THE ORIGIN OF THE CONCEPT OF 0LAYAVIJÑ0NA

Religious doctrines are a complex web of teachings whose sources are far from being singular or homogeneous. If a religious tradition has a founder, whether actual or alleged, the received teachings of the founder become the foundation upon which the orthodoxy evolves. However, when the founder addresses questions from disciples directly, various explanations are given on different occasions within different contexts to different audiences addressing different concerns.

When the founder dies, his disciples are left with the task of assembling his teachings to preserve them for future generations. Because different teachings target different audiences in their lived situations, apparent inconsistencies emerge when such concrete contexts are left out in the pursuit of abstract doctrinal formulations. With the teacher gone, the contexts of many of the teachings go with him, and his disciples have to deal with these inconsistencies.

Although some of them can be explained away through a study of the contexts of the teachings, others are harder to clarify by appealing to the contexts alone. This latter kind of inconsistency is usually indicative of tensions among some foundational doctrines within the religious tradition. Such tensions constitute a major source of creativity, often leading to the development of various doctrinal systems.

In the context of Buddhism, the rise of Abhidharma literature in early Buddhism, which is meant to classify and analyze the fundamental teachings of the Buddha, reflects the Buddhist effort to overcome a variety of tensions in these teachings. Yog1c1ra Buddhism, which started around the fourth century C.E., inherits the Abhidharma’s systematic pursuit of a body of coherent and cogent Buddhist doctrines. As William Waldron points out, “[I]t was within

the historical and conceptual context of Abhidharma scholasticism that the Yog1c1ra school arose, and within whose terms the notion of the 1layavijñ1na was expressed” (2003, 47). In this chapter we will explore one fundamental tension within the Buddha’s teachings that Abhidharma Buddhists attempt to overcome. This will set the stage for our discussion, in Chapter Two, of the continual effort by Yog1c1ra to grapple with this critical problem in its theoretical endeavors with the postulation of 1layavijñ1na at the center.

The tension in question is between identity and change, reflected in the difficulty in conceptualizing continuity. On the one hand, for there to be continuity in change, there needs to be something— identity—that is continuously changing; on the other hand, if there is identity, change is either impossible (identity itself cannot change otherwise it would be lost) or regarded as the attribute of unchanging substance or identity; hence change is only apparent/illusory but

ultimately unreal. Such a tension is played out in the Buddhist struggle to provide a coherent account of the Buddha’s core teachings of an1tman (no-self), karma, anitya (impermanence), and pratEtyasamutp1da (dependent origination). In other words, can an1tman be reconciled with karma if there is no underlying bearer of karma? Can the impermanence of existence account for the continuity and coherence of our experience of the world as depicted by dependent origination? We will see how various schools of Abhidharma Buddhism, prior to or contemporaneous with the Yog1c1ra School, attempt to cope with this tension so that Yog1c1ra’s effort, culminated in the formulation of 1layavijñ1na, can be understood within a proper context.

In providing a philosophical account of the origin of the concept of 1layavijñ1na, this chapter is written with one primary objective in mind: to reveal the rationale for the postulation of a concept like 1layavijñ1na within the Buddhist tradition. This will provide a proper context for our detailed discussion of the formulation of 1layavijñ1na in Chapter Two, thus laying the foundation for the comparative study of Yog1c1ra and modern psychoanalysis and their respective conceptualizations of the subliminal mind later in the book.

Given this book’s intended audience (Buddhist specialists as well as non-specialists), its dialogical and comparative nature, and the existence of several fine scholarly works on some of the issues examined,1 our investigation of the origin of 1layavijñ1na will not be exhaustive; neither is an exhaustive investigation desirable for our purpose here. I will focus my attention on the theoretical endeavors that anticipate the formulation of 1layavijñ1na. In doing so, I will provide the religious and philosophical ambiance within which a concept like 1layavijñ1na is called for. At the end of the chapter, I will trace the early development of the notion of 1layavijñ1na as conceived by Yog1c1rins. Let us start with the Buddha’s teaching of an1tman, or no-self.

An1tman

Arguably, the doctrine of an1tman(P1li: anatta) is thedefining teaching of the Buddha.2 So much so that Vasubandhu in his Abhidharmakoéabh1ùyamdeclares that “[t]here is no liberation outside of this teaching [[[Buddhism]]], because other doctrines are corrupted by a false conception of a soul” (1313).3 In other words, the teaching of an1tman is taken to be the decisive element, when compared with non-Buddhist teachings, in achieving enlightenment according to most Buddhist schools, with the apparent exception of Pudgalav1dins (Personalists).4

Despite its prominence, the meaning of an1tman is not without ambiguity.5 Etymologically, the prefix an-can be translated as either “no” or “not.” Consequently, an1tman can be understood as either no-self or not-self, and indeed it has been translated in these two ways (Harvey, 7–8). The translation depends on the interpretation one adopts. When translated as “no-self,” which is accepted by most Buddhist scholars, the term suggests that the Buddha categorically denies the existence of a self or soul because that which is negated by the “no” in “no-self” is the self as an entity.6 However, if we adopt the latter translation, not-self, this suggests that the Buddha only denies what is not a self rather than denying that there is any self at all because the “not” in “not-self” negates whatever is predicated of or attributed to the self. Proponents of this interpretation of an1tmanoften cite a passage in the SaÅyutta Nik1ya(44:10; translated by Bhikkhu Ñ1âamoli, 209–210) wherein the Buddha remains silent when a wandering Vacchagotta inquires about the existence of self. Later the Buddha explains to his disciples that the assertion of the existence of the self leads to eternalism while the assertion of its non-existence leads to annihilationism (Bhikkhu Ñ1âamoli, 209–210).

Historically, the interpretation of an1tman as not-self comes from an earlier time, and this is evident in the Buddha’s own exposition of an1tmanas recorded in the Sutta Nip1taand the SaÅyutta Nik1ya, which are believed to be the earliest records of Buddha’s teachings (Nakamura, 27).7 In this connection we find, “This is not mine, this is not what I am, this is not my self” (SaÅyutta Nik1ya22:59, translated by Bhikkhu Ñ1âamoli, 46–47). The first statement denies the ownership of the aggregates, the second and the third reject the metaphysical entity called “I” or “self” as the owner denied in the first statement.8 Steven Collins, in his discussion of the Buddha’s teaching of anatta (Sk: an1tman), also approaches anatta as not-self in the early Buddhist teachings: “It is

precisely the point of not-self that this is all that there is to human individuals. . . . There is no central self which animates the impersonal elements” (82) into which personality is analyzed. In the Anattalakkhaâa Sutta,anattavis-à-vis not-self is discussed in terms of the lack of control of the five constituents of personality and in terms of the inappropriateness in regarding what is impermanent and unsatisfactory as self (Bhikkhu Ñ1âamoli, 46–47). If the Buddha only preaches the doctrine of an1tman vis-à-vis notself, this means that the interpretation of an1tman as no-self, as accepted by most Buddhist scholars, is a later development that is then anachronistically attributed to the Buddha. What is indeed puzzling, as many scholars have observed (e.g., Frauwallner, 125–126; Harvey, 7; Werner, 95), is that the Buddha in the early Suttas never explicitly denies the existence of the self. Be that as it may, the issue at stake here is this: Is the no-self interpretation of an1tman compatible with the not-self interpretation of the doctrine? After a meticulous investigation of the early Suttas,9 Peter Harvey observes:

