The eye that I see God with is the same eye God sees me with.
My eye and God’s eye are one and the same.
God is abstract being, pure perception, which is perceiving itself in itself.
— Eckhart
Reality without appearances would be nothing, for there certainly is nothing outside appearances.
— F. H. Bradley, Appearance and Reality
According to many “illuminative” philosophies, both Eastern and Western, sense-perceptions are delusive and must in some way be transcended. This is particularly true for the nondualist Asian philosophies discussed in chapter 1. For Śaṅkara, the world as perceived, although not illusory from a phenomenal point of view, must ultimately be subrated and realized to be dreamlike māyā, for only Brahman is really Real.55 In the Fire Sermon, the Buddha states that his disciples should have aversion to sense-organs, sense-objects, sense-contact, and sense-consciousness, in which case passion will fade away and liberation occurs. Such claims seem to be recommending the negation of sense-phenomena in order to experience a Reality apart from them. This interpretation is consistent with a predisposition we have inherited from the Western metaphysical tradition, Parmenides through Kant, to distinguish between the constantly changing world of phenomena that the senses present to us and an unchanging Reality “behind” them; the former is usually devalued in favor of the latter, whose nature it is the task of philosophy to determine. Plato’s Ideas (or Forms) are to be directly experienced by intellect alone, purified of any relationship with the senses, thus establishing a dichotomy that has had fateful consequences for Western philosophy and Western culture. It has been just as fateful for the Eastern tradition that this dichotomy did not occur, for as we shall see, the nondualist systems look upon the conceptualizing mind as a “sixth sense” which needs to be “transcended” at least as much as the other five — perhaps more.
The problem with this usual interpretation is that many puzzling passages, often attributable to the same sources, are incompatible with such a blanket rejection of sense-perception. In the Vivekacūḍāmaṇi, Śaṅkara makes a claim that seems inconsistent with his other views, but perhaps it is just inconsistent with others’ views of him: “The universe is an unbroken series of perceptions of Brahman; hence it is in all respects nothing but Brahman.”56 In one of the Honeyball sutras, the Buddha teaches the monk Bahiya that the end of suffering — that is, nirvana — is to be found in training himself so that “in the seen there will be just the seen; in the heard, just the heard; in smelling, touching, tasting, just smelling, touching, tasting; in the cognized, just the cognized.”57 Both these passages suggest that sense-perception itself is not the problem: rather, Reality is staring us in the face all the time, but somehow we misperceive it.
How are we to reconcile these claims — by no means uncommon, as we shall see — with those critical of the senses? I argue in this chapter that what must be transcended is not sense-perception in toto but a certain type of sense-perception which, because we are not usually familiar with any alternative type, we tend to identify with sense-perception generally. As the Buddha recommends and Śaṅkara implies, another kind of sense-perceiving can be developed that reveals Reality — or, to be more exact, is Reality. (This is complicated by the fact, to be discussed later, that this other way of perceiving might not be termed sense-perception at all, since it can be argued that the act of perception is relative to perceiver and to sense-object, both of which are lacking in this other sense-perception. As a result, what might be called “only-perception” turns out to be equivalent to no perception. This is only the first example of a paradox that recurs many times in this book.) The difference between these two types of sense-perception is the difference between dualistic and nondualistic perception. The former, perception as we normally experience it (or interpret it), is sense-perception in which there is a distinction between the perceiver and the object perceived. The latter is nondual because there is no such distinction; therefore it has sometimes been described by denying (as Buddhism does) that there is a subject perceiving and sometimes by denying (as Vedānta does) that there is an external, objective world which is perceived. In such perception there is no longer any distinction between internal (mind) and external (world), or between consciousness and its object.
This chapter develops this conception of nondual perception. The second section examines in some detail the views of Buddhism and Advaita Vedānta on the nature of perception, and on the basis of those views constructs a theory or “core model” which, with some qualification, is consistent with both nondualistic systems. In the following section I ground these generalizations about perception with an examination of hearing and seeing, using Berkeley and Hume to develop this theory into a more coherent claim. The fourth section relates nondual perception to Western theories of perception, places it within contemporary epistemology as a version of phenomenalism, considers how it fares against the objections usually raised against phenomenalism, and reports on two psychological experiments into meditation that seem to support the possibility of nondual perception.
The implication of this view is that the commonsense, apparently objective world that we usually take for granted — which is understood to be composed of discrete material objects causally interacting in space and time — is a fiction the mind creates by superimposing its thought-constructions upon perceptions. Such an approach is not unfamiliar to modern Western philosophy, for it has some affinity with the basic stance of Kant’s metaphysics. But there are two fundamental differences between such nonduality and Kantian metaphysics. First, whether this thought-construction is due completely to language acquisition and other socialization, or partly to innate faculties of the mind, the claim of the nondualist Asian systems is that this process can be undone — quite literally deconstructed or “de-automatized” — which is why their basic attitude is soteriological as much as philosophical. Such deconstruction is possible because of the second difference. One of the problems with Kant’s distinction between noumena (things-in-themselves) and phenomena (things as we experience them) is that, while maintaining that causality is a category applicable only to phenomena, he also inferred that things-in-themselves must be the causes of phenomenal appearances. Nor can Kant easily escape this inconsistency, for without some such view there is no reason to postulate the existence of things-in-themselves at all, since he believed that they cannot in principle ever be directly experienced. The nondualist is not subject to such a criticism, since things-in-themselves — what I call nondual percepts, in the case of perception — are experienced immediately upon the cessation of thought-construction. Such a view avoids the postulation of a Reality “behind” Appearance. Rather, Reality is Appearance itself, although this of course cannot be appearance as we normally understand it, which is appearance of something. The nondualist explanation turns the usual view upside down: it is our normal, commonsense understanding — in which we distinguish between physical objects and their appearance to us — that is guilty (as Berkeley and Nietzsche realized) of metaphysically postulating a reality “behind” appearance. This was so obvious to Berkeley that he was surprised when others did not accept his critique of matter, that mysterious stuff we never actually experience. Like Vasubandhu long before him, he was denying not sensible qualities, such as impermeability, but the self-existing substratum to which they supposedly adhere. In this way the nondualist presents us with the possibility of actually returning to things-in-themselves, percepts as they are, before they have been thought-constructed into the dualistic world of a subject confronting a materialized world of discrete objects.
Soon after Berkeley there lived an English engraver and poet for whom this was not just philosophy but life itself, and we shall have occasion to quote him often in the pages that follow.
The whole creation will be consumed and appear infinite and holy, whereas it now appears finite and corrupt.
This will come to pass by an improvement of sensual enjoyment.
But first the notion that man has a body distinct from his soul is to be expunged. . . .
If the doors of perception were cleansed, everything would appear to man as it is, infinite. For man has closed himself up, till he sees all things thro’ narrow chinks of his cavern. (William Blake)58
It was not until Kant that Western philosophy became truly aware of the role of the mind in sense-perception: how the mind does not just receive but interprets and synthesizes perceptions into the phenomenal world we experience. That perception involves conception is a commonplace of contemporary philosophy, although attention has shifted from Kant’s Aristotelian categories to language as the means by which this organization occurs. But Indian philosophy has been aware of this since at least the time of the Buddha. After a brief introduction to the Indian distinction between nirvikalpa and savikalpa perception, I consider the view of Buddhism and Advaita Vedānta on this issue. We shall see that Pāli Buddhism emphasizes the need to distinguish the “bare percept” from its conceptual and emotional superimpositions. A more explicit statement that such a bare percept is nondual is found in Mahāyāna Buddhism: it is part of the Prajñāpāramitā claim that perception, like everything else, is śūnya (empty); it is implied in the Mādhyamika critique of all dualities, and it is clearest in the Yogācāra assertion that subject and object are not distinct. The same nondual claim will be found in Advaita, with a subtle but significant difference. Just as Vedānta distinguishes sharply between Brahman and the phenomenal world, so it distinguishes between our usual dualistic perception and the nondual experience of Brahman, which it does not call perception at all. We will need to consider how important this disagreement with Buddhism is, whether it points to a difference in experience or merely a difference in describing the same nondual experience.
With one important exception (Ch’an or Zen, which as a school of Mahāyāna was of course much influenced by Indian Buddhism), what follows deals solely with Indian philosophy. We do not find much reference to this topic in Taoism because Chinese philosophy, being generally more pragmatic than Indian philosophy, was not as interested in epistemological questions. A similar view is perhaps implicit in the first chapter of the Tao Tê Ching, but, since it is little more than a hint, discussion of that passage is postponed until the next chapter.
One of the main ways Indian philosophy acknowledges the role of conception in perception is by making a distinction between savikalpa and nirvikalpa perception. Our usual perception is sa-vikalpa (with thought-construction), but there is the possibility of nir-vikalpa perception, which is “without thought-construction” because the bare sensation is distinguished from all thought about it. The basis of both Sanskrit terms is vikalpa, a compound from the prefix vi (discrimination or bifurcation) and the root kalpanā (to construct mentally). This distinction is found in most of the important Indian systems, Jainism and the monotheistic schools of Vedānta being the main exceptions. There is of course much disagreement over the psychology and ontology of perception, but with the exception of Advaita Vedānta (examined later) it is agreed that nirvikalpa and savikalpa are not completely different types of perception, but earlier and later stages of a complex process. For example, the pluralistic Nyāya system, as developed by Gautama, defined nirvikalpa as “unassociated with a name” (avyapadeśya) and savikalpa as “well-defined” (vyavasāyātmaka). By its association with language all perception becomes “determinate,” but this is necessarily preceded by an earlier stage when it is unassociated, a “bare sensation.” “Nirvikalpa perception is the immediate apprehension, the bare awareness, the direct sense-experience which is undifferentiated and non-relational and is free from assimilation, discrimination, analysis and synthesis.”59 We can sense this bare sensation, but as soon as we try to know it, this “raw unverbalized experience” (William James) becomes associated with thought-conception and hence determinate (savikalpa).
This summary of the dualistic Nyāya position raises two issues important to the nondualist. First, what is the role of language in this distinction between nirvikalpa and savikalpa perception? Do we hypostatize a percept into an object by naming it, thus “identifying” it as a member of a certain class of objects? And does a sense of self arise in the same way — are the concepts of I and mine used to objectify ourselves? Second, we can readily see that this indeterminate/determinate distinction is not only epistemologically interesting but also obviously has ethical implications, among others. For example, there is a relationship between perception and the problem of craving. Due to mental tendencies accumulated from the past, the mind is prone to meddle with some percepts more than others and thus to activate certain predispositions. This suggests that a permanent resolution of the problem of craving might be related to an understanding of nirvikalpa perception and the process by which it becomes savikalpa.
In this section my main concern is naturally with Mahāyāna Buddhism and Advaita Vedānta, the two most important nondualistic systems in India. But here there also seems to be an important parallel with Yoga, which of the six orthodox Indian systems is the one most concerned with describing the path to liberation. Patañjali’s Yoga Sūtra discusses the various stages of samādhi (yogic meditation) in great detail, and one could make a strong case that his four preliminary stages of samprajñāta samādhi actually “undo” savikalpa perception in order to return to the bare nirvikalpa percept. This suggests that, despite the overtly dualistic Sāṅkhya metaphysics that Patañjali adopts, the deeper asamprajñāta samādhi might actually be nondual, in the sense that the meditator is no longer aware of any distinction between his own consciousness and the object of meditation. Sāṅkhya metaphysics will be discussed further in chapters 5 and 7, but we now return to our main concern. Since Vedānta as a system is antedated and influenced by Buddhism, we consider the latter first.
Because the Buddha’s concern was almost exclusively soteriological, it is not surprising that the Pāli sutras present no single, developed theory of perception. However, they contain a wealth of epistemological material, much of it relating directly to perception, and, although the terms nirvikalpa and savikalpa are not used, the distinction between perception with and without thought-construction is clearly critical.60
A good account of this material is given in Edward Conze’s Buddhist Thought in India. Conze summarizes the analysis of perception found in the Pāli Canon into “three levels of the apperception of stimuli, to which three kinds of ‘sign’ correspond — the sign as (1) an object of attention, as (2) a basis for recognition, and as (3) an occasion for entrancement.” In the first stage, one “turns towards a stimulus”; attention is directed to what I call the “bare percept,” an act both active and passive because one chooses to turn toward it but one cannot determine what the sensation will be. In the second stage, what has been perceived is recognized, “as a sign of its being such and such a part of the universe of discourse, and of habitually perceived and named things.” So the “bare” visual percept is now seen as a woman, or a table, or whatever, with all the respective connotations. These connotations are elaborated in the third stage, which “is marked by the emotional and volitional adjustment to the ‘sign.’” The sign is now interesting to us and awakens volitional tendencies; I am attracted to the woman and wonder how I can meet her.
