Visvabandhu Tarkatīrtha’s “The Nyāya on True Cognition (pramā)”. Translated from Sanskrit and Bengali with explanatory notes

Full Text

Introduction

The paper of my Panditji was presented at a seminar on Comparative philosophy, which I organized for The Ramakrisna Mission Institute of Culture, Kolkata, in 1998. The summary of the paper in English was presented at the seminar, then I translated it with my commentary for philosophers and Indologists, as this paper contains a range of technical terms.

This paper aims to discuss the Nyāya conception of truth, which is a property of cognition. Unlike the Western concept of correspondence, which is relational in nature, the Nyāya philosophers do not claim that a proposition is true if it corresponds to a fact. This is since the Nyāya does not postulate a tertiary entity called “proposition”. Similarly, the Nyāya does not postulate ideas or images which resemble things in the world, although they postulate relational entities which are due to the relation between the things in the world and cognition. Truth is defined in terms of these relational properties of cognition.

Since truth is a property of qualificative cognition, I have mentioned the Nyāya conception of cognition, as well as the distinction between qualificative and non-qualificative cognition, which has an affinity with Russell’s distinction between knowledge by acquaintance and knowledge by description, although the Nyāya avoids both skepticism and solipsism.

Truth, both in the east and west, is a central problem of philosophy as it bridges the gap between epistemology and ontology. This is due to the fact that we would like to know whether cognition or thought represents reality. It is also related to a barrage of concepts, such as belief, knowledge, existence, fact, proposition, statement, sentence, etc. Moreover, it is also related to morality as judging something to be good or bad, just or unjust, depends on knowing something true about reality. According to formal semanticists, the word ‘true’ refers to a set of true sentences. Hence, language is a vehicle for the communication of truth.

The Nyāya philosophers have also discussed whether the word ‘true’ refers to a class-essence (jāti) which is common to all true cognitions. According to Gaṅgeśa, the founder of Navya-Nyāya, truth (pramātva) cannot be considered as a class-essence (jāti). This is due to the fact that there will be a defect called ‘sāṅkarya’ between truth (pramātva) and the class-essence perceptual apprehendedness (sākṣātva).

It is to be noted that according to the Nyāya there are several restrictions on jāti or class-essence, and sāṅkarya is one of them. In the case of sāṅkarya defect, there are two properties say F and G, such that F resides in the locus where G is absent, and G resides in the locus where F is absent, although both of them reside in the same locus. If there is sāṅkarya defect, then at least one of them cannot be considered as a class-essence (jāti). Since both the followers of the Nyāya and their opponents have accepted perceptual apprehendedness as a class-essence or class-character (jāti), but not the truth, so it is free from doubt.

In addition to the sāṅkarya defect, it is pointed out that truth is partitive in character, but a class character is non-partitive (avyāpyavṛtti). For example, an animal cannot be a cow in one part and a horse in another. If an animal has cowness, then this property pervades all the parts of this animal. Similarly, if an animal has horseness, then it pervades all the parts of this animal. But the truth is partitive in character. Suppose there is a piece of silver and mother-of-pearl. If a person says ‘these are silvers’, then it expresses the cognition that both of them are silvers. It is to be noted that this cognition is true for the silver, but false with respect to the mother-of-pearl. Hence, the truth cannot be treated as a class essence.

To avoid the above objections and many others, Gaṅgeśa defines truth as ‘tadvati tat-prakāraka-anubhavaḥ pramā’ (an apprehension or cognition is true if the ‘that’ of the that-possessor is the chief qualifier of this cognition). This means the same as ‘taddharmavadviśeṣyakatve sati tatprakāraka-anubhavaḥ pramā’ (a cognition is true if it is apprehension and the qualificand of this cognition is the possessor of the chief relational qualifier at the cognitive level). Panditji has also claimed that the property of being the qualificand and the property of being the qualifier are related by the relation of mutual determiner-determined (paraspara-nirūpya-nirūpaka).

About the late Mahāmahopādhya Pandit Visvabandhu Tarkatīrtha (1916—2006)

Well-versed in Advaita Vedānta and Mīmāṃsā, Pandit Visvabandhu Tarkatīrtha was the last surviving scholar of the Bengal School of Navya-Nyāya. He came to Kolkata in 1930 and stayed and studied with his guru Pandit Ananta Kumar Tarkatīrtha for several years. After retirement from Sanskrit College, Tol Department, Kolkata, he taught at Jadavpur University and Visva-Bharati University. After being appointed as a part-time research professor at the Ramakrishna Mission Institute of Culture, Kolkata, he supervised PhD students and wrote a book Tattvacintāmaṇi: Anyathākhyātivāda.

He taught contemporary comparative philosophers/Sanskritists, such as Bimal Krishna Matilal, Sibajiban Bhattacharcrya, and Jitendra Nath Mohanty. He was creative, as some of the problems of contemporary logic and philosophy of language were addressed from the tradition of Navya-Nyāya.

His scholarly works include the following: Prāmānṇyavādaḥ Harirāma Tarkavāgīśa, with a commentary in Sanskrit, Prabhā; Bengali works: Abhāvīya Pratiyogitā, Samavāya Lakṣaṇa Vicāra, Akhhyātivāda, Bhāratīya Darśane Duḥkha, Vyakti Vācaka Pada vā Proper Name, Anuvyavasāya, Pramār Lakṣaṇa, Jñāna Saṃvedana, Nyāyamate Katipaya Pada O Tadvācya, Svarūpa Sambandha, Pramāṇa; In Sanskrit: Indriyanānātva-Vimarśaḥ, Pratibadhya-Pratibandhaka Bhāva, Śābdabodha. The above Bengali and Sanskrit articles were published in various journals, beginning in 1954. His works have generated new creative philosophical activity in India as well as in other countries.

Translation

We desire those objects which will be beneficial to us or by means which we can avoid untoward consequences. It is to be noted that the methods for acquiring these objects should be within our means (1). For example, a thirsty person having observed water will proceed towards it. Knowledge plays an important role in our desire or intention. In this example, the thirsty person remembers that his/her thirst was quenched after drinking water in the past. Moreover, he/she has the cognition that this water is of the same type which has quenched his/her thirst in the past. He/She understands that the water, which is in front of him/her and which is of the same type as the previous one, would also quench his/her thirst. Hence having cognized it as something which will quench his/her thirst he/she proceeds towards it. This is how we engage ourselves towards the acquisition of the desired objects.

But the things, which we acquire after a lot of effort or which require a lot of sacrifices, deserve special consideration as we have to determine whether the cognitions on which our acquisition depends are reliable or not. In other words, we need to know whether the things which we think to be beneficial to us are really beneficial. Hence we need to consider whether these cognitions are true. When we come to know the truth of these cognitions, we intend to acquire even those things which require much effort or sacrifices. Hence it follows that the cognition of truth is a presupposition for our desire to acquire things even if the acquisition of those things requires a lot of effort. Therefore, the question is: What is the nature of this truth which resides in a cognition? (2)

As an answer to this question, some philosophers claim that truth is the class-essence (jāti) of true cognitions. As a class-essence of qualities resides in every quality but does not reside in anything else, so does the truth. In other words, the truth which is a class-essence resides in true cognitions only. Hence it distinguishes true cognitions from everything else and thereby becomes the differentia of true cognitions.

