NAVJIVAN RASTOGI

T H E O R Y 1OF E R R O R A C C O R D I N G TO A B H I N A V A G U P T A 2

The discussion of the problem of error represents an important facet of Indian Philosophy. It has been inspired by a basic concern for complete freedom, a concern to elucidate the nature of false knowledge and how it could be removed. The present paper confines itself to the Abhinavan answer to the problem. We propose to state the major premises of his thesis before we come to take a close and detailed look into it. Abhinavagupta's main interest seems to be evolving a model in whose terms all possible analyses of error may be classified. Abhinava shares the general consensus among the Indian philosophers that the test of validity of knowledge is its uncontradictedness. The Saiva view is known as Akhy~ti, an abbreviation of ApSrnakhy~ti, which means imperfect knowledge. Thus the Abhinavan theory of error is a corollary of the Saiva metaphysics of ignorance - e.g., ignorance is imperfection brought about by the obscuration of one's real nature, which is a synthesis of knowledge and freedom, through its own integral dynamism. Abhinava talks of three kinds of ignorance - (i) cosmic or metaphysical, (ii) empirical and (iii) epistemic or subjective. It is the third type which we are concerned with. The important thing to be noted in this connection is that the metaphysical error was deliberately modelled after the epistemic error i.e., common illusion. In answer to the question, "what is an error", Abhinava appears to hold that an error is incomplete knowledge which is incidental to the fact that the object does not continue to figure in as perfect form as it should upto the time of a final judgement. Incompleteness is a qualitative concept which consists in the lack of correspondence between objective content and its cognitive ascertainment. There are two explanations in respect of this incompleteness - e.g., (i) Incomplete cognition of a locus is responsible for erroneous cognition of its content. (ii) Want of exposure to other percipients renders it a privative, hence incomplete, judgment.

Journal of Indian Philosophy 14 (1986) 1-33. 1986 by D. Reidel Publishing Company.

NAVJIVAN RASTOGI

To the second question, "how is the error caused", Abhinava's answer may be summed up as follows: (i) The equation among the components of an erroneous situation is: illusory content = silver = Uparafijaka locus = nacre = Uparafijya unification = Melana = This is silver = Upar~ga. (ii) Falsity is caused by rest on an other locus than one's own, because the Adhikaran.a (locus) lets itself be inadequately grasped. (iii) Error consists in the incorrect unification of objective components (brightness of silver mistaken for that of silver). (iv) Lack of contemporaneity in the different objective constituents leads to disorder in unification. (v) Self-consciousness which witnesses both the illusion and its corrective cancellation is the constant factor. (vi) It is subjective (mis-)unification that leads to the private character of illusion occasionally aided by sensory and environmental defects and the like. In this connection certain incidental issues are subjected to close examination. Accordingly, by an incisive analysis of.~bh~sav~ida and its application to the phenomenon of error an attempt is made to show that (a) Unification is nothing but linking of constituent .~bh~sas to the principal ~,bh~sa. (b) Not all, but only those .~bh~sas that are not mutually opposed, get united due to the law (power) of determination which is a Saivist transformation of the Buddhist Apohawida. (c) Erroneous judgement is a synthetic judgement and depends upon the reversal of recognitive unification. 'Ubhayamelan~tmaka-prakriy~' is not exclusive to the recognitive knowledge alone, but is a necessary corollary of the concept of ~,bh~sa being applied to the field of epistemology. (d) Silver alone, to the exclusion of others, is seen in nacre because of similarity of brilliance. 'Silver' is a representative element, while 'this' (nacre) is a presentative. (e) A judgement forfeits its right to make a truth claim if there is disagreement between the perceptive content and determinate judgement. This is again a corrollary of the process of unification. (f) Due to transposed unification a logical fallacy called Vyadhikara.n~isiddha (lit. the fallacy of disproved substratum) takes place in the erroneous cognition.

THEORY OF ERROR IN ABHINAVAGUPTA

In respect of the discovery of error four questions have been raised - (1) How is the error discovered and cancelled? (2) What is the object of cancellation? (3) What is the result of cancellation? and (4) Is concellation a negative function? All the questions have been taken up together. Correction is relative to immediate past experience, it is not an absolute negation. Cancellation is always addressed to the aspect of unification (Melan~rh~a) because it is the precise point of error. The essence of cancellation lies not only in breaking the continuity but also in ensuring that the existing judgement should not have arisen at all. Cancellation as absolute negation does not fit in with the wholistic absolutism of the Abhinavan metaphysics. Negation, therefore, must be construed in terms of the affirmation of something. The relation of existence of one thing with the absence of another is the same as that of two things standing in relation of a locus and its content. It is, therefore, not the nacre-as-such, but the nacre-as-qualified-by-the-absence-of-silver that requires corrective perception. Cancellation is called a real subjective construction in the sense that the knowledge of nacre or non-silver by itself is ineffective unless the two are mutually coordinated by a dependence relation of negation, providing them a subjective berth in the form of a single knower. From the foregoing discussion the following equation may be formulated: Contrary Object of negation Negation Subsequent cognition = Element of Unification = Thisis nacre/ This is not silver. = Silver(-in nacre)

= Break in the continuity of = There has been of current judgement. no silver. Correction = (a) Restoration of identity of Vimar~a with Prak~a = (a) This is nacre which was seen as silver. (b) Restoration of the capability of being perceived by others (b) This is nacre.

NAVJIVAN RASTOGI

The next question that interests us is to fred out the nature and ontological status of the illusory object. The Abhinavan position approximates the third of the three "standard" positions. Although Abhinava is a staunch absolutist he advocates adherence to a Bhed~bheda (unity in multiplicity) position at the empirical level which is the main concern of epistemology. It appears that Abhinava makes a distinction between real and empirical (or actual) on the one hand, and between object and content on the other. The content may be 'real' but not 'empirical'; while 'serving the purposes' may be the criterion of empiricity, and 'figuring in cognition' that of reality. On this criterion the content is real. In answer to the question whether an erroneous perception is a simple judgement of a complex one, Abhinava says that one cannot lay down the absolute rule that illusion will always stem from determinate perception alone. As such it would be a complex judgement. But he later on modifies his stand to the effect that error is a determinate judgement as it involves the functions of synthesis and analysis. To sum up. There is ample evidence to show that Abhinava wanted to develop Apfirnakhy~ti as a comprehensive model for all the theories of error. However, in respect of the theories mentioned by him, he has used a nomenclature different from the usual one to reflect his understanding of these theories. It is no doubt surprising that he nowhere refers to Prabh~kara's theory of error called Akhy~tiv~ida. Due to apparent similarity of names some scholars have tended to confuse Abhinava's Akhy~ti with Prabhkara's. However, with the scanty nature of our information and paucity of material the problem is left for future study. As Potter rightly points out, a the basic concern of a seeker after truth is to convince himself that erroneous cognition can be dispelled for good so that complete freedom may be permanently realized. It is in this context that the Indian philosopher raises the question of the nature of false knowledge and the question of whether and how it can be removed. The various theories of error and negation provide answers, offered by different thinkers, to these questions. Although all these answers are at times very illuminating and fascinating, we propose to confine this paper to the analysis of the answer offered by

THEORY OF ERROR IN ABHINAVAGUPTA Abhinavagupta 4 in the overall context of his metaphysics of ignorance. The special significance of his effort lies in the fact that, true to his encyclopaedic genius, he tries to develop a basic pattern or fashion a model where the divergences o f all the various explanations of error may find a common ground, s Abhinavatupta shares the obsessive insistence o f the Indian mind to do away with the factors that in any way defile, negate or tamper with knowledge. This becomes transparent from the definition of valid cognition That (Pramft0.a) as determinately cognizing within itself an object, for which a single expression stands and which is free from temporal and spatial limitations, is the cognition (Miti), provided that it is uncontradicted. 6 Notwithstanding varying emphasis stemming from doctrinal considerations and occasional changes of precise terminology, the essence of most definitions or accounts consists in prescribing non-contradictedness or objective confirmation as a test of the validity of knowledge. Abhinava does not mince words in declaring that a Pram~na ceases to be a means of knowledge the moment it fails to subscribe to this characteristic feature.7-a The theory of error, ironical as it may seem, is known as Khy~ti, that is, knowledge.9 Thus a theory of error by implication aims at accounting for the experience of error which is a normal empirical phenomenon. However, the Pratyabhijfi~ opts for Akhy~ti i.e., non-knowledge, to describe error lo which undoubtedly is a more appropriate choice. The negative prefix (Nail) in the compound Akhy~ti has been interpreted as meaning incomplete or imperfect (Apfirna). 11 Thus for Abhinva it is an instance of a tatpuru.sa compound) 2 As postulated above, the epistemological scheme of a system must ensue from and substantiate the metaphysical design of the system. It is, therefore, understandable that the negative prefix - Nail - does not negate knowledge or its object, it conveys something positive. It means 'slight' or 'little') 3 In other words, Abhinava being an absolutist, questions the very existence of error in the negative sense, that there is an absence of knowledge or an erroneous or illusory object. This reminds us of a similar response from the other extreme represented by the Pr~bh~karas that illusion does not mean absence or negation of knowledge. All that goes by the name of error in common parlance is, in fact, incomplete knowledge. We shall enlarge upon the precise implication of such a stand in the sequal, but here one must note that this viewpoint is a logical corollary of the Saiva metaphysics of

NAVJIVAN RASTOGI

ignorance. 14 The same term Akhy~ti has been used to designate cosmic ignoranceJ 5 Knowledge or integral consciousness is the nature of the self which is a harmony of being and becoming, knowledge and freedom. Ignorance is imperfection 16 brought about by the obscuration of either aspect of the self through its own integral dynamism. This is known as ,~navamala (atomic impurity), 17 loss (i.e., concealment) of intrinsic absolutism) 8 This is known as imperfect consciousness, ApfirnammanyatL x9 This is cosmic ignorance and is responsible for the emergence of the creative order. Self-alienation (Apohana or Vikalpa) and finitude (Sarhkoca) characterize it. The Kashmir Saivism advises recourse to spiritual initiation (Siva-d~sa) to get rid of this ignorance. Next follows another kind of ignorance - intellectual ignornace (Bauddha Ajfi~ina).2 While the first sort of ignorance affects the cosmic being, the second conditions the intrinsic thought-mechanism which invariably leads either to non-ascertainment or to contrary ascertainment. It consists in not recognizing the self at all, or treating the not-self as self and self as not self. This is empirical ignorance. The system propounds four Up~yas marked by the gradual refinement of thought-constructs (Vikalpas) to steer clear through this phase. We are now in a position to describe the first stage as cosmic or metaphysical error, the second as empirical error and the third as epistemic or subjective error which consists in incomplete or inadequate apprehension of the object. Abhinavagupta while projecting the logical continuity and, in a sense, the progression of the different sorts of error has also underlined their mutual distinction by dividing them into two groups. The first two are portrayed as great illusions (Mah~bhr~nti) and the third as simple illusion 2~ (Bhr~nti). The first is universal error, error common to or involving all subjects; the second is individual error, error confined to individual subjects in their determinate reactions to empirical objectivity; the third is epistemic error, error which forms an extraordinary subjective experience. 22 It is this third form of error that is our main subject of discussion. The task is especially worthwhile because metaphysical ignorance is deliberately modelled after common illusions. 2a This attitude finds a close parallel in the Advaita Vedanta, and in the Yog~ic~ira and M~dhyamika Buddhism which approach the problem in the same spirit. To tackle the first question first, "What is an error?" The same question can be differently posed - "What is the essence of an illusion?" The phenomenon of illusion is a datum of our experience and has been

