1. Introduction
4.5 The Holistic Argument for Mental Perception from Verse 6 on
4.5.2 Exhaustive Analysis of All Kinds of Mental Perception and Pseudo-mental Perception
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
(pratyakṣa), the investigator of the cognition is able to think and talk about the kinds of perception in the sense that they are various perspectives in which one can observe in and reflect about the holistic reality (the manifoldness of perception). Meanwhile, the analyses remain within thought and language only, since they are conceptual constructions and differentiations by nature, and do not imply that there exists any real separation in the analyzed. In other words, the possibility of “differentiated pratyakṣa” has to be ruled out in Dignāga's theory. On the contrary, it becomes apparent that “differentiated pratyakṣa” is very likely the systematic demand from epistemological proceduralism with its assumed ontological ground, that not only the non-conscious existents but also the non-conscious cognitive elements already
require real differentiation by itself, while the other alternative (theoretical) ground for self-awareness is to admit the self-evident aspect in the holistic cognition.
4.5.2 Exhaustive Analysis of All Kinds of Mental Perception and Pseudo-mental Perception
Thereafter, Dignāga presented an exhaustive analysis of all the possible cases of mental perception classed into two categories, namely, (1) awareness of external objects and (2) awareness of feelings resulting from mental activities, which are not
“dependent upon,” i.e., not simply caused by, the sense-organs.
PS(V) 1.6ab. (Hattori, 1968: 27):
There is also mental [perception, which is of two kinds:] awareness of an [external] object and self-awareness of [such subordinate mental activities as]
desire and the likes, [both of which are] free from conceptual construction.
Both the (mental) awareness of external objects and the (mental) awareness of mental activities, i.e., feelings, are logically preceded and conditioned by conceptual
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
constructions; precisely because of this, they both belong to the sixth consciousness, the mental consciousness (mano-vijñāna, 意識). In the former case, without the universal criteria for us to recognize a unit (of an object) as well as for us to
repeatedly recognize things in kinds (of cognizable objects or of distinct perceptual perspectives), the cognition of the object via the former five sense-organs is
impossible (since we do not assume that they exist in units by themselves). In our present reconstruction of the line of thought, the establishment of such perception is designated as the major confrontation in PS(V)1. On the other hand, the feelings that we have, as what Dignāga himself pointed out, are not directly resulting from the five sense-organs and are also themselves the immediate appearance (the awareness), just as immediate as the five-sense sensation, so that we need no argument in order to classify it as mental perception. However, if one were to probe further, one would see that they are indeed also preceded and conditioned by conceptual constructions (as much as why Hattori interprets feelings as “subordinate”), since (1) they are always in relation to recognizable objects (either external or not, not external at least in relation to the recognizable self-unit, which is also in the form of space and time) that without the recognizable objects they themselves are also impossible to imagine, and (2) they themselves are recognizable in kinds such as desire, anger, ignorance, etc. – the former five kinds of sense perception are actually thinkable in the five distinct kinds too. Hence, if the confrontation on establishing the mental perception in the (self-)awareness of the cognized object emerges victorious (the contentful spatiotemporal awareness), feelings (the contentful temporal awareness) and actually also the theory that includes the former five kinds of perception (the contentless spatiotemporal awareness/the pure awareness of five senses) can be established, too (since that which must be preceded and conditioned by conceptual constructions can nonetheless be themselves in some “non-conceptual” sense). Therefore, it is
reasonable that Dignāga only focused on the confronted keystone issue. This account of the relation between perception and conceptual constructions would be found to be impervious to reason in the paradigm of proceduralism, since in that context,
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
conceptual constructions can only precede perception causally, and not vice versa.
Besides, the ontologically real separation of the cognized objects as well as of the feelings (at different times as well as in distinct kinds) must be assumed, and hence, the sharp distinction between the non-conceptual perception and the conceptual constructions will have to blur (the issue of differentiated sense perception) – there must be some unitary perceptual objects. On the contrary, looking from the
perspective of epistemological holism, the fact that we discursively think about perception in thought does not imply that they are ontologically, or even better, perceptually separate; perception remains a whole and as a matter of fact, it can be thought of in differentiated conceptual constructions solely on the basis that these two sources of knowledge are simply so connected that they result in a unitary cognition.
In this way, it is reasonable that perception can on the one hand remain being free from conceptual construction (as an undifferentiated whole) and on the other hand be preceded and conditioned by conceptual constructions, which are only in
epistemological, and not ontological, relation to perception. Both the establishment of the means of perception and that of the means of inference stand upon their theoretical divide as two sharply different yet valid sources of the cognition whose realization requires their necessary unification.
