1. Introduction
4.6 Appendix. Jinendrabuddhi – The Difficulty in the Sharp Distinction between the Object-
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
4.6 Appendix. Jinendrabuddhi – The Difficulty in the Sharp Distinction between the Object-aspects in Accordance with the Two Pramāṇa-s
We know that the number of pramāṇa-s (means of cognition) maintained by the standard opinion during Dignāga's time was four: aside from pratyakṣa (perception) and anumāna (inference), there were also śabda (verbal teaching) and upamāna (analogy).11 Ever after Dignāga refused to regard śabda and upamāna as independent means, the number of means of cognition has been admitted to be only two among the Buddhist epistemologists. The reason why Dignāga rejected the other two
traditional means is briefly that they should be included in the operation of pratyakṣa and anumāna. As written in NMukh: “those śabda, upamāna and so on are included in these two means (彼聲、喻等攝在此中).” In PS(V) 5, according to Hattori (1968: 78), Dignāga believed that the cognition as derived from śabda “indicates its own object through the 'exclusion of other objects (anyâpoha),'” which is “the
function of anumāna.” On the other hand, Dignāga thought that upamāna implies (1) the comprehension of words (the same as in the case of śabda), and (2) relating two things separately perceived through the operation of the mind, both of which are also the functions of anumāna. In other words, as Hattori concluded, śabda and upamāna are specifically included in the function of anumāna. Hence, we seem to be able to maintain that both in NMukh and in PS(V), Dignāga decided to refuse to regard śabda and upamāna as independent means because they are included in the function of anumāna. However, in PS(V) 1, Dignāga presented two different issues
questioning, as well as arguing for, the sufficiency of the two means theory other than those in PS(V) 5: (1) evanescent cognitions and (2) recognitions, both of which have to do with “the combination of pratyakṣa and anumāna” (PS(V) 1.2-3)12. Besides, in the commentary literature of NMukh, the discussion on the number of means of cognition is put in different contexts, Buddha-bhūmi-sūtra, its śāstra, and the
11 Buddhist opinions (the Yogācārabhūmi-śāstra and the Abhidharma theories) exclude upamāna.
12 What is written in PS(V) is “the combination of the [two] above-mentioned aspects of the objects”, viz., sva-lakṣaṇa and sāmānya-lakṣaṇa. Since the conformity between pratyakṣa and sva-lakṣaṇa, and that between anumāna and sāmānya-lakṣaṇa are “sharp,” we say “combination of pratyakṣa and sva-lakṣaṇa” here.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Yogācārabhūmi-śāstra, where the soteriological practice is the major concern (MTS, Ven. Shuiyue, 2011: 1525-54); from these resources, we can discern the admitted methodological principle for establishing the means of cognition: “ that which measures must conform with the measured夫能量者，必對所量 ”(MTS, Ven.
Shuiyue, 2011: 1516). In this appendix, I will revisit the argument for the two means theory in Dignāga with PS(V) 1.2-3, and also in reference to the related context provided in the commentary literature of NMukh, especially in Kuiji's YMRZLLS and Zenjū's MTS. As can be surmised from these two resources from PS(V) 1.2-3 and MTS, we will find that whether the functions of the other two means are included in the function of anumāna alone (in the a posteriori combination of pratyakṣa and anumāna) or in “the a priori combination of pratyakṣa and anumāna,” implies two very different understandings of Dignāga's pramāṇa theory. With the above-cited passage in NMukh alone, both interpretations could be maintained. However, in light of the details from PS(V) 1.2-3 and the related commentaries on NMukh, the
interpretation of a posteriori combination is challenged. Considering the
sustainability and coherence of Dignāga's pramāṇa theory as a whole, especially with the help of the introduced contexts, one can see that the understanding entailed in the interpretation of “the a priori combination of pratyakṣa and anumāna” will work far better than the other.
1.
In PS(V) 1.2, Dignāga responded to the self-raised question about evanescent cognition and recognition being in need of separate means of cognition (Hattori, 1968: 24):
Certainly there are such cognitions, but
k. 2c2-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
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
object]13.