THEORY OF ERROR IN ABHINAVAGUPTA

variously styled as Bhr~nti, Bhrama, Viparyaya, Viparfta Jfi~na, Akhyfiti and, of course, Adhy~sa or Mithy~ Jfi~ina by the celebrated Sarflkara. Essentially they all designate the same thing, that phase of experience which turns out to be invalid on cancellation or correction by a later experience. The Saptapaddrthi explains Bhrama as the reverse (Viparyaya) of valid knowledge (Pram~i). 24 It is Atattv~inubhava as against Tattv~nubhava. The Nydya-vdrtika describes it as wrong apprehension (Mithyopalabdhi) in which an object is taken for what it is not. 2s Viparyaya consists in cognizing an object as other than what it really is. If valid knowledge lies in knowing the object as it is, Viparyaya lies in doing just the opposite i.e., in knowing the object as what it is n o t . 26 The Veddntasdra 27 and Kum~irila 28 echo the same view. Error takes place when a thing is comprehended as something which it is not. In other words, error consists in attributing such characteristics to an object of perception as are not actually found in it. 29 It is called Viparftanirn.aya as it contradicts the real nature of its object and, as such, turns out to be a false judgement about the object through attribution of such qualities as are precluded by it (Vipari-tadharm~dhy~ropan.a). It amounts to saying that illusion or error represents an object in a form which does not belong to it. Sarhkara by using the term Adhy~isa enlarges the scope of attribution. Two other words Adhy~ropa 3o (superimposition) and Avabh~sa 31 (appearance) have also been used to convey the same meaning. Adhy~sa is nothing but the apprehension of something as something else. 32 This something is both objective and cognitive. Thus the illusion consists in apprehending an object as being other than it, either by ascribing one thing to another (such as taking a rope for a snake) or ascribing the attribute of one thing to another (such as perceiving a white conch as yellow). In both these ascriptions 33 the superimposition of the cognition of one object inheres in the other object. Unless one is swayed by scholastic blases the element of superimposition (Adhy~ropa) or attribution (~-ropa) necessarily implies a reference to a substratum or locus (Adhi.s.th~ina, Adhikarana, Adh~ira) of the object of erroneous cognition. In it one object (or universal, if we borrow the term from the Ny~iya, silverness - Rajatattva for example) is referred not to its own locus (silver) but to that of different object (or universal, so to say nacre - Sukti34). In all the three accounts there is agreement that an error is a 'false apprehension o f the object' by negation of its real identity or by predication of attributes it does not possess to it. It arises in the wake of our relating,

NAVJIVAN RASTOGI

confusing, or failing to differentiate two or more cognitive objects which happen to be unrelated in actual life. We realize our error when a gap is discovered between our cognition and activity inspired by it. This is what is known as contradiction, negation, sublation or correction. It exposes the absence of a correspondence between the content of a judgement and the objective world and thereby proves the invalidity of such experiences. Coming to the Kashmir ~aivism, if we go by the definition of valid cognition mentioned earlier, the essence of an error lies in its being subsequently contradicted. If asked what type of knowledge is open to being contradicted, or better, corrected, the pet reply of the Saivite is: one which is incomplete. No precise explanation of the notion of 'incomplete knowledge' seems to be available. Only once, as in a parenthesis, does Abhinavagupta afford us a clue to his understanding of incompleteness. He presents the phenomenon of error as proceeding from lack of such complete comprehension as is adequate and efficient in the circumstances. Probably he means to say that the twin characteristics of adequacy and efficiency are the sine q u a n o n of any definition of completeness. Abhinavagupta probably proceeds with the assumption that we all are well briefed on his position. A cryptic remark made by him, however, appears to hold the key to his understanding of error as incomplete knowledge which is incidental to the fact that the object does not continue to figure in as perfect form as it should upto the time of judgement. 36 According to Bhfiskara, Abhinava's commentator, such incomplete knowledge is found in instances like the perception of silver in a piece of nacre, a7 But how the cognition of nacre as silver constitutes incomplete knowledge has not been explained. Abhinava is also silent about the complete form (Pfirn..arfipat~) of an object. Nevertheless, in fairness to Abhinavagupta, it must be said that incompleteness (Apfirn.atfi) is not a mere quantitative concept, it seems to have acquired a qualitative connotation as well, as if Abhinava subscribed to the dictum of Marx and Engels that at a given stage quantity is transformed into quality and vice-versa, as Hence, if we reconsider Abhinava's remark, cited above, we shall learn that he holds that completeness consists in the final evaluation of an object. In other words, the object must continue to figure uninterruptedly in our determinate judgement. But the negative discovery that it is not silver but nacre interrupts the continuous flow of our previous cognitiofi with regard to the object i.e., silver. Therefore, completeness consists in correspondence or dialogue between an objective content and its cognitive ascertainment. The

THEORY OF ERROR IN ABHINAVAGUPTA

absence o f such correspondence would disrupt the continuity of cognition and render the object incomplete. 39 There have been two types of explanation in respect of this lack o f correspondence. According to the first explanation when one notices silver at the sight of nacre, the silver in fact does not exist there because the substrata of the two cognitions (e.g., silver- and nacre-cognitions) are found to be different (silver and nacre respectively). 4 On a closer logical analysis we find that the incomplete cognition of one locus is responsible for the erroneous cognition of its borrowed content. 41 This position is surprisingly close to the Naiy~yika's. It predicates silverness which is a quality of a thing (silver) to something else (nacre) where it does not exist. 42 According to the second explanation Abhinavagupta refers to the common perceptibility of the objective content of cognition, the absence of which is liable to keep our knowledge incomplete. The cognition of an empirical whole must be open to all eligible perceivers in order to justify its identity as an object. 43 The moment its doors are closed, it suffers from privation and degenerates into an error and thereby any judgement about it becomes a partial judgement. It is the reference to this lack of common perceptibility which lies at the root of the non-existence of an illusory object in its locus and the falsity of the relevant judgement .44 But as we shall see later, the word Apfirn.a does not connote the Saiva attitude fully, because the phenomenon of illusion is not a simple act of omission. It is not that we visualize a particular aspect of the object and overlook or miss another. Instead, we add something to the object. It is not that we simply miss something of nacre itself but that we also bring something o f silver to it. It is in this sense that one finds the word 'incomplete' deficient in connotation. We propose to look into this further as we move on to the next issue. The next question that crops up is - "How is an error possible?" To put it differently, one may ask - "How is error caused?" Broadly speaking 4s Indian thought seems to offer two alternatives. According to one alternative a cognitive experience is erroneous by its very nature. The M~dhyamika and Yog~cfira Buddhists must be mentioned here. For them epistemic error is imply a logical extension of metaphysical error. Others believe that knowledge is valid or capable of being valid if certain conditions are fulfilled. They thereby simply that it is likely to be erroneous under characteristic conditions. Most of the Indian schools, especially realists, subscribe to this view holding that these characteristic circumstances normally

10

NAVJIVAN RASTOGI

occupy the domain outside our knowledge. There is, however, a third position, if one can call it that, which accepts the intrinsic validity of knowledge and yet holds that causes of error are its own creations by granting empirical reality to them. This last option represents a sort of synthesis between idealism and realism and may be said to be captained by the Saivism of Kashmir. The basic question is whether one can identify any cause of illusion. To quote Murti, 'To assign any cause for the appearance is not to take it as a mere appearance, as we should, but as the appearance of something. 46 Abhinava has named his theory as Apfirnakhy~itiv~da implying that there is no error as such. It is in contrast to complete knowledge that incomplete, partial or limited cognition comes to be considered as erroneous or illusory. So far as our initial perceptual reaction is concerned, there is no error. Error is discovered only when the continuity of a judgement e.g., 'this is silver', is interrupted. It is the analysis of this discovery and the corresponding corrective knowledge that leads one to decide upon the cause (or causes) of this error. Abhinavagupta following his master Utpala picks up two stock examples for analysis - one is "this is silver" and the other is "there are two moons in the sky", the former typyfying mental errors and the latter sensory errors. 47 In both instances Abhinava finds that error is due to subjective unification of objective components (]~bh~isas) while in the latter case an optical defect also makes a partial contribution. When we perceive silver at the sight of nacre and exclaim "oh, this is silver", we judge it to be a piece of silver, which is solid, capable of being cognized as such by other percipients and of serving its purpose (i.e., efficient). Thus silver as an object represents unification of thisness, silverness, solidity, common perceptibility and efficiency in one locus. All these taken separately are objects themselves and are real because they are individually judged to be so, but their unification in the form 'this is silver' proves illusive because it is not supported by a succeeding judgement, arising from the percpetion of nacre, which takes the form "This is not that silver which is solid, open to common percipients and capable of discharging desired activity." In fact the silver which has been negated by the subsequent cognition was the proper (should-have-been) object of the previous judgement. But as this absence of silver in nacre escapes our attention, error results. ~ In other words, our judgement suffers from the failure to grasp the frontal object i.e., nacre, in its complete form due to a defective subjective unification in respect of the objective components. Thus error in judgement consists

