gives the key: “Nonexistence exists dependent on existence; without existence, whence nonexistence?” Nonexistence is not an independent thing; it is spoken by leaning on existence. What is nonexistence? Qingmu cites scripture to define it: “What existed before but does not exist now is called nonexistent.” Something existed before and is absent now—that is nonexistence.

Nirvāṇa does not come about that way—“it is not an existent dharma turning into nonexistence.” Nirvāṇa is not the hole left after some real thing is destroyed and disappears. The first gate has already shown there is no existent nirvāṇa; without the prior term that could become absent, nonexistence has nowhere to begin.

The scripture likewise speaks of nonappropriation:

“If nonexistence were nirvāṇa, why would it be called nonappropriation? No nonexistent dharma is ever named without appropriation.”

An existent dharma depends on appropriation for establishment; a nonexistent dharma does too. Both ends are tied to appropriation, so neither is that “inward quiescence.”

Between the Two Gates, Nāgārjuna Gives a Direct Answer

At this point the questioner has nothing left to grasp. It is neither existent nor nonexistent—what are you talking about? He asks directly. Nāgārjuna answers with the most straightforward verse in the chapter:

“Because one appropriates causes and conditions, one revolves in birth and death; not appropriating causes and conditions is called nirvāṇa.”

Notice that this verse does not change the subject. It does not posit another nirvāṇa somewhere else but speaks of the same person and the same aggregates. Appropriating causes and conditions is called birth and death; not appropriating them is called nirvāṇa. Qingmu explains in greater detail: failing to know inversion as it truly is, one comes and goes through birth and death dependent on the five appropriated aggregates; when one knows it truly, one no longer comes and goes through them. His final sentence matters most: “Because the five aggregates lacking intrinsic nature no longer continue, this is called nirvāṇa.”

The five aggregates are not destroyed; the five aggregates lacking intrinsic nature no longer continue. No thing is cleared away here. Only a mode of adhesion stops.

Nāgārjuna again cites scripture as evidence:

“Just as the Buddha said in the scripture: both existence and nonexistence are cut off. Thus one knows that nirvāṇa is neither existent nor nonexistent.”

Qingmu explains: “Existence means the three modes of existence; nonexistence means the annihilation of those three.” The Buddha cuts off two things: the existence of the three realms and the annihilationist view that destroys them. Even “annihilation” is among what must be cut off. Thus nirvāṇa falls on neither side.

Third and Fourth Gates: Combining the Two or Negating Both Fails Likewise

The questioner retreats again: what if existence and nonexistence are conjoined?

Nāgārjuna gives four verses in succession. First, if their conjunction were nirvāṇa, existence and nonexistence themselves would be liberation; yet Qingmu points out, “The two matters contradict one another; how could they exist in one place?” Second, both arise from appropriation and stand through mutual dependence; conjoined, they remain products of appropriation, while scripture calls nirvāṇa nonappropriation. Third, “Nirvāṇa is called unconditioned, while existence and nonexistence are conditioned.” Two conditioned things cannot be assembled into an unconditioned one. Fourth, Nāgārjuna uses a ready analogy: “The two cannot share a place, just as light and darkness do not coexist.” When light comes, darkness goes; when darkness comes, light goes. They never occupy one place.

The questioner finally retreats to the fourth cell: surely neither existence nor nonexistence will work.

“If neither existence nor nonexistence were called nirvāṇa, on what basis would this ‘neither existent nor nonexistent’ be discriminated?”

The question is sharp. “Neither existence nor nonexistence” is a discriminated name, but on what basis is it discriminated? Qingmu answers directly: what contradicts existence is called nonexistence, and what contradicts nonexistence is called existence; yet that pair “was refuted in the third proposition.” Once the lower pillars are dismantled, the fourth proposition resting upon them naturally falls.

The four gates are exhausted: existence, nonexistence, both, and neither. These are precisely the four propositions in the scripture Qingmu cited at the beginning. Scripture gave the conclusion first; Nāgārjuna supplies the reasoning proposition by proposition.

Whether the Tathāgata Is Present or Has Passed, None of the Four Propositions Is Appropriated

With the four gates exhausted, Nāgārjuna moves the same matter onto the Tathāgata:

“After the Tathāgata passes away, one does not say that he exists or does not exist, both exists and does not exist, or neither exists nor does not exist. While the Tathāgata is present, one does not say that he exists or does not exist, both exists and does not exist, or neither exists nor does not exist.”

Qingmu’s explanation lands heavily: “Since none is appropriated, nirvāṇa should not be discriminated as existence, nonexistence, and so on.” He then asks a question with almost no road back: “Apart from the Tathāgata, who would attain nirvāṇa?”

Read slowly. One who seeks nirvāṇa always retains an “I who will attain nirvāṇa.” Yet the “Examination of the Tathāgata” has already shown that even the Tathāgata is neither within the aggregates nor found apart from them. If the attainer is unfindable, attainment naturally falls empty. Thus Qingmu says: “The mark of nirvāṇa cannot be found when sought at any time or in any way.”

At any time, in any way—not merely “it cannot yet be found now.” No mark of nirvāṇa can be found at any time through any method.

