Chapter One: Buddhapālita and Bhāviveka—Reductio or Autonomous Inference?¶

Once Essay Eighteen's proof is complete, its first family dispute in India concerns the legitimacy of authorization. In that essay's notation: can unified authorization ⊢¬π be obtained only through reductio, or can it be positively established by an autonomous inference? The reductio and autonomous-inference camps take opposite sides, while both regard themselves as keeping Nāgārjuna's house in order. This chapter assembles the statements of both parties and renders the dispute in theorem form. Judgment comes in the next chapter.

One fact about the sources must first be made clear. Buddhapālita's commentary on the Middle Treatise has no premodern Chinese translation. His words enter Chinese by only one route: quotation by a critic. To refute Buddhapālita, the Lamp of Wisdom must first quote him. Bhāviveka's blade thus preserves Buddhapālita's own words. Every statement by Buddhapālita below comes from such quotation. Its hostile origin has a further advantage: these words are testimony accepted by both parties.

Buddhapālita's syntax is clear in the quotations. Explaining the alternative of production from another, he says only: “Production by another is likewise untenable. Why? Because it entails the fault that everything arises everywhere.”¹ This means that if production from another holds, anything distinct from the effect is another, so anything could arise from anything: everything arises everywhere. There is no thesis, reason, or example. The whole passage does just one thing: it pushes the opponent's proposition to its own absurd consequence. This is precisely the pure rejection permitted by Lemma 1 of Essay Eighteen: performing only ⊣, without undertaking the obligation of ⊢. Buddhapālita himself, however, did not state that last half-sentence. Its absence gave Bhāviveka a point of attack.

Bhāviveka's criticisms stand beside the quotations, and both became classic objections. The first is failure to supply a demonstration: “You now state no reasons or examples; there is only the assertion of a thesis and the attribution of a fault to another.” Merely recording the opponent's faults, without reasons or examples for one's own position, amounts to having no standing to bring a case before the court of Buddhist logic.¹ The second is the danger of reversal. Buddhapālita says that production from another entails universal arising. Bhāviveka replies that to state the fault in this way is to “reverse what is to be established and what establishes it.” If the reductio is read as establishing something, its thesis and reason can be turned back unchanged by the opponent and used for the opponent's own side.¹ Together the two objections amount to one demand: either offer a positive inference or explain why reductio is exempt from the obligation to do so.

Bhāviveka's own proposal appears in a defining verse in the Treatise on the Jewel in the Palm: “In true reality, conditioned things are empty, because they arise from conditions, like illusions; the unconditioned has no reality, because it does not arise, like flowers in the sky.”² Thesis, reason, and example are all present: a standard autonomous inference. The crucial point lies in the opening two Chinese characters, “true reality” (真性). The treatise explains: “Lest this conflict with what our own position accepts,” “we therefore qualify the thesis with ‘true reality’”; “it is with respect to ultimate truth that we establish the emptiness of conditioned things, not with respect to the conventional.” This qualifier fixes the target squarely on the reading in terms of real nature. Conventional conditioned things such as eyes, commonly known and accepted even by cowherds, are all left untouched. Readers of Essay Eighteen should recognize the procedure. It does the same work as the scope clause, excluding E1 and E2 assertions from the range of application. The means differ: the scope clause draws the boundary by type; the qualification marks each sentence with a modifier. Chapter Six will review their similarities and differences together.

Rendered in theorem form, the dispute fits into one sentence. Bhāviveka requires positive proof of ⊢¬π, testable before a public court; Buddhapālita applies ⊣ throughout without putting forward ⊢, leaving the authorization implicit. The danger of reversal falls precisely within Lemma 1's jurisdiction: ⊣A does not entail ⊢¬A. Once this half-sentence is made explicit, the reductio has no thesis to reverse. What Buddhapālita lacks is not force but this measure of self-understanding. Candrakīrti supplies it. One further quotation deserves to be kept separately. Answering on activity within conditions, Buddhapālita says: “How is there activity within conventional truth? Activity exists because self, other, and the various conditions depend upon one another.”¹ Activity belongs to conventional truth and is established through mutual dependence. This is a classical precursor of direction (a) of the preservation theorem. It also shows that the reductio camp never intended to abolish the everyday level. The two camps share more than they dispute.

The judgment in one sentence: both camps preserve the theorem itself and its aspect of preservation; their disagreement concerns only the route of authorization and the possibility of dialogue. Which proposal stands depends on what remains missing once the reductio approach is complete. Chapter Two opens the hearing.

References¶