Semantics of Nothingness: Bhartrhari’s Philosophy of Negation – I

Semantics of Nothingness: Bhartrhari’s Philosophy of Negation – I


Bhartrhari (fl. 450 CE) is one of the foremost philosophers of classical India. While there are many narratives relating the story of his life, that he was a king-turned-hermit, the author of three hundred stanzas, and so on, one thing is certain: he was the author of the masterpiece, the Vākyapadīya (VP). There are very few texts in the history of Indian philosophy that have had as penetrating an influence as this one. Although the text primarily relates to the philosophy of Sanskrit grammar, the first section on the Brahman (brahmakānda) discusses the metaphysics of, and provides the philosophy for, non-dualism, with the introduction of terms such as ‘transformation/false projection’ (vivarta) that became pivotal to subsequent philosophers, such as Śankara (700 CE). Bhartrhari’s thought can also be seen unmistakably on the works of another prolific classical Indian philosopher, Mandana Miśra (700 CE). 2 The depth to which Bhartrhari has shaped Indian philosophy has yet to be properly appreciated, as scholars are coming to recognize that even the Pratyabhijñā school of Kashmiri non-dualism is largely derived from Bhartrhari’s philosophy of language.

After the “linguistic turn” in the latter half of the twentieth century, philosophers in the West have been more open to exploring the possibility of solving philosophical problems by understanding more about language. 3 It would not be an exaggeration to say, by way of contrast, that philosophical speculation in India has linguistic origins. Early Brahmanical thinking is heavily ritualistic and relies on analyzing Vedic sentences. Classical philosophers primarily derive their conclusions from an exegetical analysis of the Upanisads or the Sūtra literature. The philosophical debate among Hindus, Buddhists, and the Jains oftentimes goes back to linguistic issues. The linguistic philosophy of Bhartrhari needs to be addressed in his milieu. His speculations about the nature of language and his analysis of Sanskrit both transcend the boundaries of language and relate to metaphysics, epistemology, and ontology.

Since understanding some of the most pivotal issues in the history of Indian philosophy, and particularly those issues involving debates about nonbeing and being, are so dependent on traditional Indian philosophy of language, understanding how classical Indian thinkers understood negation and how it functions linguistically is fundamentally important. This paper will therefore examine what the seminal grammarian Bhartrhari had to say about negation, particularly in debates with rival philosophers and schools. Reading Bhartrhari’s philosophy of negation is therefore not restricted to merely analyzing Sanskrit syntax. While he was an original thinker, many of his ideas have evolved historically, and we cannot address Bhartrhari’s philosophy without seeing it in the context of his predecessors. This, however, is not to indulge in only history of philosophy, but only to point out that history of ideas should not be ignored while exploring answers to philosophical questions.

Negation: From Patañjali to Bhartrhari

Patañjali (150 BCE) is one of the earliest scholars to explicitly describe two types of negation: prasajya and paryudāsa, 4 generally translated as nonimplicative and implicative negations. The first one is used to simply negate the existence of X (there is no X), while the other refers to negation of X in Y (a Y that is not X). This twofold schema of negation is used in Indian philosophy for morphological analysis (as in Patañjali’s Mahābhāsya), sentence analysis (primarily in the tradition of Mīmāmsā), and metaphysical analysis (both in the Nyāya tradition and in the Mādhyamika of Nāgārjuna). However, the way negation has been analyzed and applied varies from one school to another. Around the same time as Patañjali, Jaimini (200 BCE) explored primarily the sentences used in ritual injunction and systematized a framework of threefold negation, including the injunction of an alternative by means of negation. 5 Semantically, both forms of negations are expressed by the negative particle nañ, 6 and both are addressed in the semantic and morphological analysis of Patañjali and Bhartrhari. This paper is limited to the meaning of negation in the work of these two grammarians. I will also briefly engage the views of negation of both Jaimini and Śabara and analyze some of their crucial positions in an attempt to expand upon the semantic analysis of nañ.