As a critic of this view Gaṅgeśa, the author of Tattvacintāmaṇi, claims that truth cannot be considered as a class-essence (jāti). If we consider it as a class-essence, then it will be infected with a defect known as ‘sāṅkarya’. This is due to the class-essence called ‘perceptual apprehendedness’ (‘sākṣātva’) (3). In the case of 'saṅkara' defect there are two properties such that one resides in the locus where the other is absent, although both of them reside in the same locus (4). In the case of this type of defect both of them cannot be considered as class-essences. Hence at least one of them is not a class-essence. In our above example, a true inferential cognition is characterized by truth (pramātva), but not by perceptual apprehendedness (sākṣātva). An erroneous perceptual cognition is characterized by perceptual apprehendedness, but not by truth. Again, both of them occur in true perceptual cognition. Since these two properties are characterized by both having the same locus and by occurring in the locus (or loci) of each other’s absences, the sāṅkarya defect would occur. It is due to this defect the perceptual apprehendedness would be a class-essence but not the truth (5).

Gaṅgeśa has also put forward another argument against the view that truth is a class-essence. A class-essence cannot be partitive. For example, an animal cannot be a cow in one part and a horse in another. But if a class-essence is partitive in character, then the same animal would have cowness in one part and horseness in another. Hence it cannot be said to be either a cow or a horse. On the contrary, if an animal has cowness, then it pervades all the parts of it. Similarly, if an animal has horseness, then it pervades all the parts of it. Hence a class-essence does not reside in one of the parts of an individual. But the truth is partitive in character. Hence it cannot be treated as a class-essence. In the presence of a piece of silver and mother-of-pearl if a person says, ‘these are silvers’, or ‘This is silver and that is also silver’, then he/she cognizes both of them as silvers. This cognition is false for the mother-of-pearl, but true with respect to the piece of silver. Hence it has truth in one respect and absence of truth in another. Since we have to accept the same conjunctive cognition as having truth in one respect and absence of truth in another respect, the truth will be considered as partitive in character. Hence it cannot be treated as a class-essence (6).

Now the opponents of the Nyāya argue in the following way. An object, positive or negative, is non-pervasive (avyāpya-vṛtti) if it resides in the same locus where its absence resides (7). On the contrary, if it cannot reside in the same locus where its absence resides, then it is called ‘pervasive’ (‘vyāpya-vṛtti’). Hence a quality, an action, or a class-essence becomes non-pervasive if it resides in the same locus where its absence resides. But if it does not reside in the same locus where its absence resides, then it is pervasive in character. According to our previous argument both truth and its absence reside in the same locus. This is because a conjunctive cognition has truth in one respect and its absence in another respect.

The opponents claim that truth is a class-essence even if it is non-pervasive in character. But the followers of the Nyāya claim that there is no proof in favor of the view that a class-essence could be non-pervasive. Class-essences such as potness, which are established by perception, are always pervasive in nature. They are never non-pervasive. The view that a class-essence is non-pervasive cannot be substantiated by an inference as it requires a piece of evidence in its favor. Hence there is no evidence in favor of a class-essence being non-pervasive. Therefore, if the truth is considered non-pervasive, it cannot be said to be a class-essence. In this context the following rule may be stated: That which is non-pervasive can never be pervasive, for example, conjunction, sound, etc.; and that which is pervasive can never be non-pervasive, for example, potness, clothness, etc.

The supporters of the view that truth is a class-essence may claim that there is a deviation from this rule in the case of the never-type of absence of a conjunction (8). There is only one never-type of absence of conjunction. It resides in a substance as non-pervasive. This is due to the fact that the conjunction which resides in a substance is limited by a unique limitor, but the absence of the same conjunction limited by some other limitor resides in the same substance. Hence the same conjunction and its absence reside in the same substance, although they are limited by different limitors (9). For this reason the occurrence of the same conjunction becomes non-pervasive. Again, since conjunction does not reside in quality, the occurrence of the never-type of absence of conjunction becomes pervasive. Since the same never-type of absence of conjunction resides in a substance as non-pervasive but in quality as pervasive, it is evident that there is a deviation from the above rule. Hence this rule is not tenable. The following rule is to be accepted: If the same object occurs both as positive and negative, then it is non-pervasive; and if it occurs only as positive or negative, but not both, then it is pervasive. This is how we have to determine the pervasiveness or the non-pervasiveness of an object. Hence, the pervasiveness or the non-pervasiveness of an object cannot be determined in terms of the category to which it belongs. The acceptance of this law will rule out the deviation mentioned in the case of the never-type of absence of conjunction. Since we apprehend both the conjunction and its never-type of absence in the same object, it is to be considered as non-pervasive. Again, since we apprehend the never-type of absence of conjunction in quality and do not apprehend the conjunction in it, the never-type of absence of conjunction is to be considered as pervasive in character. Hence there is no deviation mentioned previously if we accept this law.

A positive entity and its absence are mutually exclusive. Usually, they do not reside in the same locus. The cases where they reside in the same locus require some explanation. The non-pervasive objects, such as conjunction, reside in the same locus. Suppose a monkey is on the top of a tree. The top of the tree has the relation of conjunction with the monkey, but we apprehend the absence of the conjunction with the monkey at the bottom of the same tree. Hence we do not apprehend both the presence and the absence of the conjunction relation with the monkey at the same place. They appear at different parts of the tree. The portion of the tree where we apprehend the relation of conjunction with a monkey is called ‘the limitor of the conjunction with the monkey’, and the portion of the tree where we apprehend the absence of the conjunction with the monkey is called ‘the limitor of the absence of conjunction with the monkey’. From this discussion, it follows that even if a non-pervasive object and its absence reside in the same locus it is limited by different limitors. But in the case of a cognition we cannot say that it has both truth and its absence. This is due to the fact that a part is a limitor and only a substance has parts. Since cognition is a quality, it has no parts. Hence it cannot have parts as its limitors. Therefore, we cannot say that one portion of a cognition has truth and another portion the absence of truth. Hence, the truth cannot be said to be non-pervasive. It is a pervasive property.

Now the opponents raise the question of whether the concept of part can be applied to cognition. That by means of which the same property-possessor can be conceived to have contradictory properties may be used as a part in this context. In the case of cognition, its objects are to be considered as parts. In the presence of a piece of silver and mother-of-pearl, the perceptual cognition, these are silvers, is partly true and partly false. It is true in respect of silver, but false in respect of the mother-of-pearl. Similar is the case with the cognition of the mountain has fire and the lake has fire. It is true in respect of the mountain having fire, but false in respect of the lake having fire. Hence the objects of cognition are the limitors for the use of truth and falsehood.

But this conclusion is also not tenable. This is because the part which is the pervaded of the limited becomes the limitor (10). Otherwise, the bottom of the tree would be the limitor even if the conjunction characterizes the top of the tree. For this season the pervaded portion of the limited is called ‘the limitor’. In our above example, there are silvers, truth is the limited, but the mother-of-pearl which is the qualificand (11) is not its pervaded. This is because in this erroneous cognition of silver, the mother-of-pearl is the qualificand and it does not have the limited truth. Hence we cannot accept the mother-of-pearl, which is the qualificand, as the limitor of truth.

Now it may be said that the limitor of truth is the property of not cognizing as the qualifier, which has not occurred in the qualificand (12). In other words, if a qualifier, say F, does not characterize the qualificand of cognition, then the absence of having the qualifier F becomes the limitor of truth. That which occurs in the qualificand of true cognition is the qualifier. That which does not occur in its qualificand cannot be treated as its qualifier. Hence a true cognition has the property of not having the qualifier which has not occurred in the qualificand. This property is to be considered as the limitor of the truth of cognition.

