Appendix II: Modern Forms of Madhyamaka—Re-Testing the Verdict¶

This volume’s verdict on Madhyamaka takes its classical form as object: Nāgārjuna, Āryadeva, Candrakīrti, and their continuations in China and Tibet. Yet the obligation to “present the strongest form before criticism” (Preface) has no temporal limit. During the past half-century, research on Madhyamaka within the analytic-philosophy tradition has produced several evidently stronger modern forms, each reinforcing one of classical Madhyamaka’s most vulnerable aspects. This appendix presents four principal approaches and asks the same question of each: must this volume’s verdict be revised? At the end, the same ruler is laid against this book’s own theorem-shaped form.

The semantic reading: raising emptiness to the meta-level

Siderits and Garfield’s “semantic reading” argues that emptiness is not a metaphysical proposition about things but a second-order proposition about truth. To say all phenomena are empty does not mean that everything is made from something called “emptiness”; it means there are no bearers of ultimate truth. Its slogan is self-reflexive: “The ultimate truth is that there is no ultimate truth.”¹ This reading blocks the two oldest blades. It does not deny conventional efficacy and therefore avoids the charge of nihilism; it refuses to establish emptiness as an entity and therefore avoids reobjectification. One point must be conceded. The criticism of reobjectification in Madhyamaka Part II, Chapter 3 loses half its force under the semantic reading. This is genuine philosophical progress.

But what does the reading do to the remaining links in the verdict? Raising emptiness to the semantic meta-level does not shorten the distance between view and craving but adds another layer. This volume’s migration argument concerning suffering says that Madhyamaka moves its operating surface from pre-reflective craving to views at the reflective level (Madhyamaka Part II, Chapters 1 and 2). The semantic reading moves the view one level further, into a theory about theories. How can the second-order insight “no theory possesses ultimate warrant” reach non-theoretical, pre-reflective attachment? The causal gap expands from “between view and feeling” to “between meta-view and feeling.” The semantic reading is candid on this point: it principally treats emptiness as therapy for philosophers’ metaphysical attachment. Yet philosophers’ grasping is one small branch within craving’s genealogy. For people who do not practice philosophy, meta-level discipline cannot reach the level where the target resides. With Volume I complete, this diagnosis has an in-house counterpart. The same phenomenon is received there under a different frame: the final formalization of ultimate truth is the invariant across all frameworks, not a higher domain of objects (Part XVIII, Chapter 7), and Appendix A of that part rules on the fourth position that the stance form is primary and the propositional notation its projection. Set against this, “the ultimate truth is that there is no ultimate truth” takes the projection for the original. Reobjectification can be disarmed without paying the price of a further meta-level: placement at the level of rules disarms it just as well, and adds no new distance between meta-view and feeling. The verdict therefore does not change but becomes more precise. The semantic reading saves Madhyamaka’s philosophical dignity at the cost of further diluting its soteriological content. The migration argument repeats unchanged and in purer form at the meta-level.

The defense of no position: the cost of consistency

The second approach defends “no position.” Tillemans and Westerhoff argue in different ways that Nāgārjuna’s declaration “I have no position” can be consistent. A negation can avoid commitment to the contrary thesis; saying “phenomena lack intrinsic nature” need not establish an intrinsically existent universal proposition. Tools from logic and pragmatics can disarm the oldest charge, self-refutation.² This volume likewise acknowledges the successful portion: no position can be defended at the level of statement. One entry must now be added: this defense has an in-house version. The layered definition of no position (Volume I writes “no thesis”) stands in Part XVIII, Chapter 5, dividing the metaphysical thesis from the meta-level statement of lexicon and logic; what Nāgārjuna renounces is the former. The practice of the consequence line and its self-understanding are walked through again in Part XIX, Chapters 1 and 2, from the transcriptions and the critical editions: the half-sentence Candrakīrti finally spoke (rejecting a proposition is not asserting its negation, so there is no thesis to invert) is the classical self-awareness of Lemma 1 of Part XVIII. And Part XIX, Chapter 7 delivers a further ruling: with the public instruments of lexicon as the shared basis, positive force and the consequence discipline are had together, and the dispute between the two camps gathers, under the theorem form, into a dispute over the inventory of resources. Within this book, the disarming of the self-refutation blade is thus no longer a borrowed victory; it is a corollary of its own theorem.