THEORY OF ERROR IN ABHINAVAGUPTA

ll

in incomplete predication of the object (this = nacre). The contrary cognition disrupts the continuity of the determinate judgement in respect of the unification from the rise of the perceptive judgement itself. 49 Thus the error lies in unification or subjective synthesis. According to Abhinava what happens is that the piece of nacre which is represented by 'this' acts as a limiting adjunct in the form of the locus of 'silver' (because of its silvery brightness and the rise of an image in memory) and is responsible for the resulting confusion in the judgement 'this is silver'. Since the two stages of former and later experiences are found to be mutually opposed owing to their lack of correspondence, the previous judgement is found to be illusory, s The same is the case when, due to an opitical defect or external pressure on the eye-ball, one perceives two moons. Here again the duality of number, moon, sky and their connection taken independently do not betrary any confusion. Error arises the moment they are mixed up. Here the portion of sky which as the locus of the moon acts as the moon's limited adjunct is confused with the sky as locus of two moons due to a sensory defect which results in such unification which is falsified by later cognition, for the judgement "there are two moons" does not continue after one moon has been sighted by normal vision. Thus it is the cognitive (mal-)unification of objective constituents that causes error. Error always consists in the element of unification in our cognition lacking correspondence of judgement with its relative limiting locus. Thus we are led to certain conclusions with regard to the cause of illusions: (i) There is the following equation of relationships between the components of an erroneous situation (a) illusory content = silver = Uparafijaka. (b) locus = nacre = Uparafijya (c) unification = Melana = This is silver = Upar~ga. (ii) Falsity (Vaipar{tya) is caused by resting on a different locus other than one's own, because the Adhikaran.a lets itself be inadequately grasped. (iii) Error consists in the incorrect unification (Melana or Yojan~) of objective components (brightness of nacre mistaken for that of silver). This leads to the appearance of something new i.e., silver. (iv) The lack of contemporaneity in the different objective constitutuents leads to disorder in unification. In other words, error is also due to the the association of subject and predicate with two different orders of time i.e.,

12

NAVJIVAN RASTOGI

presentative (nacre) and representative (silver). The operation of memory. traces left by a valid experience of real silver lends content to our illusory experience. (v) Self-consciousness which witnesses both the illusion and its corrective cancellation sl is the constant factor. (vi) It is subjective (mis-)unification that is responsible for the privative character of illusory perception, as aided by sensory and environmental defects, appearance of similarity, etc. A few observations may help give a better appreciation of the Saiva position. The wholistic approach of the Saiva absolutism treats reality as a synthesis of being and consciousness (Prak~a and Vimar~a). Thus the objective manifold, like its subjective counterpart, depends upon the Prak~a aspect for its being and upon the Vimar~a aspect for its mutual distinction and unification. Thus the 'this' aspect stands for universality and being, and 'being of such and such nature' represents its peculier character, s2 The former is known as the universal object (S~mhny~bh~isa) and is totally free from reference to time and space etc. This constitutes the basic unit of objective knowledge and is designated by a specific word. The latter is known as a particular object (Svalaksan~bh~sa) which is a sort of common locus of different characteristics which combine to give us an objective entity, s3 The operation of the means of knowledge extends to all objects. The individual character of an object is not fixed, it is dynamic and is a by-product of unification and as such admits of a variety of cognitions according to the taste or purpose or the intellectual equipment that one, that is the subject, brings to the object, s4 Thus, while the object represents the Prak~a aspect, the particularity wrought by determinate activity of the subject consisting in determinative synthesis (and, by implication, analysis) represents the Vimarga aspect. Thus the Saivist appears to take a gestalt view of objectivity. Accordingly, in the configuration of ~,bh~s~s there is always a principal Abh~sa, which is taken to be the locus of all others. Thus unification or co-localization (S~mhn~dhikaranya) is nothing but the linking of these Abh~isas to this common locus, ss Once a configuration is formed, its erstwhile different components now constituting a single whole discharge a collective function, even though continuing to retain their individual distinction. Thus the synthetic unity of Abh~sas or their individuality is a result of corresponding determinate subjective activity called Vimarga. It is geared to a particular formation governed by a particular purpose, s6 But a question may well be

THEORY OF ERROR IN ABHINAVAGUPTA

13

asked - why is it that only certain Abh~sas are mixed together and not others? To this Abhinava answers that only those ]~bh~isas, which are not mutually opposed, are united. For example the b,bh~sa of colour does not unite with that of wind, because they are contrary to each other. This opposition is a direct result of the power of determination (Niyati). s7 Shorn of mythical jargon, Niyati is determination. Every determination is a negation. Thus while a definition predicates a certain character to an Abh~sa it also distinguishes it from others by the implied negation of that character elsewhere. This is reminiscent of the Buddhist concept of Apoha with one difference. In Buddhism negation is conceptual, here it is the logical manifestation of absolutic essence. It is why it is supposed to be ingrained in the very nature of things that an epistemic object must be an ~.bh~sa or a unity of Abh~sas guided by the law of absolutic determinism, ss Thus, as a logical extension of the same argument, it is the continuity of the determinate cognition that is responsible for the unity of the principal ~.bh~sa which constitutes the locus of component ~,bh~sas. s9 This again, incidentally, brings us face to face with the metaphysical foundation of the Saiva absolutism known as 'Pratyabhijfi~', that is, the unity of various .~bh~sas, either at different points of time or space or through different channels of knowledge, is a synthetic process. This process brings out the underlying unity of the entire manifold through the cognitive unification of being and consciousness reflected in the synthetic judgemental (and also, at the metaphysical level, in the experiential) activity. It is, therefore, in the fitness of things to deduce that since recognition forms the core of all cognitive activity, erroneous knowledge too must depend upon the reversal of recognitive unification. There must be a lack of recurrence, continuity or synthesis of judgement in respect of the object of illusory experience. It will be quite logical to conclude that the "commingling process" (Ubhayamelan~tmaka Prakriy~) 6 in the system is not exclusive to the recognitive knowledge alone, but is a necessary corollary of the concept of .~bh~sa being applied to the field of epistemology covering all the Pram~nas. This is why, as we have seen, Abh~sa (ontological entity) and Prameya (epistemic entity) are not distinguished. If we closely follow the present line of thinking it becomes fairly easy to answer the doubtful opponent: there will be no talk of error if it is accepted that the determinate cognitive activity is the sole determinant of intraobjective difference and non-difference. Because, such a view will necessarily lead us to the conclusion that even nacre will turn out to be silver in reality,

14

NAVJIVAN RASTOGI

since it is so comprehended by the determinate judgement "this is silver". There being, therefore, no illusory experience, the possibility of its contradiction will not arise at all and this will obviously do away with the utility of positing 'uncontradictedness' as a necessary ingredient of the definition of valid knowledge. 61 Abhinava's answer to this is very clear: the judgement elicited by the perception of nacre as silver is shattered the moment nacre is apprehended. Since the two judgements (i.e., 'this is silver' and 'this is not silver') do not agree and the original judgement does not continue after the nacre has been seen, it is proved to be iUusory. 62 This further explains how error enters the process of unification. According to the theory of .Z,bh~sav~da, it is the principal ~,bh~sa that constitutes the substratum while all other relative .~bh~sas constitute its collective content. While we see nacre our natural judgement would be 'this is a piece of nacre', where 'thisness' represents universality and 'nacre-hood' constitutes particularity. Since a particular ~,bh~sa is always a configuration, a gestalt, the nacre must constitute the locus for other related ]~bh~sas to be united and to discharge a collective function. The other .i.bh~sas comprise present time, solidity, general perceptibility (capability of nacre to be perceived as such by other perceivers) and capacity to lead to the desired activity. It is a moot point that this judgement continues to last and is verified by the achievement of the desired result. But the whole process comes to a halt when we mistake nacre for silver, giving rise to the judgement "this is silver". Here all other .~bh~sas cooperate, but the principal ,~bh~sa i.e., nacre, is replaced by silver which now takes up the job of constituting the locus. At the same time the urge for collective functioning undergoes a change due to the change in the principal ,~bh~sa 63 which is deprived of its claim to objectivity (the capability of being perceived by other cognizers). Again it is a moot point that failure to obtain a desired result breakes the continuity of initial judgement and hence is discovered to be an illusory experience. Though the discovery is a subsequent event exposing it to truth, the confusion in the process of unification is the initial basic factor that causes illusion. However, one very relevant question remains. Why is it silver which is erroneously cognized in nacre to the exclusion of all other bright objects? Only a few scattered references give us a clue. Glittering brilliance is something that is common to both nacre and silver. When one perceives silver at the sight of nacre, the latter appears to have acquired the brightness of silver. This

THEORY OF ERROR IN ABHINAVAGUPTA

15

brightness recalls the memory of silver from a past valid cognition. This identification of the remembered object with the object of immediate perception, which is reflected by 'silver' and 'this' respectively, gives rise to the judgement 'this is silver' 64. The representative character of silver is possibly arrived at by a process of elimination - were the silver an object of direct perception in nacre, it would have caused the illusion of silver to everybody. In a slightly differently worded statement, Mahe~var~nanda, a devout successor of Abhinava from the south, remarks that while the apparent nature of nacre is cognized through perception, the same is logically constructed (determinately grasped) as silver. 6s Perhaps Mahe~var~nanda is following the line of his predecessor Abhinava who also considers nacre to be an object of perception and silver an object of determinate judgement. A judgement may make a truth claim if its perceptive content is endorsed by a determinate apprehension. It forfeits its right to do so if there is disagreement between the perceptive and determinate judgements. Here error lies in the fact that the object of perception - i.e. nacre, is falsely identified with the object of determinative judgement - i.e. silver, which is the oppoiste of the former. 66 This type of confusion is again a step towards error in the process of unification itself. Abhinava has gone far enough to examine the proposition 'this is silver' and has come to the conclusion that such a unification, though entirely subjective, 67 is logically fallacious. According to Abhinava the unifying arrangement in instances like this tends to adopt the following course. Even though silverness is grasped as a property inhering in an other substantive - i.e. nacre (than one's own, i.e. silver), and nacre is grasped as a locus/substantive of other property - i.e. silverness (than one's own, i.e. nacreness), the substantive-locus (nacre) has been responsible for the adjectival character of the property (silverness) and the dependant property (silverness) has been responsible for the substratum-like character (nacreness). This transposed unification is a logical fallacy and has been designated as Vyadhikaran~siddha or ,~ray~siddha (i.e., one whose substratum is disproved). Such a fallacy arises when, as seen above, something is portrayed as an attribute of some such locus where it does not belong. ~ Such a treatment, however, brings Abhinava very close to the Naiy~yika view of Anyath~ Khy~ti. By the way, by borrowing two terms from Sivaraman 69 which he has used in the context of R~m~nuja's Satkhy~iti, here also the condition of