[I]t can thus be said that, while an empirical self exists—or rather consists of a changing flow of mental and physical states which neither unchangingly exists nor does not exist—no metaphysical Self can be apprehended. This does not imply that it is real but inapprehensible, as the Buddha of the ‘early Suttas’ saw views on it as appropriate, ifit was real. Moreover, even nibb1nais not-Self and not related to a Self, and the Buddha did not accept that Self exists, or that it even lay beyond existence and non-existence. Indeed, the concept itself is seen as self-contradictory, for ‘Self’ is dependent on a sense of ‘I am,’ and this can only arise by clinging to the conditioned factors of personality, which are not-Self. (33)

There are a number of interesting points in this passage. First, the Buddha’s teaching does not reject an empirical self as constitutive of our everyday experience.10 Only the empirical self accounts for our sense of a self, which is nothing other than an empirical continuum vis-à-vis a series of psychophysical events mistaken as a metaphysical identity. This interpretation of the existence of an empirical self in the Buddha’s teaching is also defended by David Kalupahana (1992, 70–77). According to Kalupahana, the Buddha’s analyses of the personality as the sum of five aggregates, namely, body or material form

(r[pa), feeling or sensation (vedan1), perception (saññ1), dispositions (saãkh1ra), and consciousness (viññ1âa); six elements (cha-dh1th), namely, earth (pa•havi), water (1po), fire (tejo), air (v1yu), space (1k1sa), and consciousness (viññ1âa); and the twelve factors of “dependent arising” (P1li: pa•iccasamupp1da)—serve two purposes: the rejecting of a metaphysical self and an embracing of an empirical self.11 In other words, although the Buddha uses these three taxonomies to make the point that there is no unchanging soul over and above those changing elements of a personality, he also uses them to explain our sense of a self, which is nothing other than the empirical continuum of a series of psychophysical events constituted by those elements.

Second, and more interestingly, Harvey points out that the Buddha brushes aside the issue of a metaphysical self due to the incomprehensibility of its putative metaphysical status. It is very tempting to suggest that the Buddha does not actually rule out the possibility of the existence of a metaphysical self and that he only cautions us against our commonsensical understanding of the self that mistakes the empirical continuity as the metaphysical identity. This would be reminiscent of the well-known Kantian antinomy regarding the existence of the soul, which contends that neither the existence nor non-existence of a soul can be proved or disproved because the soul is not a possible object of knowledge. However, instead of following the Kantian way of relegating the metaphysical

question of the existence of the soul to a realm in which human rationality falters in the face of the antinomy,12 the early Suttas (e.g., the Mah1nid1nasutta in SaÅyutta Nik1ya), reveal the self-contradictory nature of the very concept of (the metaphysical) self (Bhikkhu Ñ1âamoli, 46–47) and the pointlessness in speaking of a self apart from experience (Collins, 98–103):

The self-contradictory Self-concept, then, concerns something which is supposed to be both permanent and aware of itself as ‘I.’ But to get even an illusory sense of I-ness, it must be feeling, or one of the other personality-factors, which work in unison with feeling (or all the factors), but these are all impermanent. (Harvey, 32)

All of those characteristics by which the self can be known are impermanent, and therefore any attempt to even conceive of a permanent self is hopelessly self-defeating. Consequently, it is meaningless to either affirm or deny a metaphysical self. This is why the Buddha is silent on the issue of the self. However, this does not mean that we do not even have a sense of self. On the contrary, we do have a sense of self, but it is neither substantive nor eternal. That is, the metaphysical self is not the self we have the sense of. I will revisit this issue throughout this book.

Therefore, it is clear that the two interpretations of the doctrine of an1tman, not-self and no-self, are indeed compatible. Not only are they compatible with each other, the no-self interpretation is a natural development of the not-self teaching of the Buddha. Put differently, the not-self teaching of the Buddha anticipates the later development of the doctrine of an1tmanas no-self in that the former denies even the conceivability of a metaphysical notion of the self as explicitly brought out by the latter. Hence, the accepted interpretation of an1tman as no-self should be understood as maintaining that, although it is inconceivable to have a metaphysical concept of the self, it does not reject the empirical sense of the self. It is in this sense that the term “an1tman” will be subsequently used in this book.

However, there are problems associated with the teaching of an1tman, as Hindu philosophers very quickly point out:

One difficulty of this theory, as should be immediately obvious and as was pointed out by most anti-Buddhist philosophers, is that it fails to account for the unity of self-consciousness and for experiences such as memory and recognition. . . . For the Hindu thinkers, the identity of the I is a condition of the possibility of knowledge, of social life and moral relationships, of suffering and enjoyment, of spiritual bondage and release from that bondage, or ignorance and illumination. (Mohanty, 30–31)

The critiques of the Buddhist theory of an1tmanare mainly centered around the following issues, which the an1tman theory is believed to be ill-equipped to deal with: the unity of self-consciousness; the necessary possibilities of memory and recognition; the identity of the subject for the purposes of knowing; and the identity of the person for the purposes of social life, moral responsibilities, suffering, enjoyment, ignorance, enlightenment, transmigration, et cetera. Essentially, all of the issues have to do with the question of “whose”: Without a metaphysical self, whose consciousness; whose past, present, and future; whose memory; whose recognition; whose knowledge; whose family; whose friends; whose responsibility; whose suffering; whose enjoyment; whose ignorance; whose enlightenment; and whose transmigration are Buddhists talking about?

