Part II, Chapter Two　The Truth of Origin: The Discovery of Manas and the Abyss of Iteration¶

Part II, Chapter Two, of the Madhyamaka essay concluded that Madhyamaka diagnoses the root of suffering as the view of real existence. Yet that view resides at the reflective level, while the craving and appropriation diagnosed by the baseline operate at the prereflective level: the diagnosis addresses the wrong level. This chapter applies the same test to Yogācāra's diagnosis, with a three-part conclusion. In the level it addresses, Yogācāra's diagnosis is more accurate than Madhyamaka's: manas is precisely a doctrinal registration of the prereflective level. Yet the diagnosis does not stop after identification; it continues iterating downward until it places its foundation at an untestable depth. And the accurately identified root of the illness is itself on no antidote's target list: treatment moves its work elsewhere. Accurate diagnosis is its merit; the latter two are its maladies.

I. Manas: doctrinal registration of the prereflective sense of self

First, the side deserving affirmation. The Madhyamaka essay used contemporary phenomenology's vocabulary to distinguish reflective self-clinging from the prereflective sense of self, and noted that Madhyamaka's reductio reaches only the former. Yogācāra need not borrow this vocabulary; it has its own. Manas constantly examines and thinks, prior to reflection, requiring no learning, uninterrupted even in sleep or unconsciousness in its taking ālaya as self. The sixth consciousness's self-view is intermittent and can be induced by mistaken teachings and reasoning: self-clinging arising from discrimination. Manas's self-clinging is innate and spontaneous: innate self-clinging. As Part I, Chapter One, stated, this is the first time in the history of Buddhist theories of consciousness that “the subtle sense of self constantly present” receives a specific name and a dedicated institution.¹

The comparison is ready to hand. Three tests in the Madhyamaka essay, Part II, Chapter Two—infants, animals, and people who have lost the capacity for reflection—show that the prereflective sense of self does not depend on reflective self-clinging. These cases present Yogācāra no difficulty. Infants and animals have manas and innate self-clinging; whether reflective capacity is present concerns only the sixth consciousness's self-clinging arising from discrimination. When the Geluk school answers the problem of levels with the distinction between innate and acquired clinging (Madhyamaka essay, Part II, Chapter Two), it borrows precisely this Yogācāra distinction. In the level addressed by diagnosis, Yogācāra is the closest of the five systems to the baseline. The operation of craving and appropriation prior to reflection is not only acknowledged here but institutionalized.

II. After registration: why diagnosis does not stop

Yet diagnosis does not close here. Manas clings to self—what does it cling to? Ālaya. Why does it cling constantly? Beginningless habitual impressions. Where are they stored? Ālaya. Why does ālaya contain defilement? Because seeds and manifestations have perfumed one another from beginningless time. Every question receives an answer, every answer demands another question, and the chain goes downward until it stops at “beginningless” (anādi): the element from beginningless time, the common support of all dharmas.²

The role of “beginningless” in this structure must be clear. It is an endpoint of explanation, hardly an explanation itself. The chain stops advancing, not because the question is answered, but because the system announces that no further question is needed here. The baseline uses “beginningless” too, but the two are different things. The baseline's beginninglessness is a necessity under dependent origination's worldview (Volume One 1.5): everything arising has conditions, and no link can serve as the first. “Beginningless” is therefore a structural feature of the chain's grammar, not historical information. Hence the Buddha says ignorance's “first beginning cannot be known,” but never says that ignorance results from perfuming accumulated across infinite time. Ignorance has no history of manufacture; it has always been there, sustained by present conditions. Scripture says ignorance too has nourishment, taking the five hindrances as its food. The baseline's diagnostic chain is therefore short. Suffering depends on craving and ignorance; ignorance is not knowing the four truths. This is a functional identification, with no further inquiry into ignorance's storage place or manufacturing history (Introduction, Chapter Two). Yogācāra's chain is long. Beneath ignorance it builds an institution, manas; a storehouse, ālaya; inventory, seeds; and a process, perfuming, before arriving at the same word. Yet on arrival, “beginningless” has changed grammatical role. It no longer states the structural fact that “the chain has no first link,” but becomes the warehouse's age. Habitual impressions are stock perfumed and accumulated across infinite time; ignorance now has a manufacturing history and a place of safekeeping. Both chains stop at the same word, but are no longer saying the same thing. The intervening extra layers explain more and commit to more. Each is a new theoretical entity needing defense, and all their ultimate defenses converge on one point: without this posit, such-and-such a phenomenon cannot be explained.

