Why Did Bodhidharma Go to the East?
Buddhism’s Struggle with the Mind in the World
If someone asks the meaning of Bodhidharma coming from the
West,
It is that the handle of a wooden ladle is long, and the
mountain torrents run deep;
If you want to know the boundless meaning of this,
Wait for the wind blowing in the pines to drown out the
sound of koto strings. [Koan 18, tr Heine]
This question—why did Bodhidharma come from the West?—is ubiquitous in
Chinese Ch’an Buddhist literature. Though some see it as an arbitrary question
intended merely as an opener to obscure puzzles, I think it represents a genuine
intellectual puzzle: Why did Bodhidharma come from the West—that is, from India?
Why couldn’t China with its rich literary and philosophical tradition have given rise
to Buddhism? We will approach that question, but I prefer to do so backwards. I
want to ask instead, “why was it so fortuitous for the development of Buddhist
philosophy that Bodhidharma went East? I will argue that by doing so he gave a
trajectory to Buddhist thought about the mind and knowledge that allows certain
issues that are obscure in Indo-Tibetan Buddhism, despite their centrality to the
Buddhist critique of Indian orthodoxy, to come into sharper relief, and hence to
complete a project begun, but not completable, in that Indo-European context.
1. China and India on mind and world
Why did Bodhidharma go to China? This koan is an old one, and I am sure that the
answer I will give would disturb even the most realized roshi. I intend a very broadbrush
examination of the trajectory of Buddhist thought about the mind grounded in
Bodhidharma page 2
the hermeneutical principle that it is often best to read texts not simply against the
horizon of their composers and initial readers but against the horizon the tradition
they engender. In the present case I will argue that by looking at a surprising endpoint,
the Ch’an/Zen tradition in China and Japan, and in particular at the natural
reformulation of some Buddhist ideas about the philosophy of mind when Buddhism
entered China, we can see in sharper relief some of the conceptual innovation
towards which earlier Indian Buddhist philosophers were groping—innovations of
which it is fair to say they were not entirely aware. Buddhism from the start was
striving, I will argue for the transcendence of a representational model of thought,
and this required transcending Indian ways of thinking about thought.
As I say, in this essay I paint with a very broad brush indeed, and will not be pursuing
the close textual approach I favor. That will follow. The picture I am after, in its
broadest outlines is this: Buddhist philosophy of mind developed in an Indian milieu
in which many of the background assumptions about the relationship between
language, thought and the objects of language and thought are much like those that
inform most of Western philosophy. On the other hand, the outlook regarding the
self, the mind and the nature of knowledge that develops in the Buddhist tradition is
in important respects antithetical to that conceptual matrix. That antithesis makes it
hard for Buddhism to articulate its new vision, and results in a certain amount of
obscurity. On the other hand, when Buddhism enters China, it encounters a
conceptual matrix that is in some respects more conducive to its own vision of these
matters, and the translation of Buddhism into the Chinese language and conceptual
environment clarifies some of what Buddhist philosophers were attempting to say.
Here I am relying on Chad Hansen’s (1992) idea that there is a fundamental
difference between the way that Chinese philosophers saw the role of language in
human life and the way that Western philosophers see that role, as well as his view
that this difference entails a dramatic difference in the way in which the mind and
the explanation of behavior are conceived in the two traditions. Briefly (and
Bodhidharma page 3
dogmatically, for now—see Hansen’s book for the details) because the classical
Chinese scholars saw language principally as a written medium that could support
different pronunciations, in which the written characters directly represent the
world, they did not take representation to be the mode through which the mind
normally engages the world. Language, to use Hansen’s metaphor, on this
understanding, guides human behavior and itself represents the world. But
importantly, language on this view does not have the role of conveying, or of
representing thought, and the representational function is conceived as entirely
external—in the graphs—not internal to the mind. (Hansen 1992, pp 32-40)
As a consequence, classical Chinese thought as it was developed prior to the advent
of Buddhism in China, was preoccupied not with the relation between an inner and
an outer world, nor with the nature of mental states, nor with problems about truth
or the nature of reality. Instead the preoccupation was with the status of
Taos—guiding discourses. And a Tao is any sequence of graphs, from a single graph
to a lengthy text. There is no privileged unit of discourse such as a sentence. So
truth does not emerge as the major question to ask about texts. Rather the Chinese
tend to ask, are they successful or unsuccessful, constant or variable, universal or
relative? Minds and propositional attitudes, and in particular mental representations
are strikingly absent from Chinese philosophical discussion, as is logic or theory of
inference.
All of this suggests that Roger Ames’ maxim that the real divide between East and
West is not the Tigris or Euphrates, but the Himalayas has more than a grain of
truth. While Buddhist philosophy has a decidedly critical edge, and often sets itself
in sharp opposition to the tenets and frameworks of the orthodox Indian
philosophical schools, Buddhist philosophy as it developed on India and later in
Tibet was saturated with a broadly Indian (and hence a familiarly Indo-European)
way of taking up with the world: the sentence was the unit of discourse; truth was
Bodhidharma page 4
hence at center stage. Writing was important to be sure, but in classical India,
writing was phonetic, and was taken to represent speech, which in turn was taken to
express thought. This brought in train the view that linguistic meaning is parasitic
on that of thought, and that the mind’s relationship to the world is one of
representation. All of this forces questions about the accuracy of the mind’s
representation of the world, about the truth of ideas, about justification, and about
the relation of the mental to the physical. Does all of this sound familiar? Could this
be why Indo-Tibetan Buddhist philosophy is such a comfortable step for those of us
who want to venture beyond our own tradition, while the East Asian traditions seem
a bit forbidding?