Consider the cognition of this as silver. In this true cognition silverness is present in the qualificand this. The properties, such as nacreness, which are not present, have not occurred as qualifiers in this cognition. For this reason this cognition has the property of not having the qualifiers which have not occurred in the qualificand. Hence it has this property as the limitor of truth. But a false cognition has the property of having the qualifier, which has not occurred in the qualificand. This property would be the limitor of falsehood. The cognition of a is F is false if F has not occurred in a. Therefore, this cognition has falsehood which is limited by the property of having the qualifier which has not occurred in the qualificand. Consider the cognition of this is silver in the presence of mother-of-pearl. In this cognition, silverness is the qualifier which has not occurred in the mother-of-pearl. Hence this cognition has the property of having the qualifier which has not occurred in the qualificand. Therefore, this property is to be considered as the limitor of falsehood.

But this explanation is also not plausible. This is due to the fact that the acceptance these limitors would be sufficient for the use of truth or falsehood. Hence there is no need to consider truth as a class-essence.

Those who consider truth to be a class-essence claim that all true cognitions have only one class-essence called ‘truth’. The same truth is used differently depending upon the second term of true cognition. Since there are different true cognitions, there are different uses of the same truth. Consider the true cognition of this is silver. In this true cognition, there is the use of the truth which is qualified by silverness which is the second term of this cognition (13) (rajatatva pratiyogika pramātva). Similarly, in the true cognition of this is a mother-of-pearl, there is the use of truth which is qualified by nacreness which is the second term (śuktitva pratiyogika pramātva). Hence in a true cognition, such as this is silver, there is either the use of the unqualified truth (śuddha pramātva) or the use of the truth qualified by the property of having silverness as the second term (rajatatva pratiyogikatvaviśiṣṭa pramātva), although the truth is a class-essence. Hence there cannot be the use of truth which is qualified by nacreness which is its second term (śuktitva-pratiyogikatvaviśiṣṭa-pramātva).

Now an objection may be raised in the following way. If truth is one property, then the truth which is qualified by having silverness as the second term would be identical with the truth which is qualified by having nacreness as the second term. Hence, in the true cognition of this is silver, if there is truth qualified by the property of having silverness as its second term (rajatatva-pratiyogitva-viśiṣṭa-pramātva), then there must be the truth which is qualified by having nacreness as the second term (śuktitva-pratiyogikatvaviśiṣṭa-pramātva). This is due to the fact that they are not different (14).

In reply, the supporters of the view that truth is a class-essence claim that the problem of oneness of truth can be solved in the way the Nyāya-Vaiśeṣika solves the problem of oneness of the relation of inherence. According to the Nyāya-Vaiśeṣika inherence is one entity. The inherence relation which relates the color to earth, water, and fire, is the same as the inherence relation which relates touch to air or a sound to ether. Now an objection has been raised in the following way. If the inherence of color is the same as the inherence of touch, then the inherence of color must be in the air as well. In other words, we have to claim that air has a color. To avoid this type of undesirable consequence, the Nyāya philosophers claim that the relation is the property of being the locus of inherence qualified by having color as its second term (rūpa-pratiyogikatva-viśiṣṭa-samavāya) (15). This resides in the earth, water, and fire. Since the relation of color resides in these objects, we cognize color in these objects. Air also has the relation of inherence, but the property of being the locus of inherence which is qualified by having color as its second term resides only in the earth, water, and fire. It does not reside in air or ether. We accept the property of being the locus of inherence which is qualified by having color as its second term. This is due to the fact that we cognize earth having this property, water having this property, and fire having this property. Since we do not cognize the property of being the locus of inherence qualified by having color as its second term in air and other objects, we do not accept the presence of this property in these objects.

This is how we have to explain the truth. The relation of truth qualified by having silverness as its second term (rajatatva-pratiyogitva-viśiṣṭa-pramātva) is the property of being the locus (adhikaraṇatva) of truth qualified by having silverness as its second term. This property resides in the true cognition the qualifier of which is silverness (rajatatva-prakāraka-pramājñāna). Hence the use of truth qualified by having silverness as its second term is present in the true cognition, the qualifier of which is silverness. It cannot be used anywhere else. This is how our use of truth can be explained. Hence we can accept truth as the class-essence which resides in all true cognitions.

Now the following objection has been raised against this view. It is claimed that in the true cognition which has silverness as its second term (rajatatva pratiyogika pramājñāna) there should be cognition of truth which has nacreness as its second term (raṅgatva pratiyogika pramātva buddhi). The property of being the locus (adhikaraṇatva) limited by the property of being existence qualified by being other than quality (guṇānyatvaviśiṣṭa-sattātvācchinna) does not reside in quality. Hence there is no cognition of a quality possessing existence, which is qualified by being other than a quality (guṇānyatvaviśiṣṭa-sattāvān-guṇaḥ). But existence has the property of being the occurrent which is determined by a quality (guṇa-nirīpita-vṛttitva). This is due to the fact that quality has existence. It is to be noted that existence (sattā) and existence qualified by being other than a quality (guṇānyatvaviśiṣṭa-sattā) are not different. This is due to fact that the expression ‘the existence qualified by being other than a quality’ (‘guṇānyatvaviśiṣṭa-sattā’) means the same as ‘the existence which has the same locus as difference from quality’ (guṇa-bheda-samānādhikaraṇa-sattā) (16). The existence qualified by having the same locus as difference from a quality (guṇa-bheda-samānādhikaraṇyaviśiṣṭa-sattā) is not different from pure existence.

If we claim that pure existence is different from this type of qualified existence, then pure existence will not be available as existence is always qualified by some property or the other. Hence pure existence cannot be exemplified as it is not different from qualified existence. For this reason, we have to accept the cognition of existence qualified by being other than quality in quality (guṇe-guṇānyatvaviśiṣṭa-sattā), although we do not have the cognition of a quality possessing existence which is qualified by being other than a quality (guṇānyatvaviśiṣṭa-sattāvān-guṇaḥ). This is due to the fact that in the cognition of a quality possessing existence qualified by being other than a quality (guṇānyatvaviśiṣṭa-sattāvān-guṇaḥ), the property of being the substratum (adhikaraṇatā) limited by the property of being existence qualified by other than a quality (guṇānyatvaviśiṣṭa-sattātvācchinna) does not become the object or content (17). The rule for the substratum is the following:

If there is a difference in substratum, then there is a difference in the property of being the substratum.

Now it is claimed that there is no cognition of a quality possessing existence qualified by being other than a quality (guṇānyatvaviśiṣṭa-sattāvān-guṇaḥ). For this reason, we cannot accept in quality the property of being the substratum (adhikaraṇatā) which is limited by the property of being existence qualified by being other than a quality (guṇānyatvaviśiṣṭa-sattātvācchinna) (18). But in the cognition, corresponding to the sentence ‘the existence qualified by being other than a quality resides in a quality’ (‘guṇe-guṇānyatvaviśiṣṭa-sattā’), the pure existence has the property of being the superstratum (ādheyatā) determined by the property of being the substratum residing in a quality (guṇa-niṣṭha-adhikaraṇatā-nirūpita). Since pure existence is not different from the qualified existence, the property of being the superstratum residing in pure existence (śuddhasattā-niṣṭha-ādheyatā) also resides in the qualified existence. This is due to the fact that there is only one superstratum. Hence there is only one property of being the superstratum. For this reason, the followers of the Nyāya, such as Mathurānātha, have accepted the cognition of existence qualified by being other than a quality residing in a quality (guṇe-guṇānyatvaviśiṣṭa-sattā).

Now, this technique may be applied to our discussion of true cognitions, as there is a class-essence truth (pramātva) in every true cognition. It is claimed that there will not be the true cognition of silverness as its second term possessing nacreness as its second term (rajatatva-pratiyogika-pramātvavatī-raṅgatvapratiyogika-pramā), although truth is one (19). But there would be cognition of the truth, whose second term is nacreness, residing in the true cognition the second term of which is silverness (rajatatva-pratiyogika-pramāyāṃ-raṅgatvapratiyogika-pramātvam). But the Nyāya philosophers do not accept the presence of truth whose second term is nacreness (raṅgatvapratiyogika-pramātvam) in the true cognition whose second term is silverness (rajatatva-pratiyogika-pramātvam) (20). Hence the view that there is one truth in all true cognitions is not tenable.