The Summit: Nirvāṇa and the World Have Not the Slightest Difference

〔T30 0036a04〕The height of the whole treatise rests in these two clauses:

“Nirvāṇa and the world have not the slightest difference; the world and nirvāṇa likewise have not the slightest difference.”

Nāgārjuna says it twice, once from nirvāṇa and once from the world. Both directions are stated so that no one reads it as one-sided.

Qingmu’s explanation has only three steps, each grounded in the text. First: “Because the five aggregates continue, come, and go through causes and conditions, they are named the world.” The world is not a place but a provisional name for the continuing movement of the five aggregates. Second: “The intrinsic nature of the five aggregates is ultimately empty, unappropriated, and quiescent,” a meaning already explained in preceding chapters. Third: “Because all dharmas neither arise nor cease,” the two sides do not differ.

Thus two objects are not being leveled. The so-called world is nothing apart from that sequence of aggregates; the intrinsic nature of that sequence is originally ultimately empty, unappropriated, and quiescent. Within the world, one sees continuity; seen as it truly is, one sees quiescence. The object viewed was never exchanged for another. What changes is whether it is seen.

Nāgārjuna carries the point to the “limit”:

“The reality-limit of nirvāṇa and the boundary of the world— between these two limits there is not a hair’s breadth of difference.”

Qingmu’s reason is “because they are equal and unfindable.” It is not that measurement reveals exact equality; when ultimately sought, the “limit” indicated by the four names world, nirvāṇa, reality-limit, and nonarising-limit was unfindable from the start, so not a hair’s breadth of difference can arise.

These two verses are the furthest point reached in this treatise. Here they are transmitted solely according to Nāgārjuna’s verses and Qingmu’s explanation, without further adjudication.

Views Depend on Something; When Their Supports Are Empty, Views Cannot Stand

“Views of existence and nonexistence after passing, views of whether the world has a limit, and views of permanence depend respectively upon nirvāṇa, the future, and the past.”

Qingmu lists three groups: the four views about the Tathāgata after passing, the four about the world’s limit, and the four about the world’s permanence—“three groups totaling twelve views.” He also names their supports: the first four depend on nirvāṇa, the next four on the future, and the last four on the past.

The passage shows that views do not arise from nowhere; each stands on a support treated as truly existent. Once that support proves unfindable, the view erected upon it naturally cannot stand. Qingmu therefore concludes that the world and nirvāṇa are equal and undifferentiated.

The Conclusion: Proliferation Is Exhausted, and the Buddha Has Spoken Nothing

The final three verses gather everything at once:

“Because all dharmas are empty, what could possess a limit or no limit, both a limit and no limit, or neither a limit nor no limit? What is one and what is different? Where are permanence and impermanence, both permanence and impermanence, or neither permanence nor impermanence? Dharmas are all unfindable and all conceptual proliferation ceases; there is neither person nor place, and the Buddha has spoken nothing.”

In Qingmu’s explanation one syntax appears repeatedly: “What has a limit, and who has a limit?” He first asks about the dharma and then the person. All sixty-two wrong views require both a “what” and a “who” to stand. Neither can be found in ultimate emptiness; therefore “all sixty-two wrong views of this kind are unfindable within ultimate emptiness.”

Thus “every acquisition ceases and all conceptual proliferation is exhausted.” Yet the exhaustion of proliferation is not nothingness—Qingmu immediately says one “comprehends the true marks of dharmas and attains the path of peace and security.” The exhaustion of proliferation is the entrance to this path, not the blankness at its end.

Finally he looks back over the whole treatise: “From the ‘Examination of Conditions’ onward, dharmas have been discriminated and sought.” Existence is unestablished, nonexistence unestablished, both unestablished, and neither unestablished—“This is called the true marks of dharmas; it is also called suchness, Dharma-nature, the reality-limit, and nirvāṇa.”

Only here is the name nirvāṇa settled. It is not a thing realized and obtained, but another name for the true marks of dharmas. Suchness, Dharma-nature, and the reality-limit all speak of the same matter.

Thus the chapter’s final line is so decisive: “There is neither person nor place, and the Buddha has spoken nothing.” Qingmu explains it as “the Tathāgata has never, at any time or place, taught anyone a fixed mark of nirvāṇa.” What is blocked is “fixed mark.” The Buddha has spoken nirvāṇa and taught people to move toward it. What the Buddha has never taught is a fixed, unchanging mark of nirvāṇa that can be held in the hand.

That is precisely the kind of mark the questioner sought at the beginning. At the chapter’s end, his hand is empty; yet not one part of his practice of cutting afflictions and extinguishing the aggregates has been canceled. Only the thought that “there is a real thing to cut and a real place to reach” has been canceled.

Carried Forward: Jizang’s Mirror

The meaning above belongs to the chapter itself. What follows is a side illumination from the Sui–Tang Sanlun lineage—a mirror is not the treatise, but another way of reading the same verses.

Reading the chapter’s four propositions, Jizang’s Commentary on the Treatise on the Middle first asks: is neither existence nor nonexistence used as medicine, or does such a thing truly exist in principle? If it is medicine, the medicine disappears when the illness is exhausted. If it truly exists, one could equally assert that existence and nonexistence truly exist. He also says that nirvāṇa transcends the four propositions; the propositions are only gates indicating nirvāṇa and must not be taken as nirvāṇa itself.