But herein lies the tension that intrigues me now: Buddhist thought, as I have
indicated, is self-consciously critical of that milieu, and often its critiques are
maddeningly obscure. I think that part of the reason for that is that the insights to
which Indian and Tibetan philosophers were trying to give voice were just too un-
Indian, too un-Tibetan to be expressed clearly in the available philosophical idioms.
We can see things improving as Bodhidharma goes East and finds a tradition coming
to the world with presuppositions surprisingly close to those of Buddhist high
theory.
2. Direct Perception and Inference
Buddhist epistemology, especially as it develops in the work of such figures as
Dharmak¥rti and Dignåga, but also as it emerges in the more refined work of the
Madhyamaka tradition as it assimilates their framework, draws a sharp distinction
between the status of the deliverances of perception and those of inference.
Perception, for Buddhist epistemologists, is a source of authoritative cognition, while
inference—while it might be a useful conventional instrument—is never fully
Bodhidharma page 5
authoritative. If I see a fire in the kitchen, I know that there is a fire in the kitchen; if
I see smoke, I may surmise, but do not know.
We must immediately temper and explain this claim. For it is well known to any
scholar of the epistemological tradition of Dignåga and Dharmak¥rti—the tradition
that dominates Buddhist epistemology both in India and in Tibet—that Buddhist
epistemologists accept two warrants for knowledge, viz., perception and inference.
By what warrant am I reducing that number to one? Let me be more precise: Indeed
inference is considered by Dignåga, Dharmak¥rti, and by the majority (though not
all) of their Tibetan exegetes and followers to be a source of authoritative cognition.
But it is considered authoritative for a very different reason and in a much more
limited sense than is perception. Perception is authoritative in this tradition because
it gives us access to real entities, and these will turn out to be particulars. And
perception is a warrant both for ordinary beings and for enlightened beings. More of
this below.
Inference, on the other hand, is a warrant despite the fact that its object—the
universal—is, for any Buddhist philosopher, non-existent, and despite the fact that
inference is always characterized as deceptive. The reason that inference is
countenanced as an epistemic warrant is simply that it is pragmatically useful; that
using it gets us to our epistemic goals and supports legitimate activity. It is never
taken to deliver the truth, and it must be abandoned by enlightened beings. It is very
much a second-rate patch to cover for the limitations on our perception. For this
reason it is fair to say that although each of perception and inference is authoritative
in some sense, only perception is authoritative in a full sense. (Dreyfus 1997, esp pp
299-315)
All of this hinges not on skeptical worries about the fallibility of inference, as one
might expect if one grew up on a diet of Western skepticism, but rather on a
nominalist construal of universals, and a theory of the role of universals as mediators
Bodhidharma page 6
of inference. One way of appreciating this point is that the withholding of full
authority to inference in this framework (despite important nods to its conventional
utility and indeed indispensability as a provisional tool for ordinary human beings,
including sophisticated philosophers) applies not only to inductive, but to deductive
inference as well. (Ibid., 142-153, Tillemans 1999, pp 117-150)
Consider the following inference: All produced phenomena are impermanent; this
pot is produced; therefore it is impermanent. Why is it a relatively defective way of
knowing the impermanence of the pot? Precisely because it depends upon the fact
that the universal of impermanence contains as a subordinate universal that of
produced phenomena; and that the universal of produced phenomena comprises,
inter alia, this pot. The logic is conceived of as categorical. If universals, and the
relations among them were real, as, for instance, many of the Indian Buddhists’
interlocutors, such as the Nyåya held, these inferences could count as fully
authoritative, for then one could in fact perceive the actual categorical relations that
mediate them. On the other hand, if these universals turn out not to be real, then of
course the relevant relations that mediate the inference in question are also unreal,
and hence cannot be perceived.
Universals, however, are all, from a Buddhist standpoint, unreal, for they are mere
conceptual, or linguistic imputations based upon the arbitrary aggregation of
particulars; they have no causal powers. We can quickly get into arcane doxography
here, as we worry about the status of particulars, and I would prefer not to. Some
Buddhist schools reject the independent status of particulars, while some take them
for granted. The general point remains: No matter what you think about the status
of particulars, from a Buddhist framework, universals are nonexistent in reality, and
the fact that they appear real to a mind performing inference means that they are
inherently deceptive, and hence can never mediate knowledge in the full sense.
Hence, whether inference is inductive or deductive, it can never be fully
Bodhidharma page 7
authoritative. At best it can be a useful expedient to guide worthwhile cognitive
activity.
The object of perception, on the other hand, according to all Buddhist
epistemologists, is particular. Hence, for veridical perception (and here again, to get
things precise, an extensive doxographic interlude would be necessary), there is in
fact a genuine object of perception contact with which guarantees the epistemic
authority of the cognitive state. Andindeed, a Buddha sees the particular impermance
of thepot.
It is important to see just how restrictive this standard of highest epistemic authority
is: Particulars are momentary. Continua, including continua of mind or of physical
objects, are, strictly speaking, universals comprising innumerable particular
moments. They are never perceived. This is why so much of our ordinary
experience is deceptive according to this philosophical framework. We think that we
see enduring pots, tables and human beings. But inasmuch as there are no such
things, we are deceived. On the other hand, we do authoritatively perceive moments
of such things. We just don’t know it; typically, for those of us who are not fully
enlightened, the experience to which we have access fails to be truly authoritative;
our truly authoritative experience is beyond our access. We go through life believing
that we are perceiving when in fact we are inferring.