Karel Werner has ably demonstrated that “[n]either Hinduism nor Buddhism posits an abiding, unchanging, purely individual soul inhabiting the personality structure and therefore the Upaniùadic assertion of the 1tman and the Buddhist arguable negation of the atta do not justify or substantiate the view, still perpetuated in some quarters, that Hinduism believes in a transmigrating soul while Buddhism denies it” (95). In his observation of the early Upaniùads, what transmigrates is the subtle body with 1tman as the controlling but uninvolved power; in Buddhism it is the mental body, n1mak1ya, that structures the

personality that transmigrates (73–97). Consequently, the above accusation against Buddhists that they—in the absence of the recognition of a metaphysical self such as the Upaniùadic notion of 1tman—cannot account for “the possibility of knowledge, of social life and moral relationships, of suffering and enjoyment, of spiritual bondage and release from that bondage, or ignorance and illumination” (Mohanty, 31) is somewhat misplaced because even within the systems of the Upaniùadic tradition the metaphysical self, 1tman, does not perform such functions as a unifying and participating subject or agent. In this connection, it is interesting to call attention to the fact that similar concerns have indeed tempted certain Buddhists to seek solutions in the

direction of a substantive self. An obvious example is Pudgalav1dins (Personalists), who, like V1tsEputrEyas and the S1mmitEyas within the Buddhist tradition, advocate the existence of “something called a pudgala,(‘person[hood]’) from an ultimate point of view, as a real thing” (Williams & Tribe, 125, original italics). The pudgala is regarded as neither identical to the aggregates nor different from them. Such a view is forcefully refuted by Vasubandhu in the last chapter of his Abhidharmakoéabh1ùyam. However, as Paul Williams and Anthony Tribe observe, Pudgalav1dins are struggling with genuine philosophical problems here:

The V1tsEputrEya-S1mmitEya tradition may have had a particular interest in Vinaya matters, in which case their concern with personhood could have been significant in terms of an interest in moral responsibility. It is indeed persons who engage in moral acts, and attain enlightenment. For moral responsibility there has to be some sense in which the same person receives reward or punishment as the one who did the original deed. It is persons who have experiences of love and hate. All this, as Pudgalav1da sources make clear, has to be taken as given. (126–127, original italics)13

Nevertheless, the Pudgalav1dins’ effort is rejected by the majority of Buddhists because it violates the orthodox teaching of an1tman. In the course of pursuing more satisfying solutions to the problem of “whose,” Buddhists embark on a truly radical path.

What makes mainstream Buddhist solutions radical is that they reject the very legitimacy of questions concerning “whose.” These inquiries arise only within the linguistic convention and as such it is not necessarily reflective of the reality, whatever that may be. In fact, Buddhists, in recognizing the thoroughly conventional nature of language, are seeking to challenge the very way of thinking behind the linguistic structure that gives rise to the question of “whose.” I will use the theory of karma, due to its immediate relevance to the origination of the concept of 1layavijñ1na,to illustrate how the above issues concerning “whose” can be addressed.

Karma and An1tman

Karma means action that gives rise to the world, according to Vasubandhu’s Abhidharmakoéabh1ùyam. But what is action? “It is volition and that which is produced through volition” (551). In keeping with this definition, there are two types of action, namely, volition and the action it produces. Furthermore, the Abhidharmakoéabh1ùyam argues that volition is mental action, and that which arises from volition, willed action, is made up of bodily and speech actions. In other words, volition and willed action give birth to three kinds of action: mental, bodily, and speech.14 The Abhidharmakoéabh1ùyam holds that these three account for, respectively, the original cause, the support, and the nature of action. To be more specific, all actions have their origins in the mind, the body provides a physical support for all actions, and speech is the ultimate action by its nature (552).15 The uniqueness of the Buddhist theory of karma is its steering clear of the physicalist interpretation embraced by the Jainas and its emphasis on the mind in the generation of karma.

The main theoretical function of karma in classical Indian spiritual traditions, both orthodox Hindu schools and heterodox Buddhist and Jain schools, is to account for the cycle of transmigration. One of the most controversial issues in this connection is the alleged inconsistency between the theory of an1tman and that of karma, both of which have been accepted by Buddhists. Given the Buddhist endorsement of both an1tman and karma, who is it that performs actions and enjoys the results? How can the no-self, an1tman, reap the results of “her” own action? Who is responsible for the actions, and who is going through the

transmigration? Without a self as a responsible agent, there would be nobody who transmigrates. The issue at stake is the bearer of action. Although Buddhists can deny that there is a substantive self that transmigrates, they still need to account for the karmic continuity that is not limited to a single lifespan: I will refer to this as reincarnation as opposed to transmigration, which implies a substantive self that transmigrates from one life to another. The reason that an1tman is taken to be incapable of accounting for reincarnation is due to the fact that an1tman tends to be interpreted negatively—as rejecting any sense of self whatsoever. As we have seen previously, an1tman does not deny the possibility of any self but only the substantiality of such a

self or a metaphysical self. In fact the notion of an1tman actually opens up the possibility for an empirical, individual, and dynamic self that is manifested as the continuum of a series of psychophysical events constituted by the five aggregates or the twelve links of dependent origination. The notion of 1tman does not account for such a self at all.

This explains, at least partially, why Buddhists reject the notion of 1tman while accepting the theory of karma. As Genjun Sasaki argues, anatta/an1tman “is nothing other than kamma [Sk: karma] in its nature” (34). In other words, “the conceptions of both kamma and anatta refer to one and the same fact, differently viewed” (40). The bearer of karma, if we are to look for one, should be located within, not without, karma or action itself. Such an interpretation proposes that there is no need to posit a subject or agent separate from karma becausesuch an agent quathe karmic continuum is obviously not an unchanging entity, and it can be understood as nothing other than an1tman. This echoes our previous discussion that the teaching of an1tman allows for an empirical self, and such a self is thoroughly karmic in nature.

This explanation effectively solves the above dilemma for Buddhists concerning the relationship between karma and an1tman because it demonstrates that karma is an1tman and an1tman is karma.16 For Buddhists actions and their bearer are identical, and any attempt to separate one from the other would result in the reification of the agent as the permanent and unchanging self, or eternalism as it is referred to in Buddhist literature. In other words, action, in its bodily, mental, and verbal expressions, is the person and there is no person hiding behind her actions.17

One major consequence of Buddhists’ rejection of the notion of a substantive self is that they are able to turn their attention to a more primordial question concerning the ground upon which our sense of identity arises in the first place. This rejection is significant in that it brings to the fore the issue of continuity, which remains largely concealed by the notion of a substantive self. When the metaphysics of identity is thrown out, the challenge posed to Buddhists is how to account for continuity without appealing to substance. Granted that the self is nothing but a series of continuous momentary psychophysical events due to the instantaneity of all that are constitutive of the self and that our postulation of an eternal self is the result of mistaking continuity as identity, Buddhists still need to answer a crucial question: Can they account for continuity?