It has already been mentioned that there are implicative and nonimplicative forms of negation. The implicative negation affirms something (y) by means of negating one entity (x). By contrast, the nonimplicative type of negation simply negates a purported fact. Sanskrit grammarians often cite the following verse to identify these forms of negation:

If the negative particle corresponds with the final term [in a compound], this should be known as implicative negation. If the negative particle corresponds with the verb [in a sentence], this should be considered the nonimplicative negation. 7

Fritz Staal writes nonimplicative negation as ~F(x) and implicative negation as F(~x) (1962; see also 1988, 260). The most oft-cited example of implicative negation is abrāhmana, where the term is not used in negating a brāhmana but in the affirmation of someone else who bears some of the characteristics of a brāhmana. While all classical discussions on negation can be categorized as word-negation and sentence-negation, the position of Bhartrhari favors sentence negation, as words are not independently meaningful in his paradigm of the non-divisible sentence, sphota. In this metaphysics, whether expressions are made in words or sentences, they all stem from speech (vāc) identified with the Brahman, and no form of negation can negate this foundation. Evidently, even the word used for negation is nonetheless a word. 8

Bhartrhari’s Analysis of Sentential Negation

The word ‘na’ is used in the Sanskrit language for both implicative and nonimplicative negation. In compounds, the particle ‘nañ’ or its derivative (an, if followed by a vowel) expresses the concept of negation. Commentators suggest that twofold negation is implicit in Pānini’s (fourth century BCE) Astādhyāyī (AA). 9 George Cardona synthesizes this position by accepting two types of negation in Pānini’s rules, where one is constructed with the nominal following the negative particle in the compound (e.g. “non-X” or F(~x)), and the other is linked with the verb (e.g. ~F(x)). He further explains that the first is the positive rule, as it provides operation in the domain restricted by the negative particle and the second negates operation, thereby stopping an operation that has already been given by other rules (Cardona 1967).

In Indian philosophy, grammarians primarily focused on morphology, with words being their immediate concern. Mīmāmsakas, the ritualist philosophers from classical India, fulfilled the need of contemplating upon sentential meaning. In order to advance the analysis of sentence negation in Sanskrit grammar, I will briefly explore examples from the Mīmāmsā school, although a detailed study of this aspect would require a much larger space. Below are three examples they give of sentential negation:

  1. One should not eat kalañja. 10
  2. [The phrase] ‘ye yajāmahe’ is cited during sacrifices except for the Anuyājas. 11
  3. [The sacrificer] does not hold the sodaśi vessel in the Atirātra.

The first is an example of imperative negation, which I will set aside, as it demands a separate treatment. Something positive is derived from the second sentence, while the third sentence simply negates a fact. In the Sanskrit language, the way negation is linked, whether with the antecedent term or with the verb, determines which type of negation is used in a sentence. For instance:

phalam nāsti |

S neg. V.

  1. There is no fruit.
  2. [This] is not a fruit.

In the above example, the sentence can be understood either way. In the first translation, negation corresponds to the verb, negating the existence of fruit. In the second translation, negation corresponds to the subject, while the object under consideration, such as a plastic replica, is not a fruit. This distinctive understanding evokes the classical debate between the particularists, those who maintained that sentence meaning is gleaned from the meanings of the words in the sentence, and the wholists, who maintained that sentence meaning is indivisible.

Whether the sentential meaning is derived from words that independently express meaning, or whether meaning is a collective or unitary expression of a sentence, is one of the classical debates involving multiple schools in Indian history. The wholists, such as Bhartrhari and the Prābhākaras, and the particularists, such as Naiyāyikas and the Bhāttas, wrestled over what the term ‘na’ negates. Even when we engage the position of the particularists who state that negation relates to specific terms in a sentence, there are various ways in which the negative particle can be analyzed. For instance, (1) what we negate is the cognition of the existence of what has been negated; (2) negation affirms the falsity of cognition; (3) negation not only denotes itself but also its substratum; and (4) negation in a sentence indicates the sense communicated by the word with which the negative particle is linked.