What, then, do we make of the extensive Buddhist literature on inference, and on the
extensive use of principles of logical inference in Indo-Tibetan Buddhist literature?
Here we enter the realm of upåya—of skilful means—in its epistemological home.
Some of us are not fully enlightened. We need a ladder to get us to that stage where
our knowledge is all constituted by direct perception of reality. That ladder is
inference. Through inference we can cultivate a preliminary understanding of the
character of the object of enlightened direct perception. Then we’ll know it when
we see it, instead of being deceived into believing that we are seeing the real thing
Bodhidharma page 8
when in fact we are fabricating it through conception. For this reason, despite its
ultimately deceptive character, inference properly performed is a necessary expedient
on the path, but an expedient that, like the raft, is to be discarded when it serves the
purpose of leading us to the ground of direct perception. And, just as when one is
choosing a ladder or a raft, it is essential—even if the tool will be but a temporary
expedient—to be able to tell the difference between a sound and a shaky ladder or
between a seaworthy and a leaky raft, it is essential in doing philosophy to be able to
tell the difference between valid and invalid inference.
Thus we see an initial epistemic tension built into the Indian and Tibetan Buddhist
epistemological framework: a theory of inference is needed to sort out invalid from
valid inference; the only candidate theory is a categorical logic that is indeed
embraced. But the categorical logic in turn runs afoul of Buddhist ontology, and
must ultimately be rejected. The grounds for its rejection are inferential, and that
inference must be rejected. Epistemic consistency is difficult to achieve. But we see
Indian and Tibetan Buddhist philosophers groping—beginning with an
epistemology that takes for granted an account of the meaningfulness and utility of
inference and of general terms that demands either universals or linguistic
devices—for some way both to transcend those demands, and to justify their
transcendence using those very devices. Bodhidharma could be forgiven a bit of
perplexity.
3. Non-conceptual thought vs conceptual thought
Parallel motivations and tensions emerge when we examine Buddhist critiques of
conceptuality. Repeatedly in Mahåyåna literature we encounter admonitions that
conceptual thought fabricates its objects; that it falsifies reality. The cognitive states
of enlightened beings are regularly contrasted with those of unenlightened beings
along this axis: we unenlightened beings conceptualize, and thus fabricate, or falsify
Bodhidharma page 9
our objects. The insight of buddhas is non-conceptual. What does this mean, and
why is it so important?
I think the key word here is fabrication. (prapañca, spros pa) Buddhist epistemology is
grounded in the idea that the goal of epistemic practice is the realization of truth,
and truth is a correspondence between the knowing cognitive state and the object it
knows. The immediate object of conceptual thought is always, as we have seen, a
universal, an aspect under which something is conceived. To put the point in
contemporary epistemological terms, for an unenlightened being, seeing is always
seeing-as. But to see-as is to see falsely—to assign a character to the object of
thought that it must lack, precisely because no such characteristics are found in
reality.
But Buddhas know things. After all, they are omniscient. So they must know nonconceptually,
and indeed this is how their knowledge is characterized in all Buddhist
treatises on the subject. But omniscience requires that they know all objects of
knowledge, including the epistemic states of unenlightened sentient beings. That,
after all, is a prerequisite for their compassionate and skilful intervention in order to
facilitate the enlightenment of all sentient beings. But that requires that they know
that p, for all true p, and that means that they must know that things are... are, what?
Here is the second major tension. For Indian and Tibetan epistemology,
developing in the orthodox matrix of Sanskritic thought, the proper content of
knowledge claims are propositions, and a proposition consists in the attribution of a
property to an object or sequence of objects; and it is hard to see how that can be
done without appeal to concepts, and hence to universals. But Buddhas must be able
to do so. And so we end up with a rhetoric of inconceivability when we look for any
account of the immediate, non-conceptual knowledge of matters of fact. And of
course this is doubly unsatisfactory, for here the enterprise is simply to characterize
conceptually, for beings like us, what our own epistemic goal is. But unlike the case
Bodhidharma page 10
with perception vs inference, we can’t even claim that our cognitive activity can
present us with an indirect grasp of its own goal. Not only can a Buddha not say
directly what he knows, but we cannot even say indirectly what he knows, or what we
are trying to achieve. We can begin to see why Bodhidharma may have been
consulting a travel agent.
4. Non-duality and duality
Closely connected to these issues is the distinction we find throughout Madhyamaka
and Yogåcåra texts on the philosophy of mind and epistemology between dualistic
and non-dualistic awareness. Again, this distinction divides mundane from
enlightened consciousness. Ordinary minds cognize their objects dualistically,
distinguishing subject from object, object from its complement in the objective
domain, and characterized basis from characteristic. The mind of an enlightened
being is free from these dualistic appearances. A buddha’s mind does not represent
the object as distinct from itself; a buddha’s mind does not represent objects as
distinct from all else—that is, it does not objectify; a buddha’s mind does not
distinguish characterized from characteristic.