One clear difficulty in this regard within the early Buddhist framework is, as William Waldron observes, “the incompatibility of the continuity of effect within a strictly momentary analysis of mind, or, in terms of cognition, how any past effects can remain to influence present cognition and knowledge” (1990, 9). How can Buddhists explain continuity within their framework, which only acknowledges the momentariness of the psychophysical events that constitute the empirical self? Without continuity, the momentariness of dharma alone is hopelessly inadequate in explaining the possibility of any meaningful experience.18

Put differently, without accumulation of some sort (including memory, habit, etc.), momentto-moment events cannot themselves provide anything that assembles those momentary episodes into a coherent structure, thus nullifying the very possibility of experience. This thorny problem concerns the issues of momentariness and continuity, and for this we need to turn to the two leading non-Mah1y1na Buddhist philosophical schools, Sarv1stiv1da and Sautr1ntika. Both exerted important influences on the development of the Yog1c1ra School in setting the parameters for the Yog1c1ra’s effort in tackling this issue, as noted by Erich Frauwallner: “Of the Mah1y1na schools, it was especially that of the Yog1c1ra which attempted to develop its doctrines into a system modeled on those of the HEnay1na schools” (131). Sarv1stiv1da and Sautr1ntika are two of the most prominent HEnay1na schools.19 Momentariness and Continuity:

Sarv1stiv1da versus Sautr1ntika

The theory of the momentary nature of conditioned dharmas, despite its orthodox status in Buddhist philosophical discourse, did not appear at the beginning of Buddhism in the Buddha’s recorded teachings. As Alexander von Rospatt notes, “It does not fit the practically oriented teachings of early Buddhism and clearly bears the mark of later doctrinal elaboration” (15). The starting point of this theory is the Buddhist teaching of impermanence and change of the empirical

world, and its precise formulation is arrived at gradually (Stcherbasky, 109; Rospatt, 153).20 In Rospatt’s words, “[J]ust as the momentariness of mental entities follows from the denial of a permanent Self and from the observation of the fleeting nature of mental events, so the momentariness of all forms of entities follows from the denial of a substance underlying change and from the conviction that things always change” (11). In other words, the perceived momentariness of mental entities paves the way for the establishment of the doctrine that all conditioned entities are momentary, given the Buddhist teaching

of the non-substantiality of all entities, mental as well as physical. Th. Stcherbatsky nicely summarizes the Buddhist teaching of momentariness: The theory of Universal Momentariness implies that every duration in time consists of point-instants following one another, every extension in space consists of point-instants arising in contiguity and simultaneously, every motion consists of these point-instants arising in contiguity and in succession. There is therefore no Time, no Space and no Motion over and above the point-instants of which these imagined entities are constructed by our imagination. (84)

The momentary point-instant is the primary dharmathat is ultimately real—in the sense that it is irreducible to any entity more basic than it—and the compounded entity observable in the empirical world is only secondary and derivative, according to the Abhidharmikas. This point-instant is timeless, spaceless, and motionless “in the sense of having no duration, no extension and no movement[;] it is a mathematical point-instant, the moment of an action’s efficiency” (87), and efficiency, not substance, defines existence in Buddhism (89). In Rospatt’s observation, this point-instant, kùaâa, “may in some contexts

be understood as a precisely defined unit of time (e.g., 1/75th of a second), while in others it may (at least in certain compounds) refer to the momentary entity itself” (94). I will not dwell on the development of the theory of momentariness in early Buddhism because it does not directly pertain to our discussion here. What concerns me is not so much how a compound entity can be analyzed into a series of discrete momentary entities, but rather the other way around, namely, how discrete momentary entities can become a compound entity qua a continuum. Hence the question before us is the question of continuity vis-à-vis momentariness.

As noted at the end of the last section, early Buddhist philosophers struggled to reconcile the doctrine of momentariness with that of continuity entailed by the theory of karma. With the rejection of a substantive self, what is it that continues from one moment to the next? If nothing continues, how can karmawork? Apparently, within such a theoretical framework, there is a discrepancy between what William Waldron calls the synchronic analysis and the diachronic analysis of dharmas: “the relations between the diverse dharmas within each single mind-moment, [and] the causal relations between succeeding moments” (1990, 150).

According to Abhidharma Buddhists, all phenomena are transient and momentary; when the effect is born, the cause should have already perished. However, if the cause has indeed ceased to exist at the moment the effect is born, the effect cannot be caused in the strict sense because that which has perished does not have any causal power. As Waldron summarizes, “[T]his exclusive validity accorded to the synchronic analysis of momentary mental processes threatened to render that very analysis religiously vacuous by undermining the validity of its overall soteriological context—the diachronic dimension of samsaric continuity and its ultimate cessation” (2003, 56, original italics). In other words, for Buddhist religious practices to be meaningful their effects must be preserved in some

way so that the spiritual goal of liberation from suffering is possible.

The problem for Buddhists is this: Is it possible to account for continuity without positing some persisting entity? This may be called “the problematic of continuity” in Buddhist philosophical discourse. Continuity presupposes change. Without change continuity would be meaningless because it simply becomes identity. Any meaningful sense of continuity has to be continuity in change, or continuous change.

What then is this continuous change? Historically, continuity has been conceived of in three ways. First, a continuous change may be a continuing alteration of the state of being of one and the same object. In other words, change is predicated upon a substance; it is the change of something without which even the concept of change is impossible and that something is the substratum that grounds change or subsumes change as its attribute. In this case, change is change of properties of an unchanging substance (e.g., the duration of a desk from its creation to destruction). The second scenario, unique to Buddhists, is to

“formulate a theory of immediate contiguity (samanantara)and grant causal efficiency (arthakriy1-k1ritvaor paccayat1) to the immediate preceding dhamm1” (Kalupahana 1975, 72–73). Lastly, continuity is conceived in such a way that it rejects the idea that time consists of a series of instantaneous moments. In this conception of continuity, a moment is not instantaneous, but rather it contains the structure of the immediate past, the present and the immediate future within itself.21 In other words, a moment is itself conceived as a continuum that is a horizon containing what immediately precedes and immediately follows instead of as a discrete instant. As we will see, the first view is adopted by Sarv1stiv1da,22 the second by Sautr1ntika, and all three views, with important

modifications, are adopted by Yog1c1ra. In its advanced stages, Yog1c1ra postulates a grounding but changing consciousness, 1layavijñ1na, consisting of forever changing seeds whose subliminal existence warrants a contiguity between succeeding dharmic moments. I will briefly examine the positions of the two non-Mah1y1na schools relevant to the emergence of a concept like the 1layavijñ1na formulated by Yog1c1rins.23 Let us start with Sarv1stiv1da. This is one of the earliest Buddhist philosophical schools of great prominence, famous for its doctrine that things in three stages of time (i.e., past, present and future) all exist. By advocating this rather counterintuitive position, Sarv1stiv1dins hope to accomplish two

objectives: to account for all objects of consciousness and to lay to rest the problem that if all things are momentary, causality will not be possible. In the case of the first objective, Sarv1stiv1dins take the view that:

A consciousness can arise given an object, but not if an object is not present. If past and future things do not exist, there would be consciousness without an object; thus there is no consciousness without an object. (Abhidharmkoéabhaùyam, 807)