He gives particular weight to the verse “Nirvāṇa and the world have not the slightest difference,” saying that worldly and supramundane persons and dharmas equally transcend the four propositions. If one hears the earlier argument and thinks only the supramundane transcends them while the worldly remains within them, this is still the dual view of world and transcendence; a dual view is birth and death. He then says that in the Way they were never two and in relation to conditions they were never one. For the final verse, he treats all dharmas and all times as objects contemplated and all ways as gates of contemplation, reaching conditions exhausted within contemplation, contemplation exhausted within conditions, and both conditions and contemplation quiescent.

This mirror joins the chapter to the Sanlun lineage’s architecture of refuting the wrong and revealing the right. Its classifications are Jizang’s additions, not the words of Nāgārjuna’s verses or Qingmu’s prose. The readings are recorded side by side without adjudication, each returned to its own place.

By the time readers reach this point in the Treatise on the Middle, they have probably grown used to a certain manner of argument: a claim is set up and then brought down from every side. Going does not work; coming does not work. Arising does not work; ceasing does not work. After twenty-five chapters, one develops a kind of wariness and greets every apparently stable statement with suspicion.

Then one turns to this chapter, and suddenly not a single thing is refuted.

Beginning with “Sentient beings, covered by delusion, set the three formations in motion for what comes afterward,” the twelve links are counted straight through in order until “grief, sorrow, and all kinds of suffering and distress,” and then gathered back in reverse until “in this way it rightly ceases.” Across this entire long set of verses there is not one counter-question. Negations are not wholly absent—“if the one who clings does not cling,” “what the wise do not produce,” “this then does not arise”—but they all occur within the affirmative account of cessation in reverse order. They are not used to refute one alternative after another. Qingmu’s long explanation is plainer still. He speaks through each link in turn: karma, receiving a body, demerit and merit, continuity. At the end he even identifies the source: these meanings are “extensively explained in the Abhidharma and the sūtras.”

Why does a treatise known for refutation suddenly speak this way?

The answer lies in the chapter’s first sentence.

Someone Has Come Asking for a Path He Can Actually Walk

The chapter begins with a question: “You have explained the path of ultimate meaning through the Mahāyāna. I now wish to hear how the śrāvaka teaching enters the path of ultimate meaning.”

Notice the tone. This person has not come to win an argument. He does not say, “Your ultimate meaning is wrong,” or “The Mahāyāna cannot be established.” He acknowledges that what the respondent has taught is the path of ultimate meaning: “You have explained the path of ultimate meaning through the Mahāyāna.” He accepts the entire road traveled in the preceding chapters. What he asks is something else: Can that same ultimate meaning also be entered from the side of the śrāvaka teaching?

This is a student’s question, not an opponent’s. Most voices raising objections in earlier chapters brought forward some substantially existent thing and waited for it to be dismantled. This person is different. What he holds is the Dharma he has long practiced—the Four Truths, the twelve links of dependent origination, the thirty-seven factors of the path. He wants to know whether his road reaches the place the respondent has described.

The respondent therefore changes his tone as well. Since the question asks how the śrāvaka teaching enters the path of ultimate meaning, he answers through the sequence already present within the śrāvaka teaching. He establishes no new doctrine and replaces none of its terms. He simply counts through the familiar twelve links from the beginning.

The deepest respect one can show a question is to answer in words the questioner can understand. The very form of this chapter demonstrates that respect.

A Chain That Begins with Not Knowing

The verses open: “Sentient beings, covered by delusion, set the three formations in motion for what comes afterward.”

The starting point is delusion—not evil, but delusion. Beings do not first conceive a wicked intention and then fall into birth and death. First their eyes are covered. Qingmu makes the line still more direct: “Because ordinary beings are blinded by ignorance.” His word is “blinded.” A person whose sight has failed does not deliberately choose the wrong road; he cannot see the road. That one word sets the tone of the chapter. The twelve links are not an indictment but a case history.

With sight obscured, one acts. What action? The “three formations,” which Qingmu explains as bodily, verbal, and mental karma. Notice too the words “for what comes afterward.” These formations do not arise without purpose; they are set in motion for the sake of the body that comes later. Beneath all one’s striving lies an “afterward.”

Once formations arise, “in accordance with their formations, they fall among the six destinies.” “In accordance with” matters here. No one above is sentencing you to a destination. Your own actions carry you there: where the action goes, the person goes. Qingmu adds one stroke: “The formations set in motion are superior, middling, or inferior.” Because actions differ in grade, the bodies received differ as well. This is Qingmu’s addition for the reader; the verse itself says only “fall among the six destinies” and does not grade them.

Next comes consciousness: “Consciousness receives a body in the six destinies.” Consciousness receives one body within those six. Then, “because consciousness clings, name-and-form grows.” The decisive word is “clings.” It is not simply that consciousness arrives and name-and-form exists. Name-and-form grows because consciousness fastens to it. Qingmu says the same: “Because consciousness clings as cause and condition, name-and-form gathers.” It grows only when something sticks.