We can see why this must be so, and so we can trace both the roots of and the nature
of the error embodied by dualistic awareness as it is understood from this
perspective. Once again, we can, if we wish, distinguish a number of different
approaches taken to spelling out the nature of these dualities in various Buddhist
philosophical systems. But in the end, the major contours will be congruent enough
for present purposes. The key term here is objectification (dmigs pa, ålambana). When
we ordinary folk see a pot we take the object of our consciousness to be a thing that
exists externally to our mind, and that to be a relatively enduring thing, cognized by
a mental episode whose temporal duration might be quite brief, and whose
fundamental nature is very different (mental, vs physical, for instance, hence the
Bodhidharma page 11
distinction between it and me. (We also, on this view, objectify the self, and the same
story could be told from the subjective side). In fact, the pot we see is a momentary
pot stage; and in fact, we see it as a continuing pot. But that is a purely psychological
construct, and so exists not externally to us, but in our minds. The basis of the
distinction we draw vanishes on this view. A buddha, objectifying neither pot nor
self, merely is aware (non-conceptually, of course) of a momentary pot-stage in a
moment of pot-perception.
We distinguish the pot from the cloth on which it rests. That distinction is drawn
on the grounds that the pot and the cloth are different kinds of things—that they are
distinct continua. But neither continuum has any independent reality, and any joints
at which we carve the world we experience depend upon the application of falsifying
concepts. So such dualities are illusory. This is not to say that the buddha sees a
seamless reality—this is not some kind of mystical monism—rather that among the
myriad particulars, no kinds appear to a buddha as more than merely conventional.
We say that the pot is blue. When we do so, we distinguish the particular pot from
the universal blueness that it instantiates. That is, after all, the basis on which we can
cognize, and utter, the conventional truth, “this pot is blue.” But upon analysis, we
have a hard time maintaining this distinction. The pot cannot be posited coherently
as a bare particular; the universal cannot be discovered apart from its instances. Part
of what it is to be this pot is to be this instance of blueness; part of what it is to be blue is
to be the color of this pot. That non-duality is non-dually cognized by a Buddha.
All of this is orthodoxy, and to anyone who has spent time on Buddhist
epistemology, these points are pedestrian. Nonetheless, none of it is easy to say, or
to make coherent. What does it really mean for a mind to be one with its object; for
a thing to be one with its complement; for a particular to be one with a universal
(especially for an existent particular to be one with a non-existent universal)? It is no
wonder that in the Vimalak¥rti-nirdeßa-s¨tra, when Mañjußr¥ is asked to comment on
Bodhidharma page 12
the many bodhisattva’s accounts of the nature of non-duality he faults them all for
trying to express the inexpressible, and less wonder that Vimilak¥rti, when asked to
comment on that comments, remains silent. Silence on this matter is recommended.
But the recommendation is far from silent. The tensions mount. Bodhidharma
books a place in a caravan.
5. Awareness vs representation
Let us consider one last slice between benighted and awakened consciousness, that
between the non-representational awareness of a Buddha and the representations of
an ordinary being. This dichotomy is often used to explain all of the rest we have
scouted; and it will provide, I think, a way to understand the insights Buddhist
philosophy of mind is striving to articulate in India and Tibet, as well as the reasons
for the difficulties in articulating them. It will also provide the pivot point for seeing
why these insights were more available, and perhaps even prosaic, to Chinese
appropriators of Buddhism and those they in turn influenced.
The world of ordinary experience is often characterized in this literature as
conditioned by representations, (rnam pa, åkåra; tshad nyid, nimitta) and the world of
a buddha, by contrast, as beyond representation. The point is straightforward, and
indeed is revolutionary in Indian philosophical thought. It is even difficult for some
Western philosophers to wrap their minds around, though it is gaining a certain
currency in these post-modern times. When ordinary beings perceive or conceive
the world, we do so, on this view, through the mediation of mental episodes or
processes that are representational in character. These episodes take reality and represent
it; our minds engage directly with these mediating representations, and not
with the representeds. The representations are taken to be real, to be external, to be
mind-independent; but they are not, and they veil the real objects from the mind.
Bodhidharma page 13
This representational theory of mind and the problems it raises should sound
familiar, but there are, as we have seen, a few twists. While mediation, per se, is a
problem in the Indian and Tibetan context, as it is in early modern Western
philosophy, in the Indian context, it is precisely the fact that this mediation is
conceptual that is taken as problematic, for the concepts come from the subjective
side, and it is taken that no representation is possible without conceptualization. But
the other side of the coin is this: while most early modern philosophers, despite the
concern for the distorting or distancing potential of a mediate view of access to the
world reckoned this an inescapable predicament of cognition, Indo-Tibetan
Buddhist philosophers regard it as a soluble predicament. Since a buddha sees reality
as it is, and since that is impossible through representation, a buddha has nonrepresentational
cognition.
And so we are back to conundrum. For despite the repeated scriptural assurances
that reality is beyond representation, and that an enlightened cognition of reality is
non-representational, there is no account of just what non-representational cognitive
access to anything could be. This, of course, is not surprising. For from the
standpoint of Indian or Tibetan Buddhist philosophers, thought is a paradigm of the
representational. For that which is beyond representation to be conceivable would
be straightforwardly contradictory.
6. Transformation of the basis
Buddhist soteriological concerns reflect the tensions we have been exploring. By the
Fifth Century of the Common Era, well into the heyday of Mahåyåna philosophical
activity, Vasubandhu and his brother Asanga, at least with the mediation of
Sthiramati a century or so later, launched the Yogåcåra movement. This idealist
Buddhist philosophical system can best be seen, as refracted by the current sets of
concerns, as an attempt both to resolve some of the apparent paradoxes of
Bodhidharma page 14
Madhyamaka metaphysics through the application of the logic and epistemology
developing through the interaction of Buddhist and orthodox Indian schools, and to
develop an account of practice and soteriology that could make sense of the path
from ordinary cognition and subjectivity to the subjectivity and epistemic states of a
buddha.