In other words, a consciousness must have an object, and the object must be present when the consciousness of it arises. The reason Sarv1stiv1dins assume such a position is due to their dogma, which strictly stipulates the concomitance of an object, sense organ, and consciousness in the production of a valid cognition (Mimaki, 81). This is a causal account of cognition in that the object causes the sense organ to produce the cognition of that object. Sarv1stiv1dins contend that in the case of past or future objects, the fact that we can have a valid cognition of them itself warrants their existence as real objects. Put differently, instead of arguing that the cognition of a past or future object leads to the postulation of the existence of that object, Sarv1stiv1dins take the opposite stance that it is the existence of that past or future object that causes the cognition of it. This means that, to use contemporary philosophical terminologies, Sarv1stiv1dins reject any intentional analysis of consciousness while maintaining a strictly causal analysis.24

Even though Sarv1stiv1dins may have a point in maintaining that an object of consciousness has to exist in some sense—for example, exist as a past object to be recollected or as a future object to be anticipated—such an existence is clearly not a real existence if the real existence is defined vis-à-vis the present. What is particular about an object of the past or the future is that its very existence cannot be separated from the recollecting consciousness in the case of the past and the anticipating consciousness in the case of the future. Both past and future objects can be understood as intentional but not as real because real objects are present objects that alone have causal efficacy, independent of consciousness. The position that Sarv1stiv1dins advocate is a strict causal account of valid cognition, stipulating that there has to be a real object that causes the consciousness of it by affecting the corresponding sense organ. It is realism in its extreme form.

Secondly, by resorting to the teaching that dharmas in all three stages of time exist, Sarv1stiv1dins hope to render the problematic of continuity irrelevant in that this enables them to claim that because a dharma exist s all the time, it can always produce an effect as its cause. This theoretical move effectively bypasses the dilemma between momentariness and continuity. Two questions should immediately become apparent regarding the Sarv1stiv1da position: First, if things exist in all three stages of time, how does this square with the fundamental Buddhist position of the impermanence of conditioned dharma?25 Second, how can they account for the difference in the three stages of time

if existence is not the criterion that differentiates them? As we have seen previously, the dilemma for Buddhists concerning the problematic of continuity is their advocacy of the impermanence of dharma. If Sarv1stiv1dins advocate that things in all three stages of time exist, they have, or at least appear to have, rejected this fundamental principle of Buddhism. Consequently, the Buddhist position is, instead of being defended, abandoned. On the other hand, if things in all three stages of time exist, how do they differ? What makes the past past, the present present, and the future future? How does something in the past differ from something in the present or future? Furthermore, if dharmas always exist, the effects should likewise exist all the time. Why then can’t they generate effects all the time? Indeed, if the effects always exist, the issue of the arising and perishing of dharmas and their effects would not exist.

In their defense against the first charge, Sarv1stiv1dins argue that although dharmas exist all the time, they are not eternal because they are conditioned and only unconditioned dharmas exist eternally (Abhidharmakoéabhaùyam, 806).26 Although Sarv1stiv1dins are obviously aware of the precarious situation they put themselves in by advocating such a position, they have chosen to stay with it. But their defense is rather weak, to say the least.27

However, it is the second question, namely the difference between past, present, and future, that preoccupies most of the efforts of Sarv1stiv1dins.28 After some rounds of struggle, Sarv1stiv1dins settle on the concept of efficacy:

The answer given was that this difference is based on the efficacy (k1ritram) of the things, which is not yet present in the future, appears in the present and has disappeared again in the past. This explanation sufficed for a time and was unproblematic as long as efficacy was defined simply as activity (vy1p1ran). . . . In the meantime, a doctrine had been created which assumed in the case of things in the different stages of time a changing property (bh1van) in addition to their unchanging essence (svabh1va), which was neither different nor not different from the essence of the things. Then efficacy was equated with this property. (Frauwallner, 206)

In other words, what differentiates things in the three stages of time is the efficacy of their causal power. What was efficacious in the past, is in the present, and will be in the future. A past dharma is defined by its no-longer efficacious causal power, a present dharma by its efficacious causal power, and a future dharma by its not-yet efficacious causal power. This efficacy is regarded as a changing property of the unchanging essence of things from which it is neither different nor non-different.

In defending their position in this way, Sarv1stiv1dins must answer this question: By separating efficacy of a dharma from the dharma itself in classifying the former as a changing property and the latter as unchanging, aren’t they advocating a substantialist stance with regard to the dharma that blatantly violates the fundamental Buddhist position of impermanence? This is indeed very significant because “for the first time in the history of Buddhist thought, the Sarv1stiv1dins accepted a bifurcation of elements as having substance and characteristics” (Kalupahana 1975, 63). As we have just seen, the Sarv1stiv1dins’ answer that these dharmas are not unconditioned dharmas, hence not eternal, is not convincing.

Another question is this: What determines how the property of efficacy works in relation to the dharmas? In other words, why is it that sometimes the dharma is efficacious and sometimes not? What determines the timing of its efficacy? Sarv1stiv1dins deal with this question by linking the timing of efficacy with the human mind through which the status of a dharma, be it past, present or future, is determined:

What determines the presence of a dharma at a certain moment in a certain mental stream (sant1na) is the presence or absence of the pr1pti, the “possession,” of that dharma at the moment when that “possession” drops into one’s own mental stream. This is to say, rather tautologously, that it is present when it has “possession” of that dharma. The pr1pti itself, however, is a dharma non-associated with the mind (citta-viprayukta-samsk1r1),the contrary of the relation of the mental dharmas (caitta) with mind, and thus not in contradiction with the quality of that mind-moment. (Waldron 1990, 190–191)

In other words, Sarv1stiv1dins think that dharmas exist all the time, and it is our mind, through the function of a dharma called “possession,” that determines whether it is past, present, or future. Thus a dharma is efficacious when its efficacy is possessed by “possession.” This dharmic entity is not associated with the mind because if it were dharmas of past, present, and future would be always efficacious, thus nullifying the very purpose of postulating the entity “possession.” Obviously, what Sarv1stiv1dins have done in explaining the difference between the dharmas in three stages of time is to introduce another dharma,

the dharma of possession, which eventually determines whether the other dharmas are causally efficacious or not. The postulation of this dharma, possession (pr1pti),29 however, is apparently an ad hoc explanation of the efficacy of a dharma because it cannot really explain how this possession works without resorting to either tautology, as Waldron points out in the passage quoted above, or infinite postulation of other dharmas (e.g., the possession of possession of dharma).30

Although the theory of efficacy is designed to differentiate things in the past, present, and future, it can also be regarded as an effort to keep Sarv1stiv1dins committed to the Buddhist doctrine of momentariness in that the efficacy only lasts a moment, the present. But the separation of attributes from substance leaves Sarv1stiv1dins vulnerable to the charge, as we have already seen, that they postulate an unchanging essence of things in direct violation of the impermanence doctrine generally accepted in Buddhism. No wonder Sarv1stiv1da is regarded by many other Buddhist schools as heretical (Kalupahana 1975, 149).