But this volume’s criticism of no thesis never rests upon logical consistency. It concerns normative content (Madhyamaka Part II, Chapter 5): by what means does a system that successfully preserves no position direct practice? Once the blade is removed, the question remains and becomes harder. The more thoroughly no position is preserved, the thinner the practical content the system itself can supply, and the more the whole path must be borrowed—from Abhidharma meditation and curricula of doctrinal classification. This is structurally identical to Candrakīrti’s situation. Systematization is not betrayal but necessary compensation for no position. The cost that Part XIX, Chapter 2 records against Candrakīrti is the closing of the gate of dialogue; the cost recorded here is the borrowing of the path: two entries on the two faces of one coin. A philosophical quietism capable of defense grants a blank authorization in the truth of the path. Victory in consistency does not revise that verdict.

Perspectivism and the Gelug reading: answering with a curriculum

The third approach is represented by Garfield. The conventional world is retained in full, emptiness is a way of seeing convention rather than another reality, emptiness itself is empty at the summit, and ethics and practice are housed within a sound conventional truth. One genealogy of this reading must be recorded explicitly: it is a contemporary rewriting of Gelug interpretation. Garfield studied with Geshe Yeshe Thabkhas and translated Tsongkhapa’s Ocean of Reasoning with Geshe Ngawang Samten, as the note to Madhyamaka Part II, Chapter 3 records.³ Genealogy is not a fault, just as Candrakīrti’s inheritance from Nāgārjuna is not a fault. But its status as the “standard modern reading” must be registered as an extension of a particular tradition of doctrinal judgment rather than a neutral scholarly starting point. The registration now has a theoretical name. The closing of Volume I, Part XIX, Chapter 6 finds that the readings of the two truths converge, under the theorem form, into one reading plus a pedagogical parameter; to take one tradition of doctrinal classification as the default reading is to misfile a pedagogical parameter as a matter of objects and principle (for the two further cases of doctrinal classification, see Parts XXI and XXII). Volume I, Part XII, Chapter 3 reserves a section for the history by which it became the default picture in English-language scholarship.

Its effect on the verdict again has two sides. Full preservation of the conventional is the strongest reply to criticism that Madhyamaka “dismantles too much.” This volume accurately credited Tsongkhapa’s work in the Madhyamaka part. Yet the causal mechanism remains unchanged. The Gelug route’s answer from seeing emptiness to ending craving has always been a curriculum (the paths of persons of three capacities, the sequence of calm-and-insight, and discipline under a teacher), not an output of the argument. How change in perspective rewrites pre-reflective craving remains answered through pedagogical arrangement rather than causal explanation. Verification likewise remains unchanged, with one modern observation added: modern Madhyamaka studies test the validity of arguments, not the occurrence of cessation. This is not a failure of scholars; philosophy is not professionally responsible for liberation. Only one displacement must be prevented: using the philosophical flourishing of modern Madhyamaka studies as evidence that Madhyamaka’s road of liberation works. They are different achievements. The more impressive the transcript from the debating hall, the more important to remember Madhyamaka Part II, Chapter 6: drift from cessation toward debate is exactly the form in which this criterion has become institutionalized as professional philosophy today.

Dialetheism: the paradox filed as a discovery

The fourth approach carries no small voice. Garfield and Priest, reading the Mūlamadhyamakakārikā together, argue that there are true contradictions at the limits of thought (speaking of the unspeakable is the standing example), and that Nāgārjuna is the pioneer of this discovery; the price is revising logic itself, taming explosion and redefining negation.⁴ Volume I, Part XVIII, Chapter 7 holds the formal record of the dialogue, and only its conclusions are registered here. First, the reading is honored as a reasoning opponent: the pressure it sees is real, for statements about a limit do seem forced to cross it. Second, the disagreement lies in the remedy. Volume I resolves by layering (the hierarchy clause plus typed rejection), and the logical machinery stands unmodified; with the presupposition-cancelling reading in hand, no verse forces the acceptance of a true contradiction, and the sorting of closing formulas shows that no family of verses closes on “the contradiction is true.” Third, dialetheism is filed as the standing reserve, a complete alternative plan to be activated should layering fail somewhere; no such place has yet been found.

