1. Introduction
4.1 The Two Pramāṇa-s
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
4 The Ontological Neutralism in Dignāga's Epistemology – Critical Reconstructive Interpretation of Related Passages in NMukh and PS(V)
4.1 The Two Pramāṇa-s
One of Dignāga's major contributions to Buddhist Epistemology is his resolute restriction of the number of pramāṇa-s to two: perception (pratyakṣa) and inference (anumāṇa), together with which their new, ontologically neutral definitions are given.
These two in this restriction are redefined as a set of mutually defined concepts;
while the latter refers to the epistemic function of the differentiation in concept which belongs particularly to the consciousness of manas, the former refers to the epistemic function to directly apprehend the knowledge that is devoid of concept, in the
consciousness in general. Dignāga gave the reason as to why two and only two pramāṇa-s should be established: there are only two recognizable kinds or aspects in the object of cognition, the particular (svalakṣaṇa) and the universal
(sāmānya-lakṣaṇa); meanwhile, the rest of the criteria for valid cognitions can be included in these two.1 In the Dharmakīrti school, the relation between these two pramāṇa-s is often introduced in the proceduralistic model, i.e., a time difference is assumed between the reflection in anumāṇa and the being given direct sense-experience in pratyakṣa; when the cognitive faculty is effected by the object in causal relation in time, we have the first direct experience/knowledge of it, namely, the inexpressible particularity of that object at that one moment, and then in the function of manas, the inexpressible particularity is reflected upon in the universals (concepts) at the
subsequent time point. On the contrary, in the Xuanzang school, the relation is often introduced in the holistic model, that no time condition should be given to this
1 This resolute restriction and the reason remind us of the very first page of Kant's Kritik der reinen Vernunt:
especially the lines: “[O]bjects are given to us by means of sensibility, and it alone yields us intuitions; they are thought through the understanding, and from the understanding arise concepts. But all thought must, directly or indirectly, by way of certain characters, relate ultimately to intuitions, and therefore, with us, to sensibility, because in no other way can an object be given to us” (A19/B33).
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
relation. The scholars often comment that the former is inclined toward the Sautrāntika and the latter represents the Yogācāra position. The opinion in the
former would lead to several difficulties, for example, Candrakīrti's challenge against Dignāga's other original idea, self-awareness (svasaṃvitti), that is suspected to lead to the problem of infinite regress, and Jinendrabuddhi's concern that the sharp
distinction between the function of the two pramāṇa-s may invite self-contradiction in Dignāga's own system, especially when examined in the case of re-cognition. In the Appendix, we will see how epistemological proceduralism leads to these
difficulties against Buddhist epistemology itself and how these difficulties would not remain in epistemological holism, in contrast to the more detailed philological
analyses. Here, we have to attempt to build the picture of the ambiguous figure of the Yogācāra Dignāga first.
NMukh(T. 1628, vol. 32):
為⾃開悟唯有現量及與比量。彼聲、喻等攝在此中，故唯⼆量2。由此能
了⾃、共相3故︔非離此⼆別有所量4、為了知彼更⽴餘量。
The valid means of cognition for attaining cognition in oneself are only perception (pratyakṣa) and inference (anumāṇa). Those verbal teaching (śabda), analogy (upamāna) and so on (maintained by the other schools) are included in these two means, so there are only two valid means of cognition.
[First, we only establish two for] they are [already] sufficient to cover the particular and the universal characteristic [of the cognized object respectively];
[second,] it is not the case that aside from these two [characteristics] (of the cognized object), there is any other [aspect] of the object to be cognized (prameya) so that any additional means of cognition needs to be established.
PS(V) 1 (Hattori, 1968:24):
2 Katsura (forth-coming): 唯二量 dve eva pramāṇe.
3 Katsura (forth-coming): 自共相 sva-sāmānya-lakṣaṇā; ⾃相 sva-lakṣaṇā= the particular characteristic; 共相 sāmānya-lakṣaṇā=the universal characteristic.
4 Katsura (forth-coming): 非離此二別有所量 na hi svasāmānyalakṣaṇābhyām anyat prameyam.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Now,
k. 2a-b1. the means of cognition are [immediate and mediate, namely,]
perception (pratyakṣa) and inference (anumāṇa).
They are only two, because
k. 2 b2-c1. the object to be cognized has [only] two aspects.
k. 2 c3-d1. There is no [need for admitting an] other separate means of cognition for [cognizing] the combination of the [two] above-mentioned [aspects of the object];
k. 2d2-3a. Nor [is there any need for a separate means of cognition] in the case of recognizing (abhijñāna) [an object] again and again;
k. 3b1. BecasueBecause [if a separate means of cognition were to be accepted as necessary, then] there would occur the fallacy of infinity (aniṣṭhā).
The mutual definition of the set of concepts and the reason for them are quite clear.