Of course, the whole sequence usually occurs so quickly that one is unable to distinguish one stage from another. So we take this tripartite series of distinct impersonal processes to be one simple mental event: seeing an attractive woman. Normally we are not aware of what it is like to experience just the first stage, for we never have experienced just that by itself. But to build upon sense-perceptions in this way is undesirable, according to the Buddha, and in the Majjhima Nikāya he describes various methods for “restraining the senses.” Conze summarizes:
The task is to bring the process back to the initial point, before any “superimpositions” have distorted the actual and initial datum. The seemingly-innocuous phraseology of the formula which describes the restraint of the senses opens up vast philosophical vistas, and involves a huge philosophical programme which is gradually worked out over the centuries in the Abhidharma and the Prajna-paramita. “He does not seize on its appearance as man or woman, or its appearance as attractive, etc., which makes it into a basis for the defiling passions. But he stops at what is actually seen.” Taken seriously, this must lead to an attempt to distinguish the actual sense-datum from the later accretions which memory, intellect, and imagination superimpose upon it. . . . “He seizes only on that which is really there.” . . . This is the starting-point of the considerations which in due course led to the concept of “Suchness” [Tathatā], which takes a thing just such as it is, without adding to it or subtracting from it.61
The second and third of Conze’s stages of apperception describe how a “bare” nirvikalpa percept becomes savikalpa, and the process of “restraining the senses” is the means by which this apparently simple mental event may be broken up into its three impersonal component processes, thus deconstructing the savikalpa perception back into a nirvikalpa one. The second state, “recognition,” obviously includes the application of language to what is immediately presented by the senses. The third stage, emotional and volitional response, will usually become expression of craving (tṛṣṇa). How these two factors interact requires some discussion.
According to Buddha, tṛṣṇa is the cause of our suffering, but the term refers not just to sensual desire but to attachment in general, whether to sense-experience or nonsensuous mental events. The above analysis of perception suggests that the fundamental problem with such craving is epistemological, since it distorts one’s perception of things. However, such attachment seems limited to what is immediately presented to the senses. “I” can “seize on” a particular appearance only because that appearance is now appearing. How can I grasp at something that is not present any more? If there is some way to represent the appearance, I can retain and refer to “It.” Such “grasping at a distance” is enabled by a system of re-presentation, that is, a language. But language also widens the gulf between the I and the grasped-at objects, because when the percept again appears, the re-presentation (“urg,” let us say) does not disappear as having no more function. It still represents the appearance. Now we know what the appearance is: it is “urg” (name) or “an urg” (particular instance of a universal). Now I experience the appearance through the representation, which is as it were superimposed upon it. The problem is that the more successfully a system of representation functions, the less likely we will be able to distinguish the representation from the appearance.
The above analysis presents a plausible view of how language functions, but it is naive and inadequate by itself. It is not really the case that the presented world is divided up into “objects” which we later represent. Rather, we divide up the world in the way that we do — that is, learn to notice what there is — using a system of representation. This is the point of the distinction between nirvikalpa and savikalpa perception: savikalpa determinations are not simply “added on” to nirvikalpa percepts, but they determine what the world is for us. John Searle, a contemporary philosopher of language, explains this well:
I am not saying that language creates reality. Far from it. Rather, I am saying that what counts as reality . . . is a matter of the categories that we impose on the world; and those categories are for the most part linguistic. And furthermore: when we experience the world we experience it through linguistic categories that help to shape the experiences themselves. The world doesn’t come to us already sliced up into objects and experiences: what counts as an object is already a function of our system of representation, and how we perceive the world in our experience is influenced by that system of representation. The mistake is to suppose that the application of language to the world consists of attaching labels to objects that are, so to speak, self-identifying. On my view, the world divides the way we divide it, and our main way of dividing things up is in the language. Our concept of reality is a matter of our linguistic categories.62
Like Kant, Searle doubts that it is possible to experience “things-in-themselves” apart from linguistic categories, but the linguistic approach seems to leave the door open in a way that Kant did not: since language is learned, isn’t it possible to “unlearn” it, as the Buddha’s program for “restraint of the senses” suggests? If so, and if Searle is right that language determines “what counts as reality,” then the world experienced in such a way would be very different indeed from the world as we normally perceive and understand it. If we take Searle’s phrase literally, then the nirvikalpa elimination of language implies that the category of real would no longer be applicable to any particular — just as Mahāyāna, Advaita, and Taoism insist.
Language must also be related to the third stage of apperception, which involves expressions of craving. In order to crave something I must be able to distinguish the object of my craving from other things, and in order to do this most successfully a system of representation is necessary. For example, it may be possible to crave a particular taste without being able to identify it, but such craving is more likely to be satisfied if I can re-present that flavor as “chocolate,” Searle’s account implies that it is doubtful whether I would even notice the subtle distinctions between types of chocolate without the vocabulary to represent those distinctions, just as I am likely to see only “snow” in Alaska, whereas an Eskimo would see a specific one of a dozen representable types of “snow.” The vast number of possible conceptual distinctions can thereby increase and refine our cravings. This does not mean that craving is dependent upon our concept-formation. The general view of the nondualist philosophies is rather that our system of representation is at the mercy of our desires and in fact evolved to help us satisfy and elaborate them. The motivation behind the particular way we “divide up” the world through language (hence transforming nirvikalpa into savikalpa percepts) is, fundamentally, our craving. This does not obviate Searle’s view. We do not first perceptually “pick out” objects and only later name them; rather, we learn to notice them by naming them, and the motivation behind that naming was originally the assistance it gave in satisfying desires. That is not contradictory to the nondualist view of perception, for what is important to the nondualist is that the association of percept with craving can be broken.
The passage from Conze quoted earlier seems to imply that stopping at the “bare” nirvikalpa percept is the goal. However, Conze’s understanding of the “initial datum” stage is that it is still dualistic: “He seizes only on what is really there.” As we shall see, the Mahāyāna view is that I can “let go” of the “seizing” too — that is, even the “I” can be let go — and what is experienced then is the original thing-in-itself, a nondual percept. The Abhidharmic view differs only in that the thing-in-itself is not explicitly nondual but seems to be a set of objectively existing dharmas. Conze does not see this because he follows other commentators and understands Buddhism to recommend a rejection of sense-experience. “Buddhism goes even further [than condemning sense desire] and regards even sense-perceptions as baneful.” But instead of supporting this with an analysis of the Pāli sutras, he immediately relates this “distrust of sense-objects” to the European Neoplatonic tradition, quoting Saint Gregory and Saint Dionysius. In a footnote, he deals curtly with the fact that someone might respond with the injunction of Seng-ts’an (the third Ch’an patriarch) that we should “not be prejudiced against the six sense-objects.” His answer is that Seng-ts’an is referring to a different and more advanced stage. “In terms of the five levels [which he has earlier] distinguished, we are here with the doors of deliverance on the third, whereas Seng-ts’an speaks of the fourth.”63 The quotation in question is from Seng-ts’an’s Hsin Hsin Ming (Awakening Faith in Mind), the relevant lines of which are:
If you pursue appearances
You overlook the primal source
If you would walk the Single Way
Do not reject the sense domain
Accepting the world of senses
Conforms with true enlightenment64
Seng-ts’an himself draws no distinction between any such levels, nor do the many other Mahāyāna sources which, as we shall see, could also be cited to criticize Conze’s rejection of sense-perception. Conze would have difficulty justifying his view with the Mahāyāna texts, but from the Pāli Canon he could (although he does not) cite the Fire Sermon and the Sermon on the Marks of No-Self. Such passages do seem to reject sense-experience, but they must be set against many others in the Pāli canon that recommend not loathing or disgust but equanimity.65
Perhaps the strongest canonical evidence against Conze’s rejection of the senses is in one of the Honeyball sutras where cognizing the “bare” percept is equated with “the end of duḥkha” (suffering).
Then, Bahiya, thus you must train yourself: “In the seen there will be just the seen; in the heard, just the heard, in the muta [the sense impressions from smelling, tasting, and touching], just the muta; in the cognized, just the cognized.” That is how, O Bahiya, you must train yourself. Now, when, Bahiya, in the seen there will be to you just the seen, in the heard . . . just the cognized, then, Bahiya, you will have no “thereby” (na tena); when you have no “thereby,” then Bahiya, you will have no “therein” (na tattha); as you, Bahiya, will have no “therein” it follows that you will have no “here” or “beyond” or “midway-between.” This is just the end of duḥkha.66
Traditional commentaries on this passage mention a number of conflicting interpretations, but it seems to be advocating a return to nirvikalpa perception to reach “the end of duḥkha” — which is the most common Pāli description of nirvana. The sutra continues by reporting that Bahiya, upon hearing this, attained nirvana almost immediately. Other passages which advocate equanimity toward the senses suggest that the return to “the first stage of apperception” is a necessary part of the meditative path, but this passage goes further to imply that such a return is sufficient for the attainment of nirvana. It is tempting to speculate on the meaning of na tena and na tattha and give them a nondualist interpretation: “If in the seen there is just the seen, then, O Bahiya, you will make no inferences on the basis of that ‘seen,’ and you will not see an object ‘therein.’”
Passages such as these also shed a new light on the Buddha’s repeated exhortation against “compounds” (saṁskāra), found even in his last words: “Impermanent are all compound things; attain perfection through diligence.” After the Buddha passed away, the Abhidharma (higher dharma) developed his preference for the “non-compound” (asaṁskāra) into an ontology which classified everything that can be experienced into a fixed number of simple elements (dharmas). All compounds (for example, the five skandhas or “aggregates” that compose the self) may be deconstructed into these basic elements. This remains the most common interpretation of the saṁskāras, but perhaps the Buddha was actually making an epistemological point, criticizing compound savikalpa sense-experience in favor of the noncompound nirvikalpa “bare” percept.
It must be emphasized that no passage has been referred to in the Pāli Canon which explicitly asserts the nonduality of perceiver and perceived, although I have tried to indicate some implications in this direction. What these passages add up to is the claim, not that perception must be “transcended” (as Conze maintained), but that we should return to the initial stage of perception, the unconstructed nirvikalpa percept. Since resting with this is “the end of duḥkha” and since the anātman doctrine of Buddhism denies any self, it would seem that such percepts must be nondual in the sense that there is no separate consciousness aware of them. We find clearer assertions to this effect in the paradoxical expressions of Mahāyāna, which uses the term śūnyatā to suggest strongly, and in some cases to state explicitly, that perception is nondual.
Prajñāpāramitā. Śūnyata is perhaps the most important term in Mahāyāna, but it is not easy to translate. It comes from the root śū, which means “to swell” in two senses: hollow or empty, and also full, like the womb of a pregnant woman. Both are implied in the Mahāyāna usage: the first denies any fixed self-nature to anything, the second implies that this is also fullness and limitless possibility, for lack of any fixed characteristics allows the infinite diversity of impermanent phenomena. It has been unfortunate for Anglo-American Buddhist studies that “emptiness” captures only the first sense, but I follow the tradition.67
The term is used in both Pāli and Mahāyāna Buddhism, but differently. Śūṇyatā in Pāli Buddhism generally means, first, that this world of saṁsāra is empty of value and should be negated in favor of nirvana; and second, that both saṁsāra and nirvana are empty of any self because all compounds are only clusters of dharma-elements. In Mahāyāna, śūnyatā means that the true nature of the world (tathatā) is empty of all description and predication; and that even all the dharma-elements are empty of any self-existence because all “things” are relative and conditioned by each other. The first Mahāyāna sense of śūnyatā is already familiar to us from the distinction between nirvikalpa and savikalpa perception. The second goes beyond the Abhidharmic critique of compounds and entails, among other things, the nonexistence of any self-subsisting object “behind” a percept.
Mañjuśrī: “What is the root of the imagination which constructs something that is not actually there?”
Vimalakīrti: “A perverted perception.”
Mañjuśrī: “And what is the root of the perverted perception?”
Vimalakīrti: “The fact that it has no support.”
Mañjuśrī: “And. what is the root of that?”
Vimalakīrti: “This fact, that it has no support, it has no root at all. In this way all dharmas are supported on roots that have no support.”
In itself, a perverted perception is śūnya because it has no support, which means that it refers to nothing else, neither a perceived object nor a perceiver. Such claims, which work out the implications of “the restraint of the senses” that Conze mentioned earlier, are common in the Prajñāpāramitā literature:
Moreover, Subhūti, a Bodhisattva, beginning with the first thought of enlightenment, practices the perfection of meditation. . . . When he has seen forms with his eye, he does not seize upon them as signs of realities which concern him, nor is he interested in the accessory details. He sets himself to restrain that which, if he does not restrain his organ of sight, might give occasion for covetousness, sadness or other evil and unwholesome dharmas to reach his heart. He watches over the organ of sight. And the same with the other five sense-organs — ear, nose, tongue, body, mind.
. . . he remains the same unchanged, neither elated not cast down, neither grateful nor thwarted. And why? Because he sees all dharmas as empty (śūnya) of marks of their own, without true reality, incomplete and uncreated.
This passage accords well with Pāli Buddhism until its last sentence, when it goes further to explain that the equanimity of the Bodhisattva is due to his seeing all dharmas (including percepts) as śūnya, without any reality of their own and referring to nothing else besides themselves. That is the experience of tathatā, the “suchness” of things.