This volume’s two questions are asked all the same. Mechanism: the use of true contradictions lies wholly at the limits of thought, one level farther out than the semantic reading’s meta-level. The semantic reading at least practices therapy on philosophers’ metaphysical attachment; the limit paradoxes stand farther from the clinic still, with no assignable route to pre-reflective craving—nor have Priest and Garfield ever presented them as a mechanism of liberation. Verification: the testing ground lies inside the logical system, and what is tested is the technical soundness of a non-explosive machine, not the occurrence of cessation. Neither column receives a new entry; the verdict does not change. The value of this route for this volume lies elsewhere: it shows that the textual pressure of Madhyamaka is great enough to make contemporary logicians consider rebuilding their machinery—evidence for the reality of the pressure, not for the passability of the road.

Philology and logical reconstruction: a supplementary record

Beyond analytic approaches, two scholarly currents must be added because this volume benefits from their work. In Japanese scholarship, Kajiyama Yūichi studied the forms of Madhyamaka argument and Buddhist epistemology, while Katsura Shōryū systematized Buddhist logic and argumentation. His new translation of the Mūlamadhyamakakārikā with Siderits joins semantic reading to philological labor.⁵ In Chinese scholarship, Wan Jinchuan’s lectures introduce Madhyamaka thought, while Ye Shaoyong’s comparative edition of Sanskrit, Tibetan, and Chinese recensions of the Mūlamadhyamakakārikā establishes a new standard of textual precision. These contributions belong in the academic record and travel separately from the soteriological verdict. The verdict that Madhyamaka soteriology contains a structural gap does not diminish their value; their clearer establishment of Madhyamaka’s texts and arguments does not fill that gap.

This book’s own form: the same ruler

With the four entries registered, one form remains to be measured with the same ruler: this book’s own. The theorem-shaped reconstruction of Volume I, Part XVIII (the range clause, the hierarchy clause, the Preservation Theorem, and the formal system of Appendix A) is now itself a modern form, of one family with the four above. The obligation to “present the strongest form before criticism” applies reflexively, and this volume’s two questions are asked of it as of the others. The answers are already written in Volume I’s own verdict. Mechanism: Part XVIII, Chapter 9 states that the completed form of the theorem is not a conclusion held but a state in which nothing needs to be held any longer; for the stretch of road from seeing emptiness to the ending of craving, that part closes with a handover of tools (an inventory delivered, not a mechanism explained), and Chapter 8 adds an injunction that the achievements of the argument may not be requisitioned as evidence for faith. Verification: the premise audit contains not one item of scriptural authority, and the proof’s force stands entirely in the public court of lexicon; what can be tested is therefore only the validity of the argument—the same sentence this appendix uttered above in observing modern Madhyamaka studies, and Volume I did not exempt itself from it. The difference between the in-house form and the four above therefore does not lie in being philosophically stronger. It lies in having written “philosophical achievement does not underwrite liberation” into the body of the verdict: the distance is not discovered by audit; it was registered in advance. The division of labor between the two volumes is thereby aligned: Volume I proves and hands over the tools; this volume audits the promises that lie beyond them.

Conclusion

The four approaches each reinforce one of classical Madhyamaka’s weakest points. The semantic reading blocks nihilism and reification; the defense of no position blocks self-refutation; the Gelug reading retrieves conventional truth and ethics; dialetheism rewrites the paradox at the limit from scandal into discovery. After every achievement is acknowledged, this volume’s two questions remain unchanged: mechanism (how does seeing reach pre-reflective craving?) and verification—on what basis and by whom is attainment certified? The verdict does not change, but one qualification is added. Modern forms make Madhyamaka harder to refute philosophically and make it promise less soteriologically. The more perfect its philosophical form, the more respectable, and greater, the distance from the original question “how does suffering cease?” The measure holds for this book’s own form as well, with the nature of the reading reversed: there the distance is not discovered but registered—Volume I’s verdict is the first signatory of this sentence.

References¶