The Probletic od Whole-Part

·期刊原文
THE PROBLEMATIC OF WHOLE-PART AND THE HORIZON OF THE ENLIGHTENED IN HUAYANa BUDDHISM

By Tao Jiang

Journal of Chinese Philosophy v. 28 n.4 (December 2001) pp.457-475

Copyright 2001 by Journal of Chinese Philosophy

--------------------------------------------------------------------------------

P.457

The issue of the whole-part relationship has been a contentious subject in Indian philosophical discourse since its early stages. Generally speaking, there are two leading positions concerning the nature of the whole, from which the issue of the whole-part relationship stems. First is the reductionist position, which contends that the whole is nothing more than the parts put in a certain order; hence, the part is more fundamental than the whole, since the whole can be reduced to the parts that constitute it. Second, there is the essentialist position, advocating that the essence of the whole cannot be simply reduced to the parts, since we do not find the whole in any of the parts; hence, the whole is more primary than the part. Most Indian Buddhists subscribe to the first position, whereas Hindu realists adhere to the second.

When Buddhism spread to China, however, the issue took an interesting turn in the hands of Chinese Buddhists. In this article, I will examine this turn by investigating two representative Buddhist positions on the issue of whole and part. One is found in the Abhidharma literature, the early Buddhist philosophical treatises, wherein the core Buddhist notion of anātman (wu wob), no-self, is explained by appealing to this relationship. As we will see, the whole-part relationship discussed there is distinctively reductionist. Another major deliberation on the whole-part issue is found in the literature produced by the Chinese Huayan Buddhist school, wherein the reductionist approach is nowhere to be seen. In fact, the Huayan philosophers tried as hard as they could to fend off any reductionist mode of thinking regarding the whole-pan relationship. Be that as it may, their position in no way corresponds to the essentialist stance. This article investigates the unique Huayan theory of part and whole, thereby offering a possible way out of the reductionism-essentialism dilemma.

the whole-part relationship in the formulation of Anātman

The fact that the whole-part issue is prominent in the Buddhist discourse can be discerned in the exposition of anātman, the doctrine of

--------------------------------------------------------------------------------

P.458

no-self. Undoubtedly, the notion of anātman has always been, or at least been regarded as, one of the core doctrines preached by the Buddha, and hence it has almost become a "trademark" that distinguishes Buddhism from other religious traditions. According to the Abhidharma Buddhists, a person is a unity of five skandas, or aggregates (wu yunc): form/body (sed), sensation (shoue), perception (xiangf), predispositions (xingg), and consciousness (shi h)', there is nothing called the self, or soul, over and above these five aggregates. The doctrine of anātman was formulated as a rejection of the Brahmanic notion of ātman, the self. It was at the center of the debate between the Buddhists and the Brahmanic philosophers throughout the ages in India.

Let us examine how the notion of anātman is taught in the Abhidharma literature. One of the most famous elaborations can be found in the Milindapañha, in which the Buddhist sage Nāgasena instructs a suspicious Greek king, Milinda, on the way to understand anātman.1 Nāgasena skillfully uses the example of a chariot to show the king that the word "chariot" is simply an expression for the axle, wheels, chariot-body, pole and other parts, placed in a certain relation to each other. None of these components by itself is the chariot; over and above these components there is no chariot.

Let us take a closer look at Nāgasena's example. When the word "chariot" is used, it supposedly points to an entity called a chariot. After careful examination, however, what we find is nothing more than the wheel, the axle, the pole, and so on, put in certain order. Put differently, there is some element of a chariot that is irreducible to its components. On the other hand, it is obvious that none of its components can by itself be regarded as a chariot, nor can simply any cluster of pieces of wood. This example highlights the tension between the reductionist position and the essentialist position. Nāgasena's explanation is a typical reductionist position; He repudiates the existence of the chariot by reducing the chariot to its parts. It is helpful to remember that there is nothing in the world called "chariothood" that exists by itself, over and above the components that constitute the chariot. Through the analysis of the chariot, Nāgasena shatters our commitment to the essentialist position; that is to say, he effectively drives home the Buddhist point that the chariot does not possess a self-nature. However, Nāgasena's example is hopelessly unsatisfactory in assisting us in understanding the reason why we call the cluster of those pieces of wood a chariot. To put it another way, even if our commonsense understanding of the chariot is flawed for its essentialism, to simply do away with such a commitment does not help to explain the fact that none of its components can by itself be equated with the chariot. Given that a chariot is irreducible to its components and that there is nothing over and above the compo-

--------------------------------------------------------------------------------

P.459

nents, a natural question arises: What is it that makes a chariot a chariot, different from the sum of its components?

Intuitively, we can differentiate two kinds of wholes―strong and weak.2 A strong whole is an entity whose parts cannot exist independently from the whole, whereas a weak whole is the contrary. In other words, a strong whole consists of parts that are dependent upon the whole, whereas a weak whole is made of parts that are independent of the whole.3 Clearly, a chariot is a case of the strong whole, for none of its parts enjoys an independent status; a mere cluster of pieces of wood would be a case of the weak one. Unfortunately, Nāgasena treats the chariot as a weak whole. In fact, a critical oversight in Nāgasena's argument is this very failure to make a distinction between a strong whole and a weak one. That is, Nāgasena's contention implies that a piece of wood can exist independently and meaningfully as a wheel, a pole, an axle and so on, and if we follow that line of thought, a chariot will simply be equated with a cluster of pieces of wood.