16

NAVJIVAN RASTOGI

error may be classed in two categories. While the mental transposition of an object from actual locus to the other is the 'epistemic' condition, the constitution of an ,~bh~sa as a unification of several objective components furnishes the 'constitutive' condition. The next phase of our enquiry pertains to the discovery of error. Four questions naturally occur e.g., (i) How is the error discovered and cancelled? (ii) What is the object of cancellation? (iii) What is the result of the cancellation? and (iv) Is cancellation negative function? We shall try to fathom Abhinava's mind on these problems. As we have seen, the judgement 'this is silver' is found r o b e erroneous on the strength of a subsequent cognition taking the form 'this was not silver'. Implicit in the judgement is the viewer's belief "there was no silver, only I have been misled ''7 . Thus the predicate 'not silver' refers to the same subject, which was predicated by 'silver'. In fact 'this is not silver' should have been the appropriate judgement at the time o f the initial experience, which is now restored by the later experience. 7~ Thus the subsequent cognition supplements the incomplete perception of the previous movement. As the cancellation temporally comes later, our corrective judgement involves a subtle reference to time, although it rejects the truth claim of a previous judgement as such. 72 In other words, the "this-was-notsilver" cognition seeks to cancel the "this-is-silver" cognition which has for its object silver that came to appear at a given point of time in the immediate past. It does not cancel silver absolutely at all points of time and space. Thus the correction is relative to the immediate past experience; it is not absolute negation, it does not even negate silver-cognition, if any, in the present time. 73 The nature of the remover or negator (B~idhaka) is essentially cognitive i.e., it is a different cognition 74 which takes the form - 'this form is not silver' or 'there is no second moon in the sky'. It may be recalled that according to Abhinavagupta an object is a synthetic unity of universality and particularity, where the universality refers to its general being as Prak~a and particularity to its peculiar attributes as Vimarga comprehended by a name e.g., silver. This particularity again is an outcome of the union of various objective components, which is ipso facto a cognitive phenomenon. Cancellation is always addressed to the element of unification (Melan~ria~a), because it is the precise point of error. 7s This is achieved by removing the continuity of erroneous judgement e.g., "this is silver" or "there are two moons". The continuity o f judgement has been described as

THEORY OF ERROR IN ABHINAVAGUPTA

17

correspondence, 76 continuity 77 and recurrence 7a on different occasions. The main thrust of the cancellation is to uproot its presence. The cancelling judgement points out the gap between the cognitive judgement and its objective reference 79 "here there has been no silver" and "there are no two moons". However, discontinuity of judgement as constituting cancellation should not be understood in absolute terms. In a sense, each subsequent congition always replaces the current cognition and thereby removes the continuity of an existing cognition by implication. After reading a book if one diverts his attention to writing, the previous judgement 'I am reading a book' will naturally be discontinued by the following judgement "I am going to write". But, shall this constitute negation? Abhinava's answer is a clear "no". He says that the essence of cancellation lies not only in breaking the continuity but also in ensuring that the existing judgement should not have arisen at all. a That is why, according to him, the spontaneous corrective judgement is not that "this is not silver", but that "there was no silver here". That is why Asarhv~ida is supposed to be more expressive and, in fact, is sometimes posited as a cause of the discontinuity of cognition. Without fear of being contradicted it may now be concluded that the break in the continuity of cognition arising from the lack of correspondence between the cognitive judgement and its objective counterpart leads to the discovery of error which involves the correction or cancellation of error, al The phenomenon of cancellation is not as simple as it appears. Our encounter with nacre may give rise to two types of judgement e.g., (1) 'there has been no silver', and (2) 'this is a piece of nacre'. In the Ny~ya terminology the two instances represent the two cases of Vi.say~pah~ira and Phal~pah~ira respectively, a2 According to Abhinava negation is a positive function because it does not cancel, but corrects. "This is nacre" is the corrective phase of our perceptive judgement "there has been no silver". It is this type of approach that explains the concept of Apfirnat~ in respect of knowledge. Cancellation as absolute negation does not fit in with the wholistic absolutism of the Abhinavan metaphysics. Negation, therefore, must be construed in terms of the affirmation of something. The relation of the existence of one thing with the absence of another is the same as that of two things s3 standing in the relation of a locus and its content. Thus 'this is a piece of nacre' is a positive mode of expressing the negation of silver i.e., 'there has been no silver'. Lest capriciousness take over, the dependence relation must function between the two. Nacre-cognition is the locus of non-silver-cognition or,

18

NAVJIVAN RASTOGI

better, nacre is the substratum of non-silver, a4 It is not the bare silver as such that is being negated by the perception of nacre, it is that silver which was identified with the nacre-locus. Thus the object of corrective judgement is to place the nacre in correct perspective. It is not the nacre-as-such, but the nacre-as-qualified-by-the-absence-of-silver that one perceives. 85 Thus Abhinava discovers a sort of reasoning between the two experiences. The silver-(in nacre)-experience is a datum which is not in dispute. The point of dispute is the invalidity of silver in the silver-(in nacre)-experience resulting from the discontinuity of the previous perceptive judgement on and after the perception of nacre. Thus the perception of nacre acts as the reason (minor term - Hetu) and the invalidity of silver-(in nacre) judgement as the S~dhya (major term), s6 Two incidental observations: (i) The discontinuity of judgement by the cancelling judgement involves a reference to the consciousness of relevant objective efficiency, and (ii) Like error, its correction also is a subjective function. To elaborate: according to Abh~isav~da, an ,~bh~sa by definition is constituted, inter alia, by a reference to the expected fruitful activity. The contrary judgement arising from the sight of nacre, being itself an ~,bh~isa, weakens the grip of earlier reference to the desired objective efficiency (of silver) and, as such, provides an additional fillip towards cancellation of the previous judgement, s7 This is in tune with the Abhinavan stand on empirical reality as constituted by the twin criteria of continuity and usefulness as and also in tune with the concept of ~,bh~sa as a 'whole' made so by the taste, intellectual equipment and volitional activity of the subject, s9 In fairness to the ~aiva absolutist, however, a word of caution is necessary. The capacity to bring about a desired result is, in itself, determined by the relative cognizing subject and, as such, plays only an ancillary role so far as it supports the principal factor in the context i.e., Vimarga. By itself it is not a sole decisive factor in any corrective judgement. 9 Correction or cancellation is a constructive activity on the part of the subject. It is not simply a synthesis or repetition, it is a real construction in the sense that the previous cognition is distinguished from the subsequent one which, because of former's failure to maintain consistent predication, cancels the former. While the constructive activity in error is purely mental hence confined to the cognizing subject alone, it is bidimensional in correction (i) it rejects the previous 91 judgement and therefore breaks the continuity
-

THEORY OF ERROR IN ABHINAVAGUPTA

19

and (ii) it demolishes the barriers of exclusive subjectivism by opening the gate to other perceivers. 92 It is called a real subjective construction in the sense that the knowledge of nacre or non-silver by itself is ineffective unless the two cognitions (this-is-silver and this-is-not-silver/this-is-nacre) are coordinated mutually by a dependence relation of negation by providing them a subjective berth in the form of a single knower. From the foregoing discussion we may now devise the following equation as deemed to have been envisaged by Abhinava B~dhaka (Contrary) B~dhya (Object of negation) B~idh~ (negation) B~dh~-parih~ra (correction/ cancellation of error) = subsequent cognition (Uttara Par~imar~a OR Vimar~ntara) Element of unification (MelanLfa~a) Break in the continuity of current judgement (Vimar~nuvrttinirmfilana) (a) Restoration of identity of Vimar~a with Prak~a (Pfirnat~). (b) Restoration of capability of being perceived by others (Pramatrantaravisayatv~bh~ivanivrtti). = This is nacre/ This is not silver. Silver(-in nacre)

There has been no silver. This is nacre which was seen as silver. This is nacre.

To close the point under discussion we would like to quote Sharma "This view that error is actually a part of truth, in the sense that its correction is not a matter of cancellation but of supplementation, is essential to the absolutist's position", 93 and with Potter we would like to assert "Correction does not involve rejecting either object or even a relation but rather in appreciating the inadequacy of this 'interpretation' as a piece of silver. ''94 Nacre-cognition is thus not a negation of silver-cognition, it is an improvement upon and fulfilment of silver-cognition. 'What is the nature of an illusory object', or to be more exact, 'What is the ontological status of the object' is a question that has generated the greatest

20

NAVJIVAN RASTOGI

heat among the Indian philosophers. In fact, the various theories of error, as we have already noted, are the accounts of the nature and status of objects in erroneous knowledge. We propose to make a reference to these theories in passing without critically evaluating them just to situate the Abhinavan position with greater clarity. In a sense all the Indian systems may be divided into two broad groups - one affirming the reality of the object of erroneous cognition while the other denying it. Between these two extremes various systems seek their place with varying shades of doctrinal difference. A third pattern may also be conceived which partly subscribes to reality of the content and partly denies it. The Akhy~iti of Prabhakara, Satkhy~ti of the S~rhkhya, Viparfta-khy]ti of Kum~rila, Anyath~khy~ti of the Nyfiya-Vai.~e.sikaand Saiva Siddh~nta, Akhyfiti-sariavalitayath~rtha-khy~ti of R~im~nuja form the first group so far as the reality of the object is concerned. For them both the terms of the erroneous judgement (silver and nacre) are real. The confusion between the two seen in an illusory perception results either from non-grasping of their difference or grasping of their similarity or from misplacement or misconstruing of them. The other group consists of Asatkhy~ti of the M~dhyamikas, Atamakhy~ti of the Yog~c~ras, Anirvacanfyakhy~ti of the Advaita Ved~ntin which either totally deny the reality of the two terms or ascribe to them dependent reality which in ultimate analysis turns out to be false. The third group may include the Apohavhdins like Dharmak?rti, Bhed~ibhedav~dins, some Jainas, and philosophers like M~dhava and Srikum~ra propounding Asatkhy~ti or Abhinava-anyath~khy~ti and Sadasadvilak.sa.nakhy~ti respectively. According to them, either one of the two terms is real the other being unreal, or both of them are real in one situation but unreal in different situation, or both of them are partially real and partially unreal. With such a sketchy description it is difficult to assign a definite place to the Aptirnakhy~tiv~dins like Abhinava in the above scheme. Although Abhinava is a staunch absolutist he advocates adherence to the Bhed~bheda position at the empirical level which is the immediate concern of epistemology.9s Hence he may be associated with the third category. But only so far and no further. According to Abhinava in the nacre-silver event the object of cognition is complete nacre (Pfirn.a ~ukti) which because of its incomplete grasp constitutes error. Now going by the Abhinavan metaphysics this silver (-in nacre) cannot

