Part I, Chapter 1: Emptiness and Liberation: The Core Claim¶

This opening chapter of the Madhyamaka essay has one task: to state Madhyamaka soteriology clearly in its strongest form, without explicit criticism for now. We first need to understand where Madhyamaka itself thinks suffering comes from, what liberation is, how it is reached, and how its attainment is recognized.

The account follows the standards set out in Chapters 1 and 2 of the Introduction: the five questions modeled on the Four Noble Truths—prior premise (what suffering is), diagnosis (where suffering comes from), goal (what liberation is), mechanism (how liberation occurs), and verification (how attainment is recognized, also called identification). Madhyamaka is especially suited to assessment by these standards, for an unusual reason: it claims to offer no new doctrine at all, merely to explain the Buddha's dependent origination and the Four Truths fully. If that claim holds, Madhyamaka should fit precisely into standards themselves drawn from the Four Truths. This chapter presents its own answers in the order of the five questions. Part II examines whether those answers match what Madhyamaka actually does.

Nāgārjuna's understanding of his own role: clarifying dependent origination, establishing no new teaching

Nāgārjuna (c. second century), as he understood his work, was clarifying the original meaning of the Buddha's teaching, not founding a new doctrine. The dedicatory verse at the opening of the Mūlamadhyamakakārikā states this position explicitly: the Buddha taught dependent origination (pratītyasamutpāda), thereby bringing conceptual proliferation (prapañca) to rest; all Nāgārjuna's arguments do is work through the meaning of dependent origination to the end.¹ Emptiness (śūnyatā) is its direct logical consequence, not his invention: whatever arises dependently lacks intrinsic nature, and to lack intrinsic nature is to be empty. As a statement of doctrine this is correct; as a statement of the structure of the argument, it needs an adjustment. The list of premises in Essay 18 of Volume 1, Part II contains no dependent origination. That essay's derivation starts from the three conditions for intrinsic nature and one relational premise. Dependent origination is upheld on the conclusion side, not employed on the premise side. In that essay's terms, dependent origination is what is preserved, not what is proved. None of the twenty-seven chapters tries to prove dependent origination; what they prove is that there is nowhere to stand outside it. “Emptiness” and “dependent origination” are thus two sides of the same actual situation: one states the negative (no intrinsic nature), the other what holds (arising in dependence on conditions). What Nāgārjuna criticizes is the Abhidharma reading that analyzes dharmas into minimal real units and assigns each an intrinsic nature (svabhāva). On his account, what has intrinsic nature need not arise dependently, and what arises dependently cannot have intrinsic nature. The Abhidharma doctrine of the real existence of dharmas is therefore a fundamental misunderstanding of dependent origination.

For this very reason, Madhyamaka claims to lie wholly within the Four Truths. Chapter 24 of the Mūlamadhyamakakārikā answers an imagined opponent's objection, “If everything is empty, the Four Truths cannot hold.” Nāgārjuna reverses the challenge: precisely because everything arises dependently and is empty, the Four Truths can hold; if dharmas each had intrinsic nature and cause and effect could not connect, the Four Truths would instead be empty words. This step carries enormous soteriological weight. It means that understanding emptiness is not a two-step affair—“first understand a metaphysics, then practice.” Rather, “understanding emptiness” itself is said to be correct insight into the root of suffering. We can now follow the five questions to see how this account unfolds.

The truth of suffering: the same starting point, with nonself generalized

At the level of the truth of suffering, Madhyamaka largely takes over early Buddhism's account: the three characteristics—suffering, impermanence, and nonself—remain unchanged, and suffering remains what must cease. Its distinctive move is not to rewrite this truth, but to try to derive the “nonself” among the three characteristics through analysis. On a common Mahāyāna account, early Buddhism's nonself primarily means “the selflessness of persons” (pudgala-nairātmya): no permanent, autonomous “I” can be found either within or outside the five aggregates. Madhyamaka tries to apply the same insight to the five aggregates themselves, advancing “the selflessness of dharmas” (dharma-nairātmya). Not only does the “I” lack intrinsic nature; the mental and material elements (dharmas) that make up the “I” likewise lack it.

