Chapter Two: The Three Natures—The School’s Central Semantics¶

Establishing the terms. The three natures are this school’s central semantics. Their source is fascicle two of the Sūtra Unraveling the Deep Intent: ‘First, the imagined characteristic; second, the dependently arisen characteristic; third, the perfected characteristic.’ Each of the three has its definition. The imagined means ‘the distinctions of own-nature provisionally established by the names of all dharmas.’ It arises with speech: names and conventions assign distinctions of own-nature to things, and the layer thus assigned is the imagined. The dependently arisen means ‘the dependently arisen own-nature of all dharmas: because this exists, that exists’—the mechanism of dependent arising itself. The perfected is what is disclosed by the perpetual unfindability of the imagined in the dependently arisen. The analogy for the three is a visual disorder: ‘Like the affliction of impaired vision in the eyes of someone with such a disorder.’ The hairs, flies, and wheels appearing in impaired vision resemble the dependent as it is misapprehended; what a clear eye apprehends resembles the perfected.¹

The analogy must be read precisely. The visual disorder is the pathological root of imagining; the images that appear within it are how the dependent looks when misapprehended; what a clear eye apprehends is the perfected. These are three qualities of the same visual field, not three things: impaired vision with misapprehension, impaired vision without misapprehension, and no impairment. We will use this reading twice below: in the section on neither difference nor nondifference, and in Chapter Seven’s dispute over emptiness and existence.

The compressed version in the Thirty Verses, and who does the imagining. Vasubandhu compresses the same scheme into eight lines: ‘Through this or that imagining, / various things are imagined. / This imagined nature / has no own-nature at all. / The dependent own-nature / is discrimination born of conditions. / The perfected, in that, / is its constant freedom from the former nature.’² In explaining the opening line, the commentary raises a further question: who does the imagining? It presents alternative answers. We first consider the broader position: ‘One view holds: the eight consciousnesses and their mental factors, insofar as they are contaminated, can all imagine, because false discrimination is their own-nature and they all appear as though apprehended and apprehender.’³ On this view, imagining is not the sixth consciousness’s exclusive job; every contaminated consciousness imagines in its own capacity. This position recasts imagining from an occasional misidentification into the normal state of contaminated consciousness. It shares the tendency of manas’s being ‘constantly accompanied by four afflictions’ in Chapter Four: both give error a structure.

The three absences of own-nature: the procedure for incorporating the teaching of emptiness. Once the three natures are established, a further procedure redistributes the second period’s meaning of emptiness among the three compartments. The Thirty Verses: ‘On the basis of these three natures / those three absences of nature are established. / Thus, with a hidden intent, the Buddha said / that all dharmas lack nature. / The first is absence of nature as characteristic; / the next, absence of nature as natural arising; / the last, through being free from the former, / lacks the imagined natures of self and dharmas.’⁴ The Sūtra Unraveling the Deep Intent names them ‘absence of own-nature as characteristic, absence of own-nature as arising, and absence of own-nature as the ultimate,’ and gives the reasons one by one. The imagined lacks own-nature as characteristic because ‘its characteristic is established through provisional names, not established through its own characteristic.’ The dependent lacks own-nature as arising because it is born through the power of conditions, rather than existing naturally of itself. The perfected lacks own-nature as the ultimate because what is disclosed as ultimate is free from the imagined nature.⁵

Three compartments, three kinds of ‘absence,’ each with its own object. The single sentence ‘all dharmas lack own-nature’ is split into three receipts, each settling its own account. ‘Hidden intent’ names this procedure of division. The crucial point is the second compartment: the dependent nature’s lack of own-nature negates natural arising, not dependent arising itself. This half-sentence is the pivot of the whole essay and gives rise to the dispute in Chapter Seven.

The perfected and the dependent: neither different nor nondifferent. The commentary fixes the third compartment’s position with a dilemma. The Thirty Verses: ‘Thus this and the dependent / are neither different nor nondifferent, / like the nature of impermanence and the like; / without seeing this, one does not see that.’² The commentary spells out the consequences on both sides: ‘This perfected nature and that dependently arisen nature are neither different nor nondifferent. If different, suchness would not be its real nature; if nondifferent, this nature would be impermanent.’⁶ If they are different, suchness is not the dependent’s real nature, and the perfected becomes another thing. If they are nondifferent, suchness becomes impermanent along with the dependent. Neither side can be accepted; hence the formulation “neither different nor nondifferent.”

The dilemma has the same form as Essay Eighteen’s preservation theorem: to keep something from being external to the everyday level, while preventing it from being swept along by that level’s properties, reject two extreme readings and leave a relational term. The example makes the same point. Impermanence stands in just this relation to formations: it is not something outside them, yet this does not make it identical to them. By analogy, the perfected is placed as a ‘being thus’ of the dependently arisen, rather than another layer added above it. This is the strongest internal evidence this chapter can offer for reading the perfected as a notation.

