Centre of Jaina Studies Newsletter: SOAS - University of London
This study, the focus of my doctoral dissertation, attempts to reconstruct the theory of knowledge according to Akalaṅka, a Jaina Digambara philosopher who lived in India during the 8th century C.E. (720-780?). It is based on a preliminary critical edition and a translation of Akalaṅka's Laghīyastraya and of Abhayacandra's commentary, entitled Syādvādabhūṣaṇa (13th century). The study comprises three main areas: a thematic study; material for a critical edition; and a translation of the stanzas, the autocommentary and selected extracts from Abhayacandra's Syādvādabhūṣaṇa. What follows is a summary of this research.
Cognitive validity, cognoscibility and truth
In order to understand the relations between cognitive validity, cognoscibility and truth, one should first of all endeavour to reconstruct the general definition of the means of knowledge (pramāṇa), because every epistemological inquiry depends on it. The definition of the means of knowledge elaborated throughout the Laghīyastraya (LT) enabled Akalaṅka to make his position clear in the epistemological debates in which his theories conflict with other schools as well as with other Jaina philosophers. While characterising the means of knowledge as a cognition (jñāna) endowed with certainty, Akalaṅka limits the pramāṇa to cognitive acts only and excludes doubt, error and indeterminate cognition from it. But certainty about the object of knowledge is not sufficient to make a jñāna become a pramāṇa, for reflexiveness is a distinguishing feature of the means of knowledge: a pramāṇa knows itself and its object simultaneously; even though cognitive validity requires the presence of apperception, this apperception is not by itself a pramāṇa, as Dignāga and Dharmakīrti assumed, because the existence of a determined content ensures the singularity of each knowledge.
Non-deceptiveness in the empirico-practical realm (avisaṃvādatva) is indeed a criterion of cognitive validity, but doesn't belong to the pramāṇa definition. Non-deceptiveness enables the cogniser to offset partly the inability of every imperfect knowledge i.e. every knowledge other than omniscience to prove intrinsically its validity. Permitting to avoid an infinite regress for every new cognition meant to prove the validity of a previous one would be itself devoid of validation experiences reveal the falsity of an erroneous judgment insofar as they deceive the cogniser's expectation of it. Then reliability would give us the possibility to appreciate the validity of a cognition only a posteriori. However, when a cognition has already been tested, memory enables the cogniser to compare the new cognition to be tested with the previous one obtained before. In this case, the criterion of cognitive validity is provided by the coherence between the different cognitions. The need to use non deceptiveness as a criterion of validity doesn't mean that pramāṇa is a purely pragmatic notion. Such pragmatism is only necessary to non-omniscient beings, whose cognition always remains approximate, for in order to know perfectly a single thing, one should know everything. Thus, pramāṇa validity must be appreciated according to conformity of a cognition with empirico-practical activity. When one uses a means of knowledge that seems valid but has a result that is not suitable in the empirico-practical realm, it is actually a fallacious means of knowledge (pramāṇābhāsa). Nevertheless the distinction between pramāṇa and pramāṇābhāsa doesn't hold true, but depends on the viewpoint chosen by the cogniser and the use he aims at. Therefore we can understand that Akalaṅka applies the sevenfold predication (saptabhaṅgī) to the statement which establishes a connection between cognitive validity and non-deceptiveness.
The syādvāda or the sevenfold predication (saptabhaṅgī) is a structural rule permitting the combination of two contrary predicates {A, ¬A} with one another in all possible ways, so as to consider one and the same state of affairs as exhaustively as possible. The sevenfold predication is usually expressed as follows:
(S1) In some respect (syāt), x is only (eva) A;
(S2) in some respect, x is only ¬A;
(S3) in some respect, x is A and ¬A;
(S4) in some respect, x is only inexpressible;
(S5) in some respect, x is A and inexpressible;
(S6) in some respect, x is ¬A and inexpressible;
(S7) in some respect, x is A, ¬A and inexpressible.
Despite appearances, the saptabhaṅgī is based on two predicates, because the third one, 'inexpressible' (avaktavya), is nothing but a peculiar association of the predicates A and ¬A: language has no means to express accurately the complex nature of reality, namely the fact that a being is A and ¬A at the same time. Therefore one has to choose either the predicates semantism (S3) which implies that the predicates must be uttered successively but not simultaneously, or the notion of simultaneity (S4) that can only be expressed by a predicate in which semantism has been totally erased. Once these elementary rules governing the building of propositions S3 and S4 have been acknowledged, one understands that only three new propositions can be produced, for any other combination can be simplified to one of the existent propositions.