There is another alternative. It may be suggested that each true cognition has its truth. For example, there is one truth in the true cognition which has silverness as its second term (rajatatva-pratiyogika-pramājñāna), and there is another truth that resides in the true cognition which has nacreness as its second term (raṅgatvapratiyogika-pramājñāna). Hence there would be as many truths (class-essences) as there are true cognitions. Now the Nyāya philosophers claim that this alternative suffers from sāṅkarya defect. This can be demonstrated in the following way. In the true cognition which has silverness as its second term, there is an absence of truth which has nacreness as its second term. Again, in the true cognition which has nacreness as its second term, there is an absence of truth that has silverness as its second term. But in a true conjunctive cognition, such as this is silver and that is nacre, the truth which has silverness as its second term and the truth which has nacreness as its second term have the same locus. For this reason, there would be sāṅkarya defect.

From our above discussion, it follows that we cannot accept one truth as the class-essence which occurs in every true cognition. Nor can we accept the view that there are as many truths (class-essences) as there are true cognitions, which are individuated by their second terms. For this reason, it is difficult to substantiate the view that truth is a class-essence.

To avoid these difficulties, Gaṅgeśa in his book Tattvacintāmaṇi defines truth as follows: A cognition is true (pramā) if it is apprehension and the property which is the qualifier of this apprehension resides in the property-possessor which is the qualificand of this apprehension (21). (In a non-technical language, this definition may be stated in the following way:

The cognition a is F is true if it is apprehension and F belongs to a, where F is the qualifier and a is the qualificand)

This definition is applicable to the apprehension this is silver if there is a silver. This is due to the fact that silverness is the qualifier of this apprehension and that which is the possessor of silverness i.e. a genuine piece of silver, is the qualificand of this cognition (22). But this definition is not applicable to an erroneous apprehension, such as ‘This is silver’ when there is mother-of-pearl. In this apprehension, silverness is the qualifier (prakāraka) at the cognitive level, but the silver-possessor is not the qualificand (viśeṣyaka) at the cognitive level. In this case, the mother-of-pearl is the qualificand (viśeṣyaka). For this reason, this cognition is not true. Hence the apprehension in which silverness is the qualifier and the silverness-possessor, i.e. silver, is the qualificand has the status of truth. Similarly, the apprehension in which nacreness is the qualifier and the nacreness-possessor is the qualificand would be true.

It is be noted that in this definition the word ‘anubhava’ (‘apprehension’) occurs. It means cognition which is different from a memory-cognition. Now the question is: What is the need for the word ‘anubhava’ (‘apprehension’) in the definition? In reply, it is said that if the word ‘anubhava’ is excluded from the definition of ‘pramā’, then the definition would be applicable to memory-cognition which is due to correct apprehension (yathārtha-anubhava-jñāna) (23). Hence the definition would suffer from the defect called ‘ativyāpti’ (‘overcoverage’).

Now an objection may be raised in the following way: If a memory-cognition, which is due to a previous true apprehension, is also true, then what is the harm in treating it also as pramā (valid)? Why should it not be included within the scope of the definition of ‘pramā’?

In reply, it is said that the special instrumental cause of valid cognition (pramājñāna) is called ‘pramāṇa’ (24). All the schools of Indian philosophy have accepted this meaning of the word ‘pramāṇa’. If a memory-cognition is also treated as pramā (valid), then its instrumental cause, which is either the previous apprehension (anubhava) or the energized disposition (udbuddha-saṃskāra), should also be treated as another pramāṇa (special instrumental cause). This would violate the law of parsimony. For this reason a memory cognition is not considered as pramā (valid) in this technical sense of the word.

But this definition is also not adequate, as it applies to opposite erroneous cognition (viparīta-bhrama). Hence it suffers from the defect of overcourage (ativyāpti). Consider a nacre (mother-of-pearl) and a piece of silver. In an opposite error which is a conjunctive cognition, the nacre (mother-of-pearl) is cognized as silver and the latter is cognized as a nacre (mother-of-pearl). In the first conjunct, the nacre (mother-of-pearl) is the qualificand and silverness is the qualifier, but in the second conjunct the silver is the qualificand and nacreness is the qualifier. Now let us apply the definition of pramā to this erroneous conjunctive cognition. If we consider the first conjunct, then the nacre (mother-of-pearl), the possessor of nacreness, is the qualificand, and if we consider the second conjunct, then nacreness is the qualifier. Hence this erroneous cognition has nacreness as the qualifier and the possessor of nacreness as the qualificand. Similarly, if we consider the second conjunct then silver, the possessor of silverness, is the qualificand, and if we consider the first conjunct, then silverness is the qualifier. Hence this cognition has silver as the qualificand and nacreness as the qualifier (25). Since the definition of truth applies to this erroneous conjunctive cognition, it is infected with the defect of ativyāpti (overcourage).

To avoid this defect, it is claimed that the property of being the qualificand and the property of being the qualifier are related by the relation of mutual determiner-determined (paraspara-nirīpya-nirīpaka) (26).

In other words, the property of being the qualifier is determined by the property of being the qualificand, and the latter is determined by the former.

Hence this condition is to be included in the definition of pramā. Therefore, the property of being the qualifier residing in silverness is determined by the property of being the qualificand residing in the possessor of silverness. Similarly, the property of being the qualifier residing in nacreness is determined by the property of being the qualificand residing in the nacre (mother-of-pearl). But this does not happen in an opposite erroneous cognition. As regards the method for determining the qualificand and the qualifier of our cognition, the Nyāya relies on our apprehension.

Therefore, to determine the truth of cognition, we have to find out which property of being the qualifier is determined by which property of being the qualificand. Here also we have to rely on our apprehension (27).

Explanatory Notes

(1) Almost all the systems of Indian philosophy claim that desire is due to the cognition that the object of desire is more conducive to pleasure than pain. In other words, if a person desires x, then he/she has the cognition that x will not give rise to undesirable consequences or pain, but will yield desirable consequences or happiness. If there are two alternatives, say x and y, then there will be cognition that y on balance is not as good as x. This complex cognition is a causal condition of our desire for x. In technical language of Indian philosophers, these two cognitions are called ‘iṣṭasādhanatājñāna’ and ‘balavattarāniṣṭājanakatvajñāna’. Hence the desire for x follows this complex cognition. The mental effort (kṛti) for x follows the desire for x. The latter is a causal condition for the former. Again the mental effort (kṛti) is a causal condition of physical effort (ceṣṭā), which in turn is a causal condition of action (kriyā). This causal process has been mentioned in the following oft-quoted śloka (verse): jñānajanyā bhavedicchā, icchājanyā bhavet kṛtiḥ, kṛtijanyā bhavet ceṣṭā ceṣṭājanyā bhavet kriyā.

A few more mental processes are also to be included in this causal series. It is claimed that cikīrṣā or inclination to fulfill the desire is a causal condition of kṛti or mental effort. Hence cikīrṣā comes between desire (icchā) and mental effort (kṛti). Again the cognition that the object of desire can be attained through our effort (kṛitisādhyatvajñāna) is considered a causal condition of inclination to fulfill the desire (cikīrṣā). As regards the types of mental effort (kṛti), all the schools have accepted pravṛtti (mental effort towards an object) and nivṛtti (mental effort to withdraw from an object). But many philosophers have accepted jīvanayoni or mental effort for the sustainment of one’s life in addition to pravṛtti and nivṛtti. It is regarded as an invisible mental state [1. P. 406].

