Philosophy of Language in the Vaiyakarana Tradition

September 29, 2021 Authored by: Kapil Kapoor

The Indian mind has been incessantly reflecting on the nature and structure of language, its relationship to the world it talks about, and its organization as a system. The deliberations of the Sanskrit grammarians (vaiyakaranas) began with phonetics and ended in philosophy. It has been said that the grammarians went to the seashore in search of cowrie-shells (the sound of language) and ended up finding a pearl (in the form of Sabda-Brahman). 

The Indian Conception of Language

That language is central to India’s intellectual history is established by the attested existence of a long tradition of thinking. As knowledge is the supreme path to what in the later Indian world view got defined as the ultimate end or goal of human life, Moksha – liberation from suffering here and now, and as knowledge is inseparable from language, language has understandably been the central object of inquiry, and has come in for sustained and intense investigation in all Indian schools of thought. It has been studied in its two aspects – the ontological, svarupa, and the epistemological, samarthya, both us the object of knowledge and as means of knowledge. In a remarkable analogy, Bhartrhari compares Sabda, word, with a dipaka, lamp (1.44) – when it is lighted, it reveals itself and also reveals other associated objects (meanings) – it is the object to be grasped (grahya) and the means of grasping objects (grahaka). 

The Indian conception of language differs in three ways from the Western: 

  • Language is speech, not writing (script);
  • Language is a cognitive system (not, primarily, a means of communication); and 
  • Language is a constructivist system (not a representational one)

All the three Sanskrit words for language bhasavak and vani – denote the ‘sound-ness’ of language. Bhasa is from the verb-root bhas, which means ‘articulated sound’; vak is from the verb root vac, which means ‘to say something’; vani is from the verb root van, which means ‘to say something sweetly’. The most significant effect of this notion of language was the rise of phonetics as the first science in India and the sophisticated phonetic analyses achieved in the tradition. Panini’s grammar is also founded on these assumptions. It is a grammar of bhasa, spoken language, of the acceptable forms of speech. Even while deriving vaidiki, compositional language, its goal is to establish its spoken forms. For this reason, sandhi, the euphonic combination of sounds seen in speech contexts, has an important place in his grammar. No other grammar deals with sandhi in this exhaustive a manner or with such systematic rigor.

Further, as speech rests in a speaking human voice, and because the source of an utterance is always identifiable as an individual consciousness, no truth-claim is asserted about what is said. The speaking voice is an individual voice and not the Voice. In Hindu thought, unlike Hebraic, there is no one God and there is no one Voice. This enables a multiplicity of viewpoints, drsti bheda: ‘ekatvinam dvaitinam ca Pravada bahudha matah; the assumptions of monists and dualists have given rise to many viewpoints’ (1.8).

This fact is linked to the second postulate: language is a cognitive system and not just a system of communication, as modern linguistics – barring Chomsky – will have us believe. As explicated by Bhartrhari, language is the form that knowledge takes; and therefore language is indistinguishable from intelligence (samjna) and consciousness (cetana) (1.126). what Vyakarana or grammar studies and describes is Sphota-Sabda, the ‘language in the mind’, the system that is shared by all speakers of that particular language. Therefore, when Bhartrhari isolates the properties of language, such as its non-partitiveness (a-vibhaga) and non-sequentially (a-krama) or the processes of conceptualization and expression (1.46), he is making a statement about the nature and process of cognition (in the human mind). We are informed by him that while the phoneticists consider Sabda (language) as sound (dhavani) and the Jains consider language as Pudgala (atomic), the grammarians hold that language is the vivarta (apparent transformation) of knowledge (Jnana) subsisting in the inner knowing self (antaro jnata) (1.107-112). Conversely, Sabda, when grasped, results in knowledge subsisting in the inner knowing self, this time of the hearer. 