There is a particularly direct textual sample of this stance that “every question must have an answer”: the Sūtra Explaining the Profound Intent, Yogācāra's foundational scripture. Listing its questions chapter by chapter alongside the baseline's and Madhyamaka's responses to the same questions shows what has happened to the list of questions here.³

The three columns side by side should already give a clear impression. Faced with the same questions, the baseline either judges an answer unnecessary, leaving the matter undeclared, or stops after a functional identification. Madhyamaka dismantles the questions along with their presuppositions. The Sūtra Explaining the Profound Intent treats every question as a real question, each owed an answer by the system, and all answers lead to the same new theoretical entities: appropriating consciousness, the three natures, and the three periods. The sūtra's title itself declares this program: what is profound and hidden must be explained. Every silence the Buddha left becomes a knot awaiting untying. “Why?” can no longer stop, because each part used to untie a knot is itself a new, undefended posit. The downward chain examined above is the direction consciously chosen by this foundational scripture, not a later accident.

The Chinese diagram gives a specific explanatory gain beside each of the five levels: (1) distinguishing the subject–object structure and explaining the constitution of experience; (2) explaining the uninterrupted grasping of “I”; (3) explaining continuity of karmic results, so that causes and effects are not lost; (4) explaining the gradual accumulation of practice and defilement; and (5) marking explanation's limit and pointing to what cannot be observed. The baseline inset glosses craving as craving and clinging, and ignorance as not knowing the four truths.

Both images place manas, ālayavijñāna, and seeds above the observability line. The text of this chapter and Chapter 5 describes these as theoretical entities that cannot be directly observed. The line in the diagram and the account in the text are retained with this difference made explicit.

III. Iterative depth: connecting with the preface's verdict

The preface's verdict on Yogācāra was attachment to a foundation and explanatory certainty, working through iterative depth: drilling downward all the way in search of an ultimate bottom. This chapter can add an account of its mechanism on the diagnostic side. Each downward probe produces a real explanatory gain: manas explains the constancy of the sense of self, ālaya the continuity of karmic results, seeds the accumulation of gradual cultivation. Precisely these gains, delivered step by step, make the probing unable to stop. On reaching “beginningless ālaya,” the system gains a peculiar foundation. It explains everything but is itself unobservable and untestable; all its evidence for existence is “the system needs it.” The pursuit of certainty finally rests on a named unknown. This is the necessary endpoint of the impulse to “iterate diagnosis to the bottom,” not a Yogācāra oversight. At any observable level one can still ask why; only the unobservable can serve as “ultimate.”

The tradition responded to such questions, and its manner of response is itself worth recording as evidence. In establishing ālaya consciousness, the Establishment of Consciousness-Only registers objections one by one and answers each. To “this consciousness must then be self,” it replies: constantly turning like a torrent, neither permanence nor annihilation. To the question of which comes first when seeds and manifestations produce one another, it replies with simultaneous causation: “as a wick produces flame, and flame produces a charred wick,” then caps the account with “beginningless.”⁴ Note the form of response. Every objection is admitted, and every answer employs a new analogy or finer specification. None ends with “this question is invalid.” The baseline closes questions through the undeclared; Madhyamaka dismantles them through reductio; Yogācāra admits them all. Admission produces explanatory gains, and gains produce the next question. The mechanism of inability to stop discussed here is directly visible in the texts' question-and-answer form.

IV. After diagnosis: why manas receives no antidote

Diagnosis identifies the manifest agency of the illness's root as manas; its inventory form is in ālaya, as Section Two showed. The natural next question is: how is manas counteracted? Yogācāra's answer is unexpected: it is not. Verse seven of the Thirty Verses lists only three occasions when defiled manas ceases: arhatship, the attainment of cessation, and the presence of the supramundane path.⁵ None takes manas as its object or target of counteraction. The wording needs one degree of precision. Two of the three states lie on the causal path. The attainment of cessation and the supramundane path are actual states registered in the theory of practice, states practitioners can enter, not merely events viewed from outside. Entry still depends on the sixth consciousness's work; manas itself remains outside the list of that work's objects (Volume One, Part II, Essay Twenty, Chapter Four).