By now it should be clear that the solution proposed by Sarv1stiv1dins—that things in all three stages of time exist—is hardly satisfying because it gives rise to too many complications that are unacceptable within the confines of already accepted Buddhist orthodoxy.

Having briefly examined the merits and the difficulties of Sarv1stiv1da’s attempt to reconcile traditional Buddhist doctrines with a viable theory of continuity, let us now turn to Sautr1ntika. The emergence of the Sautr1ntika School is the result of a series of critiques of its predecessor, the Sarv1stiv1da School,31 especially the latter’s position on separating substance from attributes. Their rejection is based on the orthodox Buddhist ground that the state of being of an object is itself that object and there is nothing that exists apart from its states of being as a separate entity. Apparently, Sautr1ntikas do not differentiate substance from attributes or properties but rather regard them as the same. To them, the separation of the two is the result of intellectual abstraction with no experiential correlates.32 Clearly, Sautr1ntikas take the second view on continuity that we mentioned previously (Kalupahana 1975, 73) to accommodate both the momentariness of dharmas and continuity between those moments.

Sautr1ntika prides itself as a defender of the s[tras in Buddhist philosophical discourse; “Sautr1ntika” means “the school of the s[tras.”33 In defending the traditional Buddhist teaching of impermanence (anitya),Sautr1ntikas held the view that the external world has “only an instantaneous existence and . . . argued that its existence could be recognized only through inference” (Hirakawa, 119).34 Sautr1ntikas recognize only two moments of a dharma, nascent (utp1da) and

cessant (vyaya),and reject the static moment (sthiti-kùaâa) (Kalupahana 1975, 151). Thus a dharmaperishes instantaneously the very moment it comes into being, without lasting even a moment. This is clearly a sharp departure from the Sarv1stiv1da position that things in all three stages of time exist, which as we have seen runs the inherent risk of falling into an essentialist position. On the other hand, the Sautr1ntika position pushes the traditional Buddhist doctrine of impermanence to its logical conclusion by advocating that a dharma perishes instantaneously. In this way, Sautr1ntikas inevitably have to take up the thorny

problematic their predecessors have tried and failed to solve, namely, how to reconcile the conflict between momentariness and continuity without sacrificing fundamental Buddhist positions such as impermanence—something Sarv1stiv1dins appear to have done without acknowledging it. Sautr1ntikas refuse to accept a grounding substratum as Sarv1stiv1dins have done. Instead, they stipulate two immediately succeeding moments as congruous and render the preceding one causally efficacious, although congruity itself does not necessarily mean causality. Essentially, the problem comes down to this, as nicely summed up by G. C. Pande:

Motion and change become intelligible only when the notion of a continuum as an identity in difference is superimposed on separate moments. But there is no separate reality corresponding to the notion of the continuum. Since what is real is only momentary, it cannot really move or change. . . . Causality then becomes simple temporal succession. In the absence of any influence or activity between cause and effect, how can there be dependence between them or invariance in their succession? If causality were to be real it would be impossible to explain the relationship of cause and effect as one of either identity or difference. (206)

What is especially noteworthy in Pande’s observation is that some Buddhists, especially Sautr1ntikas in our case, replace causality with temporal succession. In other words, according to some Buddhists, a temporal continuum by itself sufficiently constitutes a continuous change without positing something persisting through. However, in order to establish a causal relationship that presupposes a continuous change, there has to be some necessary connection, something more than just a temporal succession between the cause and the effect, in that the latter is caused by the former. But what is this causal relationship? How can a causal relationship be established without a grounding substratum or something persisting through from the cause to the effect? Without a causal relationship, there can only be a succession of moments. If so, the relationship between the preceding moment and the succeeding one is contingent rather than necessary. To account for the missing link between the cause and the effect within the Buddhist system, Sautr1ntikas postulate the concept of seed, bEja. It represents their attempt to overcome the difficulties that Sarv1stiv1dins encounter in their effort to reconcile the traditional Buddhist doctrines with a viable theory of continuity.

In the Abhidharmakoéabh1ùyam, a seed is defined as follows:

K1mar1g1nuéaya means “anuéaya of k1mar1ga.” But the anuéaya is neither associated with the mind nor disassociated from it: for it is not a separate thing (dravya).What is called anuéayais the kleéa[[[affliction]]] itself in a state of sleep, whereas the paryavasth1va is the kleéa in an awakened state. The sleeping kleéa is the non-manifested kleéa, in the state of being a seed; the awakened kleéais the manifested kleéa, the kleéa in action. And by “seed” one should

understand a certain capacity to produce the kleéa,a power belonging to the person engendered by the previous kleéa. (770) Here Sautr1ntikas are differentiating two kinds of kleéa(affliction)— anuéaya and paryavasth1va—a distinction not recognized by Sarv1stiv1dins. The former is kleéa in the latent state, the latter in the manifest state. The former is identified with the seed and the latter is described as a manifestation of the former. In this context, “seed” is defined as “a certain capacity to produce the kleéa,a power belonging to the person engendered by the previous kleéa.” In

other words, it is a capacity, not an efficacy. As such, it is only potentially efficacious. Because a dharmais defined by its causal efficacy according to Buddhists, a mere capacity, or potential efficacy, cannot qualify it as a dharma. As a result, although we might want to think that seeds actually exist, their existence is not a dharmic existence. That is, they exist nominally or as a designation (prajñapti) that is neither existing nor non-existing. In making a distinction between anuéaya and paryavasth1va, Sautr1ntikas hope to solve the following dilemma:

If the akuéala-m[las [[[roots]] of evil volitions] are not annihilated till the attainment of arhatship and if they are incompatible with the kuéalam[las [[[roots]] of good volitions], how are we to explain the operation of kuéala-m[las or of kuéala volitions in a mundane (alukika) existence? Being incompatible they cannot operate simultaneously. Nor can they operate successively, for succession demands a certain element of homogeneity between the preceding and succeeding moment. If a kuéala-citta were to follow an akuéala-citta, then it will depend for its nature on a heterogeneous cause. This will amount to an admission of the unacceptable position that good springs out of evil or vice versa. (Jaini, 238)

Simply put, the above dilemma is this: In the mundane state that generally characterizes our existence, both evil and good volitions operate, and their relationship clearly can be neither that of simultaneity nor that of succession. They cannot be simultaneous due to their incompatibility or successive due to their heterogeneity. Sarv1stiv1dins, as expected, appeal to their doctrine of possession, pr1pti, to get out of the trap. Sautr1ntikas reject pr1pti on the

grounds that this doctrine, as we have seen, leads to infinite regress. Rather, they try to solve the problem by formulating the theory of seed. There are three kinds of seeds: seeds of evil (akuéala-dharma-bEja), seeds of good (kuéala-dharma-bEja), and seeds of indeterminate characteristic: The Sautr1ntikas maintained that the anuéayas as well as the kuéala elements (bEjas) co-exist side by side in the form of subtle seeds, but only one of them operates at one time. When the anuéayas operate (i.e., become paryavasth1nas), the mind is akuéala. When the seeds of kuéala operate the mind is kuéala. (Jaini, 240)

By relegating the otherwise incompatible elements to mere potentiality in the form of subtle seeds, Sautr1ntikas are able to work out the inherent conflict of the two elements due to their lack of efficacy in the dormant state, while maintaining the doctrine of non-contradiction within one dharmic moment and our common sense.