In the vṛitti, Dignāga explained that by the “evanescent” case, he referred to the kind of cognition in which one cognizes a thing of color, etc., as non-eternal, when one only cognizes an object with momentary existence, like a fading sound (Hattori 1968:80). In the cognition of the object with momentary existence, as he analyzed:
first, “one cognizes the inexpressible particularity (avyapadeśya=svalakṣaṇa) and the universal (sāmānya-lakṣaṇa), color-ness (varṇatva). Then, by means of the operation of the mind (manas), one relates [the color-ness] to [the universal] non-eternity
(anityatā), and express [the resulting cognition in the judgment] 'the thing of color, or the likes, is non-eternal.' Hence [for this kind of cognition] there is no need of any other means of cognition” (the underline is my emphasis).
From this response, we can see that Dignāga maintained that two conditions are required in order to make functional any expressible judgment, through which the universal (the color-ness) of a cognized object (a thing of color) can be associated with another universal (the non-eternity) in relation to the present cognized object.
First, the cognition of the object with its particular aspect and its universal aspect together must be formed already. Second, there must be the association of the two universals in manas. I believe the relation between the two conditions could be interpreted in two and only two possible ways: (1) functionally united but operative independently, in the case of which the second condition should be satisfied
individually at a separate operational and temporal phase, and (2) both united in function and operation, in the case of which both the conditions have to be satisfied at once so that both consequences of the conditions (the resulting cognition and the associations in manas in relation also to the particular in some way) make their real senses together.14 Organic systems are an example of (1): in order for the function of
13 Tasya sandhane na pramanantaram na ca.
14 Among the commentaries, according to Ven. Renyiu 仁宥法師 (2012: 288), only Ven. Fazun's “special translation”
(ibid.) may agree with the second interpretation: “unless there is determined understanding (of the prameya), (the combination made via manas) cannot be said to be valid cognition – it is neither pratyakṣa nor anumāna 未得定解
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
digestion to be executed, every digestive organ has to operate by itself and to cooperate with the other organs in the appropriate timing and relation so that the function of digestion as a whole could be performed. Sometimes, minor failure, disorder or even absence of certain parts of the system would reduce the performance of the total function, but would not cease the operation of the system. Electric
circuits are an analogy of (2): an electric circuit is composed of various components and conductive wires, through which electric currents can flow, so that the
combination of the components and wires allow certain operations to be performed.
Each component of the wire network has its own function. Taking out any component or any part of the wire network would break the flow of the electric current, and neither the performance of the circuit unit nor any single partial function of its component can remain functional. A computational program is also an analogy of (2): as long as any part of a program fails the circulations (logical allowability for any loop), both the partial functions and the major function would cease to be
executable. The execution of a circuit unit or a program implies the smooth functioning of all its parts.
With the former option (1), function1, “the cognition of both the inexpressible particularity and the universal (color-ness) of a thing of color,” and function2, “the association of the universal 'color-ness' and the universal 'non-eternity' in manas,” are required so that one is able to express the idea that “the thing of color is non-eternal.”
Nonetheless, in this case, function1 and function2 operate separately in each's own mechanism and at different time points. The association of the universal “color-ness”
and the universal “non-eternity” is simply the operation of anumāna, in manas. That is, the universal knowledge “non-eternity” is regarded to be validly subsumed under
即屬非量，俱非現、比”(Fazun 1982: 2; 緣⾃相者即屬現量，以正理推知⾊無常者即屬比量。若由意識結合，
未得定解即屬非量，俱非現、比). The other commentaries, including Shoho Takemura (1956/57), Cheng Lü (1975), Hattori (1968), Ernest Steinkellner (2005), Ernest Steinkellner, Helmut Krasser, Horst Lasic (2005), Jingqing Han 韓鏡清 (unpublished manuscript) and Chien-hsing Ho (digital publication archived at
http://buddhica.nccu.edu.tw), hold that the operation in anumāna is separate and in a later temporal phase after the obtainment of pratyakṣa. In general, this may be because “gṛhītvā” in the vṛitti, as a gerund, implies a temporal-order relation, between which the universal non-eternity is accompanied by (referred to) the particular thing of color (Renyiu 2012: 288).
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
the universal knowledge of “color-ness” (or even the vice versa), and the association is hence simply the function of anumāna. Thus, the association of the universals would always be the afterward processing of the obtained and accumulated a
posteriori judgments that resulted from the constant conjunction of causes and effects in all the previous resulting cognitions, very similar to the notion of understanding in the Humean epistemology (1748/1772). And the word “then” in Dignāga's vṛitti would be temporally conditional, for the association is always following the resulting cognition in time. This interpretation goes more along the line of the conclusion that Hattori drew from Dignāga's theory of anyâpoha in PS(V) 5, whereby Dignāga
decided to refute the other two as independent means because they are included in the function of anumāna alone.