THEORY OF ERROR IN ABHINAVAGUPTA

21

be rejected as false, though it is different from actual silver and nacre both. 96 But none-the-less it is real because it is revealed to be so by the cognition. Even when the erroneous knowledge is cancelled, occurrence of silver-(in nacre)-experience is never denied. During the continuity of silver-cognition silver did form the content of our knowledge. The fact o f manifestation is the guarantee of its reality. 97 Since knowledge never foresakes its natural disposition towards truth, reality must be granted to its content. 98 Now if it is real, the Buddhist phenomenalist may ask, it must serve the purposive activity ascribed to real silver. 99 Then it will obliterate difference between actual silver and silver(-in nacre). It cannot be said that unreal in the form it is being projected by the silver-(in nacre)-knowledge, because that will condemn the very thesis o f manifestation of silver in nacre-silver. Abhinava has now been trapped in a vicious circle. Either he must admit that what he means by 'real' is 'empirical' which test silver(-in nacre) cannot pass or he should accept the Vedfintin's position that this silver is a sort of Anirvacanfya silver which is real because it is experienced, unreal because it is contradicted, hence it must have a different order of existence altogether. It appears that Abhinava distinguishes between real and empirical or actual on the one hand and between object and content on the other. According to him silver(-in nacre) is real as a content o f knowledge, but it is not actual or empirical as the Buddhist seeks to establish. Similarly as a content it is real and is not Anirvacanfya as there is no confusion of two orders of existence. Thus the Vedfintin's description o f Prfitibh~sika Sat, so long as it is not interpreted as Anirvacaniya but only as a fact of experience, may be acceptable to Abhinava. Because as a datum in experience, as a content, it is not denied, it is denied as an actual object and it is where the error lies. The content may be 'real', but not 'empirical'. While 'serving the purpose' may be the criterion of empiricity, 'figuring in cognition' that of reality. However, if content fails to measure up to 'object', it is not because the object (nacre) has changed, but because silver-in-nacre has received fuller appreciation. This fuller appreciation does not negate silver-in-nacre altogether but adds to our information that it-is-nacre-that-was-viewed-as-silver. The function of Vimar~a is to restore the unity o f the real with the actual. It is precisely the import of propounding that error lies in unification of the objective components, and it is this aspect of the whole episode that is sought to be corrected, neither the content nor the object. Thus to quote Sharma loo again "The Saiva and Sfinya Absolutisms

22

NAVJI'v AN RASTOGI

represent two extreme standpoints regarding the erroneous, while the Ved~intins and Yog~c~ras assert (i.e., Adhi.s.th~ina) or deny (i.e., externality) only one of the two terms of an illusory judgement, the Saivites affirm and the Sfinyav~dins negate both the terms (this = nacre, and silver). In the Saiva view both 'this' and 'silver' are real as ]~bh~sas." 1ol We have so far not touched upon a vital problem that has been engaging the mind of almost all the realists and also the Buddhist phenomenalists. The problem has been raised with regard to the nature of judgement in an erroneous perception - whether it is simple judgement or a complex one. The question has been put differently also; is an erroneous judgement a single psychosis or a complex one? In other words, is it indeterminate experience or a determinate judgement? Let us see how Abhinava handles this problem. According to the Naiy~yika there is a difference between simple and complex judgements. The erroneous judgement is a determinate cognition of 'this' (Idam) as characterised by silverness (Vi~i.s.tajfi~nam)lo2 that leads to purposive activity. As is natural in any determinate perception erroneous cognition is also a single perception where a 'datum' i.e., sensum, mixes up with related ideas to yield one perception. However, simple judgements relating to immediate perception are immune from error. Only the determinate perception is liable to error. Kum~rila finds himself in agreement with the Ny~ya. Similarly, while Prabh~kara shares Dharmak?rti's professed thesis that error is nothing but non-grasping of difference (Bhed~graha = Apoha) he differs from him in holding that all judgements are simple, i.e. indeterminate and also that they are true. Needless to say Dharmak[rti finds the position untenable and maintains that all judgements are determinate, hence complex. For Prabh~kara 'silver' and 'this' (or nacre) are two separate simple judgements which have been condensed into one - 'this is silver', hence it is erroneous cognition. They are two true judgements, but because there has been failure to distinguish the two error follows. It, however, does not follow from this that all complex judgements are erroneous. They are labled erroneous if they fail to answer practical demands. Abhinava's stand crystalizes from his attack on Dharmaki'rti's position. Dharmakirti tries to account for the empirical world by placing extra weight on coordination (S~rtipya) as an adhesive gluing the bare point instant (Svalaksana) with the so-called universal. This relation between two extremes,

THEORY OF ERROR IN ABHINAVAGUPTA

23

one real and the other unreal, leads to determinate judgement (Adhyavasaya) which being conceptual is erroneous. As against this Abhinava takes the position: since knowledge as such is valid, it is difficult to isolate determinate cognition as being capable of being illusory from indeterminate which is not so. 13 Abhinava seems to hold that though there is no error in indeterminate judgement, all determinate judgements are not necessarily erroneouS, It is only on the inverse concomitance that if error is to be found it will be in respect of determinate judgement which is a complex psychosis. We can develop this position further, as he proceeds to assail the Buddhist logician. If determinate judgement, asks Abhinava, is responsible for the proper determination of the empirical world and if that determinate judgement is erroneous, the objective determination will become an impossibility. In fact, Abhinava refers to his own thesis that determinate judgement does not suffer from 'temperamental' illusoriness because it reveals or manifests Abhfisa. It is the manifestation or illumination that is responsible for the determination of objects (Arthavyavasth~). In fact it is presumptuous to brand illusion as indeterminate or determinate. For deciding that it is a case of illusion, its perversity (i.e. opposition to truth) must be detected by a contrary cognition. If this criterion is not adhered to, all mutually exclusive cognitions will be deemed to be illusory. I4 This opposition or perversity directly results not from conceptualization, but from imposing one thing on the different locus, ls Similarly even on the Buddhist view indeterminacy is not the essential condition that eschews illusion. Had it been so, Dharmakfrti would not have specially added the word 'non-illusory' (Abhrfinta) in his definition of perception. 16 So far as the apprehension o f silver-in-nacre is concerned, it is no doubt a determinate knowledge, but this does not lay down the absolute rule that illusion will always stem from determinate perception alone. 17 Abhinava, however, seems to modify his stand subsequently and says that error no doubt refers to determinate judgement, because indeterminate judgement refers to the bare being that figures in our immediate awareness while determinate judgement involves the function of synthesis and analysis which is no doubt a determinate activity. But only those determinate judgements are liable to err which do not endorse the data o f immediate perception and are found, as such, to be at variance with the indeterminate knowledge. ,08 Now we come to the last but very fascinating leg o f our enquiry. Abhinava's greatest contribution seems to consist in his attempt to fashion a model for all

24

NAVJIVAN RASTOGI

the theories of error. At this stage we know very little about how Abhinava proposed to give a concrete shape to it, but there is no doubt that he did have a pattern in his mind. It is really a matter of great pity that even his great follower Mahe~var~nanda and celebrated commentator Bh~skara missed the opportunity to develop it. Abhinava's effort is reminiscent of the one made by an equally great scholar V~caspati in the Nydya-Vdrtika-Tdtparya-Tikd where he tried to develop his theory of Anyath~khy~ti as providing a basic model for the various theories of error. 19 Similarly Apfirn.akhy~ti was sought to be developed by Abhinava as an all-comprehensive pattern in which all the major theories could be accommodated. In all three passages 11 are available to us from which are culled the following observations: (i) Among all the theories of error Apfirnakhy~ti represents the ultimate explanation. (ii) Error as incomplete knowledge constitutes essence of all accounts of
error.

(iii) All the five theories form a logical whole beginning with Asatkhy~ti and ending with Vipar~takhy~ti and then this whole seeks culmination in Apfirnakhy~ti after being reprocessed through Asatkhy~ti. (iv) The manifold naming of different theories only brought out their different perceptions of the problem, though this variety of approach was to be shunned in view of the ultimate account offered by the absolute-realism (Param~rthavastuv~da). (v) Out of the five theories alluded to, Abhinava has used a different nomenclature in respect of the three theories, which in all probability reflects his understanding of these theories. They are (i) Asat-khy~ti, (ii) Asthira-khy~ti, (iii) As~dh~ran. akhy~ti and (iv) Anarthakriy~k~ri-khy~ti and (v) Viparita-khy~ti. Elsewhere he refers to Anirvficya-khy~ti in addition to Asad- and Viparitakhy~tis. Thus Asthira-khy~iti 'probably' refers to .~tmaky~ti, As~dh~ranato Anirvacaniya- and Anarthakriy~k~ri- to Anyath~-. 'Probably', because we have no conclusive collateral evidence to prove indentification just at the moment. It is really surprising that in all three accounts he nowhere refers to Prabh~kara's Akhy~ti. Though on the contrary, the cluster of five theories of error (Khy~tipaficaka), referred to elsewhere, 111 does not include Viparitakhy~ti. Possibly it was a fashion to enumerate one of the Mim~fisaka views. Abhinava and his followers, 112 however, normally use the word Akhy~ti but as a shortened form of Apfirnakhy~ti. In fact, in all the three passages

T H E O R Y OF E R R O R IN A B H I N A V A G U P T A

25

cited, some m o d e r n scholars appear to have c o n f u s e d this Akhyfiti w i t h that o f Prabh~kara 113 due to the apparent similarity o f names. A critical evaluation will reveal that Abhinava's t h e o r y o f i n c o m p l e t e c o m p r e h e n s i o n is in a sense m o r e an e x t e n s i o n o f Vipafftakhyfiti than that o f Akhy~ti. We leave, however, the d e v e l o p m e n t o f this aspect for a future occasion handicapped as we are t o d a y w i t h little i n f o r m a t i o n available on the subject. 1~4

Abhinavagupta Institute of Aesthetics and Saiva Philosophy, Lucknow University, Lucknow U.P., India

ABBREVIATIONS A.B.O.R.I Bh~s.

B.S.

Epistemology I.P.K. I.P.V. (Bh.)

I.P.V.V.

K.S. K.S.T.S. Methods

M.M.

M.M.P. N.S. Ny~ya Theory Presuppositions Saivism T.A. T.A.V. T.S. V.S.