The advent of Yogåcåra, at the very time when Buddhism was being carried into
China, marks an intriguing moment in the forked history of Buddhist philosophy. In
India, an extended debate began, sparked by Candrak¥rti’s Madhyamaka reply to
Yogåcåra, between the two schools. This debate and the subsequent doxography,
defines the landscape of Tibetan Buddhist scholarship and its philosophical and
hermeneutical problematic when Buddhism crossed the Himalaya into Tibet
beginning in the Ninth Century. On the other hand, as Buddhism moved north
into China well before these debates crystallized, Madhyamaka and Yogåcåra were
seen in China not as rival philosophical schools, but as complementary aspects of a
unified philosophical viewpoint. So, while in India and Tibet, Yogåcåra and its
approach to these problems was eventually denigrated as a second-best
framework—what to do until one is mature enough to tolerate the depth and
paradox of Madhyamaka—in China, and subsequently in the East Asian traditions
that originate in China, such as those of Korea and Japan, Yogåcåra ideas flourished
and took center stage as complementary to and as consistent with Madhyamaka
(again, there is much more nuance to this doctrinal history than I am acknowledging
here, but this is close enough to right for present purposes).
Here we must be content to note a few salient features of the Yogåcåra metaphysics,
epistemology and soteriology necessary for present purposes. First, all phenomena
have three natures: At the coarsest level, they have an imagined nature; their nature
as they are imagined to be by ordinary people. That imagined nature includes their
character as existing external to the mind in space and time, and their being distinct
Bodhidharma page 15
from—that is dual, with respect to—the mind. These properties are, however,
according to this idealist school, imaginary. Second, phenomena have a dependent
nature. That is, phenomena exist only in dependence upon the mind—they are,
from this perspective, real things, but their reality is the reality of hallucinations,
things that exist only in dependence upon the mind. Realization of this nature is, of
course, a more advanced cognitive state that which only takes things as they are
imagined to be. Third, and finally, things have a consummate nature, the nature they
are seen to have form the standpoint of enlightened awareness. This is their nature as
non-different from the mind, usually put as the fact that the dependent—that is the
hallucinatory—is empty of the imagined—that is, of external reality. When one sees
the consummate nature of things, one sees that all is mental, subject-object duality
vanishes, and with it conceptualization and representation. From the metaphysical
side, the Yogåcåra argue that all things have these three natures; from the
epistemological side, the goal of practice is to realize the consummate.
Before we leave the three natures for other central features of the Yogåcåra vision
that prove so influential in Chinese Mahåyåna, it is important to focus for a moment
in the special role of the second of these three natures—the dependent. It has a dual
character, facing on the one hand the imagined, and on the other the consummate.
It hence plays an important pivot role from the standpoint of ontology, epistemology
and soteriology. On the one hand, the dependent nature is dualistic: the fact that
external phenomena are dependent on my own mind requires the distinction
between the hallucination that is my cognitive object and the mind that is its subject;
on the other hand, once the distinction between subject and object is drawn within
the domain of the dependent, that very duality collapses into the unity that is a mind
comprising a mental process. On the one hand, when I come to understand the
dependent character of phenomena, I come to see them as representations of which I
am aware, knowing them as objects, and myself as subject; on the other hand, when I
know them in this way, I know them as aspects of my own mind, and duality and
Bodhidharma page 16
representation vanish. On the one hand, to have realized the dependent is to have
realized yet another conventional aspect of phenomena, one shot through with
representation, conceptual thought, duality, etc; on the other, since it is true that all
phenomena have this character, unlike the imagined, and since the consummate just
is the absence of imagined in the dependent; to have understood the dependent is
already an aspect of enlightened consciousness. This special role of the middle
nature is hence central to understanding the Yogåcåra analysis of the possibility of
the transcendence of the mundane.
The second important Yogåcåra innovation is the reconstruction of the Madhyamaka
notion of emptiness. While for Någårjuna and his followers, the ultimate truth
about phenomena is their emptiness of essence—a fact that grounds their
transcendence of conceptual thought—for the Yogåcåra their emptiness is simply
their emptiness of subject-object duality, and the emptiness of the dependent nature
of the imagined. Realization of ultimate truth is hence realization of consummate
nature, and that in turn, as we have seen, is an aspect of the realization of the
dependent. This reconfiguration of emptiness has profound implications, and is a
crucial part of the Yogåcåra solution to the Madhyamaka paradoxes. It replaces the
notion of essencelessness with an analysis in terms of several essences, and hence
replaces a problem about how we can know objects that ultimately do not exist with a
problem about how we can know objects that exist ultimately in ways other than that
in which they appear to exist. A gulf that the Yogåcårins may well have seen as
unbridgeable within the framework of Madhyamaka becomes more tractable when
seen as akin to far more familiar appearance-reality problem.
A third conceptual innovation in Yogåcåra is the introduction of a foundation
consciousness or transcendental subjectivity that underlies all experience, including
introspective experience of the evolving and constantly changing stream of mental
episodes and processes, from which it is distinguished. This foundation forms the
Bodhidharma page 17
basis of all awareness, personal identity, the preservation of memory, the
accumulation of karma, etc. It is always subject and never object. In empirical selfconsciousness,
according to Yogåcårins, even our own awareness of ourselves is
dualistic, with the foundation consciousness acting as subject, and the stream of
mental events as object. The empirical mind is as subject to three-nature theory as
any other empirical object, and is equally imaginary, dependent, and ultimately nondually
related to the foundation consciousness.