In the latter option (2), the association of the universal “color-ness” and the universal
“non-eternity” is not regarded to be simply rules (of understanding) in anumāna, but has to be done in relation to the real object cognized in the combination of the
function of anumāna and the function of pratyakṣa on-site in the epistemic situation with a realized prameya (be it the present situation, the learned one before or the potential one)! Without the combination of the two functions already, no cognition, and hence no object, is possible. Without the object in the real resulting cognition, the association between the two universals would never be brought forth and be ascertained with necessity. That is, in order to make it possible to express any judgment like “the thing of color is non-eternal,” both the universals of “color-ness”
and “non-eternity” must already be contained in the cognition of the object, even before we shed any light of our clear awareness (the empirical reflection) upon the resulting cognition via the combination of the two means. Following this line of interpretation, the word “then” in Dignāga's vṛitti would be non-temporally
conditional. This interpretation, not proceeding along the line of the aforementioned conclusion, lends more weight to the reason that Dignāga provided in PS(V) 1.2-3, meaning that we do not have to establish another means in the case of evanescent
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
cognition because it only has to do with “the combination of pratyakṣa and anumāna,” not just anumāna alone.
2.
As for the second issue concerning recognition (abhijñāna) and the likes, including recollection, self-cognition of certain recognizable object with certain subjective attitude such as anger, desire, etc., Dignāga's arguments seemed to focus mainly on the problem of infinity. It is not necessary to postulate another means for
recognition, as he argued that (Hattori, 1968: 24-25):
k. 3b1. because [if a separate means of cognition were to be accepted as necessary, then] there would occur the fallacy of infinity (aniṣṭhā).
k. 3b2. for instance, [such mental faculties as] recollection (smṛta) and the like [would have to be recognized as separate means of valid cognition].
We could say here that Dignāga might have adopted reductio ad absurdum as the inferential method, and that assuming the anti-thesis “for the recognition a separate means is necessary” is true, the fallacy of infinity, whereby we have to admit
infinitely many separate means for the infinitely many modes appreciating given objects, would be invoked. To avoid the fallacy, we should not admit separate means for cognitions like recognition, etc.
The negative argument here, however, seems rather weak. Even if we can find infinitely many types of cognition in need of their own separate means, so long as the establishment of these means is in accordance with the establishment of pratyakṣa and anumāna on the same methodological ground, we will have to admit their
establishment, leading pratyakṣa and anumāna to lose their special, initial positions.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Hattori (1968: 81) believed that this question was not really resolved in Dignāga --
“Dignāga's theory of sharp distinction between the objects of pratyakṣa and anumāna is hardly applicable to the case of re-cognition,” and his interlocutors, the Sāṃkhya school cited in PSṬ, suggested a different and more preferable view blurring the sharp distinction. The Sāṃkhya school's theory of viśeṣa-dṛṣṭam anumāna holds that
“the particular is inferable from its likeness (sāmya) to the particular (viśeṣa) perceived before (dṛṣṭa)” (ibid.; Hattori's reference: PSṬ, 17a.3/19b.6)15.
The problem in Jinendrabuddhi's and Hattori's mind, in contrast to Dignāga's issue with Sāṃkhya's theory of viśeṣa-dṛṣṭam anumāna, is with the application of the formal formula of inference in the real epistemic situation (the inference for oneself).
Recognition is the case when one re-cognizes an object, which has been perceived by this person before, by perceiving its inferential mark (liṅga). For instance, when one perceives smoke (the mark of a fire), one has the re-cognition (pratyabhijñāna) of
“the same fire” (Hattori 1968:81). In this case, the recognition is not the function of pratyakṣa alone, because the recognition must be produced by perceiving the mark, and hence anumāna (the association of the mark, the smoke, and the object, the fire, has to be determined) is required, too. It is not the function of anumāna alone either, because what is re-cognized is the particular fire in the epistemic present situation, not the fire in general that is inferred through the mark in anumāna alone. With this understanding, Hattori would naturally think that “Dignāga's theory of sharp
distinction between the objects of pratyakṣa and anumāna is hardly applicable to the case of re-cognition,” and less plausible with regard to the issue of recognition than the theory of viśeṣa-dṛṣṭam anumāna, in which a part of the function of anumāna is to identify one particular case perceived now as similar to the particular cases
15 The traditional Sāṃkhya school holds three pramāṇa-s: perception (dṛṣṭa), inference (anumāna) and reliable authority (āptavacana) (Larson 1998: 9). Perception is the immediate knowledge arising from the contact of the sense organs with an object (ibid.), which is very different from Dignāga's idea of “in close relation to (prati-) sense faculty (akṣa).” Inference is the mediate knowledge and of the three kinds distinguished in terms of the relationship between the “mark” and “that which bears the mark, ” including the aforementioned viśeṣa-dṛṣṭam anumāna.