The Lord: . . . This prajñāpāramitā cannot be expounded, or learnt, or distinguished, or considered, or stated, or reflected upon by means of the skandhas, or by means of the elements, or by means of the sense-fields. This is a consequence of the fact that all Dharmas are isolated, absolutely isolated. Nor can prajñāpāramitā be understood otherwise than by the skandhas, elements, and sense-fields. For just the very skandhas, elements, and sense-fields are śūnya, isolated, and calmly quiet. It is thus that prajñāpāramitā and the skandhas, elements and sense-fields are not two, nor divided. As a result of their emptiness, isolatedness, and quietude they cannot be apprehended. The lack of a basis of apprehension in all Dharmas, that is called prajñāpāramitā. Where there is no perception, appellation, conception or conventional expression, there one speaks of prajñāpāramitā.68
Dharmas, because they are empty, cannot even be apprehended: this seems to go beyond denying both perceiver and sense-object to deny even the act of perception. Such a claim seems odd, but we find it also in Nāgārjuna. In his case the denial of perception is based upon the fact that our understanding of perception is dependent upon the reality of the perceiver and the perceived, both of which he also denies. For Nāgārjuna, the relativity of perceiver, perceived, and the act of perception entails the unreality of all of them, that is, their lack of self-existence. This does not, however, support the claim that we must “transcend” perception for the sake of some other kind of apprehension. Nāgārjuna is rejecting perception as we understand it, the dualistic act in which two self-existing entities are related together. This raises the question whether what we have been describing as “nondual perception” should be called perception at all. If the bare nirvikalpa sensation does not provide some knowledge to someone about something (and it cannot, since any inference is savikalpa), perhaps the term perception no longer applies and should be reserved only for thought-determined savikalpa percepts. This may explain why some texts (such as the above) deny there is perception, some assert there is nondual perception, and others paradoxically recommend perceiving without perception — which may all be different ways of describing the same sense-experience. We return to this point at the end of this chapter.
The comprehension which takes place as a result of perception does not imply an understanding of the reality (of the thing perceived). What you perceive without perceiving — that is Nirvāṇa, also known as deliverance. (Śūraṅgama Sūtra)69
Mādhyamika. The central tenet of Mādhyamika Buddhism, that saṁsāra is nirvana, is difficult to understand in any other way except as asserting the two different ways of perceiving, dually and nondually. The dualistic perception of a world of discrete objects (one of them being me) which are created and destroyed constitutes saṁsāra. Nāgārjuna describes the cessation of this way of experiencing the world in the last stanza of the Mūlamadhyamikakārikā chapter on nirvana: “Ultimate serenity is the coming to rest of all ways of taking things (sarvopalambhopaśama), the repose of all named things (prapañcopaśama).” In a footnote to his translation Sprung explains sarvopalambhopaśama: “It is not merely that ways of thinking about things change in nirvāṇa, but that the everyday way of perceiving, or ‘taking’, things ceases to function.”70
This well-known verse — as close as Nāgārjuna comes to a “description” of nirvana — emphasizes the importance of ending prapañca. The Sanskrit term prapañca (Pāli, papañca) is important in both Buddhism and Vedānta, but its meaning is controversial. In Buddhism it refers to some indeterminate “interface” between perception and thought. Several times in the Pāli Canon the Buddha mentions papañca to describe what happens in the later stages of sense-cognition, and he says that his teaching is for those who delight in nispapañca, no prapañca. The Mahāyāna Lankāvatāra Sūtra says that Buddhas are “beyond all vikalpa and prapañca.” Etymology yields pra+pañc, “spreading out” in the sense of expansion and manifoldness. This led the Theravadin scholar Ñānananda, in his book on prapañca, to define its primary meaning as “the tendency towards proliferation in the realm of concepts.”71 This is better than the ethical interpretations of the traditional Pāli commentaries,72 but there remain two difficulties with such a definition: it loses any direct relation with perception, and prapañca becomes indistinguishable from vikalpa. Both the Tibetan and Chinese Mādhyamika exegetical traditions understand the relation between vikalpa and prapañca as the relation between the mental act of conceptualization, understood subjectively, and its crystallized, objectively experienced counterpart. Thus, in the terms of this chapter, prapañca might be defined as “the differentiation of the nondual world of nirvikalpa experience into the discrete-objects-of-the-phenomenal world, which occurs due to savikalpa thought-construction.” This explains the important compound prapañca-nāmarūpa, since nāmarūpa (name and form) here can be understood to refer to the necessary relation between names and forms (the Buddha describes them as inseparable),73 that we reify forms by naming them. We shall meet with this interpretation of nāmarūpa again, implied both in Śaṅkara’s concept of adhyāsa (superimposition) and in the first chapter of the Tao Tê Ching.
It is significant that the earliest Vedāntic references to prapañca and prapañcopaśama are consistent with the above. The terms do not appear in the first Upaniṣads, such as the Bṛhadāraṇyaka and the Chāndogya, which are usually considered to be pre-Buddhistic. The two most important references are in the Śvetāśvatara, and the Māṇḍūkya. Śvetāśvatara VI.6 uses prapañca ontologically to denote the objectified universe, understood as a phenomenal world of manifoldness emanating from a creator God. Verse seven of the short Māṇḍūkya describes turīya, the fourth and highest state of experience, which is “all peace, all bliss, and nondual,” as prapañcopaśama. “This is ātman, and this has to be realized.”
The great importance of prapañcopaśama in Mahāyāna Buddhism is indicated by the fact that it is not only a term for nirvana but the preferred formulation for describing the Middle Way of Mādhyamika. In his commentary to Nāgārjuna’s Mūlama-dhyamikakārikā, Candrakīrti states and repeats that nirvana is the cessation and nonfunctioning of perceptions as signs of named things — in other words, that in nirvana perceptions do not refer to any hypostatized object “behind” the percept. “[When the wise are] cured by the balm of unmediated seeing that such things are irrefragably without substance, then they realize directly and for themselves that it is the true nature of such things not to be seen at all.”74 When we couple this with the general Buddhist denial of a self, it amounts to an assertion that nirvanic perception is nondual.
So nirvana is not even to be found “in” saṁsāra, for such a spatial metaphor is still dualistic. Rather, nirvana is the nondual “true nature” of saṁsāra. T. R. V. Murti expresses this well:
The transcendence of the Absolute must not be understood to mean that there is an other that lies outside the world of phenomena. There are not two sets of the real. The Absolute is the reality of the apparent; it is their real nature. . . . The Absolute is the only real; it is identical with phenomena. The difference between the two is epistemic and not real.75
“The reality of the apparent” does not mean a reality behind appearance but that appearance is reality itself, as we realize if we do not use appearance as a basis for vikalpa thought-construction and prapañca thought-objectification. But we must be careful about accepting any distinction between epistemic and real. In the nondualistic systems we are considering, epistemology and ontology cannot be so easily distinguished: epistemic changes in our experience amount to ontological changes as well, by revealing that things are (and perhaps always have been) very different from what we thought they were. In another well-known verse about the true nature of things, Nāgārjuna himself uses both prapañca and nirvikalpa: “Not dependent on anything other than itself, at peace, not manifested as named things (prapañcairaprapañcitam), beyond thought-construction (nirvikalpa), not of varying form — thus the way things are really is spoken of.”76
Yogācāra. Despite the above, we do not find in Mādhyamika the clear statement that nirvana is nondual cognition. That is because Mādhyamika declines to give any positive account about the nature of reality. Reality is experienced when all dualizing categories — including, no doubt, duality and nonduality — cease to function, so Mādhyamika confines itself to making a critique of those dualities: cause and effect, perceiver and act of perception, saṁsāra and nirvana, etc. In the terms of this chapter, Mādhyamika is most aware of the paradox that any claim of nonduality amounts to a savikalpa attempt to describe the nirvikalpa. But it is not surprising that this exclusively negative critique should have been followed by an attempt to characterize nirvana in a more positive manner than just “the end of prapañca,” and this we find in Yogācāra and Vijñānavāda Buddhism. It is significant, then, that the cognitive nonduality of subject and object constitutes the heart of the Yogācāra position. Passages from Vasubandhu denying the duality of perceiver and perceived are quoted in chapter 1. Here is a fuller version of his clearest statement:
As long as consciousness does not abide in re-presentation-only, so long does one not turn away from the tendency towards the two-fold grasping [perceiver and perceived]. As long as he places something before him, taking it as a basis, saying: “This is just re-presentation-only,” so long he does not abide in that alone.
But when cognition no longer apprehends an object, then it stands firmly in consciousness-only, because, where there is nothing to grasp there is no more grasping. It is thus there arises the cognition which is homogeneous, without object, indiscriminate and supramundane. The tendencies to treat object and subject as distinct and real entities are forsaken, and thought is established in just the true nature of one’s own thought.77
The most detailed discussion of perception is found in the logical treatises of the Sautrāntika-Yogācārins Dignāga and Dharmakīrti, which begin by analyzing the process of perception into two familiar moments: “the first indefinite (nirvikalpa) sensation and the following thought-construction of a definite (savikalpa) image or idea and then purposive action.” According to them, problems arise because we confuse the two moments: mental construction converts the bare sensation, itself independent of any association with language, into an object that has a name. When one thinks he perceives such an object, “he simply conceals, as it were, his imaginative faculty and puts to the front his perceptive faculty,” thus missing the fact that the object which is supposed to be immediately perceived is a creation of thought-construction.78 According to Stcherbatsky, the distinction between these two moments is “one of the foundation stones upon which the whole system of Dignāga is built: Whatever is cognized by the senses is never subject to cognition by inference, and what is cognized by inference can never be subject to cognition by the senses.” In accordance with this, Dignāga and his successors accept only these two pramāṇas (modes of knowledge): sensation, which directly cognizes ultimate reality, and inference, including all conception, which indirectly cognizes conditioned or empirical reality. The path to liberation is again a return to the bare thing-in-itself: exclusive of all its relations and characteristics, which is “sense-perception shorn of all its mnemic elements.”79
This differs from Pāli Buddhism by explicitly claiming not only that such nirvikalpa sense-perception is the goal but also that it is nondual. Stcherbatsky concludes his translation of Dharmakīrti’s “Short Treatise of Logic” with the following note:
The trend of the discussion is to show that self-consciousness is not the attribute of a Soul, but it is immanent to every cognition without exception . . . our images are not constructed by the external world, but the external world is constructed according to our images, that there is no “act of grasping” of the object by the intellect, that our idea of the object is a unity to which two different aspects are imputed, the “grasping” aspect (grāhaka-akara) and the “grasped” aspect (grāhya).
The grasping aspect constitutes the sense of self, while the grasped aspect is the sense of a self-existing sense-object. How does this differentiation occur?
From the standpoint of Tathatā, there is no difference at all! But hampered as we are by avidyā, all that we know is exclusively its indirect appearance as differentiated by the construction of a difference of a subject and an object. Therefore the differentiation into cognition and its object is made from the empirical point of view, but not from the point of view of Absolute Reality (yathātathatam). (Jinendrabuddhi)80
From the highest point of view there has never been a differentiation, which is why sense-perception has really always been nondual. This does not need to be accepted on faith, for the claim that Reality is composed of discrete moments of pure sensation is verifiable. Both Dharmakīrti and Kamalaśīla recommend that we prove this ourselves by the experiment of staring at a patch of color without thinking of anything else, thus reducing consciousness to immobility. This will give us the condition of pure sensation, although we can realize that only afterward, when we begin to think again and reflect on what was experienced.81
Ch’an (Zen). Up to now, this chapter has discussed only Indian philosophy, but we will see that Ch’an Buddhism, which synthesized Mādhyamika and Yogācāra with the Taoism indigenous to China, is consistent with the above. Stanzas from the Hsin Hsin Ming of the third Ch’an patriarch Seng-ts’an were quoted in response to Conze’s treatment of “the three stages of apperception” in early Buddhism. Huang Po is also quoted, in chapter 1: “It [the One Mind] is that which you see before you — begin to reason about it and you at once fall into error.” Another early Ch’an master, Fa-yen Wen-i, said the same thing: “Reality is right before you, and yet you are apt to translate it into a world of names and forms.”82 In the sermons recorded in the Chun Chou Record Huang Po elaborates on this:
If you students of the Way seek to progress through seeing, hearing, feeling and knowing, when you are deprived of your perceptions, your way to Mind will be cut off and you will find nowhere to enter. Only realize that, though real Mind is expressed in these perceptions, it neither forms part of them nor is separate from them. You should not start reasoning from these perceptions, nor allow them to give rise to conceptual thought; nor should you seek the One mind apart from them or abandon them in your pursuit of the Dharma. Do not keep them nor abandon them nor cleave to them. Above, below and around you, all is spontaneously existing, for there is nowhere which is outside the Buddha-Mind.83
This passage is strikingly similar to what the Buddha said to Bahiya: do not reject perceptions, but do not infer any “therein” or “thereby” from them. This too stops short of clearly asserting nonduality, but elsewhere Huang Po denies any objective reality to sense-objects:
If you understand that these eighteen realms [the six sense-organs, -objects and -fields] have no objective existence, you will bind the six harmoniously-blended “elements” into a single spiritual brilliance — which is the One Mind.
It [the One Mind] is neither subjective nor objective, has no specific location, is formless, and cannot vanish.
If an ordinary man, when he is about to die, could only see the five elements of consciousness as void . . . his mind and environmental objects as one — if he could really accomplish this, he would receive enlightenment in a flash.84
Similar passages from many other Chinese Ch’an and Japanese Zen masters could also be cited, but I confine myself to discussing the Ten Oxherding Pictures of the twelfth-century master Kuo-an Shih-yuan. These well-known pictures, which illustrate the various degrees of enlightenment using the analogy of searching for an Ox, are also explicit in claiming that what is sought is found in perception itself. The third stage, “first glimpse of the Ox,” is the first “taste” of enlightenment. Kuo-an’s commentary on this picture gives instructions on how this glimpse can be attained.