Annals of the Bhandarkar Oriental Research Institute Bh~skari BrahmasOtra-hrhkara-bh~sya Epistemology of the Bh~.tta School of Pfirvamim~fis~i l~vara-pratyabhij fih-k~rik~ I~vara-pratyabhijfi~-vimar ~in (published along with Bhfiskari) Igvara-prat yabhijfia-vivrti-vimargin ~ Kashmir Saivism Kashmir Series of Texts & Studies Methods of Knowledge Mah~rtha-mafijari Mah~r tha-mafijari-parimala N~tya~stra The Nyhya Theory of Knowledge Presuppositions of India's Philosophies Saivism in Philosophical Perspective Tantr~loka Tantr~loka-viveka Tantrashra Vedanta-s~ra

NOTES I This paper is a modified and improved version of my paper read in the International Seminar on Abhinavagupta held in October 1981 at B.H.U., Varanasi (India). 2 'Abhinavagupta was a tenth century stalwart and doyen of Kashmir Saivism, aesthetics and dramaturgy. Here we are concerned with him as a philosopher of the School of

26

NAVJIVAN RASTOGI

Recognition. His major works include the Tantrdloka ]kvarapratyabhi/a~vimar~inL i~varapratyabhi/rTdvivrtivimarkinL Dhvanydlokalocana and Abhinavabhdrati on the Nd.tya-kdstra of Bharata. 3 Presuppositions of India's Philosophies (Presuppositions), Karl H. Potter, U.S.A., 1963, p. 186. 4 No apology is needed for choosing this topic even though some good studies of the problem have already appeared. See Kashmir Saivism by L. N. Sharma, Varanasi, 1972, pp. 79-89; Pratyabhij~d Pramd.na Mimdrhsd, Ph.D. thesis by Geeta Rastogi, University of Lucknow, 1977 (unpublished), pp. 317-339. s i~vara-pratyabhz~ad-vimargini[I.P.V. (Bh.)], Abhinavagupta, published with Bhdskar[ (Bh~s.) of Bh~skarakantha, edited by K. A. S. Iyer and K. C. Pandey, Varanasi, 1983-1954, II, p. 123 ; Ikvarapratyabhi]ad-viv.rti-vimar~in~ (I.P.V.V.), Abhinavagupta, Kashmir Series of Texts & Studies (K.S.T.S.), Volume III, 1938-1943, pp. 152, 154. 6 So'ntastath~ivimar~tm~ de~ak~l~dyabhedini I Ek~bhidh~navisaye mitirvastunyab~dhit~ II I~vara-PratyabhijfiS-K~rik~ (I.P.K.), Utpala Deva, published with Bh~skari, 2.3.2. Translation Pandey's, p. 5. (Bhas., Ili, p. 140). 7 apram~narh hi pramhnalaksanaviparyayena bhavati. - I.P.V.V., III, p. 152. 8 Mark the emphasis on Ab~dhitattva (uncontradictedness) in respect of knowledge. It means that the phenomenon of error is wide enough to cover the entire gamut of knowledge, though the instances we are confronted with in an analysis of error mostly relate to perceptual error to the exclusion of non-perceptual ones namely, inferential or varbal-testimonial. Various theories thus address themselves less to the consideration of the epistemic error than to that of its subsidiary, namely, perceptual error. (See, Methods of Knowledge According to Advaita Veddnta [Methods], Swami Satprakashananda, Calcutta, 1974, p. 121) 9 From the root x/Khya - to narrate + suffix Ktin. Apte describes it as knowledge, a faculty of discriminating objects by appropriate designation. Vide, The Student's SanskritEnglish Dictionary, Delhi, 1959, p. 177. lo apfirnakhyatirfip~ akhyatireva bhr~nti~attvam.
I.P.V. (Bh.), II, p . 1 2 3 .

11 1bid.
12 It appears that, throughout the series of e r r o r - analyses, these words have been treated as instances of Tatpurusa compounds bringing the nature of their objective content into bold relief. For example, Asatkhy~ti is Asatah. Khy~tih, Anirvacan~yakhyhti is Anirvacaniyasya Khy~tih and so on. However, in a recent effort (vide, The Theories of Error in Indian Philosophy: An analytical Study, Bijayananda Kar, Delhi, 1978, p. 11) the process has been reversed by treating them as instances of Karmadh~iraya compound where the first element acts as an adjective modifying knowledge rather than as an object. Thus, it is now Asat~ Khy~tih. or Anirvacaniy~i Khy~ti.h as against Asata.h Khy~ti.h etc. But on a closer scrutiny this grammatical analysis fails in respect of .~tmakhy~ti which cannot be explained away except by imparting an additional word (.A.tmavisay~ Khy~tih.). 13 Khy~tyabh~varupatve tu rajatabh~vena jfi~nam api ayuktam eva sy~t iti bh~iva.h. Isadarthe nail na tu abhfive. - Bhfis., II, p. 123. 14 Pfirn.aprath~bhfivfid apfirnakhyfitirf~pa iyam akhy~tir eva bhr~ntih.. - I.P.V.V., III, p. 153.

T H E O R Y O F E R R O R IN A B H I N A V A G U P T A

27

IS The Pratyabhijfih theory of ignorance is a very complex affair and indiscriminate use of interchangeable key words has confounded the matter. Thus we have Ajfiana, Anavamala, Apohana, M~iyh, Avidy~, Akhy~iti, Bhr~nti, Mahfibhr~nti, Dvaitaprathfi and Vikalpa - all these terms have definite connotations, but quite often they are used as synonyms. Hence the difficulty. In fact ignorance is a multi-tier, multi-dimensional concept. Mala or ignorance as defiling or delimiting agent obscures the real and this is the essence of this concept of ignorance. 16 Ato jfieyasya tattvasya samastyena aprath~tmakam I Jfifinameva tadajfianarh ~ivas0tre.su bhhsitam. - Tantrdloka (T.A.), Abhinavagupta, K.S.T.S., 1918-1938, 1.26. 17 sv~tmapracch~danecchaiva vastubhfitastath~malah. I - T.A. 9.66. 18 sv~tantrayah~inirbodhasya sv~tantryasy~pyabodhat~ I dvidh~inavaria malamidam, svasvarfip~pah~initah I1 -- I.P.K. 3.2.4. 19 pfirn.asvarfipasya akhyfitimfitram. - Trantrdloka-viveka (T.A.V.), Jayaratha, K.S.T.S., 1918-1938, Vol. VI, p. 58. 20 dvividharia ca ajfi~nam, buddhigatarh pauru.savh ca, tatra buddhigatam ani.~cayasvabhhvaria, vipar ftani~cay~tmakarh ca. Tantrasdra (T.S.), Abhinavagupta, K.S.T.S., 1918, p. 2 - 3 ; vide, also T.A. 1.36. 21 na kevalamatra samastasarhs~ripram~trs~dhfiranyfirh mahfibhrfintau akhyfitih. paramhrthah., y~ivatpratipram~trniyathyamapi pram~trvikalparfip~y~fia mah~bhrfintau saiva tattvam iti d a r ~ y a t i . . , m~y~pram~turvy~p~ro yo vikalpmama idam sukhas~dhanamityfidih., sa rflparh yasy~h, bhr~inteh, sarflsfirasalfimatfiyfih., s~pi sarvaiva samastesu pramfitrsu pratyekarh bhedena vartamfinfi advay~khy~tirfipeti pfirvena sarhbafidhah.; na tu apfirvasya arthasya y~ khyfitirvividharfipah prak~astadrfipaprakh~anada bhrfintih. I.P.V.V., III, p. 155. 22 Pram~trantaravi.sayatv~bhavasya akhyfitireva yato bhr~ntit~, tato bhrfintiniv.rttasya khyfitiriti bh~vah. - I b i d . , III, p. 156. 23 The use of two more terms SvarOpakhyfiti and Advayakhy~ti in the Kashmir Saivism clearly betrays the imprints of such thinking. 24 Vide, The Nydya Theory o f k n o w l e d g e (Nyfiya Theory), S. C. Chaterjee, Calcutta, 1950, p. 32. 25 atasminstaditi pratyayah atasminstadbuddhih.. Nyaya-vfirtika 1.1.2, quoted, (Ny~ya Theory), p. 33. 26 atadrflpapratis.th~. - Saivism in Philosophical Perspective (Saivism), K. Sivaraman, Delhi, 1973, p. 322. 27 atasminstadbuddhih. Vedantasdra (V.S.), p. 2, quoted, Brahmas0tra-~hrhkarabhfi.sya (B.S.), ed. K. N. Misra, Varanasi, 1976, p. 11. 28 atasminstaditi pratyayo viparyayah..
-

28

NAVJIVAN RASTOGI
- Pra~astapddabhds.ya, p. 177, quoted, Critique o f I n d i a n R e a l i s m , D. N. Shastri,

Agra, 1964, p. 482. 29 anyath~santam~k~ramanyath~ grhn~ti.

- N y ~ y a - r a t n d k a r a on Sloka-vdrtika, Nir. 118, Quoted, E p i s t e m o l o g y o f t h e Bhdt.ta S c h o o l o f P ~ r v a m [ m d n s d (Epistemology), G. P. Bhatt, Varanasi,

1962, p. 96. 30 Vastunyavastvfiropah.. - V.S., p. 2, quoted, B.S., p. 11. Yfidr~arh hi jfifinasya svarfapam tadr~yevfirthe'dhyfixopayfimiti yfivat. - K d k i k d of Sucaritamiira on S l o k a v d r t i k a , Nir. 110, quoted, Epistemology, p. 96. 31 avasanno'vamato vfi bhfiso'vabhfisah. - B . S . , Int. 32 adhyfiso n~ma atasmin tadbuddhih.
- 1bid.

33 Arth~dhysa and Jfi~n~dhy~isa are the technical words for it. See, Methods, p.
124-25.

34 Here we may refer the reader to the Epistemology, pp. 9 5 - 9 7 where the author offers an excellent account of all the error-illustrations employed by Indian theoreticians. 3s Bhrfintivyavahfiro'yam t~vatyapi samucitopayogipflrnaprakhyfivirahfit. - I.P.V.V., III, p. 153. 36 tato y~vat~ purnena r~pena prakhy~tavyam vimar~aparyantam tfivat na prakhyfiti, ityapOrn.akhyfitir~p~i akhy~itireva bhr~intitattvam.
- I.P.V. (Bh.), II, p . 1 2 3 .

37 apflrn.aria jfi~nam tvatra t~ivadasti [ukteh. rajatatayfi grahanat.

-

Bhfis., II, p. 123.

38 Vide D i s c o v e r y o f M a r x , Harsha Narain, Lucknow, 1981, p. 45. 39 samyagvimar~anuvrttyfibhfisanam sarhv~danam.

- I.P.V. (Bh.), II, p .

126.