When name-and-form has grown, the six sense bases exist; when the six bases are complete, contact exists. The verses are especially precise about contact: “Faculty, object, and consciousness come together, and the six kinds of contact arise.” Faculty means the six faculties, object the six objects, and consciousness the six consciousnesses. Only when all three meet is there contact. Contact is never merely one thing touching another; three things must come together. This is a small reminder embedded within the chain: each link looks like a thing, but when opened it consists of several things converging.

From here the movement quickens: “Because of the six kinds of contact, the three kinds of feeling immediately arise”—painful, pleasant, and neither-painful-nor-pleasant. When feeling arises, “thirsting craving arises.” “Thirsting” carries enormous weight. This is not liking but thirst. A thirsty person does not merely want water; he cannot refrain from drinking. Once craving becomes thirst, the hand necessarily reaches: “Because of craving there are four kinds of clinging; because of clinging there is existence.”

What has been grasped? “Existence”—future existence, the next round. Qingmu supplies the most important stroke here: “While the four kinds of clinging are actively grasping, bodily, verbal, and mental karma produce demerit and merit,” and thereby “continue the three modes of existence in the future.” At the moment of grasping, people do not think they are making a future. They think only that they are taking hold of what is presently before them. Yet in that very moment bodily, verbal, and mental karma have moved, demerit and merit have been made, and the next round has already been joined to this one.

Thus “from existence comes birth; from birth come aging and death,” and what follows aging and death is not a blank but “grief, sorrow, and all kinds of suffering and distress.” The verses gather this entire chain into one sentence: “Through these causes and conditions alone, the great aggregate of suffering gathers.” Qingmu is briefer still: “There is only the gathering of the great aggregate of suffering.”

Both places say “only.” Only this, nothing else. There is no sufferer standing behind the suffering—only suffering gathering.

A Gate Opens in the Middle of the Chain

When the twelve links reach clinging, the verses suddenly insert a line:

“If the one who clings does not cling, there is liberation and no existence.”

Its position deserves attention. It is not placed at the end, nor in the later passage on cessation in reverse order. It appears immediately beneath “Because of craving there are four kinds of clinging; because of clinging there is existence.” While the forward sequence is still moving smoothly, a gate opens halfway through.

The verse speaks without qualification. It does not say, “suffering is reduced by one degree,” but “liberation and no existence.” If one does not cling, the segment of future existence cannot connect. If it cannot connect, birth, aging-and-death, grief, and sorrow have no way to occur. This is an entrance named by the verse itself. It goes as far as clinging without requiring us to impose a separate schedule of practice upon it.

The governing sentence for cessation in reverse order occurs in the long explanation: “The wise do not set it in motion. Because they see in accordance with reality, ignorance ceases. Because ignorance ceases, the formations also cease; because the cause ceases, the effect also ceases.” The reverse contemplation runs throughout from the cessation of ignorance. The line on liberation beneath clinging is one corner of that reverse contemplation appearing early within the forward account, not a separate point of origin.

The Difference Is Not in the Chain

The verses next give the chapter’s two most important lines:

“This is called birth and death, the root of the formations— produced by those in ignorance, not produced by the wise.”

The same chain is pushed by ordinary beings and left unpushed by the wise. Qingmu says it another way: “The wise do not set it in motion. Because they see in accordance with reality, ignorance ceases.”

Notice that there is no second chain. Scriptures and treatises do not say the wise possess another set of twelve links. The difference lies in one place alone: whether they have seen. When seeing in accordance with reality arises, ignorance ceases, and the first link loosens.

This also explains why the chapter need not refute. The refutations of the preceding twenty-five chapters break attachment—to going and coming, arising and ceasing, agent and act as things possessing intrinsic nature. None of the twelve links explained here is intrinsically existent to begin with: each exists only in dependence on the preceding link, only in dependence on conditions. “Faculty, object, and consciousness come together, and the six contacts arise” has already said everything. Contact cannot constitute contact on its own. A chain that depends on conditions at every point needs no further refutation. Count it honestly, and its dependence on conditions shows itself.

Here lies the key to how the śrāvaka teaching enters the path of ultimate meaning. The śrāvaka teaching is not exchanged for something else. Its own twelve links are simply seen through to the end.

Cessation Means Not Connecting Again

The final two lines of verse read:

“Because this ceases, that then does not arise; it is only the gathering of the aggregate of suffering, and thus it rightly ceases.”

Qingmu unfolds this into a complete sequence: “Because ignorance ceases, formations also cease; because the cause ceases, the effect also ceases.” He then reaches “cultivating the wisdom that observes the arising and ceasing of the twelve links,” and finally, “birth, aging-and-death, grief and sorrow, and the great aggregate of suffering all rightly cease in accordance with reality.”

Look closely at the verb. It is not “cut off” or “eliminate” but “does not arise.” Cessation means the next link no longer connects. When the preceding event has ceased, the following event lacks a supporting condition and therefore does not arise. What does not arise presents nothing that must be attacked.

Qingmu gives this a definition: “Right cessation is complete cessation.”

Complete—not temporarily held down, not hidden somewhere else. When the cause has ceased, the effect cannot emerge elsewhere. This differs fundamentally from what the world calls “solving a problem.” A worldly solution moves the effect away while the conditions remain, and after a while it grows again. Here the condition has dispersed, and the effect has nowhere to arise.