From the above discussion, it follows that our choice, desire, or effort is dependent upon certain beliefs or cognitions. If the cognition, say S, is causally related to the object of desire, say x, then the same cognition cannot be a cause of another object of desire, say y. Hence our choice or desire is determined by certain beliefs.

(2) It is to be noted that Indian philosophers, by and large, consider a qualificative cognition to be true or false. The Nyāya has also distinguished qualificative and non-qualificative cognitions. The Nyāya concept of qualificative cognition can be expressed by a complex expression of the form ‘aRb’. A qualificative cognition involves necessarily at least three elements, viz., a qualificand, a qualifier, and a relation of the latter to the former. This relation at the cognitive level might be called the ‘qualification relation’. The simplest qualificative cognition has as its object, say, a table together with tableness in a certain relation. This complex is expressed by the expression ‘a table’, and described by the more complex expression ‘tableness inheres in a particular table-individual’. Both tableness and the particular table are cognized as such at the level of non-qualificative cognition, but not the relation between them. It is also claimed that the cognition of tableness is a cause of the cognition of a particular table qualified by tableness. But the cognition of the relation of inherence which relates tableness to a particular table is not a causal condition of it. If the relation is also considered a causal condition, then there would be an infinite regress. For cognition of the relation would involve the cognition of its two terms (or relata). Hence both the terms would also be causal conditions. Therefore, all the three elements, viz., a particular table, tableness, and the relation of inherence, would be causal conditions of this qualificative cognition. Since the relation is cognized along with its terms, the cause of the cognition of a table qualified by tableness would be the cognition of a table qualified by tableness. Similarly, the cause of the latter cognition would be another cognition of a table qualified by tableness, and so on. To stop this regress, the Nyāya claims that the cognition of a relation is not a causal condition of a qualificative cognition. (For a detailed discussion, see [2. P. 259—263]).

Western philosophers, by and large, consider a judgment, a belief, a proposition, a statement, or a sentence to be true or false. Since a judgment involves concepts, such as subject-concept and predicate-concept, this view would postulate certain entities in addition to the things in the world. If belief is a set of judgments, then the objections against judgment would be equally applicable to this view as well. If we postulate propositions as bearers of truth and falsehood, like Brentano, Bolzano, and Husserl, then we end up with the postulation of eternal tertiary entities, which are neither physical nor mental or psychological. If we consider a statement as true or false, then a statement, being the use of a sentence, would depend on a linguistic entity. This view would be highly restrictive or counterintuitive, as deaf and mute persons, babies, and those who do not know any language will not have the idea of truth and falsehood. Hence the view which assigns truth and falsehood to sentences only would be equally restrictive or counterintuitive. In this respect, the Nyāya view has an edge over Western views, as a qualificative cognition consists of a set of relational entities, which are due to relations with the things in the world, internal or external. In the technical language, they are called viṣayitās.

(3) The property expressed by the word “sākṣātva” characterizes perceptual apprehensions, true or false. It is a unique property of perceptual cognitions only. Hence the cognitions generated by visual, tactual, auditory, olfactory and gustatory sense-organs are characterized by directness.

(4) Suppose F and G are properties, but a, b, and c are loci. In the case of saṃkara defect the following holds good:

1. a is F, but not G,
2. b is G, but not F, and
3. c is both F and G

The following example may be cited from the Nyāya ontology. Consider the relation between earthness and sense-organness. Earthness resides in physical objects such as a table, but sense-organness does not reside in these objects. Now consider gustatory sense-organ. It is characterised by sense-organness but not by earthness. Now consider the olfactory sense-organ. It is characterised by both earthness and sense-organness.

It is to be noted that, according to the Nyāya, if two class-essences have the same extension, then they are identical and the expressions for them are synonymous. For example, ghaṭatva (potness) and kalaśatva (jarness). In other words, both the expressions have the same meaning and reference. Now consider the relation between ‘gotva’ (‘cowness’) and ‘galakambalavatva’ (‘the property of having dewlap’).These two expressions have the same extension, but not the same meaning. Since a class-essence is an indivisible property, gotva (cowness) is to be considered a class-essence.

Since the expression ‘galakambalavatva’ (‘the property of having dewlap’) refers to a complex property, it cannot be treated as a class-essence. Hence we do not have two class-essences (jātis) for the same class.

An individual belongs to several classes which are expressed by natural kind terms. For example, an object is a jar as well as a substance. Hence it has both jarness (ghṭatva) and substanceness (dravyatva). Since both of them are class-essences, they do not have the same extension. The Nyāya claims that they are related by pervader-pervaded (vyāva-vyāpaka) relation. Hence jarness is the pervaded and substaceness is the pervader. Hence the sentence ‘(x) (Fx É Gx)’ would be true if ‘F’stands for jarness and ‘G’ for substanceness. In other words, one class is a proper subset of another class. In the case of saṃkara defect this condition does not hold good, as one of them is not pervaded by the other. For this reason both of them cannot be considered class-essences.

(5) On this point one may ask whether we can consider truth (pramātva) as the class-essence but not the perceptual apprehendedness (sākṣātva). As an answer to this question, it is claimed that both the followers of the Nyāya and their opponents have accepted perceptual apprehendedness as a class-character, but not truth. Hence the acceptance of perceptual apprehendedness as a class-essence is free from doubt. For this reason the Nyāya accepts the latter as a class-essence but not the former.

(6) In this context it is to be noted that the Nyāya logicians have not used the word ‘and’ in the way Western logicians have used it. A conjunctive cognition, such as p and q, would be true, according to Western logicians, if both of them are true; otherwise false. Hence if p is true but q is false, the conjunctive cognition is false. On the contrary, the Nyāya philosophers consider it to be true in one respect but false in another. They claim that if it is treated as false because of the falsity of one conjunct, then it may, by the same token, be treated as true, as one of the conjuncts is true. This is how the Nyāya would justify the claim that a conjunctive cognition of this sort as a whole cannot be treated as true or false.

(7) A property (dharma) is defined in the following way:

x is a property = Df. (Ey)(y is a locus of x).

The following diagram represents the Nyāya concept of property (dharma):

It is to be noted that a class-character (jāti) cannot be identified with the Western concept of universal. This is due to the fact that a class-character (jāti) is eternal and resides in its instances by the relation of inherence. But there is another concept of universal (sāmānya) which is much wider than that of class-character, as it includes properties that are not related to their loci by the relation of inherence. Moreover, complex objects residing in several places are also treated as universals. According to the Nyāya ontology, the ultimate particularities belong to eternal objects, atomic or ubiquitous. The particularity or the uniqueness of a non-eternal object, such as a table, is a function of the ultimate uniquenesses of its ultimate parts. They are properties, but not universals. Hence there is a distinction between the terms ‘jāti’, ‘sāmānya’, and ‘dharma’. (See also [3. P. 66—70]).

(8) It is to be noted that according to the Nyāya there are negative entities, in addition to several types of positive entities. A negative entity has a two-fold dependence on a positive entity. If x represents a positive entity, then not-x would represent a negative entity, provided x is not a universal property, such as existence or knowability. Moreover, not-x resides somewhere, which is a positive entity. The negatum x is called ‘conterpositive’ and is characterised by the property of being the counterpositive. If x is known through perception, then not-x or absence of x is also known through perception.

Regarding the types of negation, the Nyāya philosophers have accepted relational absence, and mutual absence or difference. The distinction between them can be drawn in terms of the limiting relation of the property of being the counterpositive which resides in the negatum. At the linguistic level these negations can be represented by the following forms:

1. x is not in y or x does not occur in y, not-x is in y.
2. x is not y or x is different from y.