The classification “śrāvakas realize only the selflessness of persons; only bodhisattvas realize the selflessness of dharmas” did circulate widely among later Madhyamaka thinkers, but it does not come from the Ratnāvalī itself. Nāgārjuna's treatise maintains precisely that the two vehicles realize the same emptiness. He first says that correctly knowing sentient beings to be unreal brings liberation, and immediately adds, “Bodhisattvas also see in this way.” That single Chinese character meaning “also” marks the sameness of what is seen. He further says that without removing grasping at the aggregates, grasping at the self cannot be removed; anyone seeking liberation must therefore understand that dharmas lack intrinsic nature. Finally, he states outright that the nonarising taught in the Great Vehicle and the extinction taught in the other vehicles are the same emptiness. He locates the distinction between the two vehicles in compassionate aspiration, not in what they see: bodhisattvas see this and nevertheless continue in the world out of compassion.²

This distinction matters, because the structural consequence described below follows from some classifications in later Madhyamaka, not from the intention of the Ratnāvalī itself: the śrāvaka's arhatship is demoted to “incomplete liberation.” Though suffering has ceased, subtle grasping at dharmas remains, and the arhat path becomes an initial stage of the bodhisattva path. This demotion was a later addition, with identifiable texts and authors; it was not Nāgārjuna's own position. That makes this chapter's subject all the clearer: Madhyamaka's placement of early Buddhism was worked out in several historical steps, not settled from the outset. For present purposes, we need only remember that, at the level of claims, Madhyamaka has not altered the truth of suffering. Its overt advance concerns only the direction of “nonself,” yet this advance directly rewrites the diagnosis in the next link. Whether the claim matches what it does, and whether suffering still occupies its original place within the Madhyamaka framework, will be tested in Part II, Chapter 1.

The truth of origin: suffering is rooted in the view of intrinsic existence

Where does suffering come from? Madhyamaka's diagnosis, in a phrase, is the view of intrinsic existence (svabhāva-dṛṣṭi): grasping at what lacks intrinsic nature as though it possessed it. The diagnosis can be followed along a causal chain. The deep root of all suffering is the twofold grasping at self and dharmas, which regards both “I” and “dharmas” as existents with fixed intrinsic natures. This grasping in turn supplies the epistemic foundation for the three poisons of greed, hatred, and delusion. Precisely because an intrinsically existing “I” is presupposed, there is greed for “my interests,” hatred of “threats to me,” and delusion that fails to see this mechanism at work. The three poisons drive the creation of karma, and karma drives the reproduction of suffering. The diagnosis settles on the point furthest upstream: the view of intrinsic existence.

Madhyamaka's repositioning of “ignorance” needs attention here. Early Buddhism defines ignorance as not knowing the Four Truths (Introduction, Chapter 2); Madhyamaka specifies its core content as the view of intrinsic existence, the mistaken belief that dharmas have intrinsic nature. This attempts a more precise account of “exactly what ignorance gets wrong,” rather than overturning the early diagnosis: ignorance is ignorance precisely because it reads dependently arisen dharmas as dharmas with intrinsic nature. Whether this specification remains faithful to the early accounts of suffering and its origin will be examined later. Here we first record Madhyamaka's answer.

Supporting this diagnosis are Nāgārjuna's arguments against intrinsic nature. Volume 1 has already devoted a chapter to their particular forms, such as “nonarising through the four alternatives,” so they need not be repeated here (see Volume 1, 5.3, “Madhyamaka's Methods of Argument”). For this chapter, two levels of the main point must be kept distinct. The first is the tradition's account of itself: it does not positively establish that “a thing called emptiness exists.” It merely uses reductio (prasaṅga) to expose the self-contradiction in the opponent's presupposition that “dharmas have intrinsic nature.” It establishes no thesis of its own, showing only that the other's thesis cannot be consistent. Essay 18 of Volume 1, Part II extracts a theorem from these arguments and arrives at another level. What is absent from “establishing no thesis of one's own” is an object-level metaphysical thesis—an assertion that something really has intrinsic nature. The argument has all along used metalevel resources—definitions, logic, and one relational premise—and has always been able to list them openly. The defense in the Vigrahavyāvartanī is itself such a list. That essay's verdict is also that the refutation holds universally, not merely case by case: its universal force comes from the lexical rules, not from the number of examples in the twenty-seven chapters. The difference between these levels matters repeatedly in the chapters that follow. Wherever the presupposition “Madhyamaka has only case-by-case reductio at its disposal” appears, it must be adjusted in light of this distinction.