A mere cluster of pieces of wood is, however, not a chariot. What makes a chariot different is the intrinsic mutuality―which a mere cluster lacks―between the chariot and its parts. Put differently, a part in the chariot cannot, in terms of its functionality, exist independently from the whole chariot; at the same time, the chariot is dependent upon its parts, in that the latter are constitutive of the former. Hence, there is a mutual dependency between the part and the whole in a strong whole. The case is completely different, however, when we consider a mere cluster of pieces of wood, for example, a decomposed chariot, as the whole. In this scenario, the whole, namely the cluster, is dependent upon the parts, the pieces of wood (not as axle, pole, and so forth), in the sense that the whole will not be that whole without any of its parts; on the other hand, any one piece does not have an ontological dependence on the whole, the mere cluster. That is, a piece of wood qua piece of wood has an ontological independence from the cluster of which it is a part, and thus it can meaningfully exist as a piece of wood independently of the cluster. Therefore, in a weak whole the dependency works only one way, namely, that the whole depends upon its members and not the other way around. This means that in both strong and weak wholes, the whole is dependent upon its parts, but only in a strong whole is the part dependent upon the whole due to the intrinsic connectedness among the parts belonging to the whole. Furthermore, it is an equally important fact that within a chariot, each part stands in an inherently integral relationship with the others, but that in a mere cluster of pieces of wood, such an intrinsic connection does not exist.

Nāgasena's eagerness to convince the king of the Buddhist notion of anātman blinds his eyes to the distinction between two issues; the non-

--------------------------------------------------------------------------------

P.460

existence of a permanent self-nature (chariothood) and the irreducibility of the whole to its parts in the case of a strong whole (a chariot). That is to say that a chariot's lack of self-nature cannot simply be explained by a reductionist theory of the whole-part relationship, which is applicable only to the case of a weak whole. Nāgasena tries to explain the doctrine of anātman by appealing to the whole-part relationship. But in light of his explication, it is clear that he thinks there is only one kind of whole-part relationship, in which all wholes can be reduced to their parts. This means that for Nāgasena, there is only one kind of whole in the world, namely the weak whole.

This whole-part issue may be illustrated by these three concepts: sum, unity, and whole. The sum of the chariot refers to the sum of its components; the unity designates the relatedness among the components in their constitution of the chariot; and the whole should be understood as encompassing both the sum and the unity. Put simply, sum is used when we merely have components in mind, and unity is used as far as their relatedness is concerned, while both are subsumed under the whole. Therefore, when the Buddhists teach the non-existence of self-nature, or anātman, what they have in mind is that the sum of the self can be decomposed into its parts, like a chariot that is made up of axle, wheels, poles, and the like. In sticking to such a position, however, many Buddhists remain―or at least appear to be―insensitive to the unity of the parts that is essentially irreducible to the components. In other words, many Buddhists mix up the sum with the whole by ignoring the unity; although the sum can be dissected into its components, the whole is irreducible to the parts, due to the element of unity inherent in it. Not all Buddhists subscribe to the reductionist formulation of the doctrine of anātman vis-à-vis the whole-part relationship, however, and in the following we will examine another important view within the Buddhist tradition on the whole-part issue.4

whole-part versus Li j /shi k in huayan buddhism

When Buddhism was introduced to China, the process of sinicization resulted in some significant reformulations of the traditional Buddhist doctrines. Here we explore the arguments made by the Huayan school on the issue of the whole-part relationship in the context of the li/shi relationship, which is central to the school. Huayan Buddhism was a major Chinese Buddhist school that flourished especially in the Tang dynasty, and is widely deemed to be the climax of Chinese Buddhist philosophy because it provided a philosophical foundation for Chanl Buddhism.

--------------------------------------------------------------------------------

P.461

As a vital part of the Chinese Mahāyāna Buddhist (da cheng fo jiaom) movement, Huayan Buddhism distances itself from the early Buddhist formulation while fully embracing the Mahāyāna teachings. In the case of the whole-part relationship, it shows an unambiguous recognition of the distinction between the whole and the sum, and it has a unique way of acknowledging the whole-part relationship; it teaches that the part and the whole interpenetrate each other without hindrance (wu ain). This position reveals their commitment to the orthodox Mādhyamikan (zhong guano)stance5 that the part is neither identical to nor different from the whole; but by putting a positive spin on the issue, the Huayan Buddhists successfully avoid the skeptical tendency of Mādhyamika. This takes us right into the heart of Huayan philosophy.

To better appreciate how the issue of the whole-part relationship is dealt with in Huayan philosophy, not simply as a reformulation of the Mādhyamika doctrine but as a result of ingenious insights and creativity, let us examine more closely its core teaching. Our discussion will be based on "The Mysterious Mirror of Huayan Dharmadhātu'" (hua yan fa jie xuan jingq), by the first Huayan patriarch, Du Shunr and Cheng Guan'ss commentary.6

Du Shun's famous treatise, "The Mysterious Mirror of Huayan Dharmadhātu which is at the center of the Huayan philosophical discourse, puts forward three contemplations, namely the contemplation of the true emptiness, the contemplation of nonobstruction between li and shi, and the contemplation of nonobstruction between shi and shi. These correspond respectively to the three dharmadhātus (fa jiet), the reality in the eyes of the Buddhas, characterized as "radiance, luminosity with no shadows."7 They are the dharmadhātus of li, of nonobstruction between li and shi, and of nonobstruction between shi and shi. Because the latter two are characteristically Huayan, our attention will be devoted primarily to them, although some arguments on the first contemplation will also be brought in where appropriate. This section will address non-obstruction between li and shi―the second dharmadhātus―and the next section will focus on that between shi and shi―the third dharmadhātus.