Since neither the subject nor the fundamental predicates changes throughout the saptabhaṅgī, the compatibility between Jaina logic and two main logic principles, i.e. the principles of contradiction and of excluded-middle, raises a real problem. The word syāt (sometimes wrongly translated as "maybe") doesn't introduce any modalisation in each proposition so as to express a probability; in that case Jainism would appear to be a doctrine of ontological indetermination or a scepticism. Firstly, the existence of seven propositions doesn't conflict with the principle of contradiction, for, despite the equivocity of the adverb syāt, each proposition is considered under a different respect. Secondly, the coexistence of two contrary predicates in one and the same substance can be explained thanks to an inter-expression which holds at the logical level as well as at the ontological one: in each substance the determinations characterising every other substances are inscribed. This inter-expression cannot be conceived by a mind whose faculties are limited by the power of karman. The fact that every empirical knowledge is partial results from the cognitive standpoint adopted by a finite mind. Because his capacities are limited, an ordinary cogniser sets the object to be known apart from the rest of the reality; while doing so, he treats as unilateral (ekānta) what is basically multilateral (anekānta). Therefore the omniscient being alone is able to know perfectly a substance; any other means of knowledge is imperfect and approximate.
Although the syādvāda should apparently apply to every statement (cf. LT 63), it doesn't bring the possibility of omniscience into question. Even though, according to Jaina cosmological conceptions, the highest stages of spiritual development cannot be attained any more in Akalaṅka's age (i.e. the duḥṣama era, the fifth era of the descending world-period), omniscience and salvation don't lapse. Both are attainable either in a very distant time or in another continent without cosmic cycles. Since soul has cognitive faculty as its natural property, it must absolutely be able to find this faculty again, even if some souls are not perfectible, i.e. are so much defective that an omniscient being can assert that they will never obtain salvation or omniscience. Thus, the sevenfold predication cannot apply to a statement like: "omniscience is a means of knowledge". Few statements, especially meta-statements, escape the saptabhaṅgī, because they are absolutely valid insofar as they aren't uttered from the empirico-practical viewpoint, but from the transcendental one. Nevertheless, in the empirico-practical reality, the partial and relative nature of judgements can be surpassed thanks to ersatz cognitions which perfect knowledge is diffracted in. Except the sevenfold predication, these substitutes consisting in viewpoints (naya), intentions of a word (nikṣepa) and points of investigation (anuyoga) are expressed by Akalaṅka at the end of LT in a chronological and hierarchical order: their sequence corresponds to the disciple's progressive deepening. Then Akalaṅka seems to say that the soul who has achieved this intellectual path attains omniscience and salvation; but such a statement is apparently incompatible with admitting that perfect knowledge cannot be obtained in the duḥṣama era. Which status could then be imparted to the author who follows, during the whole treatise, the thought processes supposed to lead to salvation? Could we compare him to a tīrthaṅkara? According to us, Akalaṅka aims at establishing an analogy between himself and a tīrthaṅkara, but his teachings can only deal with the mundane kinds of knowledge which the Jaina community may aspire to. What's at stake in the Laghīyastraya concerns the empirico-practical reality but not the ultimate one.
From pratyakṣa to parokṣa the empirico-practical knowledge
How to distinguish pratyakṣa from parokṣa? The ancient āgamic tradition characterised perception (pratyakṣa) as the cognition acquired by the soul, without its using any helper (sensory faculty, understanding); thus omniscience, telepathy and clairvoyance belonged to pratyakṣa, whereas sensuous cognition (matijñāna) was rejected in the indirect means of knowledge (parokṣa), as well as articulate cognition (śrutajñāna). However, following the way marked out by the Nandīsūtra and by Jinabhadra (6th c. CE), Akalaṅka propounds, in accordance with the distinction between two levels of knowledge the empirico-practical and the transcendental , a double architectonics of pramāṇa: according to the classification accepted at the empirico-practical level, sensory and quasi-sensory cognitions are parts of pratyakṣa; but Akalaṅka admits that this first classification is nothing but a concession to popular use, and strictly speaking only traditional classification is valid.