The three natures and the two truths: this chapter’s most important connection. A question and answer in fascicle eight of the commentary links the three natures directly to the two truths: ‘How do these three comprise those two truths? Know that the conventional has all three; the ultimate is solely the perfected nature. The conventional has three kinds: first, the provisionally established conventional; second, the conventional of formations; third, the explicitly disclosed conventional. Know that these correspond, in order, to the three natures.’⁷

Two things emerge from the table. First, the three natures do not stand alongside the two truths; they divide the conventional side. The conventional level is divided into three compartments, from sheer misidentification, through the mechanism of dependent arising, to what is explicitly disclosed upon dependent arising. Second, the perfected spans both columns: in the conventional column it is called the explicitly disclosed conventional; in the ultimate column it is the whole of the ultimate. This span is precisely the opening Chapter Eight will register. The same thing occupies a compartment in each column, and a slight slant in the reading turns it into two things.

The value of this connection lies in comparison. Chapter Six of Essay Nineteen reduces the various readings of the two truths to one reading plus a parameter of instruction. Its three criteria are: no added ontology, preservation of the preservation theorem, and expressible level clauses. The three-nature reading passes them one by one through this table. No ontology is added: all three compartments divide the conventional level. The preservation theorem has a place: the conventional of formations is the affirmative inventory of level E2. The level clauses are expressible: ‘the ultimate is solely the perfected nature’ draws the boundary between levels. The three natures can therefore be translated into Essay Eighteen’s notation without conflicting with the ruling in Chapter Six of Essay Nineteen. The dispute between the sides concerns how to read the perfected compartment, not the framework.

The emptiness verse of the Discrimination between the Middle and the Extremes: a prototype of residual emptiness. The school’s clearest affirmative statement of emptiness occurs in the Discrimination between the Middle and the Extremes: ‘False discrimination exists; / in it the two are wholly absent. / In this there is only emptiness; / in that, this too exists.’ The commentary is exceptionally clear: ‘“False discrimination exists” means there is discrimination into apprehended and apprehender. “In it the two are wholly absent” means that in this very false discrimination the two natures of apprehended and apprehender are forever absent. “In this there is only emptiness” means that in false discrimination there is only the emptiness free from apprehended and apprehender. “In that, this too exists” means that in that emptiness of the two there is also only this false discrimination.’⁸ The next verse brings the account to the middle way: ‘Thus all dharmas are said / to be neither empty nor nonempty; / because of existence, absence, and existence, / this accords with the middle way.’⁸

This is the prototype of residual emptiness: what is emptied out is the twofold apprehension; what remains is false discrimination and the emptiness of the twofold apprehension, each present in the other. ‘Whatever does not exist in this, in respect of that one contemplates it as empty; because the remainder is not nonexistent, one knows it as existent, as it is.’⁸ Emptiness and existence each have their place; neither empties out anything extra.

The key is that this formula makes ‘emptiness’ a binary operation: on what basis, empty of what, leaving what? All three must have a place before emptiness can be spoken of. The two sentences Bhāviveka criticizes in Chapter Seven—‘That of which it is empty really does not exist; this on the basis of which it is empty really does exist’—are another wording of this formula. Its origin is established here so that Chapter Seven’s direct treatment need not begin the account again. The actual point of disagreement can therefore be marked now. Madhyamaka asks, ‘Why is the remainder exempt from further emptying?’ Consciousness-only replies, ‘Empty that too and there will be nowhere to place anything; this falls into annihilation.’ Both sides agree there must be a place; they dispute whether that place itself must come within the scope of application.

Two cases to be passed to Chapters Seven and Eight. First, the nominalization of the perfected. A negative fact—the absence of twofold apprehension—is made into a noun, a ‘nature,’ and even supplied with the word ‘real’ (实). If this is notation, the perfected is a consciousness-only rendering of the preservation theorem. Both the section on neither difference nor nondifference and the table connecting the three natures with the two truths support this reading. If it is an ontological commitment, it must be tried as a D3-type proposition. The compartment ‘the ultimate is solely the perfected nature’ is where it is most readily elevated. Both readings have support within the text.

Second, the two readings of the dependent. One distinguishes defiled and pure forms of the dependent nature; another asks whether the dependent has own-nature ultimately. The former treats the dependent as the same continuum, capable of both defilement and purity, and goes with transformation of the basis in Chapter Six. The latter raises the dependent’s ‘existence’ to the ultimate level, conflicting with the wording in the section on absence of own-nature as arising. The latter is the issue in the dispute over emptiness and existence, to be heard in Chapter Seven.

References¶