4o Rajataikavimar[epi ~uktau na rajatasthitih I upfidhide~isarhvfichid dvicandre'pi nabho'nyathfi II

I.P.K. 2.3.13.

41 sa vyadhikaran.fisiddhasya api adhikaran~ntarasariabafidho na dos~ya, apitu taddvfiren.a fi[rayena yah. sarhbafidhah., sa eva. - I.P.V.V., II, p. 398. 42 atfidr~pye t~drOpyapratipattih.. - Pararndrthasdra-vivrti, Yogar~ja, K.S.T.S., 1916, p. 67. 43 'Etat' iti pratham~nasya pramfitrantaravisayatvasya 'vixuddho na prathate' iti pf~rnaprathhbhfivfidap~rnakhy~tirftpfiiyamakhyfitireva bhr~ntih. - I.P.V.V., III, p. 153. Cp. "anyath~bh0tameva' iti abhimatakriy~[ftnyafia pram~trantar~dhfiranaria ca."
- Ibid.

Also cp. "pramfitrantaravisayatv~bhfivasya akhyfitireva yato bhrfintitfi."

- IBM., p. 156.

THEORY

OF ERROR

IN A B H I N A V A G U P T A

29

44 Pramfitrantarasfidhfiranatvfibhfivasya yadfikhyfinarfl tanmahimnaiva asattfivaiparityavyapade[au yuktfinvayavyatir ekfibhyfim. - I b i d . , III, pp. 1 5 5 - 5 6 . 4s Vide, Saivism, p. 323. 46 A j ~ d n a , Malkani, Das & Murti, London, 1933, p. 142. 47 Ucyate aindriyakamfinasavibhramabhedena. - I b i d . , III, p. 79. 48 'idaIfi rajatarfl sthirarfl sarvapram~trsfidhfiranarfl arthakriy~yogyam' iti idamarh~e, rajatfidyarfle.su tatsammelan~rfl[e ca fibh~savimar~anabalfit na tfivat khhcit mithyfitvalfl kintu uttarak~larfl yo bhavi.syati vimar[ah 'nedarh rajatarfl vastu sthirarfl pramfitrantaragamyam abhimatakfiryakfiri iti tadvimar[avimar[aniyarh yat tatpfirvavimar[ak~lasamucitameva rfiparh tat tasmin pfirvavimar~akfile naivfimr~yate bhfivyalfi ca tenfimar~aniyena. - I.P.V. (Bh.), II, pp. 1 2 1 - 2 2 . 49 ata~ca prthak idafiatfidy~bhgsesu na kficana bhrgntih., melanfifia~e tu vimar~nuvrttinirmfllanaria vimarodayakfilfideva ~rabhya b~dhakena kriyate, iti tatraiva bhrfintibhhvah., iti siddham. - I b i d . , p. 125. s0 rajatasya uktikayg saha yadyapi eko vimargah tathfipi ~uktau rajatasya tena jfihnena yh dattfi sthitih 'idarh rajatam' iti, sfi ha, yata up~dhir~po yo deia.h' atra iti rajatacchhy~m gtmanoparafijayan guktide~ah, tasyhsariavgdfit. - I b i d . , pp. 1 2 5 - 2 6 . Sl svasarhvit atra ca Nksi.m tadaiva 'na idarh rajatam' iti v rttaparvasyaiva vimar~asya avimargNaran.arfl sarflvedyate. - I b i d . , p. 78. s2 yasya vat,it s~imarthy~it, vastu n~asukhfidikarh vyavatisthate, niyat~rfl prak~amary~d~6a n~tivartate, idamiti svarfipen.a etgdrgiti ca vi[esanabhOtanityatv~nityatvfidiyogena, talloke pramfin.am iti sthitam. -Ibid., p. 68. s3 tata ~ibhgsamfitrameva vastu svalaksanarh tu tadfibh~sasfim~nfidhikarany~bhfisar~pam fibhgsfintaram ekam anyadeva, tatra ca prthageva ca pramfin.am. - 1 b i d . , pp. 7 6 - 7 7 . s4 yathgruci yathfirthitvarfl yath~ivyutpatti bhidyate I ~ibhfiso'pyartha ekasminnanusafidhfinasfidhite II -- I.P.K., 2.3.3. ss ~bh~sabhed~dvastOngria niyatfirthakriy~ punah I s~mfin~dhikaranyena pratibh~sfidabhedin~m II - Ibid., 2.3.6. s6 ~ibhfisavimar[firthakriy~balena tathfi vyavasthgnfit iti. - I . P . V . ( B h . ) , II, p. 105. 57 yatpunargbh~Nnhfia mi~rane kfi simfi iti? tatra ucyate ye.sfim avirodhah ta eva fibhfisfi mi~ribhavanti, nahi rfapgbhgso mfirutgbhfisena milribhavati - virodhgt, so'pi ca niyati~aktyutthfipitah.. - 1 b i d . , p. 106. s8 cvamfibh~isastanmelanam ca niyamfinuprfin.itam - ityetfivadeva prameyam, et~nyeva ~game tattv~ni vaksyante. - 1 b i d . , p. 128.

30

NAVJIVAN RASTOGI

s9 nanu vimar[epi pratyfibh~safia tathaiva bhinnat~i? satyam, tathfipi tu paro yo vimar[ah, sa evfiyafia padftrthah, iti ekapratyavamar[rfipah, tena pr~nitakalpena fi samantfit khyfinarh prathanarh yasya mukhy~ivabhfisasya ekarfipabhfivfibhfisasya, ekapratyavamar[e taducito'pi hi astyekfivabh~sah., fibhfisavimar[ayoranyonyamaviyogfit, tasmfit mukhyfivabhfisfidekatvamapratihatamfiste. - I b i d . , p p . 117-118. 60 See Recognition in Pratyabhi]~d School: A S t u d y in Epistemology, Navjivan Rastogi, Annals of Bhendarkar Oriental Research Institute (A.B.O.R.I.), Diamond Jubilee Volume, pp. 848-849. 61 nanu vimar[abal~deva yadi vastfinafia bhed~bhedavyavasthfi tarhi idanfm trijagati nivrtta bhrantisarhkathah., [uktikfiyamapi satyarajatataiva apatati 'idam rajatarh' iti vim.rsyamfinatvfit, taraca bhrantyabh~ve bfidhanupapatteh, kimarthamuktarh 'mitirvastunyabadhita '(I.P.K. 2.3.2) iti. Vyabhi~rfibhave hi 'abadhita' ityasya vise.sanasya vyavacchedyarh na labhyate. - I . P . V . ( B h . ) , I I , p . 121. 62 Ibid., p. 123. 6a Bhfiskara's following observation may be found relevant - "Melan~rhe yadevedam tadaiva sthiratvfidigunayuktarfl rajatamiti samanadhikaranataya sthfipan~rupe bhfige tu" - Bhfis., II, p. 125. 64 idamiti hi satyatay~ agrasthavastuvisayam, rajatamiti tu smaryamfinarajatavi.sayam, anyatha drstarajatasyfipi [uktau rajatabhram~patteh.. - 1bid., p. 125. 6s tatra hi [ukty~ih svarOpam vyaktamevfidhyak.sikriyate, rajatatayfi adhyavasfyate ca. - Mahd.rthamafijar? (M.M.) with Parimala (M.M.P.), Mahe[varananda, ed. B.V. Dwivedi, Varanasi, 1972, p. 40. 66 uktau dr[yfiyfirh taddr[yaviparftena rajatena adhyavasfiya aikyam yada karoti, tada bhrantatvam. - I.P.V.V., I, p. 247. 6"/ k.setrajfianirmitataya fikhyfinameva ca brantiriti vitanisyate. - 1 b i d . , III, p. 152. 68 dharmasya anyadharmyfi[rayatvena dharmatayfi, dharmina~ca anyadharm~dh~ratvena dharmitaya siddhfivapi satyam tena dharm~a fi[rayena dharmata, tena ca dharmena ~i~ritena tasya dharmit~ - ityevarn y~i sanhyojan~, tasy~ asiddhirya s~ a~ryasiddhirv;i vyadhikara .n~siddhirvfi vyapadi~yate. - Ibid., II, 398. 69 Vide, Saivism, p. 329. 70 tatraiva kfile 'nedarfl rajatam abhfit' iti hi uttarah, paramaro . . . . - I . P . V . ( B h . ) , II, p . 123. Bhfiskara comments - apfirn.akhyfitirf~patvamasya [uktikfirajatajfianasya sfidhayati, tatraiva iti. hi yasmat, tatra tasmin rajataparamar~e, kale agre 'nedam rajatamabhfit iti' 'idafia rajatam nfis~t kintu may~ bhramenaiva jfiatam' ityevamuttarah, parfimar[o bhavati. - Bhfis., II, p. 123. 7~ kint u uttarakfilarh yo bhavi.syati vimar~o 'nedafia rajatarfl'..., tat tasmin pftrvavimar~a kfile naivamr.syate bhavyarh ca tenamar~aniyena. - I b i d . , pp. 121-22. 72 vimar[odayak~lfit pfarvavimar[odayasamayfit b~dhakena uttaraparamar[ena kriyate iti nedarit rajatamabhfititye vamutpad~diti bh~vah. - 1 b i d . , p. 125.

THEORY

OF ERROR

IN A B H I N A V A G U P T A

31

73 tatraiva kfile 'nedarfi rajatarh abhfit' iti hi uttarah, parfimaro, na tu uditapratyastamitfiyrh [atahradfiyfimiva idfinimeva 'idarh na' vimar[ah.
- I . P . V . ( B h . ) , II, p. 1 2 3 .