First consider the system's own reasoning, recording its alternative in full. The sixth consciousness does the work—hearing, thinking, cultivating, and the four investigations—and stores the resulting seeds in ālaya. Uncontaminated seeds' power gradually grows. When nondiscriminative wisdom becomes present, manas temporarily loses its object and has nothing to grasp. At transformation of the basis, the entire support is replaced, and manas automatically turns into the wisdom of equality. Since the guard on duty cannot be repaired, remodel the warehouse that he guards and that also sustains him. A structural misalignment follows. Diagnosis targets the seventh consciousness, the sixth executes the operations, and the eighth is the construction site: illness in the seventh, operation in the sixth, construction in the eighth. In the baseline, all three coincide in one place. Craving is both the diagnosed root of illness, the object observed, and what the practitioner directly watches weaken. Yogācāra separates the three, placing a transfer station that cannot be opened for inspection between them.

The baseline inset in the Chinese diagram has three labels around craving: it is diagnosed, it is observed, and it is watched as it weakens. The English inset replaces these with “Diagnose,” “Operate,” and “Construct”; the three roles coincide in craving in the source. The verdict “The diagnosis is precise but not actionable; the operation is real but not verifiable” concerns the Yogācāra arrangement of the seventh, sixth, and eighth consciousnesses. It appears outside the baseline inset in the Chinese diagram, although the English diagram places it inside that inset.

The plan sounds reasonable at every link. Precisely for that reason, we should not rush to affirm this “absence of counteraction,” but first examine the status of the statement “manas cannot be counteracted.” It is certainly consistent, but consistency does not mean correctness, and here consistency is all there is. Manas is an entity established for explanation. Part I, Chapter Three's chain of inference—what the system needs must exist—applies to the seventh consciousness as well. Whether an invented institution can be counteracted depends on no observation, but on the properties its inventor gives it. The system registers it as spontaneous, constantly examining, subtle, and beyond the sixth consciousness's reach, while all operations of practice can be initiated only by the sixth. It is therefore impossible to counteract; change the registration, and it could be counteracted. This is stipulation, not discovery. The architecture has already fixed the direction too. Suffering has been placed in ālaya (Part II, Chapter One), and the entire system's sole writable surface is the eighth consciousness receiving perfuming. If work could address the root of illness directly, the whole detour through the warehouse would become redundant. In this system, therefore, “manas cannot be counteracted” is a load-bearing component, not an observational report. Its job is to ensure that treatment must pass through ālaya. An invention that defines itself as immune to counteraction cannot then claim credit because “it does indeed resist counteraction.”

A mirror for this judgment is available. The baseline faced the same problem, and its solution deserves to stand alongside. In the Khemaka discourse, the ill monk Khemaka observes the five aggregates of clinging as “not self, not belonging to self,” yet declares that he is not an arhat with the influxes exhausted. “Self-conceit, self-desire, and the latent tendency to self” remain “not eliminated, not known, not abandoned, not expelled.” His view has arrived, but the deep sense of self remains: precisely the baseline version of the manas problem. In the baseline, however, “the deep sense of self does not obey commands” is an experiential fact the practitioner encounters. When one orders the sense of self about, the person giving the order is the very person wanted for arrest. The command's failure is a report after trying, not a theoretical stipulation. The discourse's treatment is “advancing in contemplation, observing arising and ceasing,” not suppression. Its analogy is cleanly washed clothing that “still has a residual scent, which must be removed by perfuming with various fragrances.”⁶ The word is literally “perfuming.” The term habitual impression (vāsanā) itself comes from the fragrance analogy; Yogācāra's perfuming uses the same metaphor. The entire difference lies in the place of registration. Khemaka's residual scent can be smelled, and the progress of perfuming can be checked in experience: self-conceit arises less often, more faintly, for shorter periods. Yogācāra moves the residual scent into a cabinet that cannot be opened. The progress of perfuming becomes a theoretical inference, its efficacy evident only upon reaching the path of seeing. With the same analogy, the baseline keeps it at the phenomenal level; Yogācāra makes it chemistry inside a warehouse. Their difference in epistemic status is fixed here too. The same sentence, “the deep sense of self does not obey commands,” is an observational report when Khemaka says it, a specification when Yogācāra writes it. The designers may have had the same experience, but once it enters a definition, its status within the system changes from evidence to axiom. An axiom is no longer tested, and can no longer claim credit for “matching observation.”