(1) represents relational absence and (2) represents mutual absence. In (1) not-x occurs in the locus y, and x is the counterpositive of not-x. The property of being the counterpositive residing in x (i.e., the role of x) is limited by both x-ness and an occurrence-exacting relation. In other words, both x-ness and an occurrence-exacting relation are modes of presentation of x. Here x-ness is the property-limitor and an occurrence-exacting relation is the relation-limitor.

In (2) y is the counterpositive, i.e., the negatum, and the property of being the counterpositive residing in y is limited by both y-ness and the relation of identity. So the relation of identity is the limiting relation of the property of being the counter-positive.

Most Nyāya philosophers have accepted three types of relational absence:

1. The relational absence of an object before its production is the not-yet type of absence (prāgabhāva). The absence of a table before its production is present in its parts. The cognition of this absence can be expressed by the sentence ‘A table will be produced in these parts’. When the table is produced, the not-yet type of absence does not exist in its part. Since it cannot exist anywhere else, it ceases to exist. This type of absence has no beginning, but has an end. Since we are not asserting the absence of all tables, but the absence of the table which will be produced, the property of being the counterpositive is limited not by a generic property like tableness, but by a specific property like a particular blue colour and As regards the limiting relation of the property of being the counterpositive, there is some difference of opinion among the Nyāya philosophers. It is claimed that since the table has not yet been produced, the property of being the counterpositive is not limited by any relation. But the old Nyāya has accepted a temporal relation as the limiting relation of the property of being the counterpositive. If the absence of the table is in its parts at t_n and the table is produced in the parts at t_n+1, then obviously the table is related to its parts by the relation of posterior existence. This temporal relation of posterior existence is considered as the limiting relation of the property of being the counterpositive. But the followers of the Navya-Nyāya do not subscribe to this view.
2. The second type of relational absence refers to an object after its destruction. It may be called ‘the no-more type of absence’ (‘dhvaṃsa’). The absence of a particular table when it is destroyed is present in its parts. Since the destruction of a particular table does not imply the destruction of all tables, the property of being the counterpositive is limited, not by a generic property, but by a specific property of the table which has been destroyed. As regards the limiting relation, here also there is difference of opinion among the Nyāya philosophers. The followers of the Navya-Nyāya do not accept any limiting relation, while the followers of the old Nyāya have accepted a temporal relation as the limiting relation. If the destruction of a particular table is the separation of its parts, then the whole table ceases to exist at time, say, t_n, when it is destroyed. If ‘ceases to exist at time t_n’ is explained as ‘existent at time t_n–1’, then the parts are related to the table by the relation of previous existence. For this reason it is claimed that the property of being the counterpositive is limited by the temporal relation of previous existence. Apart from this temporal relation the property of being the counterpositive is not limited by any other relation. A no-more type of absence has a beginning, but no end.
3. The third type of relational absence is the never type of absence (atyantābhāva), for example, the absence of colour in air, or the absence of a jar on the ground. Some of the followers of the old Nyāya do not consider the absence of a jar on the ground as a case of never type of absence. Since a never type of absence has neither a beginning nor an end, and since the absence of a jar on the ground has both a beginning and an end, these philosophers think that there is a need to accept a fourth type of relational absence. But the followers of the Navya-Nyāya as well as some of the followers of the old Nyāya think that the acceptance of the fourth type of relational absence would lead to the postulation of innumerable absences of a jar on the same ground. Each time the jar is removed, a new absence is created, and each time the jar is brought back, the previous absence is destroyed. To avoid this consequence, it is claimed that what ceases to exist when the jar is brought back is not the absence of it, but the relation of this absence to the ground. An absence is related to its locus by a self-linking relation which is to be identified ontologically with its locus. Now the followers of the Navya-Nyāya believe that this self-linking relation in the case of the absence of a jar on the ground is to be identified not with the ground as such, but with the ground when a jar is not present. Since this self-linking relation ceases to exist when a jar is brought on the ground which had an absence of a jar, we cannot perceive this absence when a jar is present on the same ground. So on the ground of parsimony these philosophers have included such examples under the third type of relational absence.

The property of being the counterpositive of a never type of absence is limited by both a property-limitor and a relation-limitor. But the limiting relation is an occurrence-exacting one. (For a detailed discussion as well as for double negation, see [4]).

(9) The Nyāya use of the word ‘limitor’ requires some explanation. The predominant use of the term may be defined in the following way:

x is limited by y iff

1. both x and y are properties,
2. x is a relational property, and
3. the property y is a mode of presentation of the object where relation property x

In this context it is to be noted that the expression ‘mode of presentation’ is used in such a way that it determines the referent(s) of a term.

(10) Here the author refers to a pervader-pervaded relation. If x is pervaded by y, then x is the pervaded and y is the pervader. For example, smoke is pervaded by fire. In other words, wherever there is smoke there is fire. In this case the contact with the monkey characterizes the top of the tree. So the top of the tree becomes the limitor and the contact with the monkey is the limited. Since the top of the tree is the pervaded of the contact with the monkey, it becomes the limitor.

(11) In the cognition, these are silvers, both the piece of silver and the mother-of-pearl are qualificands. As regards the qualificand — qualifier (viśeṣya — viśeṣaṇa ) distinction, the Nyāya philosophers claim that it is applicable to every qualificative or relational cognition (savikalpaka—jñāna). Only non-qualificative cognition (nirvikalpaka — jñāna) is excluded from its scope. A qualificative cognition has the form aRb, where a is the qualificand, b is the qualifier and R is the qualification relation which relates b to a.

The qualifier plays the role of a distinguisher. Hence it distinguishes something from other things or a collection from other collections. That which is distinguished by it is the qualificand. (For a detailed discussion on this topic, see [5]).

(12) Here the author uses the technical term ‘aprakārakatva’, which requires some explanation. The qualifier is called ‘viśeṣaṇa’, but the qualifier presented under the mode of the relation is called ‘prakāra’. When cognition is related to this type of qualifier, it has the relational property called ‘prakāratā’. The converse of this relation that characterises the cognition is called ‘prakārakatva’. Hence it is a relational property of a cognition. So aprakārakatva is the property of not cognising this type of qualifier.

(13) Here the author uses the word ‘pratiyogī’ which is the second term of a relation whose first term is called ‘anuyogī’. This category applies to any relation between two entities. Now let us define these terms. Suppose a and b are related by the relation R.

a anuyogī (the first term) iff b is related to a by the relation R. At the level of cognition the relation R is cognized in a, but not in b. The other term of the relation R, namely b, is pratiyogī (the second term). The relation of our cognition to this aspect of a is expressed by the term ‘anuyogitā’ (‘the property of being the first term’). Hence anuyogitā (the property of being the first term) is a relational property of a. Similarly, b has pratiyogitā (the property of being the second term).

In the true cognition this is silver, there is use of truth whose second term is silverness. Similarly, in the true cognition this is nacre, there is the use of truth whose second term is nacreness.

(14) Here an objection has been raised from the nature of identity relation. If truth is one indivisible entity, then the truth qualified by silverness would be identical with the truth qualified by nacreness, as both the cognitions are true.

(15) In this context, it is to be noted that although the relation of inherence is one, the relation of inherence to its loci are different. If a colour is related to its locus by the relation of inherence, then the relation is the property of being the locus of the inherence qualified by colour which is its second term. Hence the locus is the first term and the colour is the second term, and relation is the specification of inherence.

(16) Here the author uses the expression ‘bheda’, not ‘bhinna’. If x is different from y, then x has the property being different from y. Hence difference from y becomes a property of x. The word ‘bheda’ refers to a property, but the word ‘bhinna’ refers to the property-possessor.

