Chapter Seven: The Dispute over Emptiness and Existence—Bhāviveka and Dharmapāla¶

The textual scene. The traditional account of the dispute over emptiness and existence comes with a historical story. Bhāviveka and Dharmapāla lived at the same time and exchanged challenges across north and south; the story says they never met. Each of the two fascicles of the Treatise on the Jewel in the Palm contains a passage beginning, ‘A Yogācāra theorist says as follows.’ Bhāviveka records a long argument from the Yogācāra side and then answers sentence by sentence. This has the same form as the Buddhapālita case in Chapter One of Essay Nineteen: one side’s actual words are preserved in the other’s book. Transcribed by an opponent, they are nevertheless testimony accepted by both sides.¹

The two passages have different subjects. The first fascicle disputes the ‘existence’ of the dependently arisen; the second disputes the ‘real existence’ of suchness. They must be read separately: the consciousness-only side defends different things in the two places—a functional position in the first and a real existent in the second. Chapters Two and Six of this essay each left a case on record. The second fascicle’s passage strikes precisely at those two cases.

One: The first fascicle—the dispute over the dependently arisen. The consciousness-only argument has five steps, transcribed in sequence.

The first is incorporation: ‘If this means that conditioned dharmas arise from multiple conditions and do not exist naturally of themselves, and that they are established as empty in respect of absence of own-nature as arising, then it affirms the Yogācāra theorists’ meaning and accords with right reasoning.’ Your claim that dependently arisen dharmas lack own-nature and are therefore empty agrees exactly with our position. This uses the same approach as Chapter One’s three-period classification: first take over the opponent’s thesis, then fit it into one’s own hierarchy.

The second establishes the residual formula: ‘That of which it is empty really does not exist; this on the basis of which it is empty really does exist. Such emptiness is what the teacher of gods and humans truly taught.’ The third explains the meaning by aligning ‘that’ and ‘this’ with the three natures: ‘The imagined has originally had no own-nature upon the dependently arisen … “That of which it is empty” is the falsely imagined thing, whose own-nature is absent. “This on the basis of which it is empty” is the dependently arisen thing, whose own-nature exists.’

The fourth states the fear directly: ‘If this did not exist, there would be annihilation. Upon what thing would what be said to be empty?’ If what bears emptiness is itself emptied out, emptiness has nowhere to be placed. The fear is immediately extended to provisional dharmas: ‘If this did not exist, provisional dharmas would not exist either. This would become a view of nonexistence. One should neither speak nor live with such a person, who falls into bad destinies and makes others fall too.’ The fifth closes with a warning: if you say the dependently arisen also does not exist, ‘you fall into the deep pit of faults described above.’

Bhāviveka’s replies correspond point by point.

Reading the first fascicle through the theorem. Three points.

First, Bhāviveka’s dilemma is the qualifier in direct operation. Conventional existence? Direction (a) of the preservation theorem already supports the whole of level E2. What you establish, our side has already established: establishing what is already established. Ultimate existence? Produce an example with the same property. At the ultimate level, there is no jointly admitted subject bearing properties that could serve as the example’s bearer, so the inference cannot be established. Chapter One of Essay Nineteen established this qualifier. Here it operates in both attack and defense.

Second, ‘I too accept such own-nature’ should be quoted repeatedly. The two sides have no dispute at level E2; Bhāviveka himself says so. This book need not mediate on their behalf. What the consciousness-only side is actually defending is the functional position of a ‘place to put things.’ Bhāviveka’s answer supplies that position, yet the other side still refuses because it fears that conventional ‘existence’ is not solid enough. Solidity is a question of levels, not existence versus nonexistence; the fear has landed in the wrong place. Another point must be exposed. The consciousness-only argument displayed here is precisely the ‘apart from this, it cannot be found’ form registered in Chapter Three: emptiness needs a place, and if no one on the roster can provide it, the dependently arisen must. That case stated that the form can establish only the necessity of a functional position, not its bearer’s real existence. The first fascicle’s exchange is a live demonstration of this limit. Bhāviveka pays in full everything this form can claim: the functional position, conventional existence, and the basis of provisional dharmas. He pays nothing it cannot claim.