If he will but listen intently to everyday sounds, he will come to realization and that instant see the very Source. The six senses are not different from this true source. . . . when the inner vision is properly focused, one comes to realize that that which is seen is identical with the true Source.85
It is because nondual perception is the Ox that the Ox has never been astray; as the verse says, “There stands the Ox, where could he hide?” The highest degree of enlightenment is reflected in the ninth picture “Returning to the Source,” in which one realizes, paradoxically, that one never left it. It depicts a flowering branch. “He observes the waxing and waning of life in the world while abiding unassertively in a state of unshakeable serenity. This [waxing and waning] is no phantom or illusion” but is how the empty Source expresses itse1f.86 As the Prajñāpāramitā repeatedly says, form may be no other than emptiness but emptiness is also no other than form. However, the verse to this picture seems inconsistent with such a nondualist interpretation:
It is as though he were now blind and deaf. Seated in his hut, he hankers not for things outside. Streams meander on of themselves, red flowers naturally bloom red.
“As if blind and deaf” is a common phrase in Ch’an literature. Sometimes it refers to the deluded man who has no insight, but often it praises those whose seeing and hearing are completely without any sense of duality — whose seeing and hearing are sometimes described as no-seeing and no-hearing. That is why the Ch’an master Hsiang-yen could be enlightened by the sound of a pebble striking a bamboo: he heard the nondual nirvikalpa sound, freed from any thoughts about it. It is when we do not use śūnya perceptions as a basis for thought-construction that nondual streams meander on and red flowers bloom by themselves.
In concluding this discussion of perception in Buddhism, we should notice a progression or development in the concept. The main theme, that the “bare concept” must be distinguished from its conceptual and emotional superimpositions, was established in the Pāli sutras. The claim that such perception is nondual becomes explicit in Mahāyāna, first negatively in the Mādhyamika critique of all dualities as relative and hence śūnya, then positively in the Yogācāra assertion that subject and object are not distinct. With Ch’an we see that philosophical claim put into practice. How meditation can lead to such nondual experience is discussed further in chapter 6. The same points could be made with reference to the tantric practices of Tibetan Buddhism, which rest upon the identical philosophical foundation of Mādhyamika and Yogācāra. It is no coincidence that the Vajrayāna technique of visualizing a deity is preliminary to the act of becoming that deity.
As if echoing Huang Po’s declaration that the One Mind is what you see before you if you do not reason about it, Śaṇkara claims that our perception of the universe is a continuous perception of Brahman, although the ignorant do not know it. (Vivekacūḍāmaṇi, v. 521, quoted at the beginning of the chapter). Perhaps this is no coincidence, since Gauḍapāda and Śaṅkara are known to have been influenced by Mahāyāna to the extent that both have been accused of being “hidden Buddhists.” Even more important for us, it is likely Vedānta borrowed the notion of prapañcopaśama to conclude that the thing-in-itself cognized in nondual nirvikalpa experience is Brahman.87
Although he had much to say about the pramāṇas (modes of knowledge), Śaṅkara, like the Buddha, developed no theory of perception. There are differences of opinion in Advaita over exactly how an object becomes manifest to a subject. The best traditional account, although unsophisticated and generally deemed unsatisfactory, is found in the Vedānta-paribhāṣa of Dharmarāja Adhvarīndra, which presents a theory of abhedabhivyakti — that the manifestation of the object is “non different” (abheda) from the consciousness underlying the subject. D. M. Datta summarizes that view as follows:
The Advaitins view immediacy as the basic character of the Absolute Consciousness, of which the knower, the known, and the process or mechanism of knowledge are apparent differentiations due to ignorance. So, for them, immediacy is not generated by the knowing process. The self’s knowledge of an external object is empirically describable, of course, in terms of the function of the mind, or internal organ, and the sense concerned. In the light of this, the Advaitins say that in every perception the mind flows out to the object through the sense and assumes the form of the object and establishes thereby a sort of identity between the mind and the object. But this process does not generate consciousness or immediacy. It only destroys the imagined barrier between the knower (which is nothing but the basic consciousness delimited by the mind) and the object (which is also the same consciousness delimited by the objective form) by a kind of identity established between the two delimiting and differentiating factors.
So, for the Advaitin every sense perception is really the restoration of the basic identity between the knower and the known, and the allowing of the basic reality, i.e., consciousness, to reveal itself immediately.88
This Platonic view — in which “the mind flows out to the object through the sense” — is in striking contrast to the more dualistic modern understanding, according to which sense-data are processed and perceived in the brain. As such, Datta’s account seems to support the claim that perception is originally nondual. But such a conclusion would be too quick. At least two qualifications are necessary.
First, the above account seems inconsistent in maintaining, on the one hand, that basic identity is “restored” by the mind flowing out through the senses, and, on the other, that this does not generate immediacy but only destroys an “imagined barrier.” It would seem that either the basic nondual consciousness is delimited by mind and objective form, in which case there is a real barrier and restoration of identity is necessary, or that barrier is imaginary and nondual consciousness has never been delimited, which means that it is necessary only to remove the “veil of ignorance” — which sounds like what we have been describing as savikalpa thought-construction. According to Advaitins, however, this inconsistency is only apparent and is due to confusing the two standpoints or levels of truth basic to Advaita: the vyāvahārika (empirical) and the paramārthika (transcendental). That consciousness is delimited and restored is true only from the empirical standpoint; from the transcendental perspective there has never been any barrier or delimitation. Advaitins insist on distinguishing strictly between these two standpoints, which parallel (and were influenced by) the two truths of Mahāyāna, saṁvṛtajñāna and paramārthajñāna. But solving the above problem by distinguishing these two standpoints just transposes the difficulty into another dimension, for what is the relationship between the two? The sharp distinction between them — so extreme that they cannot be related back together, as we shall see — brings us to the second qualification.
In the Advaita system, it would not be correct to say that perception is or can be nondual because by definition the nature of perception (pratyakṣa) is a vyāvahārika issue only. The Advaitin does not refer to nirvikalpa and savikalpa as two kinds of perception but reserves the term pratyakṣa for the latter only. Necessarily, then, perception is always dualistic. In response to Yogācāra “mentalism,” which as we have seen denies the existence of an object distinct from our cognition, Śaṅkara defends a realistic epistemology: the object is independent of our knowledge of it, for the two are utterly different. Objects depend upon our awareness of them only to be revealed; that awareness does not constitute them. Knowledge is nothing other than the element of revelation in our experience, although the “essence” of that revelation is indeterminable since our very oneness with it means we are unable to understand it objectively. But we can see that our sense-cognition, like all cognition, is due to vṛttis, the modifications of the buddhi (the “internal organ” constituting the mind as we usually understand it). Because these vṛtti-modifications always constitute limiting conditions of one sort or another, the unlimited Brahman cannot be knowable through them. The only real (paramārthika) is Brahman, the thing-in-itself. Apparently turning the Yogācāra view upside down, Śaṅkara argues that anything dependent upon being known is unreaI, mere appearance. The empirical world of appearances is real, but only as Brahman, only because it is grounded in Brahman.89
The rest strikes a more familiar note. The ideal of pure knowledge is to know the thing as it is, without the appearance of subjective representations, and this happens only in the knowledge of Brahman, when there is no distinction left between the knower and the known. Nirvikalpa awareness (the preferred Vedāntic term is aparokṣānubhūti) is not intuition of Brahman but itself is Brahman. Then the distinction between knower, knowing, and known is realized to be delusive, but until then we must respect the empirical distinctions between them. Advaitins sometimes express this by saying that, although only Brahman is real, the empirical world nonetheless exists.
Thus the Advaitic view drives much more of a wedge than Mahāyāna does between our usual sense-perception and the nondual experience. But how significant is this difference? Is it perhaps merely a linguistic one, over how to use the term pratyakṣa? In his critique of Yogācāra, Śaṅkara misses the point because he does not understand the function of the Yogācāra arguments. It has already been mentioned that Mahāyāna accepts the same distinction between empirical and transcendental that Advaita does (perhaps vice versa is more accurate since the Mahāyāna version came first), which is important because it supports the possibility that both “two-truth” doctrines are based upon the same nondual experience. Where they do differ is on their attitude toward the empirical “lower truth.”
Mādhyamika and Yogācāra are concerned to demonstrate the self-contradictory nature of our usual phenomenal world, including the pramāṇas, because such a logical refutation is their way of devaluing dualistic experience and paving the way for the nondual experience. Unlike some later Advaitins (e.g., Śrīharṣa, whose dialectic was much influenced by Mādhyamika), Śaṅkara is not interested in refuting the pramāṇas. He is content to accept them empirically and merely to state that they have no application in the sphere of ultimate knowledge, for when Brahman is realized the pramāṇas lose all authority as a means of right knowledge.90 But Śaṅkara can do this only because he relies on śruti (the Vedic scriptures) to establish the existence of Brahman, whereas the Buddhist philosophers, not accepting such an appeal to authority, need to analyze the relation between the two levels of truth, logically refuting the lower for the sake of experiencing the higher.
This means that the epistemological difference between them is more apparent than real. Śaṅkara’s defense of a realistic epistemology does not refute Yogācāra “mentalism,” as he thinks, for what he criticizes is not the Buddhist empirical standpoint, but the Buddhist means of moving logically from the empirical to the transcendental. What is more important is the agreement of both sides that our usual empirical experience is a dualistic lower truth and that the transcendental is nondual. Advaitins often criticize Mādhyamika for dismissing this phenomenal world as “empty,” whereas for them it exists; but this is the same confusion of standpoints, for Advaita makes the identical point in a different way by insisting that only Brahman is real. The difference between them comes down to the fact that, for Advaita, the two standpoints are so sharply distinguished that they have no commerce, whereas Mahāyāna — and this work, obviously — is more concerned to understand the relation between them.
A number of aspects of the Advaita Vedānta system reveal its otherwise deep affinity with Mahāyāna on this matter. A full consideration of the most crucial issue, the relation between śūnyatā and Brahman, is reserved for later chapters, but we now turn to consider some of those other aspects.
The veil of ignorance. The claim of Datta’s passage, quoted above, is that the realization of nondual consciousness requires only that the “veil of ignorance” (ajñāna) be eradicated. As in Sāṅkhya-Yoga, this veil is destroyed by the buddhi in a mental modification (Brahmātmakāra-vṛtti) that realizes the identity between the self and Brahman. What distinguishes this vṛtti from other limiting vṛttis is that it then becomes extinct by consuming itself. “It is to the buddhi and not to the Self which is immutable that the knowledge ‘I am Brahman’ belongs” (Śaṅkara). But that knowledge is not yet the full nondual experience.
Brahman being self-effulgent, nothing is needed for its manifestation. The mental mode coincident with It removes the veil of ajñāna but does not objectify it. By no means can Brahman be a object of cognition.
. . . self-luminous Brahman is unveiled and is directly perceived as “I am Brahman”; instantly, the Self is realized as indivisible, limitless Pure Consciousness in nirvikalpa samādhi. This is illumination (prajñā), or immediate awareness (aparokṣānubhūti) beyond the distinction of the knower and the known, the self and the not-self.91
The importance of this for us is its claim that in order to realize Brahman nothing needs to be gained or added; an obscuring veil needs only to be removed. Since this veil includes the Brahmātmakāra-vṛtti of the buddhi, this amounts to eliminating all savikalpa thought-determinations. The result of this process is not a realization that the ego-self “has,” for that would still be dualistic and Brahman is nondual.
Self-luminosity. One of the most common descriptions of Brahman, used twice in the previous quotation, is “self-luminous” (svayaṁ-prakāśa). According to Surendranath Dasgupta, svayaṁprakāśa in Vedānta refers to “that which is never the object of a knowing act but yet is immediate and direct with us. Self-luminosity thus means the capacity of being ever present in all our acts of consciousness without being an object of consciousness.”92 Discussing a related matter — the importance of luminosity (photism) in Tibetan Buddhism — Tucci elaborates on what this means:
To understand these ideas better we need to recall the difference between the luminosity we are considering here and that of, for example, the sun’s rays. The sun’s rays enable us to see, in that they illuminate objects, but they do not see themselves. The luminous cognitive states on the other hand do not only illuminate what is cognizable inside and outside of us, they also illuminate themselves as objects of luminous cognition. Thus it comes about that in the cognitive process luminosity and cognition belong essentially to each other. If the luminous states as a result of adventitious defilements do not illuminate the objects proper to them, they still possess as cognitive states this power within themselves.93
Unlike Dasgupta, Tucci’s explanation refers to objects, but the point is the same, since “objects that illuminate themselves” are not really objects as usually understood. It is rather Tucci’s last sentence that illuminates where Vedānta and Mahāyāna differ, for Advaita distinguishes more sharply between self-luminous Brahman and perception which is dualistic because of adventitious defilements. Brahman is the self-luminous thing-in-itself because it is not dependent on anything else for its manifestation — unlike all supposed objects of consciousness, which, as dependent on a subject to be conscious of them, are mere appearances. But what if “I” were to realize that is true for all “my” experience right now?
Superimposition. In his discussion of the “three stages of apperception” in Pāli Buddhism, Conze twice uses the term “superimposition” to describe the relation between the initial (nirvikalpa) sensation and the succeeding thought-constructions of recognition and volitional reaction. Superimposition is a fortuitous term for this tripartite process, since one of Śaṅkara’s major contributions is that he uses it to describe the “relation” between Brahman and the phenomenal world.94 “Adhyāsa [superimposition] is the apparent presentation by the memory of something previously perceived elsewhere.” This seems to echo Dignāga’s emphasis on the role of the mnemic function in transforming nirvikalpa into savikalpa. To understand adhyāsa, there are the well-known analogies of a snake that on closer inspection turns out to be a coiled rope, and a silver coin that is actually a piece of mother-of-pearl shell. The point of the analogy is that the phenomenal world may be viewed as due to such a superimposition. Taking the pluralistic universe of material objects to be real — which constitutes māyā — is like seeing a rope as a snake. Just as we would say afterward that we had really been seeing a rope, so “we” must actually be perceiving (or “experiencing”) Brahman all the time, although we are ignorant of it — exactly what Śaṅkara says in the Vivekacūḍāmaṇi verse quoted at the beginning of this chapter. The analogy would further seem to imply — indeed, it can hardly be meaningful otherwise — that Brahman should not be characterized as transcendental to sense-perception, although of course “It” cannot be perceived as an object.