(17) It is to be noted that the limitor consists of the property of being qualified by being other than a quality (guṇānyatvaviśiṣṭatva) and the property of being existence (sattātva).

(18) It is claimed that existence is a property of a quality, and the property of being qualified by being other than a quality is a property of existence. Since we do not cognise this type of complex property in a quality, it lacks this complex property. It has existence, but not the property of being qualified by being other than a quality.

(19) This cognition has the form a is F, where ‘a’ stands for a true cognition of a nacre. In the technical language of the Nyāya ‘the true cognition having nacreness as its second term’. ‘F’ stands for ‘truth whose second term is silverness’. It is claimed that there is no cognition of this type.

(20) It is claimed that as existence resides in different objects such as substance, quality and action, so does truth reside in true cognitions. In the previous paragraph the author has discussed whether a qualified existence such as the existence qualified by being other than a quality resides in a quality. Here the author is discussing whether the truth which has nacreness as its second term resides in another true cognition.

(21) The following remark of Gaṅgeśa: tadvati-tatprakāraka-anubhavaḥ pramā has been interpreted as taddharmavadviśeṣyakatve sati tatprakāraka-anubhavaḥ pramā. A true cognition may be defined in the following way:

The cognition a is F is true if and only if a is F is an apprehension, and the F, which is the qualifier of this apprehension (prakāraka), resides in a which is the qualificand of this apprehension (viśesṣyaka).

(22) Here the category dharma-dharmī (property-property possessor) has been used to emphasize the relation of cognition to fact. Hence in the true cognition of a is F, the cognition is related to a which is the possessor of F. In other words, the viśeṣyaka of this cognition is due to a and the prakāraka is due to F. The cognition is characterized by a relational property which is due its relation to a and by another relational property which is due to its relation to F.

(23) The Nyāya use of the word ‘anubhava’ (‘apprehension’) does not apply to memory (or memory-cognition). According to the Nyāya, perception, inference, comparison, and testimony (or verbal cognition) are sources of valid cognitions. If a cognition is derived from, or caused by, perception, inference, comparison or verbal cognition (i.e. from the cognition of words which have occurred in a sentence), then it is characterized by the property of being apprehended (anubhavatva). Since a memory-cognition lacks the property of being apprehended, it is not a case of apprehension, although it rests upon some previous apprehension derived from, or generated by, perception, inference, comparison, or verbal cognition. A memory-cognition is due to mental (thought) disposition which is again due to some previous apprehension. Since cognitions are divided into apprehension and memory, ‘apprehension’ may be defined as ‘a cognition different from memory’.

Since there are four types of apprehension depending upon their causal conditions, each of them is characterized by a property which signifies whether it is derived from perception, inference, comparisons, or verbal cognition. Hence the apprehension due to perception (or sense-organ) is characterised by the property of being perceptual (darśanatva), the apprehension due to inference is characterised by the property of being inferential (anumititva), the apprehension due to comparison is characterised by the property of being comparison (upamititva), and the apprehension due to verbal cognition is characterized by the property of being verbal (śābdatva). The property of being the verbal cognition is the property of being the understanding of the relation of the (primary or secondary) referent of a word to the (primary or secondary) referent of another word.

Consider the sentence ‘a is F’. The property of being the verbal cognition charactrises the relation F to a. Hence this understanding presupposes the cognition of the referent(s) of an atomic expression. According to the Nyāya the properties of different types of cognition are cognized at the level of mental perception or higher order cognition.

Validity (pramātva) is to be defined in terms of the property of being apprehended (anubhavatva) and truth. A memory-cognition will be true if the previous apprehension which is its causal condition is valid. Hence the truth of a memory-cognition presupposes the truth of a previous apprehension. If the previous apprehension does not correspond to a fact, the memory-cognition corresponding to it would be false. Hence a memory-cognition is either true (yathārtha) or false (ayathārtha). Since the memory-cognitions do not have the property of being apprehended, they are called ‘invalid’, whether true or false. Hence the Nyāya use of the word ‘invalid’ (‘apramā’) cannot be equated with ‘false’.

(24) The word ‘pramāṇa’ consists of the words ‘pra’, ‘mā’, and ‘anat’, where ‘pra’ is the prefix, ‘mā’ is the verbal root, and ‘anat’ is the suffix. The word ‘pra’ means ‘excellence’, and ‘mā’, means ‘apprehension’. But the suffix ‘anat’ has two uses. In its impersonal voice (bhāva-vācya), it is simply used to change the form of a verb. Hence it does not signify anything more than the verb ‘mā’. If we take this interpretation, then the word ‘pramāṇa’ means ‘the apprehension which has excellence or truth’. In other words, it means a true apprehension.

But the instrumental use (karaṇa-vācya) of the word ‘anat’ signifies the special instrumental cause of a true apprehension. Since the word ‘pramāṇa’ has two meanings, the word ‘prāmāṇya’ also has two meanings. One of them is truth (pramātva) or the property of being true apprehension, and the other one is the property of being the special instrumental cause (karaṇatva) of a true apprehension.

There is a 3^rd or a secondary meaning of the word ‘pramāṇa’. In this sense, it refers to the locus of a true apprehension. [6]

A) In this context it is to be noted that the Nyāya philosophers are defining the truth of a cognition in terms of the following three properties:

1. i) tat-prakārakatva (the property of having that as its qualifier at cognitive level)
   The relation of cognition to a (relational) property (prakāra) is called ‘prakāratva’. The converse of this relation i.e. the relation of a prakāra to cognition is called ‘prakārakatva’. Hence it refers to a relational property of a cognition. Consider x is F in the relation R, where x is the qualificand (viśeṣya), F is the qualifier (viśeṣaṇa), and R is the qualification relation (saṃsarga). According to the Nyāya, when F is presented under the mode R it is called ‘prakāra’. For example, the table has a book in the relation of conjunction. Here the table is the qualificand (viśeṣya), and the book presented under the mode of conjunction is the relational qualifier (parkāra), in short, qualifier. The relation of the cognition to the qualificand is called ‘viśeṣyatā’, the relation of cognition to the prakāra (relational qualifier) is called ‘prakāratā’, and the relation of cognition to the viśeṣaṇa (qualifier) is called ‘viśeṣaṇatā’. The converse of these relations are called ‘viśeṣyakatva’, ‘prakārakatva’ and ‘viśeṣaṇakatva’ respectively. Since the definition of ‘truth’ is given in terms of the relational properties of the cognition, it refers to viśeṣyakatva and prakārakatva. In this respect the Nyāya definition is to be contrasted with the Western definitions, as they refer to propositions, statements, or sentences.
2. tatvad-viśeṣyakatva (the property of having the qualificand qualified by that at cognitive level).
3. tatvad-viśeṣyakatva nirūpita tat-prakārakatva (The latter being determined by the former).

It is to be noted that the first two conditions are applicable to true as well as to a type of false cognition called ‘viparīta-bhrama’ (a type of false conjunctive cognition). For this reason, the third condition has been used to distinguish true cognitions from a type of false ones (viparīta-bhramas).

The determiner-determined relation (nirūpya-nirūpaka sambandha) holds between the correlatives. Hence it can be defined in the following way:

x is determined by y if they are correlative properties.

In this context it is to be noted that some Nyāya philosophers have claimed that the latter is limited by the former. Hence tadvad-viśeṣyakatva-avacchinna-tat-prakārakatva is the defining property of truth.

Now it may be asked whether we can define truth in terms of tat-prakārakatva- avacchinna-tadvad-viśeṣyakatva. Since there is mutual limitor—limited relation between tat-prakārakatva and tadvad-viśeṣyakatva, if one holds, then the other will also hold.