As a consequence of establishing the theory of seeds, Sautr1ntikas are better equipped to solve the problem of continuity. That is, even though a dharma perishes in the same moment it is born according to the Sautr1ntika interpretation of the Buddhist doctrine of impermanence, its effect does not have to be manifested in its immediate successor. Rather, the effect will take a back seat and remain a seed, dormant for some time, accumulating its potency and waiting for the right conditions for it to come to fruition. In this way, both momentariness and continuity can be theoretically accommodated. In terms of the

manifested dharma, it perishes instantaneously; in terms of continuity, it is retained by the dormant seeds to be manifested later. Because one is manifested and the other is latent, their coexistence is not hindered by their homogeneity. More importantly, the seeds, working behind the scenes as it were, provide a critical connecting link between cause and effect within the boundary of Buddhist orthodoxy without relying on a persisting entity; seeds are not dharmic entities due to their inefficaciousness, but rather a designation for propensities and proclivities of our mental life.

However, the postulation of seeds has given rise to other questions that essentially concern the status of seeds “during two states of deep meditation, the attainment of ‘unconsciousness’ and the attainment of cessation, in both of which there is complete cessation of all mental activities. Since the seeds themselves are carried along in or with the citta, or more precisely, the vijñ1na, the continuity of the seeds is also brought to a standstill in these meditative practices” (Waldron 1990, 211). The problem is that if, as some Buddhist scriptures describe, in the two deep meditative states all mental activities are brought to a halt, how is it possible for the meditator to emerge from such states? As Paul Griffiths points out, the standard Sautr1ntika view on this is as follows:

The consciousness that emerges from the attainment of cessation has as its immediately antecedent and similar condition the physical body of the practitioner in the attainment of cessation. This is possible because, according to the Sautr1ntika theoreticians, mind and body, the physical and the mental, ‘mutually seed one another’—that is, each is capable of planting seeds in the others, seeds which may lie dormant until the proper time for their maturation occurs. (64) In other words, Sautr1ntikas regard the physical body as capable of holding seeds of the mental activity, which in turn provide the immediately antecedent condition for the meditator’s mind to emerge out of the deep meditation states wherein all mental activity is supposed to have ceased. However, this view is clearly problematic for the following reason:

The Sautr1ntikas also wish to preserve the necessity of an immediately antecedent and similar condition for the emergent consciousness, but by allotting that function to the ‘seeded’ physical body they are forced to loosen, almost to the point of disregarding, the requirement that the relevant condition be ‘similar.’ For it is difficult to see even a ‘seeded’ physical body, a continuum of physical events, as being generically the same kind of thing as a continuum of mental events.

(65) In order for the meditator to come out of the cessation states, two conditions are required: the operation of sense organs and the immediately antecedent mental activity (Waldron 1990, 213). The first condition concerns bodily support, the second mental support. The meditator still retains her body, but she does not have mental activity as the immediate antecedent on which her emerging post-meditative mental state can rely because it is interrupted in those meditative states. Bodily support cannot give rise to mental activities, as Sautr1ntikas claim; the two are perceived as heterogeneous and as such they cannot be in a cause-effect relationship.35 Consequently, this difficulty appears to threaten the very purpose for which seeds are postulated in the first case, namely to account for continuity in the course of instantaneous change.

To find a solution to this difficulty, Sautr1ntikas have tried to formulate an undercurrent of seeds into a new kind of consciousness (Waldron 1990, 216) that is subtle and difficult to detect, thus it is different from the forms of consciousness (vijñ1na) hitherto known in Buddhist, or even general Indian, philosophical discourse.36 Sautr1ntikas claim that the two states of deep meditation are free of all other mental activities except the subtle form of mind that is still functioning without being detected, thus providing the meditator with the crucial mental support to emerge from the cessation states. This formulation comes very close to the Yog1c1ra conception of 1layavijñ1na, but it is in the hands of Yog1c1ra Buddhists that we witness a full and systematic development of the concept of a subliminal mind, 1layavijñ1na, that incorporates Sautr1ntika’s insight regarding the subliminal activities in the form of seeds. And this is the topic to which I now turn.

Early Development of 0layavijñ1na

After examining some Buddhist endeavors predating the conceptualization of 1layavijñ1na to smooth out the inconsistencies and difficulties regarding the orthodox Buddhist doctrines of karma and impermanence (continuity and change), let us now move to the conception of 1layavijñ1na formulated by Yog1c1ra Buddhists in their effort to carry on their predecessors’ work. As William Waldron rightly observes, early Buddhists share similar presuppositions expressed in similar terminology, and they are troubled by similar problems; what is different are their solutions (1990, 141). In a nutshell, the similar problem before them is how to reconcile the tension between two central Buddhist doctrines, impermanence and karma, with the former advocating momentariness and the latter continuity.37 This is an issue of continuity in the course of instantaneous change, what I have called the “problematic of continuity” in Buddhist

philosophical discourse. Based on the theoretical advances made by their predecessors, such as the Sarv1stiv1dins and the Sautr1ntikas, Yog1c1rins have developed a more cogent and sophisticated theory to tackle this problem. This is the celebrated theory of 1layavijñ1na. As Paul Griffiths aptly observes, the Yog1c1ra notion of 1layavijñ1na is an “intellectual construct designed to account for problems of continuity in Buddhist theories of personal identity” (96), and it has “substantial and interesting effects upon the soteriology of the Yog1c1ra theorists” (ibid.).

Lambert Schmithausen has written hitherto the most comprehensive and thorough investigation on the early development of this concept in his 0layavijñ1na: On the Origin and the Early Development of a Central Concept of Yog1c1ra Philosophy. My discussion here is primarily based on his eminent book with the aim of making the ongoing presentation of the origin of the concept complete, thus laying the groundwork for an exploration of the role played by 1layavijñ1nain the work of a later Yog1c1rin, Xuan Zang. I will limit my presentation here to the very beginning of the conceptualization of 1layavijñ1na with an eye on its relationship to earlier efforts to tackle similar problems. The later development will be left to Chapter Two, where the concept of 1layavijñ1na will be dealt with in greater detail.