In one sense, superimposition is an unfortunate term, since it suggests the Platonic Form or Kantian noumenon dualistically “behind” phenomena — and that is the way it has sometimes been understood. For example: “Māyā is energized and acts as a medium of the projection of this world of plurality on the nondual ground of Brahman.”95 Perhaps “on” here is meant only metaphorically, but it is a dangerous metaphor. If Brahman is a nondual ground, the realization of which is characterized by self-luminous immediacy, then it cannot be projected upon because it is not something objective — an objection that Śaṅkara was quick to raise against his own definition of adhyāsa in the preamble of the Brahmasūtrabhāṣya. He replies that, although Brahman cannot be an object of pure consciousness, it can still be the apparent object of empirical (i.e., dualistic) consciousness. It is difficult to understand what this could mean except as asserting the nonduality of enlightened cognition.
Māyā. Since māyā is “the chief characteristic of the Advaita system” (Radhakrishnan), it is significant for the claim of this chapter that, although māyā is said to be indeterminable (anirvacya), it is nonetheless identified with “names and forms” (nāmarūpa) that in their evolved state constitute the phenomenal world.96 “All variations are due to the superimposition of names and forms by māyā on immutable Brahman.”97 This conception of māyā is so similar to the Buddhist usage of prapañca that it suggests the former, like the latter, might be understood to be due to the savikalpa determination of nirvikalpa percepts. In such statements as māyā “is an appearance only” and “the whole phenomenal world is the appearance of Brahman,” the fact that appearance must ultimately be negated may be taken not as referring to some transcendental realm apart from sensory experience but as denying the apparent “objectivity” of sense-experience, that there are discrete, self-existing objects.
Although scholars may argue about who influenced whom, the cross-fertilization of the nondualist systems is undeniable here. Not only do we find the term prapañca in the post-Buddha Upaniṣads, but māyā is also an important term in Mahāyāna. Mādhyamika understands māyā in its more original meaning of “magic trick” and then applies this to the world, whose objectivity is as delusive as the illusions of a magician. Śaṅkara makes māyā more “concrete” when he also describes it as a positive, beginningless material causal force responsible for creating the world. But, again, this may not be a significant difference; in fact, it may clarify our understanding of prapañca. To identify māyā and prapañca need not reduce māyā to our imagination: it may serve the opposite function of expanding the effects of our thought. If the objectivity of the world — that is, the objective world — is due to our ways of thinking about it, it suggests that the significance of our thought-processes is much greater than we usually believe, in a way which also makes sense out of the doctrine of karma. Śaṅkara agrees with Buddhism not only that avidyā never belongs to anyone or anything, but also (unlike Gauḍapāda) he seems to identify māyā and avidyā. If there is no self that is deluded, if on the contrary the sense of self is due to delusion, then avidyā must be “transpersonal” and is elevated to a more cosmic role; it is no longer merely our own personal ways of thinking. Thus māyā refers to the fact that the delusive objectification of the world has a “collective” dimension which transcends the delusion of particular individuals.
Four important aspects of the Advaita system have been briefly discussed: the realization of Brahman as requiring only the eradication of the veil of ignorance, the nature of Brahman as “self-luminous,” and Śaṅkara’s concepts of adhyāsa and māyā. In all four cases we see significant parallels between the Advaita position and that already developed in the Buddhist sections of this chapter. The disagreements seem to concern linguistics — for example, how to use the term pratyakṣa — more than the relationship between dualistic savikalpa perception and nondual nirvikalpa perception/intuition. But in fact the linguistic issue is only the surface of a more profound difference to which we now return.
A person’s main character weakness is often the other side of his greatest character asset. For instance, someone who shows great determination when that is necessary is also likely to be unreasonably stubborn when it is not. Sometimes the same is true for philosophical systems. The strong point of Advaita is its uncompromising solution to the problem of the relation between the Absolute and the phenomenal world (a problem which does not exist for Mahāyāna, for it understands the Absolute as nothing other than the “emptiness” of phenomena). For Śaṅkara, the problem of how the world was created does not arise because, as for Nāgārjuna, there never was a creation. But unlike Mahāyāna, Śaṅkara does away with this issue by sublating the phenomenal world into illusory māyā, which is completely devalued in comparison with Brahman. The corresponding problem that arises with this solution is the difficulty of characterizing the nature of māyā, which is neither real (since it has no existence apart from Brahman) nor unreal (since it does project the world of appearance). It is not surprising that māyā is ultimately deemed indeterminable and indefinable, which in my opinion amounts to an admission of failure. The problem has merely been pushed back one step: that which has been postulated to understand the “relation” between the Absolute and the phenomenal world cannot itself be understood. That Brahman both does and does not incorporate māyā is a problem that cannot in principle be resolved in Advaita, but conceiving of Brahman as the self-luminous nirvikalpa percept, the view defended in this chapter, provides an explanation: the difference between māyā and Brahman is the difference between percepts bifurcated into subject and object and those same percepts experienced nondualistically. This is why Mahāyāna could equate saṁsāra and nirvaṇa. According to Mahāyāna, the nirvikalpa ground of perception remains the same whether perceptions are “grasped” dualistically or realized to be nondual. But Advaita Vedānta would not accept this form of “nonduality,” preferring to characterize Brahman as nirguṇa and beyond all perception.
In the end, it turns out to be difficult to distinguish such an unqualified being as Brahman from the śūnyatā of Mahāyāna, as we shall see in chapters 5–6. But there is still an important difference. For Mahāyāna, śūnyatā is not a category distinct from phenomena but a statement about their lack of self-nature. As the Prajñāpāramitā Heart Sutra says, emptiness is not other than form. In Advaita, however, a wedge has been driven between attributeless Being and phenomena, between the higher and the lower truths. For Mādhyamika, the “two-truths” doctrine is a shorthand way of expressing a difference between two modes of experience: what we normally experience as real is, from the perspective of another experience, unreal. The two levels of truth salvage the truth of both experiences by subsuming the one below the other. From the viewpoint of nondual experience, the dualistic lower truth is untrue, but nonetheless we must all dwell in that delusion to some extent in our everyday lives — which is why it is not merely delusion. As my analysis has shown, the problem with the Advaitic position is that these two truths have been sundered; that version of the doctrine has since become reified into a orthodoxy which paralyzes the possibility of developing fresh ways to understand the old truths. Anyone who accepts the “perennial philosophy” will accept some version of the two-truths doctrine, but the challenge for philosophy is elucidating the relation between them. The simplest refutation of such a split is to realize that there must be some relation or liberation would not be possible, because we could never make the transition (or “leap”) from delusion to enlightenment. This is not to overlook the importance of the distinction between the two levels. The point of that distinction is that we must not attempt to understand the transcendental from the relative or empirical perspective. But that is not the function of Mahāyāna analysis, which rather demonstrates the self-contradictions of our empirical experience in order to undermine our commitment to it. Nor is this world liable to such an objection, for my project is not to “extrapolate” from the lower truth to the higher but to “interpolate” by using the traditional nondualist claims about the higher truth to examine our understanding of the lower.
The Eye of Man, a little narrow orb, clos’d up & dark,
Scarcely beholding the Great Light, conversing with the ground:
The Ear, a little shell, in small volutions shutting out True
Harmonies & comprehending great as very small . . .
— William Blake, Milton
The previous section established that the nonduality of perception is a central tenet of some important Asian philosophies, particularly Buddhism and (with some qualifications) Advaita Vedānta. But it did little to reduce the oddity of that claim, so incongruous with all our common sense. We need to elaborate on what such a claim might mean, although without hope of being able to grasp this matter completely. We cannot hope to understand nondual perception clearly through concepts if our usual dualistic perception is delusive precisely because it is conceptualized.
Almost everything said so far has referred to sense-perception in general. These generalizations must now be grounded in a discussion of the two most important senses. Since hearing is the easier of the two to “understand” nondualistically, it is examined first. Although nondual hearing is by no means common, music is probably the medium of most nondual experiences. The “silence” that nondual hearing reveals will help us to understand better the difference in perspective between Mahāyāna Buddhism and Advaita Vedānta, which are so similar and yet also diametrically opposed. Our discussion of nondual seeing will use the arguments of Berkeley and Hume to help us understand the nondualist critique of the visual object as material, discrete, and self-existing. In contrast, the nondual “Light-object” is a śūnya, self-luminous event.
■ ■ ■ ■
. . . music heard so deeply
That it is not heard at all, but you are the music
While the music lasts.
— T. S. Eliot, The Dry Salvages
Chapter 1, in a discussion of nonduality in Zen, included some quotations from a contemporary Japanese Zen master, Yasutani Hakuun. These were from his dokusan (private interviews) with Western students. One of those statements dealt specifically with the nature of hearing:
There is a line a famous Zen master wrote at the time he became enlightened which reads: “When I heard the temple bell ring, suddenly there was no bell and no I, just sound.” In other words, he no longer was aware of a distinction between himself, the bell, the sound, and the universe. This is the state you have to reach.
Yasutani elaborates on this in another dokusan with a different student.
Usually when you hear a bell ringing you think, consciously or unconsciously, “I am hearing a bell.” Three things are involved: I, a bell, and hearing. But when the mind is ripe, that is, as free of discursive thoughts as a sheet of pure white paper is unmarred by a blemish, there is just the sound of the bell ringing. This is kenshō [enlightenment or self-realization].98
While such nondual hearing can hardly be said to be common, neither is it confined to adherents of the nondualist Asian traditions. The lines from T. S. Eliot quoted above clearly allude to a very similar experience, and other examples could be cited. Eliot’s is especially interesting because it refers to the medium by which most nondual experiences probably occur. The experience described is unmistakably nondual. Not only is there no hearer, but there is no objective music that is heard. It doubtless records an experience that Eliot had, perhaps many times, and that I suspect many people have had occasionally. One literally becomes “absorbed” into the music; the sense of a self that is doing the hearing fades, and at the same time the music ceases to be something “out there.” Especially if the musical work is a familiar one, we normally (and dualistically) hear each note or chord in the context of the whole phrase, by remembering the previous notes and anticipating the ones to come, as if the whole phrase were simultaneously present before us and we “read” it from beginning to end. But this is an example of mnemic savikalpa determination of the nirvikalpa sound. This changes in the nondual hearing: no matter how well I may know the work, I cease to anticipate what is coming and become that single note or chord which seems to dance “up and down.” Music is the ideal medium for nondual experience, since we listen to it for enjoyment — that is to say, we listen for no other reason or intention outside itself; we do not need to assign the sounds a meaning, which is to have them refer to something else. The sound need not be a sound of something, and without any such thought-construction we have “a pure sound, a bark without its dog” (Neruda). For those of religious inclination, like Eliot, such moments of nondual hearing have a spiritual or mystical quality, but I suspect that for all those who have had them they are cherished as a “heightening of consciousness.” This is despite the fact that at the time one cannot be said to be aware of oneself “enjoying” the music, for when I do become aware of myself as enjoying, the nonduality of the experience has already faded away into dualistic hearing and it cannot be brought back by any effort of will or attention. Nondual experience cannot be repeated or produced by the self because it is something that happens to the self — the sense of self evaporates temporarily. One can only create conditions where this is more likely to occur (e.g., meditation), but even then the expectation of such an experience will interfere with its occurrence, as experienced meditators know.
There is another aspect to nondual hearing, which is brought out clearly in a letter by the recent French philosopher Simone Weil. She wrote that she was in the habit of reciting the “Our Father” in Greek each morning with absolute attention — in other words, her prayer was a meditation exercise.
At times the very first words tear my thoughts from my body and transport it to a place outside space where there is neither perspective nor point of view. The infinity of the ordinary expanses of perception is replaced by an infinity to the second or sometimes the third degree. At the same time, filling every part of this infinity of infinity, there is a silence, a silence which is not an absence of sound but which is the object of a positive sensation, more positive than that of sound. Noises, if there are any, only reach me after crossing the silence.99
It is not clear from this account whether Weil’s experience can be called nondual, but it contains an aspect that pertains nondual hearing: along with the sound there is also an awareness of that which is beyond the sound, which in the context of sound is silence, but it is a silence that is “heard” — “the sound of no sound,” we might say. (We shall meet with this curious paradox in later chapters also. For example, nondual action is “the action of nonaction” — Chinese, wei-wui-wei — and nondual thinking has been called “the thought of no thought.”) This is part of what Mahāyāna means by the “emptiness” of phenomena: when a sound is experienced as not referring to anything else (not the sound of a barking dog), then “in place of” (we might say) an awareness of the thought-constructed referent (dog), there is instead an awareness of silence. This is how one is able to “stop the sound of that distant temple-bell” (a common Zen koan); when one becomes that nondual “bong!” one also becomes aware of that silence “beyond” — that is, the “emptiness of” — the sound. (This paragraph has thrice referred to an “awareness of silence,” but of course this dualistic mode of expression should not be taken as implying that the experience of the silence is dualistic. Rather, the nondualist position is that the silence and the consciousness of the silence are not two.)