NMukh (T. 1628, vol. 32):
意地4亦有離諸分別唯證行轉。又於貪等諸自證分，諸修定者離教分別5，
皆是現量。
In the realm of the mind there is also [mental perception] that is free from various conceptual constructions and that occurs only as simple awareness (證) [of the perceptual form (行)].6 Besides, the self-awareness of desire, etc. and the intuition in the practice of meditation (cognition of yogins) that is not
4 Katsura (forth-coming): 意地 manobhūmi; mānasam.
5 Katsura (forth-coming): 諸修定者教分別 yogināṃ apy āgamavikalpāvyavakīrṇam; 皆是現量 pratyakṣam.
6 Shen (2007: 117-118): 證行=證受行相; “'zheg-sho 證受' means comprehension, and 'xingxiang 行相' means the image that appears in mind”; in agreement with the understanding of Zheng (2008: 220), Cf. Cheng Lü and Ven.
Yincang, 1928. Katsura (forth-coming): 唯證行轉 anubhavākārapravṛttaṃ=that occurs (⾏轉 pravṛtta Cf. Kośa Chap. II) as simple awareness/consciousness (唯證).
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
associated with the master's teachings7 in differentiated [forms], are both [mental] perceptions.
In this highly relevant passage in NMukh, the traditional commentators maintained that here, Dignāga merely received and represented the Mahāyāna consensus on perception that there are four kinds of perception, namely, the five-sense perception, mental perception, perception in self-awareness and yogin's intuition in practice (Shen 2007: 117 -8; Zheng 2008: 220 – 221). In this view, there seem to be a number of asymmetries between the two texts with regard to the issue of the exact number of the types of perception accepted by Dignāga.8 However, if we review this passage from the perspective of holism and also compare it with the proposed systematics in PS(V) 1, we would not say here that Dignāga merely received and represented the Mahāyāna consensus. First, obviously since all the conceptual differentiations only function in thought and reality is holistically factual, the issue here would not rest upon determining the number of the real existing kinds of perception but the analysis (the conceptual reflections) of the perceptual features of the cognition in which the conceptual constructions are necessarily involved. These four kinds of perception are not the classification of the existing cases of perception but various ways through which we can think about perception in some necessary relation to conceptual construction. All of them are brought together to support the establishment of
“mental perception.” Hence, we need not specify the number of the kinds on the behalf of theoretical demand. Second, in comparison to the proposed systematics of PS(V) 1 that the keystone task is to establish the validity of mental perception – the perception in the sixth-consciousness, which can by definition be connected either with conceptual construction toward “an [external] object” (the contentful
spatiotemporal awareness) or with the immediate awareness of the mind itself (the contentful temporal awareness about its own states), we can say that Dignāga only recognized two mechanisms of mental perception and that the rest of the cases of
7 教 āgama.
8 Franco 1993; Yao 2004.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
perception that have to do with concepts can be classified into these two mechanisms of awareness. For instance, yogin's intuition (PS(V)1.6cd) and conceptual
construction that is “brought to internal awareness,” i.e., that is “immediate” to awareness (PS(V) 1.7ab), are both classified as mental perception. Although yogin's intuition is obtained following the verbal, differentiated instructions by the masters, it is the state of the mind that is immediately known as a “feeling” and hence counts as mental perception, belonging to the kind of “feelings such as desire, etc.”9 When we cognize an external object via conceptual constructions (via inference), we cognize it not with our senses but with the conceptual constructions (the inference) relying on rules that are causally grounded; in this case, the cognition is a cognition of inference and not of perception. However, when we cognize an external object via conceptual constructions, we also directly know the conceptual constructions (the inference), via which the objects inferred are mediately known; in this case, the cognition of the conceptual constructions is not an inference, but an immediate “awareness,” i.e., the
“internal awareness” (PS(V) 1.7ab). As Hattori (1968: 95) concluded, “the cognizing of an object through kalpanā is ānumana, and not pratyakṣa. But, whether it is ānumana or pratyakṣa, the essential nature of the cognition is the same, that is, it is self-cognized.... In the process of self-cognition, there is no kalpanā”; meanwhile, this theory of the non-conceptual nature of the self-cognition is plausible, as Hattori (ibid.) himself noted as well, only upon admitting the epistemological fact of the
“twofold appearance theory.” Given such criteria for valid mental perception, being in the process of cognition, that is, being the awareness that has to be
self-evident, the erroneous cognition, cognition of empirical reality, etc. are decided not to belong to mental perception but to simple conceptual constructions:
PS(V) 1.7cd – 8ab. (Hattori, 1968: 28)
Erroneous cognition, cognition of empirical reality, inference, its result,
9 This also echoes Dignāga's redefinition of “pratyakṣa” and his separation between jñāpaka-hetu and kāraka-hetu.
Feelings are the immediate cognition in relation to the indria of manas, and they are “produced” by the teachings as jñāpaka-hetu, not as kāraka-hetu.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
recollection, and desire are not true perceptions and are accompanied by obscurity (sataimira).
These types of cognition are themselves not meant to refer to the awareness itself but to the object inferred, but the relevant awareness for inferring them are indeed true perceptions. For instance, the erroneous cognition of bugs resulting from failing eyes is an inference – via certain resembling marks (linga/hetu), the “object” cognized is inferred to be bugs, but both the awareness of the concept “bug” and the awareness of the sight itself are self-evident.