That word “complete” is the joint where the śrāvaka teaching meets the path of ultimate meaning. The forward contemplation of the twelve links explains how the aggregate of suffering gathers. Their cessation in reverse order explains how that aggregate completely ceases to gather. When it completely ceases to gather, no place remains for appropriation. The road sought by the questioner passes through at this point.

Qingmu finally gives the source with perfect clarity: “As explained extensively in the Abhidharma and the sūtras.” He does not count this explanation as a new view of his own or of the Treatise on the Middle. This is what the Buddha already taught. Here it is simply spoken once more, in order, for the person who asked.

Then Where Is Ultimate Meaning?

After reading the chapter, a question may remain. The questioner sought entry into the path of ultimate meaning, but the entire reply speaks of birth, gathering, receiving a body, demerit, and merit—every sentence concerns worldly appearances. In which sentence, then, does ultimate meaning reside?

The crucial point is that the respondent does not hang a separate ultimate meaning above and outside the twelve links. If he did, he would have built a two-story house: dependent origination below and emptiness above. The questioner would climb the stairs and still hold two different things.

Instead, the respondent speaks the twelve links all the way through. Follow them through and one sees that nowhere in the chain does anything stand on its own. Formations await delusion before arising; consciousness awaits formations before receiving a body; name-and-form awaits the clinging of consciousness before growing; contact awaits the coming together of faculty, object, and consciousness; feeling awaits contact, craving awaits feeling, clinging awaits craving, and existence awaits clinging. Trace the sequence upward and one cannot find anything that exists there by itself without depending on something else. Even of the final heap, the verses say only, “Through these causes and conditions alone, the great aggregate of suffering gathers,” and Qingmu only, “There is only the gathering of the great aggregate of suffering.” Neither will add a bearer behind it.

Ultimate meaning therefore lies not in any one sentence but in how this entire chain is seen. Seen in the ordinary direction, the same twelve links are birth and death; seen in accordance with reality, they are without intrinsic nature. The observed object has not been exchanged for another. Only the observing eye has changed. Qingmu’s line, “Because they see in accordance with reality, ignorance ceases,” says precisely this. It is not seeing another thing, but seeing what is here as it really is.

This is the reasoning by which the śrāvaka teaching can enter the path of ultimate meaning. Ultimate meaning is not another road outside the śrāvaka teaching. It is what appears when the śrāvaka road is walked to its end.

What Can a Reader Carry Away from This Chapter?

First, see clearly which link you are standing in now. Has craving arisen? Has the hand already reached? Has karma already moved? The twelve links are not a list to memorize but a measure to place against one’s life.

Second, remember the line indicated by the verse: “If the one who clings does not cling, there is liberation and no existence.” The verse joins liberation to non-clinging. Yet the governing movement of cessation remains, “Because they see in accordance with reality, ignorance ceases.” One sentence names an entrance and the other its basis; no fixed temporal order need be imposed between them.

Third, do not seek an owner for the chain. From beginning to end, verse and commentary say only, “There is only the gathering of the great aggregate of suffering.” They do not say who receives it. The moment the mind seeks an owner, the first link is joined again.

Carried Forward: Jizang’s Mirror

The foregoing is the meaning of this chapter itself. What follows places alongside it another view from the Sanlun line of the Sui and Tang. The mirror is not the treatise; it is simply another way of reading the same set of verses.

In his Commentary on the Treatise on the Middle, Jizang explains the chapter title through right contemplation of causes and conditions: by rightly contemplating causes and conditions one enters the two emptinesses, and when the twelve links cease one attains nirvāṇa. He also correlates the causal sequence of the twelve links with the refutation of the two views of permanence and annihilation. The two past causes producing the five present effects refute permanence—if something were permanent, how could it arise from causes? The three present causes producing the two future effects refute annihilation—since present causes produce future effects, how could there be annihilation? He further says that the eight present links are nothing but many conditions and therefore contain no self. As for why the verse’s “this then does not arise” is called right cessation, his answer is this: the twelve links arise from deluded ignorance, and the wise therefore do not produce them. Since they are not real, they are empty. Right contemplation in this way enters ultimate truth and is called right cessation.

This mirror joins the chapter to Sanlun’s interpretive framework of refuting error and disclosing the correct. That judgment is Jizang’s addition, not the original wording of Nāgārjuna’s verses or Qingmu’s long explanation. The readings are recorded side by side without adjudication, each returned to its own place.

Across its twenty-seven chapters, the Treatise on the Middle refutes coming and going, arising and cessation, agent and action, cause and effect. Then Chapter Twenty-Six suddenly counts the twelve links with disarming plainness. A reader might think the work has ended, only to turn to this final chapter and discover that the treatise’s author still has one thing left to do.

When this chapter is complete, the final sentence of the entire treatise is not a conclusion but an action: “I now bow my head in homage.”

A treatise renowned for refutation comes to rest in a bow. What this chapter has to say is hidden in that very contrast.

This Question Retraces the Mahāyāna Path through the Śrāvaka Teaching

The opening line says: “I have heard the Mahāyāna Dharma refute wrong views; now I wish to hear the śrāvaka Dharma refute wrong views.”