The foundation consciousness is in many ways the central philosophical innovation
of this tradition. While madhyamaka philosophers in India and Tibet criticize it
savagely as a return to the substantial self or åtman of the orthodox Indian schools,
the Yogåcårins and those in East Asia who assimilated their doctrines treated it as a
necessary posit in order to solve a host of problems, including those we have in front
of us. Why do we see things erroneously? Potentials carried by the foundation
consciousness? Why does the past determine the future? Potentials carried by the
foundation consciousness? Why can we transform our awareness? Potentials carried
by the foundation consciousness. Why do we have any experience at all? The
nature of the foundation consciousness is pure subjectivity.
But there is a world of difference between the experience of a deluded, defiled
foundation consciousness and that of a buddha, and our whole problem has been to
explain what that difference is, and how to get from here to there. And here the
Yogåcårins introduce a further important conceptual construct—the transformation
of the basis. The idea is this: so long as karmic potentials reside in the foundation
consciousness, they will ripen in the form of conscious experience, which experience
will be ineliminably representational, conceptual, dualistic, deluded. But through
hard practice, including not only proper conduct whose purpose is to purge these
potentials, but also the cultivation of an awareness of the three natures, and hence a
conceptual apprehension of the falsity of experience and of the true nature of reality,
Bodhidharma page 18
these potentials can be exhausted. At that point, there is a fundamental
transformation of the nature of the foundation consciousness. It is no longer the
repository of the potential for representation. But its nature as pure subjectivity
remains, despite its having no objects. Since all objects are illusory, a consciousness
without objects is an accurate consciousness, but one of a different order—a nonrepresentational
consciousness immediately aware of the fundamental nature of
reality. And that is enlightenment.
This allows us to approach the fourth and final conceptual innovation in this
idealistic system, one that was a kind of footnote to Buddhist philosophy in India, but
which assumed a central place in China and indeed in certain Tibetan schools—the
idea of Buddha-nature, or the innate potential for enlightenment, often referred to as
the seed or matrix of Buddhahood. The very nature of the foundation consciousness
is pure subjectivity. Its objects, we have seen, are, while illusory, adventitious. Since
they are adventitious, they can be eliminated. Their elimination is the achievement
of buddhahood. We hence all have the seed of buddhahood in us at the start. This is
our buddha-nature. Most radically, this is put as the fact that all sentient beings are
primordially enlightened, but this fact is occluded by the ignorance we have been
surveying. The achievement of buddhahood is not so much a transformation as a
recovery or rediscovery of that which is already present.
Here is another way to put this surprising point: Buddhahood is the elimination of
all error and all conditioned ignorance. Error is the mistaking of that which is
nondually related to the mind for dualistically perceived objects, or the
superimposition of conceptual construction on that which is in its own nature
unconceptualized. We erroneously take the minds we experience in introspection
and the object of perception and ideation to be real, and the latter to be distinct from
the former, which we take to be their subjects. All of that, though, is wrong; our
genuine experience is merely the subjectivity of our foundation consciousness, which
is by nature non-dual, non-conceptual, non-representational. We just don’t know it.
Bodhidharma page 19
When we recognize that fact and experience ourselves through that recognition, we
recognize our own enlightenment.
All of this Yogåcåra theory of experience and reality is meant to provide a cogent
account of the insights that are so difficult to capture from a Madhyamaka
perspective: inference is delusional precisely because inference always proceeds
through representations, and representations always mediate between a subject and
on object; inasmuch as all objects are illusory, all representation is misrepresentation,
and so all inference misleading. Perception requires a slightly
different treatment for a Yogåcåra. We need here to distinguish between perception
contaminated by dualistic representation—the perception that ordinary deluded folks
like us experience—and perception that is free from it. The point is that what is
generally taken for perception is not purely perceptual at all, but is conceptualized
and mediated. True perception is immediate, nonconceptual, non-dual. Once again,
the deeper point that the Yogåcåra are after is that in reality our perceptual
experience is already like that. We just don’t allow ourselves to know it.
As I noted, in India and Tibet, this system was regarded by most Buddhist
philosophers as only a stepping stone to a more profound view. We can see why.
For one thing, there is the reinstatement of a foundation consciousness that appears
to function as a non-empty self; for another, the madhyamaka insistence on the
reality of the conventional world seems to be undermined by the Yogåcåra idealism.
And for another, ordinary perception loses is status as authoritative. (Though this
point, again, must be made with some delicacy, given the close association of
Dharmak¥rti’s epistemology that in fact privileges the authority of perception with
Yogåcåra metaphysics. Exploring this Madhyamaka critique would, however, take as
far afield.)
A deeper reason for Madhyamaka dissatisfaction with Yogåcåra epistemology in
India and Tibet, however, is this: it appears that the central idea that our awareness is
Bodhidharma page 20
already genuinely non-conceptual, non-dual, non-representational was inconceivable
to a paradigmatically Indian way of understanding human experience. The whole
problematic of Indian epistemology—Buddhist and non-Buddhist alike—was to
understand how it is that our inner mental states can afford knowledge of the world,
and to determine how our concepts could grasp a world that from its side is
nonconceptual. So, Bodhidharma simply had to go elsewhere.