Two points must be made for the system, together here. First, a word for the treatise authors: no one puts something obviously wrong into a definition. Manas was probably written this way because this account looked broadly correct. The deep sense of self does not obey commands; anyone who has tried can verify that. The specification reads as true. Yet “looking broadly correct” has a direct consequence for the system's arrangement of work: this page can be closed. Once manas is properly defined, the authors put it aside to refine the real center of engineering, ālaya consciousness. Part I, Chapter Three, showed where that refinement goes: six arguments, six seed characteristics, qualifications for receiving perfuming. The close attacks and defenses all concern the eighth consciousness; the seventh receives only a section on “establishing a dedicated office.” It looks right, so it escapes inspection; it is stipulated, so it is settled. This acknowledgment does not shift the criticism's position but completes it. Stipulations can serve comfortably as stipulations precisely because they resemble experience. Second, the very form of substituting perfuming for treatment may be the mechanism closest to modern habit psychology among the five systems. Implicit processes are not directly commanded by will and can be rewritten only through repeated practice; “knowing” is not “changing.” Finally, the scope of what is withdrawn must be clear. Section One's affirmation is not withdrawn. The existence of the prereflective level and the operation of a sense of self there have evidence outside the system: the tests with infants and animals, and the Khemaka discourse above. What is withdrawn is only the credit being prepared for the defined property “not subject to counteraction.”

The section comes together in one sentence: Yogācāra's diagnosis and site of work are misaligned, and the misalignment belongs to the design, not faulty execution. Diagnosis points one level beneath where one can act, manas; work occurs one level beyond where one can check, ālaya. Diagnosis is accurate but cannot be acted on; work is real but cannot be checked. This misalignment is also one structural reason for the timetable of three great eons. The only lever is indirect seed storage, so the road must be measured in eons (Part I, Chapter Two). The error lies in placing the record of rewriting where no one can inspect it, not in the form of gradual rewriting. Yet that choice of site affects the entire line. The transformation of the basis promised by the truth of cessation becomes a replacement occurring out of sight; believing it means believing the whole system at once. Every exercise on the path is directed toward that unseen place and receives no feedback, leaving practitioners no basis for correcting their road. Once the register is placed in the unknowable, the credibility, verifiability, and practicability of the truths of cessation and the path are all pledged in advance. They fall due respectively in Part II, Chapters Three, Five, and Six.

V. Two points of fairness

First, as with Madhyamaka, refinement of one's own system must not be equated with admission of being in the wrong; Part II, Chapter One, of the Madhyamaka essay established this limit. Adding manas and ālaya serves real explanatory aims in each case. This chapter criticizes not addition itself but its direction: every step goes toward “a deeper substrate” rather than “a more testable mechanism.” Second, the baseline's diagnosis also faces “why does craving arise?” Its stopping point—ignorance, not knowing the four truths—is likewise no ultimate explanation. The difference is that the baseline recognizes it as a practical identification: knowing this is enough to begin practice. It builds no ontological basement for it. Diagnostic depth is not an offense; what must be checked is the ontological cost paid for depth.

VI. Conclusion

Yogācāra's truth of origin diagnoses the level more accurately than Madhyamaka: it sees the prereflective level and names it. Yet accurate diagnosis does not yield accurate treatment. No antidote acts on the identified root of illness; the site of work moves inside ālaya. Indeed, because that root is defined as spontaneous and impervious to commands, treatment must take a detour. The accuracy of diagnosis and treatment's detour are two sides of the same registration. Diagnosis also refuses to stop where action is possible, iterating into unknowable depths until it locates suffering's ultimate cause—beginningless habitual impressions within ālaya—where no observation can reach. The deeper the diagnosis, the further from testability. This and the preceding chapter's migration of suffering are two sides of one matter: suffering moves into the storehouse, and its cause resides there too. A root of illness in an unknowable place and a route of treatment that bypasses that root will decisively affect the quality of mechanism and verification. Later chapters address those matters. The next first examines the endpoint: is the cessation promised by transformation of the basis an ending or an upgrade?

References¶