The impulse behind this question is the same as in the preceding chapter. There the question was how the śrāvaka teaching enters the path of the ultimate meaning; here it is how that teaching refutes wrong views. The questioner is not asking for a second system. He wants to know whether the Dharma already in his hands can carry him all the way through.

The answer is perfectly steady. It does not begin in abstraction, but first lays out what must be refuted, and does so in exact order:

“In a past life, did I exist or not? Views of the world as permanent and the rest all depend on the past.”

“In a future life, will I be produced or not? Views of limit and the rest all depend on the future.”

Every entanglement concerning the “self” moves in only two directions: backward and forward. Looking backward, one asks whether one existed before, and views of permanence and impermanence arise. Looking forward, one asks whether one will be produced again, and views of limit and no limit arise. Qingmu completes each set as four propositions, then says: “These are called views of permanence and the rest that depend on the past,” and “These are called views of limit and no limit that depend on the future.”

Eight propositions, two directions. The author gathers every human anxiety about oneself into just two directions in time. He then asks: “For what reason are these called wrong views?”

The Self in the Past: Identity Entails Permanence; Difference Entails Annihilation

The long verse closes the issue at its outset: “A self in a past life cannot be found,” and “The self in the past did not produce the self of this life.”

Qingmu takes the plainest route. He avoids abstraction and asks about ordinary life: someone cultivates merit, is born a god, and later becomes human. If the past self were precisely the present self, then “a god is precisely a human.” Again, someone becomes a caṇḍāla through wrongdoing and later a brahmin; if the self were identical, “a caṇḍāla is precisely a brahmin.”

He deliberately gives a counterexample: “Suppose a brahmin named Devadatta in Śrāvastī is still called Devadatta upon reaching Rājagṛha; arriving there does not make him different.” Change the place and the person remains the same—exactly the picture held by one who posits a permanent self. But a change of condition is not a change of place. God and human are not one thing standing in two locations; therefore “a god is not precisely a human, nor a caṇḍāla precisely a brahmin.” To say they are one incurs the fault of permanence.

Someone retreats one step: perhaps the self is not identical with the body; the self remains unchanged while the body changes. The same person is “called a washer while washing clothes and a reaper while reaping.” Washer and reaper are not two persons, yet “the washer is not precisely the reaper.” Qingmu follows the analogy to its point of pressure: “Is the present washer different from the reaper or not?”

If not different, they remain identical and the former fault of permanence stands untouched. If different, “the washer does not produce the reaper”; likewise, a god does not produce a human, and the self becomes impermanent. Qingmu then adds the fatal sentence: “What is impermanent lacks the mark of self.”

There are only two paths by which to rescue a permanent self: one leads to permanence, the other to annihilation. Neither goes through.

What Use, Exactly, Is That Self?

The questioner now revises the claim. The self is identical, while the difference between god and human lies in appropriation: the body of five aggregates undergoes karma and therefore takes various forms, but the self itself is “neither god nor human, neither caṇḍāla nor brahmin.”

This rescue sounds clean, but Qingmu’s reply is severe. If it is the body that becomes god or human, then merit, wrongdoing, birth, death, coming, and going are all affairs of the body. “If pleasure and pain, anger and joy, anxiety and fear all belong to the body rather than the self, what use would the self be?”

Pause over this sentence. He does not say that your self does not exist. He asks: What use is that self? You have protected it too thoroughly—so thoroughly that it has nothing to do with pleasure or pain, merit or wrongdoing, birth or death. Why rescue something that touches nothing at all?

Karmic fruition never needed that self: “Because the conditioned continuity of the five aggregates does not lose merit and wrongdoing, liberation is possible.” Continuity is sufficient to bear karmic fruition and sufficient to make liberation intelligible.

The questioner then offers the strongest analogy: an owner repairing a house. Using grass, timber, mud, and plaster, the owner determines whether the house is good or poor; by doing good and evil, the self likewise receives good or ugly bodies, and it is the self that acts throughout birth and death in the six destinies.

Qingmu dismantles the analogy with care: “A householder has form, touch, and strength, and can therefore repair a house.” The owner can work on a house because the owner has form, touch, and strength. “The self you propose has neither form nor touch and therefore no power to act.” The very self you require has nothing at all—how could it act?

There follows a long series of tests, each pressed against ordinary experience. If the self were the agent, “it should not inflict painful deeds upon itself”—no one chooses suffering for oneself. If the self were the seer, and the eye sees forms, the eye should be the self; yet “if the eye were the seer, it could not apprehend sounds and other objects.” If the self used the faculties as a reaper uses a sickle, nevertheless “apart from the sickle a reaper can be found, whereas apart from body, mind, and faculties no separate agent can be found.” Beyond a sickle one finds the person who reaps; beyond body, mind, and faculties one finds no user. If an agent could exist without being apprehended by the faculties, “then the child of a barren woman could act.” If the right eye sees and the left eye recognizes, supposedly proving a separate seer, then “what the right hand has learned the left should also be able to do, yet in fact it cannot.”