At the very outset, we are confronted with two fundamental concepts in Huayan Buddhism, li and shi. Shi is usually understood as things or phenomena in the world, but the interpretation of li appears to be more troublesome. Li8 is a Chinese term employed by Du Shun to express the Buddhist notion of śūnyatā (kongv), emptiness. This is evidenced in Du Shun's explication of the dharmadhātu of li, the first dharmadhātu, as the contemplation of true emptiness, śūnyatā. As such, li in Huayan thought does not differ from Nāgārjuna's (Long Shuw) conception of śūnyatā. If we look deeper, however, some traces of distinction between the two are unveiled. The notion of śūnyatā as conceived in Nāgārjuna's

--------------------------------------------------------------------------------

P.462 TAO JIANG

celebrated Mūlamadhyamakakārikā is a purely logical concept inferred from the acute observation of the impermanence and dependent origination of the phenomena in the world.9 In other words, śūnyatā is the result of inference, not the content of perception.10 On the other hand, li as presented by Du Shun is the content of perception,11 albeit of a special kind, which we will address later in the essay. This means that the Huayan Buddhists regard emptiness as that which can be directly perceived, not simply inferred from inspecting the phenomenal world.

In light of this observation, some interpretations of li can be readily excluded. Quite often li is translated as principle, universal or noumenon, but none of these can stand close scrutiny. In the first case, li cannot be principle, because we cannot perceive a principle but only infer it. In the case of the second translation, li as the universal and shi as the particular, we will be forced to take either the realist or the nominalist position regarding the universal.12 A realist regards the universal as something existing independently of the particular; this is the position most Hindu philosophical schools advocate, but it is resolutely rejected by the Buddhists. If one is a nominalist, as some Buddhists are when it comes to the issue of the universal, one would view li as merely the result of mental constructions without correspondence to reality (whatever that is). As such, it would not be that which is directly perceived but rather inferred, similar to // as principle, and this is clearly not the li; Huayan teaches. Lastly, the word noumenon smacks too much of Kantianism, in which noumenon is not a possible object of perception; hence, it is an inappropriate rendering of li.

What, then, does Du Shun mean by li ? Since in the elaboration of the first dharmadhātu, Du Shun resolutely refutes the false view of searching for li; outside of shi, the following discussion of the relationship between li and shi will help us correctly understand li.

The second dharmadhātu is the contemplation of nonobstruction between li and shi, wherein ten possible relationships are spelled out: li pervades shi, shi pervades li; the making of shi is according to li, shi unveils li ; li removes shi, shi conceals li ; true li is identical to shi, shi is identical to li ; true li is different from shi, shi is different from li. These ten can be divided into five pairs: li and shi pervade, complete, destroy, identify with and deny each other. Among the five pairs, the first one is the most important, because "it is the general characteristic―the subsequent eight characterizations [formation and disintegration and so on] come to be based on this mutual pervasion."13 Therefore, an investigation of the first pair will be most revealing in arriving at a right understanding of this relationship:

First is the aspect of li pervading shi. This means that the pervading li is by nature undivided whereas the pervaded shi is differentiated in division

--------------------------------------------------------------------------------

P.463

and position. In each individual shi, li pervades in its totality―it is not a part [ofli] pervading. Why? Because real li cannot be divided. Therefore each individual particle contains the boundless real li completely.14

This passage points out a crucial distinction in the characteristics of li and shi: the indivisibility of li and the dividedness of shi. For the indivisible to pervade the divided, shi has to contain li in its entirety, otherwise li would become divided, violating its indivisible nature. On the other hand, the indivisibility of li in no way suggests that it is infinitely small like a basic particle of some sort; on the contrary, it is boundless, whereas shi is limited by its own boundary. Hence, we are confronted with a puzzle: How can the limitless be contained in the limited? We will come back to this question.

Second is the aspect of shi pervading li. This means that the pervading shi has boundaries whereas the pervaded li must have no boundaries. These bounded shi must be wholly the same, not partly the same, as the unbounded li. Why? Because shi, having no substance, is after all like li. Therefore a single atom pervades the universe without dissolving. As does one atom, so do all things. Ponder this.15

This passage shifts the perspective to shi in explaining the li/shi relationship, and it teaches that shi pervades li. If the previous passage is bewildering enough, this one appears to be even more enigmatic. How is it possible for the limited shi to pervade the boundless li ? The rationale given is that since shi is empty of its self-nature, or "having no substance," it is just "like li." If this is the case, since li in turn is boundless, shi should also be limitless. Therefore shi can pervade li.

Du Shun himself is fully aware of the bizarre nature of these arguments. He resorts to some metaphorical devices in trying to convey the gist of the Huayan philosophy. For example, he compares li to the whole ocean and shi to one small wave and argues that although a wave pervades the whole ocean, it does not make the wave bigger than it is. Moreover, although the whole ocean is contained in a small wave, it does not make the ocean any smaller than it is.16 The same is the case with all the other waves.17 Thus, Du Shun translates the li/shi relationship into the whole-part relationship. Now, the task here is to make sense of apparently contradictory statements: If the whole ocean (li) in its entirety is present in one wave (shi), why is it not small? And if it is not small like a wave, how can it be said that its whole is in one wave? On the other hand, if one wave extends throughout the ocean, why is it not large? If it is not vast like the ocean, how can it pervade the ocean? Put in terms of the part-whole relationship, if the whole in its entirety is enveloped in one part, why is the whole not as small as the part, or the part not as large as the whole? If a part pervades the whole, why is it not as large as the whole; otherwise, how is such pervasion possible?