These are therefore the four possible analyses of ‘na’ according to the particularists:

  1. Negation in a sentence negates the existence of the referent linked with the negative particle nañ. The sentence would be “There is no fruit” in the above example.
  2. Negation makes known the falsity of cognition. The sentence, then, is “This is not a fruit” (but a plastic replica).
  3. The negative particle denotes both negation and the substratum of that negation. In an example, “There is no book on the table,” the substratum of the negation, the table, is also referred to by this negation. In this manner, the referent of the negative particle is not the table but the book, but the table is implicitly referred to as existent.
  4. The negative particle is not independently meaningful. Since it means something by being affiliated with other words, it is therefore ‘coreferential’ (dyotaka). Negation, in this sense, is intrinsic to the meaning of the word itself, as is affirmation. For instance, the word ‘table,’ when articulated, has the potential to both affirm and negate the table. That is, negation is already there as a potential within the term, and the negative particle only brings to the spotlight what is already there as the meaning of the term.

Although Bhartrhari maintains a holistic approach, he does not reject some of the arguments discussed above. In particular, he addresses at length the fourth point. This approach highlights his broader agenda to synthesize all the existing positions as far as possible. Returning to our example, the term ‘table,’ for instance, can mean both the being and absence of table, and the negative particle simply highlights negation. This, however, is not to say that Bhartrhari surrenders to the particularists, as he rejects their viewpoint and concludes that words such as ‘asymmetry’ cannot be broken into parts and analyzed separately. This position of Bhartrhari tallies with that of Wittgenstein: “The positive proposition necessarily presupposes the existence of the negative proposition and vice versa” (T 5.5151). For Bhartrhari, both assertion and negation rest on speech (vāc), equated with the highest universal (mahāsāmānya) that involves all that exists. For Wittgenstein, both positive and negative facts are “facts.” Negation is crucial for Bhartrhari not just for comprehending sentence meaning, but also because it represents his primary strategy for describing reality: both of the terms that he uses in the first verse of VP to describe the absolute that is identical to speech are constructed in the negative form (anādinidhanam and aksaram). This negation, however, does not go all the way, because for Bhartrhari, the absolute is the Brahman, or the Śabda-tattva, a positive entity.

This analysis of the negative particle needs to be read in light of the ways classical philosophers have assigned meaning to it. There is not one single position, even among the grammarians, regarding the role of nipātas, a class of word elements of which the negative particle is a member. Whether these particles are the signifiers (vācaka), or the cosignifiers (dyotaka) is another question where the classical philosophers differ. For grammarians such as Bhartrhari, these particles appear to be merely cosignifiers. This needs to be understood within the context that ‘meaning’ in Bhartrhari’s philosophy is understood in terms of ‘primary’ and ‘secondary.’ Cosignifiers bring the secondary meaning to the spotlight. In words such as ‘asymmetry,’ if negation of symmetry were the primary meaning, the negative particle /a/ would be the signifier and not cosignifier. Grammarians such as Bhattoji Dīksita (seventeenth century), however, maintain that the particles are both the signifiers and cosignifiers. The Nyāya philosophers, in yet another variation, maintain that select particles in the group of ‘pra’ that are prefixes are only cosignifiers and the other particles in the group of ‘ca’ are only signifiers. Grammarians reject this position (VSM 1977, 56).

Although these positions may appear to be merely linguistic, philosophy in India is closely intertwined with linguistic issues such that one cannot be addressed without the other. Whether there is a primacy of the particle (i.e., negation, rather than an affirmation of something positive) or not can change the course of ritual for the ritual philosophers, the Mīmāmsakas. One of the central categories of the Advaitins is ignorance or avidyā, a negative term, and the difference in understanding leads to the position of one sub-school or the other. The school of logic, Navya-Nyāya, advances its argument to counter the Sautrāntika-Mādhyamika arguments of negation. And even within the Mādhyamika school of Buddhism, the division of Svātantrika and Prāsangika primarily rests on how to interpret negation. Therefore, linguistic speculation about negation is a gateway to enter the many schools of classical Indian philosophy.