What is the relation between the nondual sound and this silence that is also “heard”? The answer to this question reveals the difference in perspective between Advaita and Mahāyāna. In Weil’s account the two seem to be distinct: noises must cross this silence in order to be heard. Advaita, which distinguishes the attributeless Absolute from all ephemeral phenomena, would agree with that: in this case, Brahman corresponds to the “heard silence” and noise serves as an example of delusive phenomena to which we cling by superimposing names-and-forms, with the consequence that we never “hear the silence” which is always there, unchanging. The answer of Mahāyāna is slightly but significantly different. It accepts the above analysis with the proviso that noise is not merely something that conceals silence but is itself an expression or manifestation of the silence. Mahāyāna allows no duality between the silence and the sound. From one perspective, we may say that noise (or sound) is how the silence manifests itself; from another perspective, that silence is the “underside” of the sound, revealing that the sound has no “self-essence” (svabhāva). What is important is that the same nondual experience can lend itself to both interpretations — and others as well (discussed at greater length in chapters 6 and 7.)
■ ■ ■ ■
For many years snow has covered the mountain,
This year the snow is the mountain.
— Dōgen
Vision, by far the most important sense, is also the most difficult one to understand nondualistically. That our understanding of experience is dualistic may be due to the fact that vision has tended to serve as the “standard case” for perception generally and therefore as the measure for all the other senses — and for knowledge as well, which is why most languages abound in visual similes for knowing. We are inclined to distinguish the “sound heard” from the “objective sound over there,” because we follow the model of vision, which seems to require a tripartite ontology, distinguishing the one who sees from the visual appearance (which changes according to perspective) as well as from the visual object (which is understood to persist unchanged). Vision provides us with a “co-temporaneous manifold,” whereas all other senses construct their perceptual “unities of a manifold” out of a temporal sequence of sensations. The predominance of vision thus gives us a different sense of time than that of all the other senses: the present is not just the passing now but also a dimension where things can be observed to remain the same. “Only sight therefore provides the sensual basis on which the mind may conceive the idea of the eternal, that which never changes and is always present.”100 This makes possible the philosophical distinction drawn by Plato and Advaita between Being and Becoming, the former conceived of as an immutable reality that persists “behind” the deceptive world of change.
What do we actually see? This question throws us into the long-raging philosophical controversy over whether it is correct to say that we see physical objects or whether in fact there are only “sense-data” (e.g., an elliptical visual image) from which the physical object (a round plate) is mentally constructed. It is important not to settle this question linguistically by appealing to the ordinary usage of language, for whether or not the bare nirvikalpa percept can be properly said to be what we see, the issue is the relation between eye-consciousness and thought: whether and in what way the apparently objective physical world is constructed by their prapañca-interaction.
Normally — that is, in a nonphilosophical context — we know how to answer readily enough: we see pens, cups, books . . . physical objects, which have weight, color, and so forth. If we delve into the meaning of what it is for something to be a physical object, we find three characteristics which are important to the nondualist because he wants to deny them all:
Matter. An equivalent for the “physical” in “physical object” is “material.” That objects are material means that they are composed of matter. We take matter to be an independent, self-existing stuff, which is real if anything is real, but our experience of this matter is largely confined to two of its aspects: that it is the source of visual images and that it is impermeable. One material object is usually impermeable to another. The cup is solid to my touch; neither my finger nor water can penetrate through it, which is why it can function as a cup.
Self-existence. A physical object is self-existent. It has an existence of its own which is not dependent on other objects or on subjects (a consciousness that is aware “of” it), although it may be affected by them. The cup conditions other objects and is affected by them, but it still has its own existence until it is destroyed. This concept of svabhāva differs from that of Mādhyamika, according to which nothing that has self-existence could ever be changed or destroyed, but it does embody the commonsense notion. The bubble may have a very short life, but it nonetheless exists until it pops.
Persistence. A corollary of the previous characteristic is that the object tends to persist unchanged unless affected externally by something else. It is easy to think of counterexamples to this, but they do not refute the fact that this describes our usual notion of what an object is like: it stays the same unless interfered with. The cup does not change unless someone else chips it or drops it on the floor.
When the key characteristics of the visual object are specified in this way, the arguments of the nondualist against its objectivity are predictable.
Against matter. Following the example of Berkeley, the nondualist can deny that we ever see such a thing as matter or a material object; given the nature of the eyes, all we can ever see is light. As Berkeley maintained in his New Theory of Vision, the notion of matter is a thought-construct created by combining the perceptions due to sight (that is, light) and touch (impermeability, etc.).101 Strictly speaking, we can never see the impermeability of any object. That I see it as impermeable is part of the savikalpa determination of the luminous nirvikalpa percept. In his discussion of causality, Hume remarked that Adam could not have inferred from the fluidity and transparency of water that it would suffocate him.102 The nondualist would add that Adam could not have inferred from the sight of water how it would feel to the touch. Of course this relating-together of the senses must occur quite early and is now so automatized or “subconscious” that it is normally impossible to see “objects” as other than impermeable. Nonetheless, the nondualist claims that this thought-constructed “unity of apperception” can be undone.
Such undoing must include eliminating the subject of perception. From his claim that we do not “see” distance, Berkeley too quickly inferred that all visual objects are really in the mind, which he understood subjectively. He would have done better to argue as Hume did, that in experience itself there is nothing corresponding to a self:
I never catch myself at any time without a perception, and can never observe anything but the perception. . . . All our particular perceptions are different, and distinguishable . . . and may exist separately, and have no need of anything to support their existence.103
By combining these two empiricist arguments — Berkeley against the material object that is perceived, Hume against the subject that perceives — the meaning of the claim that perception is nondual emerges more clearly.
If there is only light, with no physical objects to be seen and no seer to see them, then light must be very different from what we normally take it to be and from the way phenomenalists usually describe sense-data. Our usual understanding of light is dependent upon a dualistic ontology, which relegates it to the role of medium between object and subject, mechanistically reflecting off the one into the eye of the other. But if there is no such subject–object ontology, light must be reevaluated to incorporate not only the object that it is believed to refer to, but also the consciousness which is usually believed to be aware of it. This means that visual “things” are composed not of matter but of something which we might term “Light,” and such “Light-things” are śūnya because they do not “refer” to anything else (e.g., a material substratum) when they are experienced as they are in themselves, nondually.
The many references to light in the religious and “illuminative” traditions suggest this. For example, there is the “self-luminosity” (svayaṁprakāśa) of Brahman:
The sun does not shine there, nor the moon and the stars, nor these lightnings — not to speak of this fire. He shining, everywhere shines after Him. By His light all this is lighted.
[Brahman] is the Light of lights; It is that which they know who know the Self.
They [knowers of Brahman] see everywhere the Supreme Light, which shines in Brahman, which is all-pervading like the light of day.104
A similar view of luminosity is central to the Tibetan tradition:
In the entire course of the religious experience of Tibetan man, in all of its manifestations from Bon religion to Buddhism, a common fundamental truth is evident: photism, the great importance attached to light, whether as a generative principle, as a symbol of supreme reality, or as a visible, perceptible manifestation of that reality; light from which all comes forth and which is present within ourselves.
. . . the connection between light and mind, which is defined as “nonduality of the profound and the luminous,” characterizes the state of transcendent consciousness. . . . the connection between sems [transcendent nondual consciousness] and light, and the identity of these two terms, forms the basis of Buddhist soteriology in Tibet.105
Such luminosity of Mind is inconsistent with most Pāli interpretations of Buddhism, but in the Dīgha Nikāya there is a curious passage where the Buddha says that in nirvana “there is this consciousness without a distinguishing mark, infinite and shining everywhere.” In the Aṅguttara Nikāya also the Buddha describes this consciousness as “luminous” (pabhassara) and freed from adventitious defilments.106 Many other references, both Eastern and Western, could of course be cited. There is not a lack of such allusions but rather so many that we no longer notice them and their significance is lost. We take the reference as metaphoric, but perhaps it is literal. Maybe “the Great White Light” of Bon and Tibetan Buddhism is nothing other than what light really is if it were “seen” as it is.
Against self-existence. If visually there is only nondual Light and if everything we now consider to be a material object is self-luminous, this explains why according to the nondualist traditions there are no sentient beings: there is only sentience. The concept of a sentient being has meaning only in contrast to something nonsentient. In place of this negated dualism (and its negated corollary, life versus death), “all phenomena are like a dream, an illusion, a bubble and a shadow, like dew and lightning,” as the Diamond Sutra concludes.
The Bodhisattva does not really save any sentient beings, because there are none to save. On this point Śaṅkara, Lao Tzu, and the Christian mystic Eckhart agree with Mahāyāna.
Eckhart: “All creatures in so far as they are creatures, as they ‘are in themselves’ (quod sunt in et per se) are not even illusion, they are pure nothing.” Śaṅkara: “This whole multiplicity of production existing under name and form, in so far as it is Being itself is true. Of itself (svatas tu — that is, as self-existing) it is untrue.”107 Chapter 5 of the Tao Tê Ching describes the sage as not humane, for he regards the people as “straw dogs”; there is only the Tao, itself void like a bellows. There is still some difference, sharpest between Vedānta and Mahāyāna. For Śaṅkara, creatures are true insofar as they are Being (Brahman) — a formulation Mahāyāna would not accept since it denies any Being. Both explanations may be viewed as opposite solutions to the old problem of how something (in this case, consciousness) can arise out of nothing (nonconscious matter). In fact, their answers are the only two possible solutions: either there never was a “nothing” — that is, “matter” was never nonconscious because it was always self-luminous (the Advaitic solution) — or sentient beings are still nothing (the śūnyatā of Mahāyāna). As we shall see later, these two formulations are not really opposed, for in the final analysis the choice between them becomes linguistic. “It is difficult indeed to distinguish between pure being and pure non-being as a category” (Dasgupta).108
Against persistence. We usually distinguish between the visual appearance of an object, which changes as our perspective or the light changes, and the object, which is believed to persist unchanged. If there is no physical object, then there is nothing to “stay the same” and the distinction we make between objects and their interactions — between things and events — crumbles. It might be objected that nondual Light may remain constant, but the sense of the same is different in the two cases. It is part of what we mean by a material object that its staying-the-same does not need to be explained, for it is the nature of what we understand as matter to do that unless disturbed in some way. But the staying-the-same of a Light-thing must be an active persistence or enduring; its continued presence is an act, we might say. This notion some Buddhist schools attempted to express in the claim that Reality is momentary (kṣaṇika) — a formulation which, I argue later, is only half true. Perhaps Heidegger makes the same point better when he said that “things ‘thing.’” The nondualist experience is not self-existing objects that interact causally, but empty events or processes. The cup “dwells” on my desk. The concept of an object is a “shorthand” way of accounting for the fact that certain Light-events tend to persist and change in a predictable pattern. This stability allows us to relate these events causally and form expectations. In itself, such a shorthand is obviously very useful, but when it becomes so automatized that we forget it is shorthand, then we mis-take the persisting event (e.g., the self-luminosity of Brahman) as a physical object (self-existing matter).
According to the nondualist, then, what is seen? Instead of a self-existing, material object, which passively persists unchanged, there is śūnya, self-luminous sentience, which actively dwells.
This chapter cannot conclude without showing the place of this theory of nonduality in the context of Western epistemology and indicating, although briefly, how it meets some objections that may be raised against it. We will see that, in the terms of modern epistemology, nondual perception is best understood as a version of phenomenalism. Nondual phenomenalism escapes some of the objections that have been raised against other phenomenalistic theories, but not the main difficulty: how to account for the role of the sense-organs. Responding to this problem propels us into a deeper understanding of the issues involved. But first I consider some contemporary objections to the notion of “pure sensation” without any conceptual superimposition.
■ ■ ■ ■
Although twentieth-century epistemology accepts that perception involves conception, many philosophers have objected to the implication (e.g., in sense-datum theories) that our primary level of experience is composed of “pure sensations” stripped of any reference to objects in the perceived world. They claim that such “bare percepts” do not constitute the building blocks of our lived-in world, for they are the artificial products of intellectual analysis and could never be used to reconstruct the intentional structures of conscious experience. According to this view, introspection gives us no evidence of such indeterminate percepts distinguishable from perceived objects, and it is therefore correct to say that what we do immediately intuit in perception is objects. This is in criticism of claims such as this one by Berkeley:
When I hear a coach drive along the street, immediately I perceive only the sound: but, from the experience I have had that such a sound is connected with a coach, I am said to hear a coach. It is nevertheless evident that, in truth and strictness, nothing can be heard but sound and the coach is not then properly perceived by sense, but suggested from experience.109
Heidegger disagrees with Berkeley: “what we hear in the first instance is never noises and sound-complexes, but the creaking wagon, the motorcycle. . . . It requires a very artificial and complicated frame of mind to ‘hear’ a ‘pure noise.’”110 Yet (to make Berkeley’s point), had I never seen or heard one before, I would not be able to say that what I am hearing is a motorcycle. So the issue is whether, once I am familiar with motorcycles, there is a conscious inference from sensation to perception, an inference that I can recognize through introspection. Then the negative answer of Merleau-Ponty and Heidegger is certainly the correct one: once I am familiar with “the sound of a motorcycle,” that sound is normally not distinguishable from its source; so what I hear is the motorcycle. But rather than argue over the correct use of the word hear, what is important here is that inference cannot be denied; it is simply that the inference is so automatic that it is unconscious.