7. Not thinking, thinking, without-thinking
When the Mahåyåna went East, it was thus a system that incorporated not only the
problematic of Madhyamaka and its relentlessly analytic approach to understanding
reality, but also the idealistic phenomenology of Yogåcåra and its fumbling towards a
solution to the problems that appeared insoluble within the Madhyamaka
framework. But when it hit China, suddenly these problems are transformed. As we
have seen, Buddhist philosophy was striving for something that within the Indian
framework was radical indeed: a total transformation of thinking about the
relationship between mind and world, and indeed about the nature of mind and
world themselves; a replacement of the orthodox model of an independent,
continuous knower of an independent, continuous known, with knowledge achieved
through the mediation of a set of representations with a model of direct, non-dual
apprehension of an essenceless, constantly changing reality by an essenceless
constantly changing continuum of mental states. The problem was that while
Buddhist philosophers had this revolution in mind, the very framework they were
striving to transcend made the formulation of the transcendence difficult.
But Chinese thinkers never thought of the relation between knower and object in
such representational terms. For them, the function of language was captured by the
notion of a Tao—a guiding discourse; the relevant evaluative categories for language
and for thought were effectiveness vs ineffectiveness, constancy vs inconstancy (both
Bodhidharma page 21
culturally and temporally), not truth vs falsity. The problems that concerned them
were problems about the justification of choice of Tao. Buddhism emerged as a
foreign Tao from the West, and at that a most peculiar one. Indeed when we look at
the Chinese assimilation of Buddhism we are struck by the difficulty of translating
Sanskrit texts and Sanskrit ideas into Chinese, and the degree to which the Buddhist
critique of orthodoxy in India comes to look like the Taoist critique of Confucian
orthodoxy. This is sometimes represented as a distortion of Buddhism as it is
Sinified. But we might tell a different story.
Chinese Buddhist philosophers certainly assumed the preoccupation of Buddhist
philosophy of mind and epistemology with the role of signs in mediating
engagement with the world. This, as I have been emphasizing, is a central concern
of classical Chinese philosophy as much as it is a concern of classical Indian
philosophy, even though the specific model of mediation is very different. And
Chinese Buddhist philosophers agreed with their Indian progenitors that a critique
of that mediation is necessary—that immediate experience is to be preferred to
mediated experience. But of course their reasons were different. The concern in
China was that every Tao—even a Buddhist Tao—is conventional, arbitrary,
constricting; that the world and effective engagement in it is too complex, too
variable, requires too much spontaneity of the virtuoso human, to be captured in a
finite Tao—a discourse of signs. Taoist and Chinese Buddhist philosophers agreed
that while this much could be stated by a Tao, any Tao, like any raft, must be
discarded if life is to be lived appropriately.
Now, here, too, is a paradox: as L’ao Tzu put it, no speakable Tao is a constant Tao.
[I] Even this one. And Chinese Buddhist thinkers, particularly in the Ch’an tradition,
were happy to adopt this paradox. The Vajracchedikåprajñåpåramitå-s¨tra, which was
so influential in Ch’an Buddhism, and so celebrated by Hui Neng, after all, happily
tells us that all “truth is uncontainable and inexpressible,” [7] presumably including
Bodhidharma page 22
this one. The second chapter of Tao Te Ching emphasizes the artificiality, relativity
and unreality of conceptual categories, as does the ninth chapter of the Platform
Sutra. Five colors make one blind; five tones deaf. And the pursuit of Tao, unlike
that of learning, requires decrease day by day. The Vajracchedikåprajñåpåramitå also
tells us that “bodhisattvas should leave behind all phenomenal distinctions and
awaken... by not allowing the mind to depend on notions evoked by sounds, odors,
flavors, touch or any qualities...” [14] Now the Vajracchedika, to be sure, is an Indian
text. But it is a text that had curiously little influence in India. Its real blossoming as
an important text in the Buddhist tradition occurs when it is transmitted to China
and taken up by Hui Neng in the Platform S¨tra.
But while there is an air of paradox about all of this, there is, we must acknowledge,
greater clarity about just what this self-undermining discourse is doing than could be
achieved in the Indian context: Our Buddha-nature, the seed of enlightenment
within each of this, on this view, lies in the fact that the precondition of our
following any system of signs is something beyond signs—a capacity for spontaneous
engagement that must lie within our mind as we learn signs, on pain of regress.
This, as Tsung Mi would put it is root; all of the signs; all of our conceptual activity
is mere branch. Awakening is a return to the root, not so much through a pruning of
the branches, but through a recognition of their secondary status. (Gregory 1995, p
67)
As we learn signs—as we are socialized into concepts—these come to mediate our
experience by guiding us, constricting our range of cognitive possibilities even as they
makes others possible. We come to see the world in terms of them—superimposing
a conceptual structure over a reality that is not inherently conceptualized. In doing
so, we create a subject-object dichotomy that places us on one side of the guiding
discourse, and the objects to which it is directed on the other. No mystery there. But
how are we to conceptualize liberation? And how, as Tsung-Mi, was to ask, can we
Bodhidharma page 23
solve what Peter Gregory has called the Buddhist problem of theodicy—the problem
concerning how we can be deluded if we are primordially enlightened, and how, if
delusion is part of the nature of a sentient being, we can eliminate it. (Ibid. 196)
Here is the Ch’an innovation: the distinction between action with no thinking,
action with thinking, and action without thinking. As Døgen puts it, “to study the
dharma is to study the self; to study the self is to forget the self; to forget the self is to
be actualized by all things.” (GenjØkØan 4) This device enables the mysteries and
obscurities of the Indian Mahåyåna to be shed. When we first approach a new
domain, we do not know how to think about it. We act ignorantly and ineffectively
because we act with no thinking: we have no Tao, no signs, to guide us. This is not
enlightened action; it is not even action acceptable at a mundane level. By learning a
guiding discourse, we come to learn a way of acting. But when we act through this
kind of explicit guidance, our actions will be, while generally correct and effective,
hardly expert. They will be studied, restricted. As we gain expertise and practice, we
can leave the discourse behind, we can improvise, and attain the spontaneity in a
domain that is the goal of practice. This sequence is familiar to anyone who has
mastered a language, an art, a sport, a skill. And this is the metaphor that guides
much Chinese theory about the transcendence of language.