The issues concerning us at this point are how and where the conceptualization of 1layavijñ1na actually begins. What are the salient characteristics of 1layavijñ1na in its earliest schematization as opposed to its later development? In order to trace the origin of the conceptualization of 1layavijñ1na, Schmithausen sets an ambitious goal to locate the initial passage where the concept originally appears in existent Yog1c1ra literature. Before launching the effort to look for the initial passage, if there is one to be found, Schmithausen lays down two requirements for identifying such a passage: It has to have a justifiable motive to introduce a new type of vijñ1na different from the six traditional vijñ1nas, and the choice of the name 1layavijñ1na also has to be reasonable (15). He finds one passage in the Basic Section of the Yog1c1rabh[mi:

When [a person] has entered [[[absorption]] into] Cessation (nirodha [sam1patti]), his mind and mental [factors] have ceased; how, then, is it that [his] mind (vijñ1na) has not withdrawn from [his] body? — [Answer: No problem;] for [in] his [case] 1layavijñ1na has not ceased [to be present] in the material sense-faculties, which are unimpaired: [1layavijñ1na] which comprises (/possesses/has received) the Seeds of the forthcoming [[[forms]] of] mind (pravóttivijñ1na), so that they are bound to re-arise in the future (i.e., after emerging from absorption). (Quoted in Schmithausen, 18)

This passage appears to have met the two conditions set by Schmithausen. The subject matter that this passage deals with is the issue of the mind during a deep meditative state, nirodha-sam1patti. As we saw earlier in this chapter, during such a meditative state all mental activities are brought to a halt. That is to say, there is an interruption of the otherwise continuous mental activities. Consequently, when the meditator emerges from such a state, she would have no mental antecedent as the support. Without this the emergence itself would not even be possible because the emerging consciousness has to be given rise to by

its immediate mental antecedent as its mental support in addition to the bodily support I discussed earlier. Apparently the traditional six types of vijñ1na are not able to offer this support. What Yog1c1rins have done is to propose another form of mind that is subtle and subliminal with the hope that it will both keep the integrity of the Buddhist teaching that in nirodha-sam1patti there are no mental activities and provide a continuity between the mundane state of the mind of the meditator before and after it is interrupted by deep meditation. However, in asserting that during a deep meditative state like nirodha-

sam1pattithere still exists a subliminal form of mind in the state of unmanifested seeds, Yog1c1ra Buddhists have essentially revised the description concerning mental activities during such a state from no mental activities to no detectable mental activities. This leaves room for the postulation of a concept like 1layavijñ1na, and indeed even renders such an effort imperative.

Moreover, the above passage also justifies the use of the word “1layavijñ1na”; “in it, this term would be most appropriate if taken to mean ‘the (or, if the term is new, perhaps better: a) [[[form]] of] mind [that is characterized by] sticking [in the material sense-faculties],’ in the sense of being hidden in them—a meaning which moreover would contrast perfectly with the term ‘pravóttivijñ1na,’ i.e., mind as it comes forth or manifests itself in a [[[Wikipedia:cognition|cognitive]]] act” (Schmithausen, 22). In other words, 1layavijñ1na as formulated here is a form of mind that is hidden in the material sense faculties as opposed to the mind that manifests itself in a cognitive act, pravóttivijñ1na, like the other vijñ1nas.

There are a number of points worthy of our attention once the hypothesis of this being the initial passage is accepted. First, it is obvious from this initial passage, as well as from our previous investigations, that the new vijñ1nais originally formulated primarily with an eye on the problematic of continuity.38 Because the mental continuum is interrupted during the deep meditative state, some form of continuity has to be worked out. The Sarv1stiv1din pan-realist answer has almost insurmountable theoretical difficulties as we have seen. The Sautr1ntika theory of bEja is appealing, but because the seeds are themselves

not a form of mind but simply associated with the mind and the mental activities are interrupted in the given circumstances, they should also in turn cease their activities and thus become incapable of providing mental support to the re-emergence of mental activities for the meditator from the deep meditative state. Consequently, a new form of mind, subtle and undetected, is called for. Indeed, the initial passage seems to have provided a link between the

Sautr1ntika theory of seeds and a more developed form of 1layavijñ1na in that the passage does nothing more than to “hypostatize the Seeds of mind lying hidden in corporeal matter to a new form of mind proper, this new form of mind hardly, or, at best, but dimly, acquiring as yet an essence of its own, not to speak of the character of a veritable vijñ1na” (Schmithausen, 30).

This brings us to the next noteworthy point in the initial passage, namely, 1layavijñ1nais explicitly taught to be embodied in the material sense faculties. That is, the 1layavijñ1na as it initially appears in Yog1c1ra literature has a distinctly corporeal character to it. Instead of subsuming the material sense faculties under it, it sticks in them. Interestingly enough, in this initial passage there is no mention of its being the basis of personal existence or being mistaken for the Self, functions that will come into play later on.

In order for 1layavijñ1na to assume the prominent position it comes to occupy later in the Yog1c1ra theoretical edifice, this primarily corporeal nature of 1layavijñ1na would have to be transformed. Instead of sticking in the material sense faculties, it would become their foundation. This happens in the Pravótti Portion of the Yog1c1rabh[mi (Schmithausen, 51).

In the initial passage, there is no indication as to whether this new form of consciousness is good, bad, or neutral. Furthermore, the issue of the occurrence of alayavijñ1na in ordinary states other than the nirodha-sam1patti has not yet become an explicit one (32). All these issues will become important when 1layavijñ1na starts to assume other roles in addition to its maintaining continuity for a meditator who goes through deep meditation. I will deal with these issues in Chapter Two.

What is striking—if we look back on the historical development of Buddhist attempts to solve the problematic of continuity, from the pan-realist Sarv1stiv1dins to the somewhat constitutionist Sautr1ntikas,39 and eventually to the metaphysical idealist Yog1c1rins40— is that Buddhists appear to have increasingly resorted to human subjectivity in finding a solution. Consequently, although I agree with Schmithausen that “the Initial Passage does not show any trace of

idealism or spiritualism” (32) and that “the origin of the 1layavijñ1na theory does not seem to have any material connection with the origin of the doctrine of vijñaptim1trat1 [[[cognition]] only]” (33), its further advancement in the hands of Yog1c1rins, its eventual incorporation into the grand Yog1c1ra system, and the

central role it has come to play within that system fit rather nicely into the general trend of idealization in the development of Buddhist doctrines, culminating in the Yog1c1ra School itself. In this respect, it is interesting to note that scholars have generally distinguished two schools of Yog1c1ra, the

“classical” and the “new.” The classical school is represented by the early Yog1c1rins such as Maitreya, Asaãga, Vasubandhu, Param1rtha, and Sthiramati, and the new school by Dharmap1la and Xuan Zang. The new school is considered more idealistic in its orientation.41 This means that even within the Yog1c1ra School, the idealization trend continues. In the next chapter we will see how Xuan Zang’s formulation of 1layavijñ1na reflects this trend.