The most arresting test concerns saliva. Someone sees another person eating fruit, salivates, and calls this the self. Qingmu says it occurs “through the power of memory, not the power of self”; indeed, the episode turns against the self: “A person among others is ashamed when saliva appears, yet it forces its way out beyond control—know from this that there is no self.” If you cannot even command your own saliva, how could there be a self that commands everything?

After refutation, Qingmu gives the positive account: “The five aggregates merely combine, and the provisional name ‘self’ has no determination.” It is like this: “As beams and rafters combine to make a house, apart from them there is no other house.” A house is genuinely useful, yet no separate thing called house can be found apart from beams and rafters. Thus “the self exists only as a provisional name and has no determinate reality,” and “there is appropriation alone, with no separate appropriator.”

The long verse settles the matter here:

“The present self is not apart from appropriation, nor is it appropriation itself; it is neither without appropriation nor wholly absent: this is the settled meaning.”

All four clauses must be seen. Not identical, it does not fall into permanence; not separate, it does not fall into annihilation. There is neither an absence of appropriation nor simply nothing at all. That final “not wholly absent” is the reader’s guardrail: refuting self does not replace it with a vacant annihilationist view.

Two Directions in Time, the Same Illness

The author now changes tempo and presses swiftly forward.

If the past and present selves differ, “there is annihilation and karmic fruition is lost”; moreover, “one acts and another receives.” If the self was formerly absent and now exists, “the self is a constructed dharma and is also without cause.” Thus all four past propositions are refuted together:

“Concerning the past, views that the self existed or did not exist— whether combined or uncombined, none of these is tenable.”

For the future set, the author does not even assign new faults: “Views of this kind are all the same as those concerning the past.” Qingmu says it more directly: the faults of those four propositions “should be stated here just as with the past.” The same illness, turned toward another direction in time, remains the same illness.

So the inquiry returns to god and human. “If a god is precisely a human, one falls into the extreme of permanence,” because “a permanent dharma is not produced.” What is permanent cannot be born again; if a god is not born among humans, there is no reason to call that god human. Conversely, “if a god differs from a human, this is impermanence,” and worse, “there is then no continuity.” Qingmu adds the decisive line: “Where there is continuity, one cannot speak of difference.”

Someone tries a compromise: half god and half human. Qingmu blocks it in one sentence: “It incurs the fault of two marks in one body.” Nor can the fourth proposition, “neither permanent nor impermanent,” be preserved. It arises only in dependence on the preceding three; when they fail, it has no footing.

The claim that birth and death are beginningless fails in the same way.

“If a dharma had determinate coming and determinate going, birth and death would be beginningless; yet in reality this is not so.”

Hidden beneath “beginningless” is a thing with determinate coming and going; yet “when this dharma is investigated with wisdom, nothing can be found from which it comes or to which it goes.” Remove the foundation, and the structure collapses.

The four propositions were never four separate things. They are a chain of claims leaning upon one another.

Limit and No Limit: No Boundary Can Be Found on the Five Aggregates

The future set of views changes its vocabulary and asks whether the world has a limit.

The author first blocks both extremes with a fact already at hand:

“If the world has a limit, how could there be a future life? If the world has no limit, how could there be a future life?”

—yet “in reality there is a future life.” A limited world leaves nowhere for future life; a limitless world equally fails to explain it.

The author then gives the chapter’s most beautiful positive image:

“The five aggregates continually follow one another, like the flame of a lamp; therefore the world should be called neither limited nor limitless.”

Qingmu explains with utmost simplicity: “A lamp flame exists through the conjunction of conditions. While the conditions are not exhausted, the lamp does not go out; when they are exhausted, it does.”

The lamp-flame analogy must be seen precisely. A flame is not a thing that burns; it is conditions continuing. While the conditions continue, so does the flame; when they end, it ends. You cannot draw a boundary across a sequence of continuity, because there was never a thing there that could be cut.

The author then presses each claim once more. If former aggregates cease and later aggregates do not arise conditioned by them, there is a limit. If former aggregates do not cease and later aggregates arise without depending on them, there is no limit. Qingmu gives “limit” a concrete definition: “Limit means the final body.” A limit would be the very last body. Yet former aggregates cease and condition the arising of later ones, so neither extreme can take hold. He also clarifies the scope: “There are two kinds of world: the world as land and the world of sentient beings. This is the world of sentient beings.” The boundary examined here concerns the continuity of beings, not the dimensions of the cosmos.

Qingmu then cites a verse from the Four Hundred Verses: “True Dharma, one who teaches it, and one who hears it are rare.” His explanation works in both directions. “Because the conditions for attaining true Dharma are absent, wandering through birth and death has no limit”—without encountering true Dharma, the cycle does not end. Yet “because one sometimes hears true Dharma and attains the path, it cannot be called limitless”—people do hear and awaken, so limitlessness cannot be asserted. Rare does not mean unobtainable; this sentence is left for the practitioner.

The remaining two propositions fare as before. Half limited and half limitless incurs “two marks in one dharma.” Neither appropriator nor appropriation can be partly destroyed and partly undestroyed, because that carries “the fault of the two marks, permanence and impermanence.” The fourth proposition, neither limit nor no limit, again has nothing to support it: “No limit is posited in opposition to limit, as short in opposition to long.” Whatever stands through mutual dependence loses its footing when what it depends on is refuted; “because there is nothing on which it can mutually depend,” it cannot be established.