Nothing in this picture is, strictly speaking, antithetical to the Indian Buddhist
theory that preceded it. Indeed, I think that it represents a startling completion of
the project begun in India. But note the enormous difference. Here the story can be
told shorn of worries about the status of universals; shorn of a representational
model of mind; shorn of worries about the reality of subject and object. The
intuition that a truly enlightened way of engaging with the world both requires the
use of language and its transcendence is vindicated in a way that also reveals the
insight that a representational model of subjectivity was wrong all along. But
Bodhidharma page 24
because that model was never presupposed in the first place, the road is smoother
and the paradoxes less daunting.
We can also see why Yogåcåra ideas that seem a bit mad in the Indian context seem
prosaic in the Chinese context: The transformation of the basis is simply the
transcendence of the need for a recipe; the imagined nature is the nature that things
have according to any Tao taken as definitive, a focus on the branch as primary; their
dependent nature is the fact that any character they are experienced as having
depends on some Tao—some framework—and seeing that these things are so
dependent on our conventions allows us to transcend those conventions and to see
things as consummated, to return to the root. When we do that, we transcend signs,
we transcend objectification, and we do not experience ourselves as subjects opposed
to our objects. This is the immediacy of experience in the context of effective
perception and action that constitutes liberation, the affirmation by all things
because of our effective, practiced engagement with them.
And of course in a perfectly ordinary sense this kind of engagement is perceptual, not
inferential, and so, from a revised Indian Buddhist perspective, driven by
authoritative, rather than by deluded, or obscured, cognition; it is not mediated by
representations, but rather direct; it is non-dual. All of these desiderata so hard to
articulate from within the Indian context fall out very nicely when the role of
language and of signs is seen from the Chinese perspective.
8. Selflessness and the reworking of the narrative of consciousness
A central tenet of Buddhist philosophy is that there is no self. This doctrine is
generally and most directly interpreted as the thesis that there is no substantial ego
underlying our experience—that we are nothing but a sequence of causally linked
psychological and physical events and processes. And this is both highly plausible
and an important contribution of Buddhist philosophy. But the current investigation
Bodhidharma page 25
allows us to push the doctrine of no-self even further. To be sure, the ontological
dimension of self that is the target of this idea is the substantial ego. But there is also
an epistemological dimension, and this dimension is also illuminated as we look at
the Indian Buddhist critique of the self from the vantagepoint of the East.
One way of contrasting the belief in a self with the attitude of selflessness s is in
terms of self-awareness, or as it is put in the Buddhist tradition, self-grasping. In this
sense to have a self is simply to grasp oneself as an entity, as a subject standing
opposed to an object. So in an important sense the goal of the realization of
selflessness can be conceived as the goal of achieving a state of consciousness in
which the awareness of oneself as a distinct subject no longer figures. Non-duality in
this sense is simply non-dual awareness, that is, action without-thinking. Just as
action without thinking is an achievement depending on prior thought, selflessness
in this sense is an achievement depending on careful cultivation of self.
We hence see a reconfiguring of the entire narrative of self-consciousness when
Buddhism moves to East Asia, but a reconfiguration that represents not so much a
deviation from the trajectory of Buddhist thought in India and Tibet; rather it is a
completion of that trajectory that would have been difficult or impossible in the
Indo-Tibetan context. For in order to complete the Buddhist response to the Indian
representational, conceptual account of epistemic subjectivity and action it was
necessary to find a way of talking about subjectivity and action that never
presupposed that very paradigm as a background, a way of talking about subjectivity
in which representation and conception are not seen as inextricable from even
relatively normal cognition, and in which language is seen ab initio as an external
guiding device. And that required Bodhidharma to travel to the East. The flip side
of this record, of course, is that given this conceptual milieu, the problematic of
Buddhist philosophy could never have originated in China. The right puzzles would
Bodhidharma page 26
never have risen to salience. For Buddhism to come to China, Bodhidharma had to
come from the West.
Bodhidharma page 27
References
DØgen (1985). Moon in a Dewdrop: Writings of Zen Master DØgen (K Tanahashi,
trans). Berkeley: North Point Press.
Dreyfus (1997). Recognizing Reality: Dharmak¥rti’s Epistemology and its Tibetan
Interpretations. Albany: SUNY Press.
Gregory, P. (1995). Inquiry into the Origin of Humanity: An Annotated Translation of
Tsung Mi’s Yüan jen lun with a Modern C ommentary. Honolulu: University
of Hawai’i Press.
Hansen C. (1992). A Daoist Theory of Chinese Thought: A Philosophical Interpretation.
Oxford: Oxford University Press.
Heine, S. (2002). Opening a Mountain: The KØans of the Zen Masters. New York:
Oxford University Press.
Hui-neng. (1990). The S¨tra of Hui-neng, in A. F. Price and W. Mou-lam (trans),
The Diamond Sutra and the Sutra of Hui-Neng. Boston: Shambala.
Tillemans, T. (1999). Scripture, Logic, Language: Essays on Dharmak¥rti and his Tibetan
Successors. Boston: Wisdom Publications.