74 tatra anyathfibhfitafia vimar[fintaram unmfllakamudeti. 7s melan~rfi~e tu vimar~nuwttinirmSlanarh . . . . . 76 77 78 79 - 1.P.V.V., III, p. 152. b~dhakena kriyate.
- I . P . V . ( B h . ) , II, p. 1 2 5 .

sarhvfidah sthairya~fi or sthiratfi anuvrttih (a) atra rajatafia.., tasyhsarhwid~t, samyagvimar~finuvrttyfibhfisanarh sariavfidanam. - I . P . V . ( B h . ) , II, p. 126. (b) dvitvfibh~sacandrfibh~sayorapi melanfibh~se vimar[finuw.ttivy~vartanarh vficyam tadetadfiha dvirfipe candre'pi, na kevalafia rajata eva - 1bid., pp. 1 2 7 - 2 8 . (c) dvicandr~di tu tath~bh~sam~namapi uttarak~larh pramfivy~p~rfinuw.ttirfipasya sthairyasya unmfilanena 'dvicandro nfisti' ityevafiarfipena asatyam. - Ibid., p. 34. ao evakfirfirtharh svayameva sphu.tfl(aroti anuw.ttibhahgakaranena hi unmfilanarh prfiya[o vimarsa[ya, b~dhakena tu utpattivirodha eveti spas.tayituria 'nanu' ityfidinfi. - I.P.V.V., II, p. 79. 81 Up~dhide[e'tra rajatamitide~arOpatayfi uparafijakasvabhfiv~yfim [uktau asamvfid~dunm~lanfit satpratipaksabhfitavimar~asthairyfit na idam rfiparfi rajatamiti, sarhyagvadanaria bhfisanaparfimar~an~tmakam sarhvedanam, tadvipak.so'sarhvfida.h. - lbid.,p. 151-52. 82 Vide, Nyfiya Theory, p. 36. 83 iha bh~ve eva bh~iv~intarasya abhfiva iti vyavahartavya.h iti ayarh t~vat aparityfijya.h pr~titika.h panth~i.h, tatra bh~vasya bhfiv~ntare.na ya fidh~ry~d~rabhfiva.h sa eva bh~vatadabhfivayo.h. - I.P.V. (Bh.), I, p. 377. 84 rajatasya yal] [finyat~khyo dharmastadvi[is.t~yfih. ~uktik~y~h yat prathanametadeva pOrvameva pramfitrantar~i .mira avisayo'yamiti asya rfipasya prathanam. - I.P.V.V., III, p. 156. as rata eva hi [uktyfidestacchfinyatvena prathanfit, na tu rajatasya abhfivarfipa.tayfi prathanfi.t tadeva hi svfitantryena upapfidyate rajat~bh~samfitrasya anyade~divi~is.tarajatfibhfisamfitrat~dfitmyapratipannasya abhfivayogfit.
Ibid.

86 nanu asti e.sfi pratipattih.- tat rajatajfi~narh tadeva pram~narh na abhfiditi adhunaiva may~ jfi~yate ~uktijfifin~nubhav~t hetoriti . . . . . ~uktijfifinasvasariavedan~vadhau rajatajfi~nasvasarhvedanat~ api m~yopar~gayoghpagam~t sthitaiva . . . . . tatastasya rajatajfi~nasya dharminah. s~dhyam apram~nfitmakam ananuw.ttidharmavimarakatvarh n~ma yat, tat uktik~jfi~nasatyatay~ s~dhyate. - Ibid., II, pp. 4 0 0 - 4 0 1 . 87 Suktikhjfi~narh svasafiavedanapatitafia, iuktikay~ arthakriy~i krt~ iti jfi~narh yatah. upaciyate, tatah ~uktik~ijfi~narh balavat bhfitam, arthakriynupravela~fny~bh~vaparidurbalarh tat svasafiavadanapatitameva rajatabodharh b~dhate tadvimar~amunmSlayat. - Ibid., III, pp. 4 0 3 - 4 0 4 . 88 kriyfisarhbandhas~imfinyadravyadikkhlabuddhayah. 1 saty~h, sthairyopayogbhy~mek~nek~ry~ match II - I.P.K., 2.2.1.

32

NAVJIVAN

RASTOGI

a9 yathfiruci yathfirthitvarfl yathfivyutpatti bhidyate I fibhfiso'pyartha ekasminnanusafidhfinasfidhite II - Ibid., 2.3.3. 90 svarflpabhedfit hi sarhbhfivyetfinyatvarh, na ca svarfipam arthakriyfikfiritvam, ityuktarh vak.syate ca bahu~ah, svarfiparh ca pratyavamar~abalfidekameva b~hyfintarfid~vapi iti. - I . P . V . ( B h . ) , II, p. 120. 91 iha ~ukty~ t~vat rajatasya na kficit b~dh~ nfima kriyamfinfi dr.syate, ~uktijfi~nena rajatajfifinam bfidhyate ityapi na yuktam, svasmin vi.saye ~tmani ca svarflpe dvayoh jfifinayoh parinis.thitayoh, vi[rfintayoh anyonyarh virodhasya abhfivfit . . . . . 'uktikfi iti' 'na rajatamiti' jfifinasya unmfilanatfi tadiyavirmar~tmakapramfirfipavy~pfirfinuvartanavidhvarhsalfi kurvat pramfitari pratis.thfirh bhajate. - Ibid., I, pp. 3 6 5 - 3 6 8 . 92 pramfitrantaxavisayatvfibh~vasya akhyfitireva yato bhrfintitfi, tato bhrfintiniw.ttasya khyfitiriti bhfivah.. - I.P.V.V., III, p. 156. 93 KAshmir Saivism, pp. 8 6 - 8 7 . 94 Presuppositions, p. 219. 9S gha.tagat~bh~sabhed~bhedadrst.ireva ca param~th~idvayadrstiprave~e upfiyah samavalarhbaniyah.
- I . P . V . ( B h . ) , II, p. 1 2 9 .

96 nahi rajatajfifinena satyarajatasya [uktervfi vyavasthfi kficiditi - I.P.V.V., I, p. 243. 97 bh~sam~ntaiva satt~ tadasacca khy~ti ceti vy~hatam.
- 1bid.,

p. 245.

98 bhrfintabodhe'pi svasafiavedanamabhrfintarfi sarvatreti ghatayati.

-lbM.,

III, p. 73.

99 tena pratibhfisarfi cedasti, tat tadanus~riny~ arthakriyayfi api bhfivyam pratibhfisam~ne rope yadi asad syfit, pratibhfisam~nataiva katharia sy~it.
- 1 b i d . , I, p. 245. K~shmir Saivism, page 89. 1ol Explanations with in brackets are ours. 1o2 Knowledge of the qualified and not the qualified knowledge (Vi~istasya jfifinam). 1o3 sarvatra ca adhyavasfiyfideva arthavyavasthfi sa ca bhrfinta eveti . . . . . prcehfimah. kimanena vastu prakfi[yate na v~. yadi t~vat pralaigate bhr~ntatvamanupapannam. - I.P.V.V., I, p. 242. 104 saugatamate'pi .t~k~k~o bhr~ntitattvarfi dar~ayan nirvikalpake t~vat na bhr~ntit~, nfipi sarvatra vikalpe iti ghatayati . . . . yasyfi eva hi sarhvido bfidhakapramfinena viparftani.sthatvam~vedyate saiva anavasthitirftpatvfit b h r f i n t i r b h a v i t u m a r h a t i . . . anyathfi . . . . . pa.tavipar ftagha.tasafiavedanfidi avadhirya sukhasarhvedanamapi bhr~ntam sy~diti. -Ibid., pp. 2 4 3 - 4 4 . los nanu vaiparityarh tattadde~vas.tambhak~.taria v~stavaria grhyate. - 1 b i d . , p. 244. 1o6 ekacandrade~tvas.tambhastu dvicandrasyeti vastutas tadviparito'sau nirvikalpaken~pi ~ h ~ t a iti dvicandre nirvikalpakarh kasm~t na bhr~ntam, tath~pi hi tannivrttaye eva pratyak.salaksane'dhikarh bhr~ntapadam pr~ksepi. - 1 b i d . , p. 244.

too

T H E O R Y OF E R R O R IN A B H I N A V A G U P T A

33

to7 k~maria rajatabhrame pflrvarh [uktih. anyade~vas.tambhena na pr~kh~ate iti aviparitaiva nirvikalpake bhfitfti tatsarhvedanamavikalpakam abhr~ntam, rajat~dhyavasfiyastu rajate [uktide[~vastambhena bhavan vastu to viparito bhavatiti bhramatvaria tatra adhyavas~yasyaiva, na tu niyamo'yamupapadyate adhyavas~yasyaiva bhramarapateti.

- Ibid.

lo8 atah. prathamfine nirvikalkapavisaye r~pe tat vaiparityarh yena adhyavasiyate tadeva adhyavasfiyar~pafa sarvatra bhramajfi~ne bhrfintamiti yuktam, nanu nirvikalpak~ntarameva tathfi astu. nety~ha 'tacca iti'. avikalpakametadeva yat prath~m~trarflpatvam sarfiyojanaviyojan~divy~pfirastu vikalpanam, tadfiha vikalpfitmanastu iti bhrfintitvarh yuktamiti sarhbandhah, tasya hi prathamfinarfipap~titvamasti . . . . . ~uktau .drsy~yfirh taddr~yaviparitena rajatena adhyavasfiya aikyafil yad~ karoti, tad~ bhr~ntatvam. nile tu dr.sye nlenaiva vikalpyena aikye k~ bhrfintatfi. 1 b i d . , pp. 246-47. lO9 Vide, Nyfiya Theory, pp. 3 7 - 4 0 . 11o (a) ityap~rnakhy~itirfip~ akhy~tireva bhr~ntitattvam tadva~ena hyasadviparit~nirv~cy~dikhy~itayo'pi ucyantam.
I.P.V. (Bh.), p. 123.

Bhaskara comments on the word ~.di - "~di~abdena ~tmakhyfitigrahanam" - Bh~s., II, p. 123. Here Bh~iskara discusses the five theories of error in brief, but fails to see in them evolution of a logical pattern. (b) 'ath~pi'-ity~din~ 'astyevarh'-ityantena granthena asatkhy~tim asthirakhy~tim as~dh~ranakhy~tim anarthakriy~k~rikhy~tiria viparitakhy~tirh ca kramena paryavas~yya 'kevalamasau' ity~dinfi t~mapi asatkhy~tau vi~ramayya akhy~itiparam~rthat~ria nirv~hayisyati. - I.P.V.V., III, p. 152. (c) evam akhy~tirfipatve bhr~nteh sthite'satkhy~tiviparitakhy~tin~madhey~d.h~namapi pratibh~sapratyavamar~abal~iy~tam, tath~ param~rthavastuv~dena na sahyamity~ha 'tadevam' ity~di. - 1 b i d . , p. 154. 111 ~tmakhy~tirasatkhy~tirakhy~tih. khy~tiranyath~ I tath~'nirvacaniyakhyatirityetatkhy~tipaficakam II 112 akhy~tiv~dapadaviprasth~ne'pi. - M.M.P., p. 46. 113 abhinavaguptap~d~ api t~men~makhy~timev~fig~urvanti . . . . . . . evaria ca pr~yo gurumatasarfimato'khy~tiv~da ev~bhinavaguptamahe~var~nandhdinam safiamatah. - M.M., introduction, p. 'Tha'. 114 I am thankful to Prof. H. P. Alper of the Southern Methodist University, Dallas for going through the paper and suggesting various improvements.