Third, language is a constructivist system. As all cognition (Bodha) takes the form of language, the reality that is cognized by us is necessarily a linguistic construct. Language is not a system that ‘names’ some pre-existing reality, as was the claim of the representational view of language before Saussure and Derrida ‘happened’ under the influence of Sanskrit language. The constructivist view asserts that language constructs the reality that we claim to be ‘out there’. The grammarians say that it is through naming that objects, both outside the mind and inside, are cognized as separate or different from each other, creating for us iti amnaya, ‘this enumerable universe’ (1.120).

Bhartrhari: The Grand Synthesist 

These assumptions about the nature of language inspired a long line of thinking about the relationship between language, thought, and reality, and culminated in Bhartrhari’s grand synthesis of the tradition in his Vakyapadiya.

The three divisions of language – substance, form, and the potential of words to denote/connote – are respectively studied as phonetics, grammar, and the philosophy of word-meaning, under the two aspects of object (grahya) and means (grahaka). It is possible to see, in the long tradition of linguistic reflection, a growing understanding of the nature of meaning from the Vaidika concept of Sabda as a potent sound with power to create or destroy, to the etymologists’ notion of Sabda as something that has a meaning grounded within a particular discourse, to the Buddhist requirement of locating meaning outside the text, in the consciousness of the predicator, to the Mimamsaka argument – in consonance with their liturgical purpose – that meaning is the action or performance enjoined, to the Naiyayika position that words refer to objects, Vastu or artha, to, finally, the grand synthesis articulated by Bhartrhari – meaning is something constituted in the mind and then related, variously, to an outside object in Ukti, application. A cline of references is set up by Bhartrhari: Sphotartha (conceptual meaning) – Sabdartha (verbal meaning) – Vacyartha (expressed meaning) – bauddhartha ([mentally] grasped meaning) Vastvartha (referential meaning). This is how meaning, as determined by a number of conditions and factors of use, is constructed in usage, Ukti.

We may restate here the growing understanding in this debate. The Rig Vedic etymological (e.g. usas-usa) illustrate the etymological view of word-meaning. For the phoneticists, with their assumption of sound being ontologically cosmic, Sabda as a sound-aggregate has the cosmic power to create and destroy. On the other hand, Kautsa, a materialist sceptic, argued that Vedic mantras are meaningless – they are marked by tautology, contradiction, opacity, and impossibility, among other things. The etymologist Yaska countered the two challenges: (i) That of the phoneticists, who said that mere enunciation of hymns is efficacious, by upholding the primacy of knowledge and dismissing their disregard of meaning (1.18), and (ii) of the materialist sceptics, by rejecting their position and arguing in substance that no utterance is meaningless, that there are degrees of opacity, identified by him as various textual indices, and that these can be handled by a system of determination of meaning (1.16-17, 2.1), a system that finally matured as the Mimamsa system. The Mimamsa commentator Shabaraswami continued this tradition by taking the position that words (or utterances) in a text have exactly the same meaning that they have in actual usage and therefore their meaning is to be fixed by reference to usage. 

The etymologists’ assumption of a determinable meaning is questionable by the Buddhists. From the perspective of their foundational principles of flux and momentariness, words are merely names for momentary mental impressions of momentary objects. They question Yaska’s concept of textual meaning and ask: what relationship actually obtains between the word and the world? They say that words are names for constantly changing objects. This is most evident in the use of propre names – we continue to call a person by the same word/name from childhood to old age, though he or she undergoes visible and substantial changes over time. Every individual, including the speaker and the hearer, thus constructs objects whose reference can be grasped only by exclusion, by isolating them from other objects, by negation. 

This is a method already in use in the Upanishads. The method of ‘neti-neti’, not this, not this’ has been used, for example, in an effort to define Brahman in the Kena Upanishad

The Mimamsakas are aware that Vedic mantras are differently, even contradictorily, interpreted and that this is the ground for the Buddhist correlation between the flux in meaning of Vedic language and the impossibility of categorical knowledge (of reality). Therefore, they characterize the language of Vedic hymns as metaphorical, rupaka. How do we then determine meaning? Mimamsa suggests principles such as coherence, sequence, type-token relationship between the expresser and the expressed, and conformity to usage. Meaning is in the act, and therefore the focus is on injunctive sentences (unlike the Vedanta focus on declarative sentences). 