--------------------------------------------------------------------------------

P.464

Du Shun's answer to these questions is both tactful and interesting. The reason that we are trapped in the above perplexities is that we mistakenly assume that li and shi are identical to each other when we learn of their nonobstructed interpenetrations. To state that they interpenetrate each other does not amount to saying that the two are identical. He says, "Li and shi are relative to each other; they are neither one nor different. Therefore they can completely contain each other without destroying their original status."18 This is conspicuously reminiscent of Nāgārjuna's contention. Du Shun, however, attempts to reinterpret the Mādhyamika neither/nor argument in terms of the whole-part relationship:

First, in the relativity of li to shi there are four points: (1) because real li is not different from shi, the totality of real li is one shi; (2) because real li is not one with shi, li is eternally boundless; (3) because non-identity is non-difference, the boundless li is totally in one atom; (4) because non-difference is non-identity, li of one atom has no bound or limit.19

Here, the metaphor is changed by comparing shi to an atom. This passage is explained from the perspective of li and the one that follows it from the perspective of shi. Since the two passages are in the same vein of argument, I will skip the latter one and focus on the quoted passage. Du Shun's illustration of the position of neither identity nor distinction is apparently done with an eye to bringing the limitless and the limited together, and his method of explanation is unique. He first clarifies what it means to be nonidentical and nondistinct, respectively; then he equates the two from both positions (nonidentity is nondistinction and vice versa), and gives the meanings of such equations. More concretely, to say that li is not distinct from shi suggests a mutuality and inseparability between the two, in that li can be grasped in light of shi. However, to claim that li is not identical to shi is to grant li some other sense of reality, irreducible to shi. In asserting that nonidentity is nondistinction, Du Shun makes the point that the irreducibility of li to shi is the very mutuality and inseparability between the two; hence, the boundless li that is nonidentical to the limited shi can be enveloped in the limited shi that is not distinct from the boundless li. In asserting that nondistinction is nonidentity, the master illuminates the idea that the mutuality and inseparability between li and shi is the very irreducibility of li to shi; hence, the li of the shi that is not distinct from the shi is the boundless li itself that is not identical to shi. The overall picture here is the theme of the second contemplation: the nonobstructed interpenetration of li and shi.

Some other issues remain unresolved, however. If the boundless li is indeed wholly in one shi, is there li in other shi or not? If yes, it means that there is li outside that shi and, hence, it is not in its entirety in that shi; if no, li becomes strangled in one shi and is thus limited, which leads to the conclusion that it does not wholly pervade all things.

--------------------------------------------------------------------------------

P.465

In responding to these questions, Du Shun tries to nullify our commitment to the world view wherein the boundaries of all things are clearly demarcated. In our commonsense understanding of the world, everything has an inside that is clearly marked off from the outside. From Huayan's perspective, however, such a world view is the result of our ignorance of reality, wherein the boundaries between the interior and the exterior of things are blurred. As a result, one is in another, another in one, everything in everything else, everything else in everything. When the inside and the outside become irrelevant, the above contradiction is naturally dissolved.20 This will be the subject of the next section.

Be that as it may, an interesting issue turns up: Does the distinction discussed earlier between a strong whole and a weak one still hold up within the Huayan scheme? Obviously, all the distinctions between the two ultimately lie in our intuition of what a thing is. Normally we can call a chariot―not a cluster of pieces of wood―a thing (the cluster of wood is a cluster of things). Hence, when we talk about a "thing," it refers to a strong whole―a chariot, for example, "Thing" does not apply to a cluster of pieces of wood, which is a weak whole, since a piece of wood does not exist independently as an axle, a pole, and so on, although this is not to deprive it of its status as a mere piece of wood. At the bottom of our intuitive understanding of a "thing" is that it is a strong whole, having a clear boundary that distinguishes it from the other things, whereas a weak whole does not have a clearly marked boundary.21 In light of the Huayan insight, boundaries that separate one object from another fade away, and so the distinction between a strong whole and a weak one, based on commonsense intuition, simply vanishes.

Many may be expected to dismiss the Huayan position as trivializing the part-whole relationship by annulling the distinction between a strong whole and a weak one. Many may raise objections such as the following: What is achieved in blurring the boundaries between things? What is accomplished in doing away with the difference between the strong and the weak wholes?

To respond to those challenges, we have to reexamine Huayan's theory of li, the whole, that is, what the whole is in Du Shun's system. In nullifying the distinction between a strong whole and a weak one, Du Shun apparently rejects our common understanding of the whole-part relationship. Close scrutiny of his theory of li will reveal that he actually maintains a radically different notion of the whole, li.

Ordinarily, any theory regarding the whole-part relationship is grounded on the observation that the part is constitutive of the whole. Our earlier discussion of Du Shun's teaching has suggested to us that in Huayan Buddhism, however, such a constitutive relationship between shi qua part and li qua whole does not exist. More concretely, a theory

--------------------------------------------------------------------------------

P.466

regarding part-whole relationship is normally premised upon the seemingly self-evident fact that a part is a part of a whole and that a whole is a whole of parts. In the case of the Huayan doctrine, however, shi does not constitute li, neither does li constitute shi, but rather the two interpenetrate each other without hindrance. In other words, li is not the entirety of shi, otherwise, li becomes divisible, which would violate its indivisible nature. Hence, it is not a constitutive relationship but rather that of interpenetration.

Given this, another question emerges; If the relationship between li and shi is that of interpenetration, should li ; be something other than shi which interpenetrates shi, since interpenetration is interpenetration between two things? Not so, Du Shun resolutely responds, because li is śūnyatā itself, or emptiness, as he spells out in the first Dharmadhātu. This means precisely that there is nothing out there that interpenetrates shi. This apparently rules out any substantive interpretation of li, or the whole. That is, li qua whole is not a thing at all. What could it be? I would like to propose that li can be interpreted as activity itself; that is, li is an activity without an acting agent,22 whereas shi is a thing qua being.