Reliable authority includes teachings in the sutra and from trust-worthy masters. More details on Jinendrabuddhi's ideas about the contrast will require further research.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
perceived before by their likeness (which itself is also direct knowledge) – hence the particular could be the object of anumāna.
The theory of viśeṣa-dṛṣṭam anumāna does remind us of Wittgenstein's theory of
“family resemblances (Familienähnlichkeiten)” (Philosophische
Untersuchungen/Philosophical Investigations, PI: 67) and Quine's theory of
“stimulus meaning” (1960). Wittgenstein thinks that words confine their applicable cases not because there is something (the observable marks) in common (essential) to all of the possible cases, for when one looks at these concrete cases, one “will not see something that is common to all” (PI: 66). Rather, one sees “(concrete) similarities, relationships, and a whole series of them” in these cases of the same kind (ibid.), just like the family resemblances such as “build, features, colour of eyes, gait,
temperament, etc. etc. overlap and criss-cross in the same way,” via various combinations of which one is recognized to be a member of a family and among which not a single concrete resemblance has to be shared in common by all of the family members (PI: 67). A word (the name) has a “direct (direkte) … relationship with several things that have hitherto been called with the word; and “this can be said to give it (the word) an indirect relationship to other things we call the same name”
(ibid., the braces are my addition). Wittgenstein certainly would not ponder this matter with pramāṇa-s but with language use; nonetheless, he would agree that one has to be able to identify a case of a group of things in perception alone, where the perception of marks (liṅga) has to play a role in associating an object with its name
“directly,” independently from the consideration of the “general forms of
propositions and of language (allgemeine Form des Satzes und der Sprache)” (PI:
65). Following this line of thought, with the apparent absence of conceptualization, a direct relationship between a word and its present object – in the present
epistemic/naming situation – could be extensively and indirectly used to relate the word to its other cases which one has perceived or will perceive, so that one has to call all of the present case, the cases perceived before and the cases one will perceive,
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
the same name. Given this, a word acquires its indirect relationship to all of its
applicable cases – this is where the conceptualization starts. Quine (1960) developed a similar methodology to treat the relationship between word and object. With regard to behaviors, a native speaker, upon being stimulated directly by a given object X, would assent or dissent to the sentence “(this is) X;” through this, the direct
relationship between the cognitive stimulus by X and the word “X” could be tested and established as the “stimulus meaning” of the term “X” (1960: 29), so that further propositional, theoretical combinations of these simple terms with “logical
connections” could form more complicated sentences (theories/theoretical meanings) (1960: 11) .
Wittgenstein and Quine could in a way be aligned with the Sāṃkhya school
temporarily with regard to their shared intuition that there must be something in the direct knowledge enabling us to associate one particular to another particular, so that the indirect knowledge and language become workable upon such a basis. This implies, in the context of the epistemologists, that the particular can also be the object of the means of indirect knowledge. Furthermore, the indirect knowledge and
language derive their foundation from, and thus are shaped by, the direct experience.
Though Wittgenstein, Quine and Sāṃkhya must have different metaphysical
positions, the Humean picture of cognition must be commonly admitted to a certain extent by them: direct experience is the foundation of indirect knowledge, meaning that impressions caused by the contact between the objects and the sense organs group themselves with similarities in the human mind, forming rules of
understanding at a later time point.
In both Enquiry Concerning Human Understanding (EHU) and A Treatise of Human Nature, Hume obviously distinguished two forms of perception: impressions and ideas (EHU 2.3). Perceptions are understood in Hume as the basic items of the
In both Enquiry Concerning Human Understanding (EHU) and A Treatise of Human Nature, Hume obviously distinguished two forms of perception: impressions and ideas (EHU 2.3). Perceptions are understood in Hume as the basic items of the