The truth of cessation: conceptual proliferation ceases; nirvāṇa too is empty

What is liberation? For Madhyamaka, the positive content of the truth of cessation is the cessation of conceptual proliferation (prapañca). Once the view of intrinsic existence has been seen through, the conceptual weaving that imposes intrinsic nature on the world subsides. The mind no longer spins idly in the slots prepared for intrinsic nature—“being/nonbeing, identity/difference, coming/going.” The “stilling of conceptual proliferation” praised in the dedicatory verse of the Mūlamadhyamakakārikā is this endpoint.

One feature of Madhyamaka's truth of cessation needs particular emphasis: it judges nirvāṇa itself to be empty as well. Chapter 25 of the Mūlamadhyamakakārikā, devoted to nirvāṇa, concludes that nirvāṇa too lacks intrinsic nature. In this respect, nirvāṇa and saṃsāra do not differ: hence their nonduality. This differs from early Buddhism's silence about “what nirvāṇa is” (Introduction, Chapter 2). Early Buddhism offers no positive description; Madhyamaka goes further, explicitly denying that nirvāṇa has any intrinsic nature to grasp at, and even denying the grasping that takes “nirvāṇa to be a thing one can realize.”³ In its own understanding, this does not reduce nirvāṇa to nothingness, which would fall into the annihilationist view that Madhyamaka strenuously avoids. It rewrites “nirvāṇa” from a really existent goal to be reached into the nonattainment that follows the complete release of grasping. Once the goal is set in this way, it interlocks tightly with the mechanism in the next link: if the endpoint is “no longer grasping at anything as having intrinsic nature,” the path to it is naturally “seeing through all intrinsic nature.”

The truth of the path: seeing emptiness breaks the two kinds of grasping

How does liberation occur? The core of Madhyamaka's mechanism is a single statement: seeing (darśana) that dharmas lack intrinsic nature destroys the epistemic foundation of grasping at self and dharmas. Once their lack of intrinsic nature is seen as it is, the foundation is pulled from under the view of intrinsic existence. The three poisons lose their foothold; the creation of karma then stops, and the reproduction of suffering is interrupted. This sequence fits the diagnosis exactly. The diagnosis locates the root of the disease in the view of intrinsic existence; the mechanism dissolves that view by “seeing emptiness,” answering it point for point. This correspondence between diagnosis and mechanism is one of the standards used in Introduction, Chapter 2 to assess the schools. On this point, Madhyamaka appears to meet the standard closely.

Two points should be supplied on Madhyamaka's behalf to avoid oversimplification. First, how does “seeing emptiness” amount to “breaking self-grasping”? Nāgārjuna's answer is that self-grasping is essentially a mistaken judgment with epistemic content: it judges what lacks intrinsic nature to possess it. It is a judgment that argument can overturn, not a mere emotion. When reductio refutes that judgment, grasping loses its epistemic support. Second, Nāgārjuna does not equate “understanding emptiness intellectually” with completed liberation. His texts repeatedly stress a process requiring prolonged cultivation (bhāvanā). The right view obtained through argument is only a starting point; meditative contemplation must turn it from an accepted proposition into a direct seeing here and now.