Third, tested by the criterion of resistance to reformulation, the Establishment of Consciousness-Only argues throughout through dependent arising, moments, and dependence on conditions. It offers no resistance to E2 reformulation; π is absent. This book therefore judges that most of the first fascicle’s dispute dissolves in the form of the theorem. The mechanisms the sides jointly defend far outnumber their disagreements.

Two: The second fascicle—the dispute over suchness. This passage has a different weight from the first fascicle and requires separate handling.

The consciousness-only side first concedes one point, then defends another: ‘There is no further ultimate above the ultimate. Suchness itself is the ultimate of dharmas. Thus it is reasonable to say that suchness is empty with respect to the ultimate, but unreasonable to say that suchness is not really existent.’ The concession is ‘suchness may be called empty’; the defended point is ‘the claim that suchness is not really existent does not hold.’ This division between concession and defense has precisely the form of Chapter Two’s recorded case: the perfected as notation can be conceded, while the perfected as a real existent must be defended.

Bhāviveka replies along four routes.

The argumentative route gives two inferences: ‘That suchness is not the true ultimate, because it is an object of apprehension, like visible form and the like’; ‘wisdom apprehending suchness is not true supramundane nonconceptual wisdom, because it has an object and is conditioned, like cognition of worldly objects.’ Whatever is an object cannot be ultimate. Wisdom apprehending suchness both has an object and is conditioned, placing it in the same class as cognition of worldly objects.

The scriptural route cites four passages in sequence: ‘The wisdom-eye sees nothing at all’; ‘here even wisdom does not operate, much less names’; ‘the Tathāgata’s awakening cannot be directly observed’; and ‘here no dharma can be seen or retained in memory.’ All four point to one thing: at the top level there is nothing seen, operated upon, or held.

The reductio route gives a sharp reply: ‘If it is called empty because this is absent above this, shepherds and the like also all know that there is no further clothing or silk upon clothing or silk. They too should be called seers of true reality.’ If ‘no further ultimate above the ultimate’ suffices for emptiness, then shepherds know that there is no silk upon silk, and all shepherds have seen true reality. This reply targets the idle spinning of a diluted word ‘empty.’

The fourth route is the objection concerning the non-Buddhists’ self, the chapter’s heaviest blow. Bhāviveka’s words: ‘If suchness is said to be beyond speech yet really existent, this is simply the non-Buddhists’ self called suchness through a difference of name and conception.’ He immediately sets out the opponent’s predicament in full. The self posited by non-Buddhists, ‘though really existent, all-pervading, permanent, an agent and experiencer, is free from discrimination; because it is not a domain in which language operates and is not an object of discriminating understanding, it is called free from discrimination.’ They too say that speech does not operate there and mind does not realize it, and therefore call it a self. If so, ‘what difference is there?’

Bhāviveka then supplies his own affirmative position: ‘Know that suchness is only the permanent cessation of all discrimination, not a really existent nature, neither separate nor existent.’ The Dharma body of transformation of the basis is ‘by original nature unarisen, by original nature permanent.’²

Reading the second fascicle through the theorem. The blow lands with exceptional precision on this book’s criteria.

First, ‘beyond speech yet really existent’ is the standard form of D2. Inexpressibility grants no exemption. Chapter Three of Essay Eighteen has ruled that statements such as ‘neither one nor different’ remain predications. The presupposition is unchanged; their quality depends entirely on the level where their use places them. Being beyond speech merely moves predication outside verbal expression; it cannot remove the commitment.

Second, the objection concerning the non-Buddhists’ self is formally a classical version of the π test. To describe something as free from all discrimination yet really existent is to hide π outside verbal expression. What the fourth rejection in Appendix A of Essay Eighteen rejects is precisely π concealed beneath rejection. Bhāviveka makes the same move here with the same diagnostic method: look not at what it says, but at what it leaves where speech cannot reach.