To help us grapple with this interpretation, let us put shi/li in syntactical terms: The relationship between shi and li seems to correlate with the subject-verb relationship. Li qua activity appears to be expressible as the verb and shi qua being, as the subject. Therefore, the part-whole relationship between shi and li is now transformed into a subject-verb relationship. As a result, the relationship between shi and li becomes a predicative one,23 not the constitutive one discussed above. The predication of shi by li leads to the conclusion that shi is predicated by śūnyatā, thus shi is empty in its nature. This is the orthodox position in Mahāyāna Buddhism. Accordingly, li should be understood as the self-negating activity of the world. Through this activity, the world is constantly regenerated.24

Within the structure of subject and verb, normally the subject is considered to be more primary than the verb, since the verb is understood to be subsumed under or "owned" by the subject, in the sense that it is the activity of that subject, but not vice versa. Such a prioritization of the subject seems to coincide with the way we experience the world, or at least what we believe to be the case; that is, the world as we experience it consists of things that are in certain states or engaged in certain activities. However, the world thus experienced and understood is one made up of things isolated from each other―not in the sense that there is no connection among them but that the connectedness is secondary or derivative of the being of individual objects. In other words, in the world of the subject, there are first "naked" things, and then they are somehow related to other things. A world thus conceived can only make concep-

--------------------------------------------------------------------------------

P.467

tual sense without actual experiential correlates. In the actual world, there is never a moment that a subject can be separated from its verb(s). We always experience a unity of subject and verb, never a naked being.

The observation of the inseparability between subject and verb corresponds nicely with our earlier discussion on the relationship between shi and li. That is, in the case of the shi/li relationship, there is no li that is apart from shi and no shi apart from li ; in the case of the subject-verb relationship, there is no subject that is apart from verb and no verb apart from subject.

To interpret li as a verb is to successfully avoid the dilemma between the reductionist and the essentialist positions on the nature of the whole. What is common to the reductionist and the essentialist positions on the natures of part and whole is that both embrace the being-centered understanding of the part and the whole. That is, both the reductionist and the essentialist regard part and whole as beings. The only difference between them is that the reductionist accepts the constitutive theory of the part-whole relationship, whereas the essentialist separates parts from the whole. If our perspective is now shifted to the activity itself, the verb instead of the subject, the being-centered nature of both part and whole would become secondary. With the attention on the verb, the constitutive relationship between part and whole is gone, too, since it is rather counterintuitive that the verb by itself can have components. Therefore, from the perspective of the verb, the reductionist theory, which reduces the whole to the parts, will not arise at all. On the other hand, Huayan philosophers would reject the essentialist theory, since essentialism would presuppose a substantive understanding of the whole, which is not applicable if the whole is the activity, the verb.

In a word, the whole in Huayan Buddhism is not a substantive whole but a predicative one; hence, it is indivisible, boundless, inseparable from shi while also irreducible to it. Li may be qualified as the ultimate verb, the absolute self-negating activity of the cosmos; thus it is nothing other than śūnyatā itself. This is how Huayan reinterprets śūnyatā as li―by making it function as a verb.

However, as we shall see in the next section, the Huayan endeavor to reinterpret śūnyatā as li is not simply syntactical in nature, but rather points to another dimension of possible human experience, unshackled by the ordinary domain. As we noted earlier, li is not the result of inference but is perceivable, according to Huayan Buddhism. After all, Du Shun's whole presentation of the three dharmadhātus is premised upon his direct experience of these dharmadhātus', his experience originates from his meditative practice, known as "the Ocean-Seal Samādhi" (hai yin san mei z) in Huayan. By appealing to a syntactical device, we are in a better position to explore what might be involved in that experience,

--------------------------------------------------------------------------------

P.468

and to that extent the syntactical explanation can do us a great service. Shi is predicated by li, śūnyatā , or, to put it in the Huayan language, there is a mutual interpenetration between // and shi. This would lead to the consequence that one ski should interpenetrate other shi, due to the fact that there is nothing beneath shi that obstructs one from interpenetrating the others. This is a good place to turn to the third dharmadhātu, the dharmadhātu of nonobstruction between shi and shi.

nonobstructed interpenetration BETWEEN shi AND shi

The third dharmadhātu in Du Shun's scheme is where shi and shi interpenetrate each other without hindrance. The formulation of this dharmadhātu is characteristically a Huayan seal. Here the element of li recedes to the background and shi takes center stage. There are ten aspects of interpenetration between shi and shi: li resembling shi; shi resembling li ; shi containing li and shi without obstructions; noninterference between universality and particularity; noninterference between extension and restriction; noninterference between pervading and containing; noninterference between subsuming and entering; communion without hindrance; mutual inherence without interference; universal merging without obstruction.

According to Cheng Guan's commentary, the first three aspects of this contemplation are at the core of the third dharmadhātu.25 Hence, we will concentrate on them. Among the three aspects, the first two, namely the mutual conformity of li and shi, "are the overall meaning and can produce the following eight aspects."26

Li that conforms to shi is manifested as shi, limited as shi, differentiated as shi, greatness and smallness, unity and multiplicity as shi, and so on. In the latter aspect of shi conforming to li, shi is not just pervading as li, it also means that as li, shi is signless, unhindered, neither inside nor outside, and so on.27

Clearly, the aspects of the mutual conformity between li and shi are a reiteration of the second dharmadhātu dealt with earlier. Cheng Guan offers a justification as to why they are placed at the beginning of the third dharmadhātu:

These two still include nonobstruction of li and shi, but because there are these two, there can be the meaning of nonobstruction of shi, so they are included in the realm of shi.28

Cheng Guan deems the mutual conformity between shi and li as belonging to both the second and the third dharmadhātu. As such, it builds a bridge between the two, which means that the nonobstruction between shi and shi grows out of the nonobstruction between li and shi.