The Final Turn: Who Produces These Views, Where, and When?

After the śrāvaka teaching has refuted each gate, the author suddenly looks up and asks a question wholly unlike the preceding ones:

“Because all dharmas are empty, views of the world as permanent and the rest— where, when, and by whom are such views produced?”

Qingmu makes the change of level explicit: “The preceding used the śrāvaka Dharma to refute views; now the Mahāyāna Dharma speaks.”

The preceding refutations addressed the content of views: permanence fails, impermanence fails, limit fails, no limit fails. This question addresses the arising of a view: “Dharmas have from the beginning been ultimately empty; in such empty nature there is neither person nor dharma.” If so, where, when, and by whom could these views arise?

Qingmu makes each of the three terms concrete: “Place means land; time means days, months, and years; who means a person. These constitute the substance of views.” Place, time, and person are the three pillars on which views depend. But the person has already been refuted—“because self is refuted, no person produces this view.” Even visible form taken as a place can be refuted, much more time and direction.

Most important is the criterion he gives: “If views existed, they would have determinate reality; if determinate, they could not be refuted.”

Something possessing intrinsic nature could not be moved by refutation. Yet every one of these views has already yielded under examination through many causes and conditions—and that very susceptibility shows that it never possessed intrinsic essence. Therefore: “Views have no determinate essence—how could they arise?”

This turn does not add one more refutation; it gathers refutation itself back in. The preceding twenty-six chapters refuted the dharmas being grasped. This question reveals that the person who grasps, the place of grasping, and the time of grasping likewise lack intrinsic nature. Once the inquiry reaches here, refutation itself has nowhere to stand.

Therefore the End Is an Act of Homage

And so the final four lines of the entire treatise are:

“The Great Sage and Lord Gautama, moved by compassion, taught this Dharma; he cut off every view, and I now bow my head in homage.”

Qingmu states their number plainly: views are “briefly five and extensively sixty-two,” and the Buddha “taught the Dharma in order to cut off these views.”

The causal order matters. It is not that the Buddha first possessed an ingenious view to transmit. Sentient beings were first entangled in sixty-two views and could not escape; moved by “compassion,” the Buddha taught. The beginning of teaching is compassion, not victory. Without this sentence beneath the dense inquiry of twenty-seven chapters, it could look like mere virtuosity in debate. With it, one sees that the entire treatise has been dismantling a net, not weaving another one.

The author’s own position is settled in the same sentence. He does not say, “Now I understand,” or “Now I have realized.” He says, “I now bow my head in homage.” Qingmu gives the reason: the Great Sage and Lord Gautama “possesses immeasurable, boundless, inconceivable wisdom.”

When every view has been cut off, the wisdom that cuts them off is not one’s own—this is the final action the author leaves to the reader. When refutation reaches what cannot be further refuted, one does not remain standing. One bows.

What Readers Can Carry Away from This Chapter

First, watch the direction in which thought moves. Think backward about what one was and forward about what one will be: each direction generates an entire set of views. The failure of all eight does not forbid thought about past and future. It shows that no fixed self waits there to be found.

Second, do not protect the self too thoroughly. “What use would the self be?” is a sharp question: a self unrelated to pleasure and pain, merit and wrongdoing, is useless even if rescued. Karmic fruition does not depend upon it, nor does liberation—“because the conditioned continuity of the five aggregates does not lose merit and wrongdoing, liberation is possible.”

Third, remember the lamp flame. No boundary can be drawn upon continuity. There is no need to fear annihilation, and no need to search for a first point of birth and death.

Fourth, when refutation reaches its end, remember to bow. “He cut off every view” is followed by “I now bow my head in homage”—not by anything else.

Carried Forward: Jizang’s Mirror

The foregoing is the meaning of this chapter itself. What follows looks alongside it through the Sui–Tang Sanlun tradition. The mirror is not the treatise; it is only another way of reading the same verses.

In his Commentary on the Treatise on the Middle, Jizang preserves one degree of room within conventional truth. He says that claiming there is wholly no self in conventional truth is itself a wrong view. A provisionally named self is not absent, yet there is neither a real entity nor a real efficacy belonging to it. This carries the verse’s final phrase, “neither without appropriation nor wholly absent,” one step further in the direction of conventional truth.

He reads neither limit nor no limit through continuity. Because earlier and later aggregates condition one another, they are not annihilated, and hence there is no limit; because they arise and cease moment by moment, they are not permanent, and hence there is no limitlessness. He identifies this passage as the correct śrāvaka view.

For the final question, he establishes three roots: the person producing views, the place where views arise, and the time at which they arise. The chapter shows all three to be ultimately empty; without a root of views, views do not arise. Jizang thus judges that, after refuting ordinary people’s views, the chapter washes away the views of the two vehicles and of bodhisattvas with the water of Mahāyāna ultimate emptiness, rendering wrong and right equal and great and small nondual.

This interpretive structure is Jizang’s addition, not the wording of Nāgārjuna’s verses or Qingmu’s extended commentary. The readings are recorded side by side without adjudication, each returned to its own place.