In order to justify the association of two kinds of perception empirical and transcendental in a single pramāṇa, Akalaṅka underscores a common characteristic: clarity. This characteristic doesn't concern the nature of the thing itself, for one and the same thing may be apprehended by different means of knowledge according to the circumstances. In this way Akalaṅka rejects categorically the dichotomy between objects of perception and objects of inference, as it was assumed by Dignāga and Dharmakīrti. Understanding whether clarity concerns the object of representation itself or the way an object is apprehended by knowledge is more difficult. Both hypotheses aren't equivalent because clearly cognizing doesn't only demand to cognize an object clearly, but also requires the second part of the cognitive process, i.e. apperception, to be clear. A cognition that is clear only as far as the object of representation is concerned is nothing but a case of fallacious perception. The degree of clarity produces a second opposition: whereas sensory perception is endowed with partial clarity, the three kinds of supra-sensory perception have a total clarity. This degree depends on whether there are conceptual constructions in the cognitive process or not. Total clarity is a feature of transcendental perception, i.e. the cognition by which the cogniser apprehends an individual in his ecceity.
But if clarity is a good criterion on which to distinguish sensory perception from supra-sensory perception, it isn't sufficient to establish exactly what sensory perception is according to Akalaṅka. In order to understand the nature of this kind of cognition, it should be firstly determined which stages belong to its process, and then conceived at what moment of a continuum devoid of breaks ignorance becomes knowledge. Akalaṅka views sensory cognition as a succession of four moments: apprehension (avagraha) of the form of an object, inquiry (īhā) into one of its property, determination (avāya) of this property and retention (dhāraṇā). The moment immediately following the contact between a sensory faculty and an object, the moment which consists in the mere intuition of an existent, doesn't belong to perceptive process yet. This intuition (darśana), in contradistinction to apprehension coming just after, isn't conceptual and, on this account, doesn't enable the cogniser to make a decision in the empirico-practical realm. For the existent apprehended by darśana is the mere positing of a thing, so that this cognition is devoid of content. In the empirico-practical reality, a cognition can be non-deceptive only if it is conceptual.
Sensory faculties are not sufficient to change sensation into perception, because they aren't able to produce concepts by themselves; this function is peculiar to understanding (manas). Thus manas activity is similar in the case of sensory perception (indriyapratyakṣa) and of quasi-sensory perception (anindriyapratyakṣa), which consists in recollection, recognition, inductive reasoning and inference. Moreover manas ensures a continuity between these two kinds of perception, both characterised by a partial clarity. The pre-eminence of sensory faculties or of manas is enough to determine the precise nature of perception. Understanding also plays a very important part in indirect cognition (parokṣa), insofar as it makes activity of language possible parokṣa is based on. In that way, there is a strict parallelism between the subdivisions of anindriyapratyakṣa and those of parokṣa; language distinguishes two kinds of cognition one from another: prelinguistic cognition comes within anindriyapratyakṣa, whereas the cognition endowed with the same cognitive content belongs to parokṣa from the moment that is put into words. Then a principle of continuity appears which is functional between the kinds of mundane cognitions (indriyapratyakṣa, anindriyapratyakṣa and parokṣa), as well as between the subdivisions of each kind; this principle may assume the shape of a chronological succession (in the case of the four moments of indriyapratyakṣa) or the shape of genetic relations between the cognitive validity of the four sorts of anindriyapratyakṣa. The same principle of continuity, of which one of the main manifestations consists in the ontological inter-expression, also works between the various empirico-practical pramāṇa and the transcendental ones.
To sum up, the validity of a cognition can never be absolutely asserted by a non-omniscient being; a finite mind is only able to determine whether a cognition is valid or not according to a pragmatic criterion: the nondeceptiveness in the empirico-practical reality. That's why using the sevenfold predication makes sense solely when judgements are uttered from an empirico-practical point of view; the sevenfold predication is nothing but an ersatz cognition, like the seven viewpoints, that enables an imperfect soul to get over the limits belonging to all kinds of empirical knowledge, so as to reach a higher stage of knowledge. For, there is no absolute break between empirical and transcendental cognitions. Akalaṅka aims at underscoring continuities between the different kinds and levels of knowledge: for instance, the notion of clarity establishes a link between both kinds of empirical perception (sensory and quasi-sensory) and transcendental or supra-sensory perception, whereas language is sufficient to change a cognition from the quasi- sensory perception to the indirect means of knowledge. Such continuities also reveal Akalaṅka's special position in the history of Jaina philosophy: he embodies a point of convergence between the traditional inheritance and a new rational way of thinking, that makes him still interesting for modern scholars.
Anne Clavel wrote her PhD dissertation on Akalanka's theory of knowledge at the University of Lyon (France), where she has taught Sanskrit and Indian Culture.