--------------------------------------------------------------------------------

P.469

Since this dharmadhātu points to the relationship between shi and shi, what is at stake here is apparently the understanding of shi. How should we understand shi in the context of Huayan philosophy? As our foregoing discussion of the li/shi relationship has made clear, any effort to interpret shi will be confronted with the following dilemma: If shi were simply things or phenomena as interpreted previously, it would be something that is in contradistinction to li, a view rejected by Huayan; if shi were to be identified with li, shi would lose its original status as shi, which would mean there would only be the world of li, and the interpenetration between the two would be meaningless. To use Cheng Guan's words, "If we see only in terms of shi, then they obstruct one another; if we see only in terms of li, there is nothing which can mutually obstruct."29 That is, if we see only in terms of shi, nonobstruction is an impossibility. On the other hand, if we see only in terms of li, nonobstruction does not arise in the first place, since to begin with, there is no starting point for obstruction.

The formulation of the third dharmadhātu sets out to resolve the dilemma of the seeming incompatibility between the nonobstructed interpenetration and the retention of the original status of both li and shi, or between the nonidentity and the nondistinction of li and shi. Since the theorization of the dharmadhātu in Huayan is based upon that which is perceived, not inferred, a legitimate question arises: In what state can one perceive the impossible, namely that li and shi both interpenetrate each other without obstruction and yet keep their original status, and that the two are neither identical with nor distinct from each other.

This state is not an ordinary kind of perception, but rather a special kind of meditative experience. This meditative state is what Huayan calls the "Ocean-Seal Samādhi" mentioned above. It is "not just any samādhi, but the sāgara-mudrā samādhi, the 'samādhi which is like the images in the ocean,' which was the samādhi in which the newly enlightened Buddha beheld the entire universe as one living organism of identical and interdependent parts."30 Let us try to sort out conceptually what might be involved in that meditative state, assuming that we accept the validity of the experiential claim.

Obviously, if we are to work our way out of the dilemma between the nonidentity and nondistinction of li and shi, there has to be a way to drop the contradistinction between li and shi. The third aspect of Du Shun's formulation offers a valuable way in which to overcome this contradistinction, that is, shi contains both li and shi without obstruction:

Third is the aspect of shi containing li and shi without hindrance. This means that because shi and li are not one, while keeping one thing intact it still has the capacity of universal inclusion. It is like one atom: its form is not large, yet it can contain the boundless universe. Since shi, such as lands and so on, are not apart from the universe, therefore all of

--------------------------------------------------------------------------------

P.470

them appear in a single atom; as in a single atom, so it is in all things. This is because shi and li fluidly interpenetrate, being neither one nor different. There are four points in all here: (1) one in one; (2) one in all;" (3) all in one; (4) all in all. Each has a reason. Ponder this.31

The explanation furnished here recalls our deliberation in dealing with the dharmadhātu of interpenetration of li and shi without obstruction. A key distinction at this stage is the realization of shi containing li and shi. In line with the third dharmadhātu , the emphasis here is squarely placed on the side of shi, and more importantly, shi here is no longer the shi that is in contradistinction to li, but rather contains both shi and li. Thus, shi undergoes a perceptual transformation. It is indicative of this fact that in the third dharmadhātu, shi is no longer the shi perceived in the everyday mode as phenomena, each of which stands in isolation to the others. Rather, there is a perceptual transformation, so that shi is perceived to contain both /; and shi in the samādhic state, wherein it stands in its original relatedness with the others in their shared participation in the self-negating activity ( li ). Such an epistemological32 transformation inevitably yields a new picture of reality.

What kind of reality is envisioned here? It is summed up by the four points listed by Du Shun, of which Cheng Guan offers a detailed exposition:

Take the first point, "one in one": because the first "one" does not lose its characteristics, it has a body that contains; because from the perspective of li it is not different from the second "one," it can contain the second "one." Meanwhile, since from the perspective of li;' the second "one" is not different from the first "one," in accordance with the contained li, it is in the first "one," because there is no li apart from shi.

Second, as for "one in all," because all do not lose their characteristics, they have bodies that contain; since from the perspective of li they are not different from the one, they can contain the one. Because from the perspective of li, the one is not different from the all; in accordance with the principle of its own oneness, the one is in the all.

Third, as for "all in one," because one does not lose its characteristics, it can be that which contains, while because from the perspective of li, it is not different from the all; it can contain all. From the perspective of li, the all that is contained is not different from the one, so in accordance with the li inherent in all, it is in the one.

Fourth, as for "all in all," because the first "all" do not lose their characteristics, they have bodies that contain; since from the perspective of li they are not different from the second "all," they therefore contain the second "all." Because from the perspective of li, the second "all" are not different from the first "all," in accordance with the li of the second "all" they are in the first "all." Therefore, in the conclusion he [Du Shun] says each has a reason.33

The "one" here refers to one shi and "all" to all shi. Apparently, these passages are intended to give us some hint as to how the interpenetration between shi and shi takes place, and the picture drawn here is that

--------------------------------------------------------------------------------

P.471

of a nonobstructed interpenetration of shi and shi. Accordingly, the first passage deals with the relationship between two individual shi, the second and the third, the relationship between an individual shi and all shi, and the last, all shi simultaneously. The former two passages refer to one shi, with the first from the perspective of another shi and the second from that of all shi, in examining how they interpenetrate each other; the latter two passages shift the focus to all shi, with the third proceeding from the angle of one shi and the last from that of all shi.34