Bhartrhari addresses negation at length in two different sections of his writings, first when establishing sentence meaning (VP 2.240–45), and second in the last section of his magnum opus when addressing compounds (VP 3.14.248–315). In the first instance, Bhartrhari deals with the issue of negation when addressing sentence meaning, maintaining that meaning cannot be reduced to single words and that a sentence gives a unitary meaning and must be read as a whole. In Sanskrit, one can place the negative particle at the end. For instance, aśvatthaś chedanīyo na (one should not cut the ficus tree). Bhartrhari argues that, if each word were to independently give rise to meaning, the sentence could be considered complete before the negative particle appears, enjoining one to cut the tree (VP 2.240). If words were to independently convey meaning, in sentences such as vrkso nāsti (tree, there is not), one would have in mind first the existence of tree as affirmed by the term ‘tree,’ and the negative particle would deny its existence (VP 2.241). It then would mean that a single sentence gives rise to two contradictory concepts. Punyarāja (1000 CE) adds in his commentary an interesting argument: if something exists, it cannot be denied, and something that does not exist does not need to be negated. Either way, the negative particle is meaningless. 12 This argument is given to negate the particularist’s position that each word in a sentence gives meaning independently. If the meaning is given by sentence holistically, on the other hand, the aforementioned consequence will not ensue.

One can argue that the positive terms in a sentence give rise to the object in the mind, and the negative particles negate only what exits in the mind and not the external reality. Bhartrhari at this point states that it is not the cognition but the real object that is denoted by the negating term (VP 2.242). Punyarāja’s commentary upon this verse is crucial:

The particle nañ negates the meaning expressed by the word. A concept is not denoted by a word. The word denotes an external object. Concept, dependent upon [external] object, cannot be referred to by a word. This being the case, how can this concept {sā} be negated by [the particle] nañ? 13

Embedded within this position is the thesis that language describes reality. The opposite position is that language only expresses our concepts and therefore cannot describe the thing-in-itself. For Bhartrhari, the word principle itself is the Brahman, the absolute. Language, in his metaphysics, has a higher status and is capable of describing the object, not just its concept. Bhartrhari is explicit in the following verse: “If the [negative particle] nañ establishes that the concept that arises [by hearing negation] is false, [in that case] the [negative term] nañ will have a separate operation [and] how can [its] absence be comprehended?” (VP 2.243) With this meaning of negation, when one says ‘not a tree’ one would be only negating the idea of a tree, not the tree as such. Punyarāja also adds in the commentary that since the objective of the verb is to simply negate existence, the negation in ‘nāsti’ is of the nonimplicative (prasajya) type. Bhartrhari rejects the argument that a negative particle does not correspond to any object or the substrate that it is negating. The argument is that negation always accompanies something that is being negated. If negation were generic, in the sense that what it negates is the substrate and not the particular object, the terms that accompany negation would lack referential meaning other than the substrate of negation (VP 2.244). Whatever its position in the sentence, on this view, the negative particle would be referring to both negation and its substrate and the words that accompany the particle would be irrelevant. Another alternative is that instead of finding an independent reference for the negative particle, it is read as coreferential. This would allow one to escape from the above dilemma and the meaning could still be broken into words. The question is, does this particle nañ refer to its meaning directly (vācaka) or is it coreferential (dyotaka)? If the particle is merely a coreferent, this would mean that both ‘tree’ and ‘negation’ would be identified by the first term, ‘vrksa,’ and the particle nañ would only be coreferring to what has been established by the first term, or this negative term would simply be dangling, having no independent meaning of its own. It would then be merely delimiting, that it is not-tree (VP 2.245). Bhartrhari thus reaches the conclusion that sentential meaning is indivisible and words do not have independent meaning.

To be Continued …

The paper appears as a chapter in the book “Nothingness in Asian Philosophy” edited by JeeLoo Liu and Douglas Berger and has been republished with author’s permission.

Featured Image: Nature

Disclaimer: The facts and opinions expressed within this article are the personal opinions of the author. IndiaFacts does not assume any responsibility or liability for the accuracy, completeness,suitability,or validity of any information in this article.