This is the skeleton of Madhyamaka soteriology and its strongest defense of itself. It claims that its mechanism has the same structure as early Buddhism's: “dissolve ignorance through right view, thereby removing the conditions of suffering.” It merely seeks to define the content of ignorance precisely as the view of intrinsic existence, and that of right view precisely as the view of emptiness. If this holds, Madhyamaka follows the old path of “knowing and seeing things as they are” more thoroughly, without starting a separate enterprise. This “epistemological optimism”—the belief that correct cognition can penetrate and alter the mechanism that produces suffering—is the load-bearing wall of the entire Madhyamaka soteriology. Whether it can stand is Part II's central question.

Verification (identification): is seeing emptiness direct realization, or a right view reached by inference?

The final question is how to determine that “seeing emptiness” has occurred, and who is qualified to determine it. This is precisely where dispute within the Madhyamaka tradition continues to this day, and it is the question of verification to which this volume pays particular attention.

The dispute centers on whether the “seeing” that realizes emptiness is direct perception in the sensory sense (pratyakṣa, direct realization), or inference in the reasoning sense (anumāna, right view obtained through argument). Both alternatives face difficulties. If seeing emptiness is merely a right view reached by inference, it does not fundamentally differ in form from understanding a principle of physics: it is intellectual acceptance, whose persuasive force depends wholly on the quality of the chain of argument. It then becomes difficult to explain why this should produce a transformation as deep as liberation. If seeing emptiness is direct perception, the question becomes “direct perception of what?” Emptiness is not an object; how can it be directly experienced? The Abhisamayālaṃkāra attempts to provide a systematic architecture for this direct realization, matching stages of the path one by one with the bodhisattva grounds (bodhisattva-bhūmi). In doing so, however, it develops “direct realization” into an extremely elaborate theory of the path of seeing (darśana-mārga), whose relation to actual meditative experience remains deeply disputed within Tibetan Buddhism.⁴

The disagreements extend to “how to reach right view.” The Svātantrika school of Bhāvaviveka (c. sixth century) maintains that right view must be established by valid independent syllogisms, then deepened through meditation. The line of Candrakīrti (c. seventh century) holds that reductio suffices, with no need to establish an independent argument; the name Prāsaṅgika is a later retrospective designation, not Candrakīrti's own. Tsongkhapa (fourteenth–fifteenth centuries) offers a synthesis grounded in Prāsaṅgika, while emphasizing that “conventional valid cognition”—cognition valid at the conventional level—is indispensable to the stages of the path.⁵ On the surface these are disputes over argumentative method. Beneath them lies the same problem of verification: how is the step made from a proposition accepted on the strength of an argument to a direct seeing here and now, and who confirms that it has been made? Later in this volume we will see that it is precisely at this link that Madhyamaka's standards of verification show signs of becoming more elite and specialized. That comes later; here the problem is simply set out.

Recapitulation: a soteriology claiming the same structure as the Four Truths

In the first row, the Chinese figure says that arahatship is downgraded to an intermediate station. The second row specifies ignorance more precisely as reading dependently arisen dharmas as intrinsically existent. The fifth heading means “Verification (identification).” “Footer” is an added English layout label, not part of the Chinese figure's closing claim.

Taken together, Madhyamaka's answers to the five questions run as follows. The truth of suffering remains unchanged, but “the selflessness of dharmas” is intended to encompass and govern “nonself.” The truth of origin locates suffering's root in the view of intrinsic existence. The truth of cessation takes the cessation of conceptual proliferation and the emptiness of nirvāṇa as its endpoint. The truth of the path breaks the view of intrinsic existence by seeing emptiness, thereby removing the conditions of suffering. Verification rests on the still unresolved dispute over whether seeing emptiness is direct realization or inference. At every link Madhyamaka claims merely to carry early Buddhism's “knowing and seeing things as they are” through to the end, thus retaining the structure of the Four Truths and even pursuing it more thoroughly than the śrāvaka path.

This chapter's task ends here: to present these claims faithfully in their strongest form, without explicit judgment. Their weight can nevertheless already be noted. Madhyamaka places almost the whole weight of liberation on the single epistemic act of “seeing through intrinsic nature.” The consequences of this change for soteriology are incalculable. It enabled Madhyamaka to reconstruct soteriology comprehensively, and created a paradigm for its later transformations.

References¶