All shi in their sum are not a possible object of perception, however. The question then is: How can anyone perceive the mutual containing between one shi and all shi as well as all shi simultaneously containing all shi (the second, third, and fourth points above)? In line with the earlier observations that the shi here is no longer that which is in contradistinction to li, but that which contains both li and shi, and that li is a pure activity perceived in the samādhic state, it is possible to venture the following explanation of all shi: All shi that are perceived in their totality are not perceived in their physical, being-centered, appearance but as the self-negating activity (li) that is shared by them. This is the underlying reason that shi can interpenetrate each other. What is at stake in the present dharmadhātu―the nonobstructed interpenetration between shi and shi―is as much about one shi or all shi as objects as it is about their interpenetrative activities. This is the very rationale furnished by Cheng Guan in his elucidation of Du Shun's vision of the mutual containing of shi. To wit, on the one hand, they all retain their own characteristics as shi, that is, they all have their own body, which is what contains. On the other hand, they share the same li, the self-negating activity, which is what is contained; by containing li, they contain the others as well.

concluding remarks

We have examined two Buddhist formulations of the part-whole relationship: the reductionist explication of the Buddhist doctrine of anātman that appears in the early Buddhist literature, and the interpenetrative understanding of the issue espoused by the Huayan patriarch Du Shun and his commentator Cheng Guan. It should be clear by now that the reductionist tendency in early Buddhism is reversed by the Huayan Buddhists. Early Buddhists appeal to the reductionist mode of thinking in driving home the point that there is nothing called the "self" in the world that exists independently, over and above those elements that constitute it. The claim is based on the commonsense idea that the part-whole relationship is constitutive. When the issue gets into the hands of the Huayan philosophers, such a constitutive relationship between part

--------------------------------------------------------------------------------

P.472
and whole is dismissed. Instead, Huayan interprets the part-whole relationship in light of the subject-verb structure, rendering it a predicative rather than a constitutive relationship. In this way, Huayan manages to avoid the pitfalls of both the reductionist and the essentialist positions regarding the nature of the whole. Consequently, the reductionist―and hence, substantive―mode of thinking of Nāgasena gives way to the predicative―and thus organic and holistic―mode of thinking of Du Shun.

It is important to recognize, however, the fact that the Huayan philosophers do not abandon the doctrine of anātman, but rather further it by working out what is involved in the experience of anātman. According to them, what is at stake is the possibility of knowing all. In the process, they avoid a purely negative formulation of anātman. Hence, their challenge to anātman is not doctrinal, but methodological. Only when one becomes all-knowing can one truly embrace and verify the principle of anātman, since in such a state there is nothing that can obstruct anything from being known completely. In other words, the one who knows and that which is known are no longer two. This is the nondual position maintained by the Mahāyāna Buddhist tradition, to which Huayan belongs. A person who embodies such an experience is a Buddha, an enlightened one.

From the perspective of an enlightened one, according to the Huayan teaching, the world is itself the very world within which all beings share a common experience and a common destiny. The natural soteriological conclusion that can be drawn in terms of the whole-part relation vis-à-vis the Huayan li/shi scheme is that the suffering of one is the suffering of all, and the enlightenment of one is the enlightenment of all. We share the experience of suffering, but we are together headed toward a common destiny, namely, enlightenment. Because of the infinite interpenetration of all, however, suffering and enlightenment likewise interpenetrate. This is tantamount to saying that suffering is no different from enlightenment and enlightenment is no different from suffering, echoing the well-known Mādhyamika motto: nirvāṇa is saṃsāra and saṃsāra is nirvāṇa.35That is, any moment can be either an enlightened or a suffering moment. Enlightenment is right here, right now. We can clearly see why the Huayan system is believed to have provided the philosophical foundation for Chan Buddhism in China.

endnotes

1. A Sourcebook in Indian Philosophy, edited by Sarvepalli Radhakrishnan and Charles A. Moore (Princeton, NJ: Princeton University Press, 1957), pp. 281 -284. [Back]

2. This is Brentano's distinction, as revised by David Bell in his Husseri (London: Rout-ledge, 1990),p. 18.[Back]

--------------------------------------------------------------------------------

P.473

3. Independent and dependent parts are Husserl's terminologies used in his Logical Investigations. As Kit Fine summarizes, "If one object is founded upon another, then any object containing the first but not the second is dependent both absolutely and relatively to any independent whole which contains the two given objects. . . . Independent parts of a whole are parts not founded on any part of the whole" ("Part-whole," in The Cambridge Companion to Husserl, edited by Barry Smith and David Woodruff Smith [New York: Cambridge University Press, 1995], p. 468).[Back]

4. There were some very important developments on the whole-part relationship within the Indian Buddhist tradition, and their sophistication far exceeded the formulation by Nāgasena (for example, Dharmakīrti's arguments). However, since the primary objective of this article is to trace how the Huayan Buddhists reoriented the discourse of the whole-part issue given their Buddhist inheritance, and since it is questionable how much impact Dharmakīrti (Fa Chengi) had on the development of Chinese Buddhism historically, we will not go into those later developments within the Indian Buddhist context.[Back]