A few remarks deserve particular attention here. First, the term “pramāṇa” is ambiguous; it refers to the means, criterion, condition or faculty via which valid cognition is produced respectively on the one hand, and it also refers to the valid cognitions themselves or their kinds on the other hand. In both the Dharmakīrti school and the Xuanzang school, since Dignāga's idea of self-awareness (that
cognition cognizes itself while cognizing an object) is admitted, the ambiguity can be preserved. In the position of the Sautrāntika thinkers, one may admit this because given that there are externally real means of cognition, what matters is that they have to at least manifest their respective forms in the representation (akara) and not the non-cognizable status of the externally real means; hence, even when there is assumed the non-cognizable real causal relations warranting the correspondence between the non-representational object (via the non-cognizable sense faculty and the non-cognizable faculty of understanding in manas) and the resulting cognitions, the sense of knowledge and the understanding have to “appear” in cognition (i.e.,
svalakṣaṇa and sāmānya-lakṣaṇa). Thus, together with admitting the idea of
self-‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
awareness, one can admit the ambiguity. In the position of the Yogācāra thinkers, one admits this simply because ultimately (in this position) speaking, there is nothing but representations.
Second, their subtle redefinition awaits further scrutiny. In NMukh alone, Dignāga provided the etymological explanation of pratyakṣa: “pratyakṣa is so named because it occurs in close connection with (prati) each sense faculty (akṣa); 現現別轉故名現 量” (Hattori, 1968: 77). Ouyang (1930), following the Tibetan translation of
Nyayapraveśa, also determined the meaning of “現現別轉5” to be “consciousness comes into existence in closely (non-mediated) dependent connection to each sense faculty 識依個別根⽽有也.” This definition is definitely not the common
understanding of this term during and after Dignāga's time. In Candrakīrti's criticism of Dignāga's idea of “mental perception (mānasa-pratyakṣa, Hattori 1968: 93),” we can glimpse the common understanding of the term among the Buddhists and Nyāya thinkers: “prati” means “toward”, and then pratyakṣa is “the function of each sense-organ (akṣa) toward (prati) its object” (Hattori, 1968: 76-77; MacDonald, 2014 vol.2:
273-277 ; Liu, 2009: 57-58). This ties in with the proceduralistic story in the Nyāya and the Sautrāntika whereby external objects affect the sense organs for real to cause the production of sense perception, which is followed by the mental process of
understanding (conceptualization) in the mind at a subsequent time point. The fact that Dhamrakīrti stopped using the term “mānasa-pratyakṣa” may imply his
agreement, to a certain extent, with this criticism of Candrakīrti (Arnold, 2010: 340);
we can obtain support for this suspicion in Dhamrakīrti's obvious assumption of a time difference between the five-sense perception and mental determination, and in his use of causality as the warrant for valid knowledge. Candrakīrti's criticism can be further consulted in PSP 71.10~75.1(Arnold, 2005b: 449-461; MacDonald, 2014: vol.
1 268-275, vol.2: 247-288; Liu, 2009: 47-51 ). Essentially, Candrakīrti believed that Dignāga's establishment or acceptance of “mental perception” was forced because of
5 Katsura (forth-coming): 現現別轉 akṣam akṣam prati varttate. Katsura's translation: because it occurs in connection with/depending upon such and such sense-organ.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
his problematic definition of perception: since perception comes to follow faculty, we need mental perception because we have the faculty of the mind; however, following the mainstream definition, unlike the five sense organs having their separate object-realms, the mind has no real object-realm and is merely some “internal” and
“abstract” process and resolution “inside itself,” so we should not establish “mental perception” to avoid the asymmetry. However, in the position of epistemological holism or the position of the Yogācāra, there would be no problem at all accepting Dignāga's establishment of “mental perception”; following Dignāga's new definition of “pratyakṣa,” there is no asymmetry, and Dignāga is correct to establish “mental perception.” First, as much as we entitle the direct knowledge of the former five senses “pratyakṣa” because they come into existence in immediate connection to the sense faculties, we entitle the direct knowledge that comes into existence in
immediate connection to the faculty of the mind (manas, the sixth sense faculty)
“pratyakṣa” too; on the occasion when we are referring to the means/knowledge via which the object in space (in spatial form) is brought into our consciousness (in thought), since what is now referred to is no longer the knowledge's immediacy (the immediate connection to manas) but its relation to the object, i.e., the inference, we entitle this different domain “anumāṇa.” This is not a difference in ontology, but a difference in epistemology; anumāṇa is epistemologically different from pratyakṣa, not ontologically (as opposed to the insertion by the proceduralists of an ontological difference, or time difference, between them). They are two different ways to
approach the same object, which is the result of their unified cooperation. Second, when sticking to the consensus that manas is necessary in the realization of any real knowledge, i.e., when the cooperation of the two pramāṇa-s is admitted (in
epistemological holism) as the necessary condition for any resulting cognition, there should be no intermediation (such as time) between the former five senses and the sixth sense, adding to which is Dignāga's strict restriction that there are two and only two pramāṇa-s simply because there are two and only two aspects in any cognized object. What results is that (a) all the sense faculties at the level of the condition of
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
cognition are holistically “in immediate connection” and (b) the object of the former five consciousness and the object of the sixth consciousness are one and the same object, which can be thought of in various aspects (approached in five kinds of spatial forms, in temporal forms or in thought). With these two immediate relations (a) and (b), the asymmetry pointed out by Candrakīrti no longer remains; the pair of the two means tie in with the two aspects of the same object, and sāmānya-lakṣaṇa is ideally left to its own without taking any ontological status, as most of the rational
epistemologists so agree. To conclude this in a simplified way, only when the object-aspects are regarded as ontologically separate and when the means are assumed to function in different ontological statuses (different time and space) can there be the so-called problem of asymmetry.