Third, the gap in wording between the two passages must be recorded clearly. In the first fascicle, the consciousness-only side defends a functional position, matching the wording of the Establishment of Consciousness-Only. In the second it defends real existence, out of step with that text. When explaining the relation between the perfected nature and the dependent nature, the Establishment of Consciousness-Only uses relational expressions such as “neither different nor non-different” and “neither identical with nor separate from” (Chapter Two).³ This book reads these as wording on the side of notation. Measured against the rhythm of the two-part package established in Chapters Four and Six, the second fascicle’s passage thus has another identity: the only scene in this essay where the second part is absent. A thing is added—suchness really exists—but the declaration that it is not a thing does not arrive. The commentary’s ‘neither different nor nondifferent’ is precisely a way of supplying the second part. This passage, however, explicitly calls ‘suchness is not really existent’ unreasonable, effectively refusing delivery of the second part. The objection concerning the non-Buddhists’ self therefore lands exactly: the π test sounds at precisely the point where the package is split. This chapter accordingly grades the evidence. The first fascicle’s passage agrees with the commentary, has high credibility, and can serve as testimony accepted by both sides. The second is transcribed by an opponent and differs from the commentary’s wording. It is registered only as ‘there was indeed this reading within the consciousness-only tradition,’ not as the school’s settled position. The convenience and risk of transcription appear together in this passage.

Fourth, even on the most conservative reading, the second fascicle’s value is sufficient: the two openings in Chapters Two and Six were not invented by this book; a classical opponent challenged them on the spot. In the sixth century, someone was already pointing to the perfected’s nominalization and the uncontaminated realm’s ‘inconceivable, wholesome, permanent,’ saying this was the non-Buddhists’ self under another name. This book’s entry merely continues that record. As before, it does not decide the verdict on their behalf.

Three: Material from Dharmapāla’s side, for future treatment. Dharmapāla’s Commentary on the Mahāyāna Expanded Hundred Verses is a direct consciousness-only response to the Madhyamaka tradition. Its target, however, is Āryadeva’s Four Hundred Verses, on a different axis from this chapter’s line of confrontation. A future detailed treatment can open a separate section, placing Dharmapāla’s diagnosis of ‘misapprehended emptiness’ beside Bhāviveka’s dilemma. This essay does not develop it.

Four: Verdict and handoff. Three lines.

The first fascicle’s dispute: judged a misidentification of levels, largely dissolving in the form of the theorem. The evidence is Bhāviveka’s explicit ‘I too accept such own-nature’ and the Establishment of Consciousness-Only’s complete lack of resistance to E2 reformulation. Once the point at issue is narrowed, the two sides’ ‘existence’ is not on the same level.

The second fascicle’s dispute: judged a real point of disagreement and the essay’s only instance of direct opposition between the two sides at the same level. All remaining disagreement falls on one item: are suchness and the perfected notation or real existents? In the language of the two-part package, is the second part to be supplied? Chapter Two’s recorded case and Chapter Six’s case of the uncontaminated realm converge here and pass to Chapter Eight for an overall verdict.

The quality of transcription: the first fascicle serves as mutually accepted testimony; the second is registered as one position. This follows the treatment of the Buddhapālita transcription in Chapter One of Essay Nineteen, for the same reason: words preserved by an opponent have evidentiary value, and the version an opponent selects has a bias of selection. The two must be assessed separately.

A closing sentence on rival classifications of the teachings. The three-period account incorporates the teaching of emptiness as the second period’s hidden intent; the two-truth reading incorporates the three natures as conventional mechanisms. The two techniques of incorporation reach a stalemate, while Bhāviveka’s ‘I am not so; how could I mistakenly affirm the Yogācāra theorists’ position?’ rejects incorporation itself. Under the ruling in Chapter Six of Essay Nineteen, the dispute over the direction of incorporation carries no new logical content; it carries routes of instruction. This chapter has already judged the logical part.

References¶