5. The Mādhyamika argument with regard to the relationship goes like this: There are four possible relationships between the chariot and its components, that is, the chariot is identical to the components, different from them, both identical to and different from the components, or neither identical to nor different from the components. The first possibility would give rise to the position Nāgasena maintains, which is reductionist in orientation; the second to the Hindu realist-essentialist position regarding chariothood, since it suggests that the chariot exists somewhere outside of its components—this position was rejected forcefully by Diṇnāga (Chen Nap), a Buddhist logician whose orientation is that of a nominalist; the third contains a logical incompatibility because identity and difference are mutually exclusive, and hence cannot predicate the same subject matter at the same time in the same way; and the last one seems to be the best way to capture the relationship between the whole and the parts without the risk of falling into the traps of reductionism or essentialism. Even the last possibility falls short of being a satisfying answer, however, because it inevitably results in the mystification of the chariot. [Back]

6. In quoting from this text, I will use Thomas Cleary's translation in Entry into the Inconceivable: An Introduction to Hua-yen Buddhism (Honolulu: University of Hawaii Press, 1983), but because his rendition contains misinterpretations of certain basic terms, some revisions are necessary. All the pagination of this text corresponds to the pagination of Cleary's translation. [Back]

7. Paul Williams, Mahāyāna Buddhism: The Doctrinal Foundations (London: Routledge, 1989), p. 123. [Back]

8. We have to be careful not to confuse li in Huayan Buddhism with that in neo-Confucianism. It should be noted that although Zhu Xi was no doubt knowledgeable about Huayan's li, his li-qi scheme appears to have been formulated to reconcile the contradiction arising out of the classical Confucian debate on the issue of human nature while sanctioning Mencius's position. In this sense, li seems to be a primarily moral concept that takes on a metaphysical dimension and transcendent quality when it is integrated into the grand notion of the Great Ultimate (taiji u). Through such conception of li, Zhu Xi makes the universe a moral one. As such, his li bears very little resemblance to the li formulated in Huayan philosophy. Li in Huayan Buddhism is the self-negating activity of the cosmos, or emptiness, to use an orthodox Buddhist term. It is supposed to be perceived by an enlightened being in a deep meditation state known as the "Ocean-Seal Samādhi." In Huayan's conceptualization and schematization of li, the issue of its moral nature does not come into the picture. [Back]

9. Although śūnyatā itself is not another view, but rather something pointing to the ineffable (illocutionary), if we take Jay L. Garfield's interpretation in The Fundamental Wisdom of the Middle Way: Nāgārguna's Mūlamadhyamakakārikā (New York: Oxford University Press, 1995), pp. 354-359. Even though Garfield's translation has used a lot of Tibetan sources, this particular point still holds ground within the Indian Buddhist discourse, and I will not explore the details here―to talk about the notion of śūnyatā at all is an inferential effort. However, Huayan philosophers endeavor to spell out the very ineffability itself. In other words, their description of emptiness is based on the perception of non-obstruction between li and shi and between shi and shi, not a simple assertion of its ineffability. [Back]

--------------------------------------------------------------------------------

P.474

10. I am using the term perception in the broadest sense, including any experience that does not involve the discriminatory function of the mind. [Back]

11. I am here presupposing a Diṇnāgian type of epistemology in approaching the issue of emptiness li. In Diṇnāga's epistemological system, there are only two valid means to acquire knowledge: perception and inference. In using the term content, I hope to avoid the subject-object dualism implied if the term "object of perception" is used. [Back]

12. An idealist position, in the Platonic sense of the term, is not available in the traditional Buddhist discourses. [Back]

13. Cleary, Entry into the Inconceivable, p. 108. [Back]

14. Ibid., p. 91; translation modified. [Back]

15. Ibid., p. 92; translation modified. [Back]

16. The metaphor of ocean and wave may recall two important philosophical categories in traditional Chinese thought, tix (that which is) and yongy (that which functions), but this is not the way Du Shun uses the metaphor in the current context, wherein the whole-part relationship is the concern. [Back]

17. Cleary, pp. 93-94. [Back]

18. Ibid., pp. 95-96; translation modified. [Back]

19. Ibid., p. 96; translation modified. [Back]

20. Ibid., pp. 97-99. [Back]

21. The boundary is that which holds together a strong whole, especially in activities, whereas the boundary of a weak whole does not hold to the same extent. [Back]

22. This is similar to a Chan phrase, "seeing without a seer." [Back]

23. Shi is always predicated by li in the sense of predication by a verb, and not any other kinds of predications. [Back]

24. We can probably detect some Daoist influence in Huayan's formulation of li, for Dao is also a cosmic activity that constantly regenerates itself through its very self-negation;

hence, Dao can give rise to one, one to two, two to three and the world. [Back]

25. Cleary, p. 115. [Back]

26. Ibid., p. 111. [Back]

27. Ibid., p. 112; translation modified. [Back]

28. Ibid., pp. 111-112; translation modified. [Back]

29. Ibid., p. Ill; translation modified. [Back]

30. Francis H. Cook, Hua-yen Buddhism: The Jewel Net of lndra (University Park, PA: Pennsylvania State University Press, 1991), p. 73. [Back]

31. Cleary, p. 113; translation modified. [Back]

32. I am aware of the nuances involved in using the word "epistemology" in describing meditative experience, for epistemology in the ordinary sense presupposes a duality between the subject and the object, whereas in a high meditative state such a duality no longer exists. However. I am using the term "epistemological" simply to describe that-as-perceived as opposed to that-in-itself. In the samādhic awareness, the distinction between that-as-perceived and that-in-itself has already evaporated. In a word, that-in-itself here is used simply as a logical term; what I am concerned with is the transformation of the mode of perception from the ordinary to the samādhic. [Back]

33. Cleary, pp. 114-115; translation modified. [Back]

34. Fa Zangaa offers a famous example of ten mirrors simultaneously reflecting each other to illustrate the point. [Back]

35. Paul Williams, Mahāyāna Buddhism, p. 69. [Back]