Normally . . . perceptual consciousness seems intuitive — that is, without interpretation and quite unanalyzable; except in perceptual reduction its content almost always consists of ostensible objects. All the same, psychological evidence shows that there is a range of subjective processes. . . . Perceptual consciousness is introspectively a whole but must be supposed to be a product of a range of selective, supplementary, integrative or organizational, and quasi-interpretative processes acting on a supposed basic sentience. But — and this is the point — both processes and sentience are unconscious and so may plausibly be regarded as cerebral activities or adjustments of the nervous system. However, since we cannot as yet give any precise neurological statement of these processes, we have to describe them as if they were conscious, basing the description on the difference between the input to the senses and the finished product, but this product (perceptual consciousness) does not reveal within itself the processes that may be supposed to form it. (R. J. Hirst)111
Here the philosophy yields to psychology, and it is significant that one recent scientific study concludes that in perception the conceptual element plays an even larger role than sensation:
Perception seems to be a matter of looking up information that has been stored about objects and how they behave in various situations. The retinal image does little more than select the relevant stored data. . . . We can think of perception as being essentially the selection of the most appropriate stored hypothesis according to current sensory data. (Richard Gregory)112
Evaluated according to our everyday experience, this is not implausible. Our minds are usually so preoccupied with various intentions that we do not so much observe objects as infer their presence on the basis of the most cursory glance. Another way to describe such “intentional perception” is to say that normally observation is selective.
It needs a chosen object, a definite task, an interest, a point of view, a problem. . . . A hungry animal . . . divides the environment into edible and inedible things. An animal in flight sees roads to escape and hiding places. . . . Generally speaking, objects change according to the needs of the animal. (Karl Popper)113
But what happens if one simply observes without any task, point of view, or needs — as occurs during some types of meditation? If one were able to let go of all intentions, might one come to perceive in a very different way and realize something hitherto unnoticed about those perceptions? It may be granted that the inferences we undeniably make are unconscious at the moment we make them, for they are not observable through normal introspection; but that does not imply that they must remain unconscious and that there are no techniques by which they may be brought to consciousness. We know from psychoanalysis that it is possible to re-expose to consciousness memories and emotional responses which have been long repressed. There is no reason to assume, as Hirst does, that the same is not true of perception. This of course does not settle the issue but makes it one that can be resolved only empirically — that is, experientially — a turn which is agreeable to the nondualist, who invites us to realize this for ourselves in meditative samādhi.
Most of the scientific research into meditation and samādhi has been concerned with its physiology, but I am acquainted with two scientific experiments whose results seem to support the possibility of nondual perception. These experiments were conducted by Dr. Arthur J. Deikman, of the Austen Riggs Medical Center in Stockbridge, Massachusetts, and he reported on them in the April 1963 and February 1966 issues of The Journal of Nervous and Mental Disease (from which I quote below).
In both experiments the subjects (four in the first, six completed the second) sat in a simple room with subdued lighting and were asked to concentrate on a blue vase ten inches high; the number and length of the sessions varied. After excluding those percepts readily explicable in terms of such familiar concepts as after-images, phosphenes, and stabilized retinal images, there remained certain unfamiliar phenomena experienced by all the subjects. The color of the vase shifted to a deeper and more intense blue, frequently described as “more vivid.” “The adjective ‘luminous’ was often applied to the vase, as if it were a source of light.” “It was as if light were coming from it.” “It started radiating. I was aware of what seemed like particles . . . [that] seemed to be coming from the highlights there and right at me.” Another effect was an instability in the shape of the vase. Its size and/or shape changed; it appeared to become two-dimensional; and there was “a diffusion or loss of its perceptual boundaries.” Often the vase appeared to move: “rocking,” “drifting,” “wavering.” “The outlines of the vase shift. At that point they seem almost literally to dissolve entirely . . . and for it to be a kind of fluid blue . . . a very fluid kind of thing . . . kind of moving.” “. . . Things seem to sharpen and there is a different nature to the substance of things. It’s as though I’m seeing between the molecules . . . the usual mass of solidity loses its density or mass and becomes separate.”
Solid material such as myself, the vase and the table . . . seem to be attributed then with this extra property of flexibility such as in its natural, fluid state. It’s almost as though we are, myself and the vase and the door, a form which has lost its fluidity the way water loses its property of fluidity when it’s frozen.
All Deikman’s subjects reported that the vase lost its solidity and rigid boundaries, becoming more fluid and formless; yet, paradoxically, this made it seem even more vivid and real to them. Subjects often used the term “feeling” to describe these experiences, meaning not touch or emotion, “but rather perception that cannot be located in the usual perceptual routes of sight, hearing, and the like.” The phenomena observed were not replicable at will. “On different occasions subjects would try to repeat an experience they had had and usually found this very difficult, if not impossible. Indeed such attempts were found to be an interference in the concentration process.”
The phenomenon most significant for us is also the one most interesting to Deikman, for it is the first of the individual phenomena discussed in the first report: the “merging experience” of Subject A, who “from the very beginning reported striking alterations in her perception of the vase and her relation to it.”
“One of the points that I remember most vividly is when I really began to feel, you know, almost as though the blue and I were perhaps merging, or that vase and I were. I almost got scared to the point where I found myself bringing myself back in some way from it. . . . It was as though everything was sort of merging and I was somehow losing my sense of consciousness almost.” This “merging” experience was characteristic of all her meditation sessions, but she soon become familiar with it and ceased to describe it as anything remarkable. Following the sixth session she reported, “At one point it felt . . . as though the vase were in my head rather than out there: I know it was out there but it seemed as though it were almost a part of me.” “I think that I almost felt at that moment as though, you know, the image is really in me, it’s not out there.” This phenomenon of “perceptual internalization” did not recur although she stated that she hoped it would.
In later sessions Subject A described a “film of blue” — later, a “mist” and then “a sea of blue” — that developed as the boundaries of the vase dissolved, covering the table on which the vase sat and the wall behind it, giving them all a blue color. She experienced some anxiety in that “it [this sea of blue] lost its boundaries and I might lose mine too. . . . I was swimming in a sea of blue and I felt for a moment that I was going to drown . . .” Her anxiety seems similar to that often experienced by Zen students just before self-realization (as we shall see in chapter 6). The Zen solution is to “let go” and merge with that sea, which is the ego-death that leads to enlightenment. Deikman adds: “despite the anxiety it occasioned, she felt that the experience was very desirable.” Deikman mentions a later session, conducted after the end of the experimental series, in which Subject A “reported that a diffuse blue occupied the entire visual field and that she felt merged completely with that diffuseness.”
Deikman’s second paper reports on another instance of “breakdown in the self-object distinction”:
It was also as though we were together, you know, instead of being a table and a vase and me, my body and the chair, it all dissolved into a bundle of something which had . . . a great deal of energy to it but which doesn’t form into anything but it only feels like a force.
Subject B in the first study experienced a different sequence of perceptions that Deikman describes as “de-differentiation” and then “transfiguration.” Looking out of the window after his sixth session, he was unable to organize his visual impressions:
I don’t know how to describe it, it’s scattered. Things look scattered all over the lot, not being together in any way. When I look in the background there is much in the foreground that is kind of drawing my attention. . . . [later:] The view didn’t organize itself in any way. For a long time it resisted my attempt to organize it so I could talk about it. There were no planes, one behind the other. There was no response to certain patterns. Everything was working at the same intensity. . . . I didn’t see the order to it or anything and I couldn’t impose it, it resisted my imposition of pattern.
Deikman comments that this description “suggests that the experience resulted from a de-automatization of the structures ordinarily providing visual organization of a landscape (30–50 feet).” But during the next day’s session, Subject B’s perception of the landscape “might be termed ‘transfigured.’” He mentioned very few objects or details but instead talked in terms of pleasure, luminescence, and beautiful movements. For example:
. . . the building is a kind of very white . . . a kind of luminescence that the fields have and the trees are really swaying, it’s very nice . . . lean way over and bounce back with a nice spring-like movement . . .
. . . It’s a perception filled with light and movement both of which are very pleasurable. Nobody knows what a nice day it is except me.
Subject B later added: “It was coming in to me in a sense, I wasn’t watching myself watching. . . . the antithesis of being self-aware.”
In evaluating these results, Deikman considers a number of hypotheses that might be advanced to account for the phenomena: projection, hypnagogic state, hypnosis, sensory translation, sensory deprivation, and unconscious suggestion (but “the very striking phenomena reported were quite unexpected to the experimenter”). He rejects these in favor of “de-automatization”:
Hartmann explicates the concept of automatization as follows: “In well established achievements they (motor apparatuses) function automatically: the integration of the somatic systems involved in the action is automatized, and so is the integration of the individual mental acts involved in it. With increasing exercise of the action its intermediate steps disappear from consciousness . . . not only motor behaviour but perception and thinking too show automatization.” “It is obvious that automatization may have economic advantages in saving attention cathexis in particular and simple cathexis of consciousness in general. . . .” . . . de-automatization is the undoing of automatization, presumably by reinvesting actions and percepts with attention.
To this may be added Deikman’s summary of the implications of the first experiment.
The meditation procedure described in this report produces alterations in the visual perception of sensory and formal properties of the object, and alterations in ego-boundaries — all in the direction of fluidity and breakdown of the usual subject–object differentiation. The phenomena are consistent with the hypotheses that through contemplative meditation de-automatization occurs and permits a different perceptual and cognitive experience. . . . De-automatization is here conceived as permitting the adult to attain a new, fresh perception of the world by freeing him from a stereotyped organization built up over the years and by allowing adult synthetic and associative functions access to fresh materials, to create with them in a new way that represents an advance in mental functioning. . . . The struggle for creative insight in all fields may be regarded as the effort to de-automatize the psychic structures that organize cognition and perception.
In his second study, Deikman concludes:
If, as evidence indicates, our passage from infancy to adulthood is accompanied by an organization of the perceptual and cognitive world that has as its price the selection of some stimuli to the exclusion of others, it is quite possible that a technique could be found to reverse or undo, temporarily, the automatization that has restricted our communication with reality to the active perception of only a small segment of it. Such a process of de-automatization might then be followed by an awareness of aspects of reality that were formerly unavailable to us.
If automatization provides a satisfactory account of the perceptual process, then the claim of Merleau-Ponty and Heidegger that what we do immediately hear is a motorcycle turns out to be not only consistent with but even implied by the claim that nirvikalpa perception is nondual. From the perspective of the nondualist, it is necessarily true of the phenomenal everyday world that we do not distinguish between the sound and the visual object, for that is part of what we mean by the avidyā (delusion) of dualistic perception. Then the difference between the nondualist and most Western epistemologists is not over the thought-constructed nature of phenomenal experience but over what de-automatization (if that is possible) leads to — whether a merely inexpressible sensation of no interest, or another mode of perceiving that reveals something otherwise overlooked about the nature of sense-perception. Deikman’s experiment suggests the latter.
Much of twentieth-century Western philosophy has been concerned with this issue. For example, Husserl realized that all our explicit experience of objects takes for granted an “unthought background” of practices and relations to other objects — a “basic horizon,” as it were — and Wittgenstein came to the same conclusion regarding the functioning of language. The usual response is epitomized by Husserl’s attempt to analyze that horizon phenomenologically, which, however understandable the effort may be, would amount to bringing that background into the foreground, a feat no less extraordinary than levitating by pulling on one’s shoelaces. For the nondualist traditions, this analytic approach is self-defeating, since the prapañca-attempt to grasp that background objectively also hypostatizes the subject in his act of constituting the object. In such a manner we will never be able to experience the nondual ground underlying them both. But if the “unthought background” was once “thought” — if that horizon is a sedimented set of beliefs, inferences, practices, and so on, that were once conscious — then the possibility of de-automatization opens up a completely different approach. Again, the best way to settle the issue is not logically but experientially.
■ ■ ■ ■
By definition, nonduality escapes the main problem for most Western theories of perception: how I can ever have knowledge of things if those things are separate from my mind. For example, direct realism, which claims that we immediately perceive physical objects, cannot account for how the subject (understood as mental) can reach something outside it and completely independent of it. It also cannot explain error and illusion — why some perceptions are veridical and some not, how the plate can be both round and oval. Illusion is not a problem for the nondualist, for the nirvikalpa percept is neither veridical nor nonveridical; like the sense-datum, it cannot appear to be anything other than it is. Questions of error and illusion arise only with savikalpa determination — that is, in the phenomenal world.
Representational realism and causal theories wreck on the same rock: having driven a wedge between what is actually experienced and the object represented or causing the experience, they cannot thereafter bridge that gap to establish the independent existence of perceived objects. Nonduality as developed in this chapter might be viewed as an “idealist” theory of perception, for it denies the existence of objects independent of the mind. But we must remember to distinguish such nonduality from subjective idealism, which reduces the object to the subject, whereas our theory of nonduality denies the one as much as the other. It is as wrong to say that the object is “in” the mind as to say that consciousness resides “in” all physical objects. So subjective idealism is no better a label than realism. I think that nondual perception is better understood as a version of phenomenalism: if we accept that (as Mahāyāna insists) emptiness does not exist apart from form, then there are only nondual appearances.