Now the Nyāya philosophers claim that if truth is defined in terms tat-prakārakatva- avacchinna-tadvad-viśeṣyakatva, then the limitor includes both the qualifier and the relation which is the mode of presentation of the qualifier. But if it is defined as tadvad-viśeṣyakatva-avacchinna-tat-prakārakatva, then the limitor will be the qualificand only. Hence on the ground of simplicity, the latter is preferable to the former.

B) Now let us discuss why truth cannot be defined in terms of tadvad-viśeṣyakatva (the property of having the qualificand qualified by that at cognitive level) only. It is to be noted that this property is present in a false cognition as well. This is due to the fact that the referent of ‘tat’ (‘that’) has not been specified.  Consider the cognition of the lake has fire. If we substitute lakeness for ‘that’, then the cognition has the lake qualified by lakeness as its qualificand. Since this property characterises a false cognition, it suffers from over coverage (ativyāpti). For this reason, the word ‘tat’ cannot be used as a variable for any qualifier. It stands for the chief qualifier of a cognition.

In the cognition of a blue table, the word ‘that’ refers to a blue colour which is the chief qualifier. Hence it satisfies the condition tat-prakārakatva (the property of having that as its qualifier at cognitive level). Since this cognition has tadvad-viśeṣyakatva (the property of having the qualificand qualified by that), which resides in the cognition of the table, it satisfies both the conditions of true cognition.

As mentioned before, these two conditions are not sufficient to define truth, as both of them are present in an opposite false conjunctive cognition, such as the mountain has water, and the lake has fire. To eliminate this type of false cognition, the Nyāya introduces the third condition in the definition of truth.

C) Now let us discuss whether pramātva (truth) can be defined in terms of prakāravat-viśeṣyakatva (the property of having the qualificand possessing the qualifier at cognitive level). According to this view a true cognition is qualified by the converse of the property of being the qualificand possessing the qualifier. Consider the cognition of a is F. This cognition is related to the complex object or fact a possessing F. Hence the object of this cognition, namely, a possessing F, has viśiṣṭa-viṣayatā, which is a relational property. The converse of this relational property called ‘viśiṣṭa-viṣayitā’ would qualify this cognition. If we define truth as prakāravat-viśeṣyakatva, then we have to accept viśiṣṭa-viṣayitā, in addition to prakārakatva (the converse of the property of being the qualifier) and viśeṣyakatva (the converse of the property of being the qualificand.

It is to be noted that the viśiṣṭa-viṣayitā is present in desires which can be fulfilled (saṃvādī-icchā). Most Nyāya philosophers have accepted viśiṣṭa-viṣayitā’, which is the converse of viśiṣṭa-viṣayatā, but not as a defining property of truth. This is due to the fact that the definition of truth would violate the principle of simplicity, as we have accepted an additional property in the definition of truth. Moreover, this definition would suffer from over coverage, as it is present in desires which can be fulfilled.

D) As regards the (T) sentences of Tarski, such as ‘p’ is true ≡ p, the Nyāya claims that the meaning of “ ‘p’ is true” is not the same as that of ‘p’, although they are equivalent. In other words, ‘p’ would simply generate the cognition of p, but the cognition generated by “ ‘p’ is true” would include both the sentence ‘p’ and the meaning of ‘true’ as its objects. The meaning of ‘true’ would be the qualifier (viśeṣaṇa) of the sentence ‘p’. Hence it means ‘the cognition generated by the sentence ‘p’ has the property of having things or objects as they are (yathārthatva) cognised to be’.

E) In Western philosophy, the supporters of the correspondence theory define truth in terms of correspondence with a fact. Hence the questions are asked about the nature of this correspondence and the nature of propositions or statements which are truth-bearers. But the Nyāya philosophers would avoid these questions as they have defined truth in terms of three properties of a cognition.

F) As regards the relation between a true cognition and the fulfilment of our desire (icchā) or mental effort (kṛti), the Nyāya philosophers claim that every true cognition has the property of being a cause for the fulfilment of our desire or effort (saṃvādī-pravṛtti-janakatva). But this property cannot be equated with truth, although we infer the truth of a cognition utilizing this property. For example, we have a true cognition of water. Then we have a memory cognition of a general proposition, such as, if it is water, then it will quench our thirst. So we remember a sentence of the form ‘(x) (Fx ⊃ Gx)’. Then there will be cognition that this object, i.e., water, would also quench our thirst.

Now the following questions may be asked: How do you know the truth of the universal proposition? How do you know that water had quenched our thirst in the past?

According to the Nyāya philosophers the truth of the universal proposition or the truth of the previous cognition would be known by some other cognitions which are non-dubious. In other words, the truth of a cognition, say p, is known by the non-dubious cognition of q, and the truth of the latter by some other doubt-free cognition, say r, and so on. There is a regress, but not opposed to the relation of cause and effect. This type of argument is free from circularity, although there is a causal regress, which stops at non-dubious cognition. It avoids the following two extreme views: i) the truth of p cannot be known unless we know the truth of q. Similarly, the truth of q cannot be known unless we know the truth of r, and so on. ii) To know the truth of p we have to know the truth of q, and in order to know the truth of q we have to know the truth of p or some cognition which will entail p.

From this discussion it follows that the Nyāya view is more plausible than many other views in philosophy, as it avoids both circularity and skepticism.

G) From the above discussion of the Nyāya philosophers it follows that truth is a complex divisible imposed property (sakhaṇḍa upādhi) consisting of three properties. It is neither a class character nor is it a simple indivisible-imposed property. There are as many truths as there are true cognitions. In this respect it is similar to the semantic conception of truth, as the word ‘true’, according to Tarski, refers to a set of true sentences. But the Nyāya philosophers can explain the truth of each of the true cognitions in terms of three properties of the cognition. The Nyāya concept may be presented in the following way:

The cognition x is F is true if and only if (i) F is cognised as the qualifier (along with its relation to a), (ii) x as the possessor of F is cognised as the qualificand, and (iii) the latter is the limitor of the former. Since ‘x’ and ‘F’ are variables, ranging over relational properties of a cognition, we can explain the truth of cognitions which are substitution instances of the above definition.

(25) In this context it is to be noted that the Nyāya philosophers have applied the definition of truth to the entire conjunctive cognition, not to each of its conjuncts separately.

(26) It is to be noted that the category of determiner-determined holds between correlative properties. Hence if x is determined by y, then y is also determined by x. If there is mutual determiner-determined relation, then there is mutual limitor—limited relation as well. But the limitor-limited relation does not imply the determiner-determined relation. Consider the cognition of a jar. In this cognition jarness is the limitor of the jar as well as the property of being the qualificand residing in the jar. But there is no relation of determiner-determined between jarness and jar, or between jarness and the property of being the qualificand.

Here the author has used the category of determiner-determined relation, but other authors, including the author of Bhāṣāparicchedaḥ, have used the category of limitor-limited. The latter category signifies the qualificand-qualifer relation (viśeṣya-viśeṣaṇa sambandha) between them, but not the former.

(27) The Nyāya conception of truth cannot be equated with the Western correspondence theory of truth. This is because the Nyāya philosophers define truth in terms of certain properties of a cognition, not in terms of the relation of correspondence between a cognition and a fact. Since it is different from the classical correspondence, coherence and the pragmatic conception of truth, as well as all the varieties of deflationary conception of truth, here we come across a new animal in our zoo.

About the authors

Jaysankar Lal Shaw

Author for correspondence.
Email: jaysankar.shaw@gmail.com
ORCID iD: 0000-0002-1605-6020

PhD, Professor

References