Having come to this conclusion, I hasten to add that nonduality must be distinguished from other phenomenalisms (e.g., sense-datum theories) which take for granted a naive understanding of the subject. Despite Hume, most versions tend to question only the ontological status of the object and fail to realize that in perception the nature of the perceiver is just as problematic. Because of this, the nondualistic theory of perception avoids some of the difficulties that plague other phenomenalistic accounts. A good example is the issue of solipsism: the view that nothing exists except the self — which thus can be aware only of its own experience — is a problem that lurks for all theories denying objectivity. Traditionally, to be argued into solipsism is equivalent to being checkmated at chess, but the nondualist escapes the mating net. Like subjective idealism, phenomenalism seems to imply solipsism because it isolates the observer by deconstructing other sentient beings into his own sense-data. But such a reduction is not objectionable to the nondualist since the subject is also deconstructed into “sense-data.” This avoids the problem of all the data of consciousness becoming private: I may be the only one in the universe, but only because I am the universe.
The phenomenalist must also answer difficult questions about the status of the unmediated “sense-data” that the “self” is believed to “have.” Are they physical or mental? Spatial and temporal? How long do they last? The nondualist response is that such questions presuppose there are such “things” as “sense-data,” but to understand them as something presented to a subject means the nirvikalpa percept has already been processed into objectified savikalpa. Something about nondual perception is always indeterminable by intellectual analysis, for the presupposition of all such analyses is the dualistic need to objectify what in this case cannot be grasped objectively. One question that is meaningful is whether nondual percepts are physical or mental. The nondualist answer is that they can be neither because they are prior to the delusive bifurcation of mind from matter, which suggests comparisons with the “neutral monism” of William James and Bertrand Russell.
A third and more contemporary objection to phenomenalism transforms the phenomenalist’s ontological claim into a conceptual thesis about language. Since (according to the argument) we cannot determine the nature of perception empirically, what is at stake must be what we mean when we talk about physical objects. Phenomenalism then becomes the claim that statements about physical objects are (or should be) sets of conditional statements about “what we would see if . . .” But it is impossible to convert statements about physical objects into hypothetical ones without losing an important part of the meaning. The nondualist answer to this is, first, that nondualistic phenomenalism in the Asian traditions is a claim which can be and is settled empirically every time someone becomes enlightened. Nonduality is not a theory about language but about how the world is experienced without the superimpositions of language. Second, it may readily be granted that our normal beliefs about physical objects extend beyond any translation into nondual percepts, for that additional belief in the self-existence of the perceived object (and the perceiver) constitutes the delusion that needs to be overcome.
This Life’s dim Windows of the Soul
Distorts the Heavens from Pole to Pole
And leads you to Believe a Lie
When you see with, not thro’, the Eye
That was born in a night to perish in a night
When the Soul slept in the beams of Light
— William Blake, The Everlasting Gospel
One difficulty for phenomenalism that nonduality does not escape is accounting for the causal processes apparently involved in the physiology of perception. When physical objects are deconstructed into sensations (or bundles thereof), experience seems to be fragmented: the Light-objects in this room persist only while there is cognition “of” them, and they immediately disappear when my head turns the other way — to reappear when I turn back. In order to avoid this implausibility, phenomenalists sometimes postulate what Bertrand Russell termed sensibilia — “objects which have the same metaphysical and physical status as sense-data without necessarily being data to any mind” — which Russell at one time considered to be “the ultimate constituents of matter.”114 Sensibilia of some sort are implied by the nondualist claim that objects are self-luminous. But if nondual sensibilia are self-luminous, why are sense-organs necessary at all? And their necessity is hard to deny: if one has no eyes one cannot see at all, dually or nondually.
This objection is too obvious to have been overlooked. Although the nondualist’s answer is implied by everything else discussed in this chapter, it still comes as a shock, revealing more clearly than anything else how alien nondualist perception is, not only to Western epistemology but to all our common sense. For the nondualist bites the bullet and denies that physiological processes are causes of perceptions. More baldly, sense-organs are no more necessary to perception than sense-objects are, because both are śūnya. As the Heart Sutra says, “there are . . . no eyes, ears, nose, tongue, body or mind.” A philosophical exposition and defense of this view is found in the third chapter of the Mūlamadhyamikakārikā, where Nāgārjuna refutes the reality of sense-organs, sense-objects, and the act of perception by demonstrating their relativity to each other. Since they are all śūnya, to believe that we perceive with the sense-organs is a delusion. Rather than dismiss such a conclusion out of hand as nonsense (and at first encounter it could have seemed no less so to Nāgārjuna and his contemporaries), we should consider whether our usual understanding of the physiology of perception does in fact prove dualism, or whether that understanding is the delusion which perpetuates our sense of dualism.
To begin, let us remind ourselves that in nondual perception there is no awareness that one is seeing with the eyes or hearing with the ears. According to the ninth Oxherding Picture, the perceiving of an enlightened person is “as though he were blind and deaf” in the sense that “he absorbs himself so unselfconsciously in what he sees and hears that his seeing is no-seeing and his hearing no-hearing.” To be simultaneously aware of the sense-organ would mean that attention is divided, hence the experience is dualistic and the Light-object (for example) could not be completely self-luminous. This view is equally agreeable to both Mahāyāna and Advaita, but Advaita quite understandably wants to distinguish between such transcendental experience and our usual perception, in which sense-consciousness is dependent upon the contact between organ and object. But the only way we can avoid splitting experience into two radically different types, thus severing saṁsāra from nirvana, is to make the extraordinary claim that we do not actually perceive with the sense-organs even now.
How could anyone dare to suggest such a thing? The crucial point is that the necessity of eyes for visual perception (for example) is not something immediately experienced (nirvikalpa) but is an inference (hence savikalpa) — however unavoidable that inference may be every time I close my eyes. Wittgenstein made a similar claim in the Tractatus:
Where in the world is a metaphysical subject to be found?
You will say that this is exactly like the case of the eye and the visual field. But really you do not see the eye.
And nothing in the visual field allows you to infer that it is seen by an eye.115
Of course Wittgenstein is not arguing for nondual perception, but his quotation may nonetheless shed light on Nāgārjuna’s otherwise peculiar refutation of the sense-faculties: “Vision (darśana) does not see itself. How can something which does not see itself see other things?”116 This odd argument is open to several interpretations, perhaps none of which is immediately convincing. The obvious response is that it is only because the eye does not see itself that it can see other things — an objection that Nāgārjuna immediately considers. But I think that Wittgenstein points to what Nāgārjuna is getting at: since the eye does not see itself seeing other things, how do we know that we see with the eye? It is circular to make my seeing dependent upon my eye, when the conclusion that I see with my eye is an inference dependent upon my seeing. We can never immediately see that it is the eye that is seeing, but only infer it in various ways (e.g., from looking in a mirror). However deeply automatized such a basic inference may be, still it is nothing more than a savikalpa thought-construct. This argument also implies something else important to the nondualist: that we have never had any dualistic sense-experience. The sense of duality can only be thought-constructed by juxtaposing one nondual experience (e.g., an eye opening) with another (the experience of a self-luminous Light-object).
Then to “transcend” all savikalpa-determinations is also to “transcend” the sense-organs, but — we cannot stop yet — to transcend the sense-organs becomes equivalent to transcending sense-perception altogether. Our understanding of sense-perception is so relative to sense-organs and sense-objects that if those are completely denied then the concept of perception loses all meaning. Perception thus inflated becomes perception denied: if (as this chapter argues) there is what might be called perceiving-only (without sense-objects and sense-organs), then there is no such thing as perception, and never has been.
Though there is no being of an object because of the knowledge of perception-only, through this knowledge that there is no object, “perception-only” is also refuted. When there is no being (of an object), perception is not possible, so these are alike in this way. (Vasubandhu)117
So we end up with what was denied at the beginning of this chapter: the necessity to transcend perception. But our route has been a backward one; on this account, the way to transcend perception is to nondually become it. We are not to negate perception for the sake of some other faculty (e.g., intuition) but to realize that what we have understood as perception (the act of relation between sense-organ and sense-object) is in fact something very different. In other words, what transcends perception is nothing other than the true nature of perception itself.
What does such nondual perception/nonperception leave us with? According to Mahāyāna and Advaita the world is māyā because it is like dreams and a magic show. As the Diamond Sutra concludes:
All phenomena are like a dream
An illusion, a bubble and a shadow,
Like dew and lightning.
Thus should you meditate upon them.
Other Prajñāpāramitā texts compare perception to a mirage, for nothing is ever created or destroyed. Māyā, dreams, mirages, and magic all have the same characteristic of seeming to be different than they are, of presenting us with something that appears real when it is really śūnya. While sleeping we may dream that we are “in” a body and using its sense-organs, but they are not actually necessary for the dream experience. If that were also true for our “waking” lives, it would explain the nondualist claim that the universe is Mind. It is also consistent with the visualization exercises in Tibetan Vajrayāna Buddhism. It is common to meditate first on a physical object (e.g., a deity depicted on a thangka mandala) and then to develop the ability to visualize it in detail in the mind. Enlightenment occurs when the student realizes that the physical object in the visual world and the mentally visualized object in his or her mind are not essentially different from each other. Nondual perception, in refuting the self-existence of Light-objects, implies that the physical object is no more real than the visualized one. Denying the bedrock of objectivity removes our grounds for distinguishing one from the other: According to our sympathies, either this is a self-refuting absurdity or it points to the root of subject–object, mind–body dualism: the sharp but delusive distinction we make between physical objects and mental events. Perhaps material objects are only thoughts that have been concretized in some way.118 Such speculations are hardly original, but the claim of nondual perception gives us a different mode of approach to them.
Yet to follow this logic too far and completely deny the role of the sense-organs would be one-sided, to say the least. Like Advaita, Buddhism also stresses the sense-organs’ phenomenal (saṁvṛti, “lower truth”) necessity. The doctrine of pratītya-samutpāda (interdependent origination), which explains all phenomena by relating them in a causal continuum, identifies sensation as the effect of contact between sense-organ and object. The crucial problem then becomes how to understand the connection between these lower-truth, cause-and-effect relationships and the higher-truth claim that nondual experience is unconditioned (nirvikalpa, tathatā) without bifurcating the two as Advaita does. Expressed in this way, the question becomes part of the larger issue of causality. (A full examination of it is reserved for chapter 6.)
The problem we face in trying to understand the role of sense-organs cannot be distinguished from the more general difficulty of understanding nondual perception philosophically. Because our usual understanding of experience is dualistic, we can “think” nonduality only in one of two incompatible ways. Either we conceive of consciousness materialistically, as panpsychically residing “in” physical objects, or we idealistically reduce the object to an image “in” the mind. It is the first conception, in which the object somehow incorporates consciousness, that falters before the causal processes of the sense-organs. The second conception, in explaining the sense-organs too as objectified mental experience, reduces the material sense-organs to mental percepts that are no more privileged than any other percepts, thus escaping the difficulty. This is not to claim that the second conception is valid whereas the first is not. Both are inadequate because they are based upon dualistic categories of understanding, which unfortunately philosophy cannot hope to escape completely. But the second conception does seem to shed more light upon this problem. Like most philosophical answers, it also raises another question: if it is true that sense-organs are not necessary, then why have they materialized? What has caused their objectification?
This chapter concludes by offering a speculative answer to that question.
The problem of sense-organs could be overlooked until now because the approach used has been almost completely “mentalist,” the second of the two conceptions above. It is thought-construction that transforms nirvikalpa into savikalpa perception, and so on. But even a purely mentalist analysis can be accused of taking for granted a Cartesian-type (now commonsense) mind-body dualism, for both the idealist and the materialist starting-points presuppose the very dualism they try to eliminate. As the previous paragraph implies, such a mental–physical dualism is a corollary of the subject–object duality being denied. The claim of subject–object nonduality is more consistent with different approaches to the mind–body problem, such as the neutral monism of James and Russell or Spinoza’s double-aspect theory, according to which mind and body are different aspects of the same substance. Adopting such a double-aspect approach would require us to consider the process of thought-construction-and-projection from the material side as well. Does it have a physical correlate?
Man has no Body distinct from his Soul; for that call’d Body is a portion of Soul discern’d by the five Senses, the chief inlets of Soul in this age. (William Blake)119
As soon as we ask the question in this way, something falls into place. For what if not the sense-organs function to condition sensations? To think that the sense-organs must merely receive sensations passively, and that thought-construction occurs only in the brain, is an assumption that, however deeply engrained, seems to presuppose some form of mind-body dualism. A Spinozan double-aspect account raises the possibility that our sense-organs are objectifications of our prapañca — of our tendencies to thought-condition sensations. This is consistent with the view of Tibetan Buddhism, according to which the body is understood as what might be called materialized karma-potential and the sense-organs are those parts of the body where such saṁskāras (karmic tendencies) tend to concentrate. That does not mean one’s saṁskāras are fixed, for body as well as mind changes, but it does suggest that vikalpa and prapañca may be more deeply ingrained and more difficult to overcome than a purely mentalist analysis would suggest — just as the Advaita conception of māyā as materialized avidyā suggests.
If this speculation is correct, then as the sense-organ changes so the world will change. “The eye altering alters all. . . . The sun’s light when he unfolds it, depends upon the organ that beholds it.”120 Such a claim is not original to Blake, but is part of the Neoplatonic tradition:
For one must come to sight with a seeing power akin and like to what is seen. No eye even saw the sun without becoming sun-like, nor can a soul see beauty without becoming beautiful. (Plotinus)121
That belief is consistent with the tantric emphasis on the body as the means of liberation, as the microcosm of the macrocosm. The Buddha said that the whole world is in this fathom-long body. More recently, Merleau-Ponty has argued that the human body and the perceived world form a single system of intentional relations, that to experience one is to experience the other, and that the body’s presence to the world is what enables things to exist.122 In other words, a nondualist understanding of perception also seems to imply the nonduality of body and mind.