The Mimamsakas set up the principles of ‘meta-rule’, ekavakyata (syntactic unity), and ekarthibhava (semantic unity) for interpretation of utterances. Patanjali (c. 2nd cent. BCE), indirectly characterizes meaning as a mental construct by defining Sabda as that which when articulated brings to the hearer’s mind the cognition of an object. 

Bhartrhari is the great synthesizer in this tradition that includes the Rig Veda as well as Patanjali’s Mahabhashya. He understands meaning as an abstract mental entity, Sphota, somewhat like the mental construct of the Buddhists, which gets related to the outer object when its analogous word is used to denote the object. Because of its nature as manasa Vyapara (mental activity), meaning is highly variable, depending as it does on individual cognition. 

The mapping of the denotation is meditated by the speaker’s Vivaksa (intention). For the hearer also, meaning is something figuring in the mind. For both the speaker and the hearer, the cognition constitution of meaning rests on given conditions and factors (see note 6).

Language, Meaning and Reality

The relationship between language and reality is the core issue in the theory of meaning. In the celebrated Mahayana text Lankavatara-sutra, ‘the Bodhisattva Mahasattva said this to the Blessed One: “Pray tell me… the signification of two things, expression and expressed… for the purification of all beings.”

One answer to this question comes from Sanskrit literature. Kalidasa, in the very first sloka of his epic Raghuvamsa, says: 

Vagarthaviva samprktau vagartha-pratipattaye;

Jagatah pitarau vande Parvati-paramesvarau.                

For the right understanding of words and their meanings, I bow down to Parvati and Parameshwara, the parents of the universe, the perpetual relation between whom is as intimate as that between words and their meanings. 

It is the relationship of nama (name, conception) and rupa (form, physical object) that is nitya, perpetual. We go to the Brihadaranyaka Upanishad for the categories of nama and rupa: ‘This universe, before coming into being, was Unmanifest (avyakta). It became manifest (vyakta) through the conjunction of nama (construct) and rupa (form). 

Therefore, even at this time this Unmanifest object becomes manifest (gets expressed) as “has this nama and this rupa”.

The Buddhist Vaibhasika system also characterizes objects as having the dimensions of rupa and namaRupa is defined as that through which objects are figured out. It is further derivately defined as that which takes a form, i.e. objects or entities that occupy physical space or space in the mind.

It is understood that the empirical world is constituted of nama, name, and rupa, form/entity. While ‘form’ denotes ‘matter’ or ‘determinate entity, ‘name’ refers to the configuration in the mind. Buddhism identifies these two dimensions (rupa and nama) of all objects that come into being as pertaining to matter (bhautika) and awareness (cetana) respectively. Rupa being bhautika belongs to the domain of senses (indriya) and the objects of senses (visaya). Nama being caitasika belongs to the domain of feeling/experience (vedana), intelligence (Samjna), ingrained impressions (samskara), and reasoning (vijnama). 

Sabda (language/word) encompass nama. Whether it incorporates rupa also and, if yes, to what extent, is a subject of inquiry. The relationship of these two dimensions to the totality or complete knowledge of the object is also a major subject of inquiry in the problem of relationship between language and reality. 

The status of the phenomenal world in relation to language is an important problem in the theory of meaning. As the Lankavatara-sutra says, the Brahmanas and Shramanas predicate ‘existence’ on the basis of linguistic usage, while in fact existence and non-existence are both constructs of the mind; 

In all things there is no self-nature, words too are devoid of reality… the ignorant and the simple-minded [are] not knowing that the world is what is seen of Mind itself… Blessed One, the philosophers explain birth from being and non-being, while, according to the Blessed One, all things coming into existence from nothingness pass away by causation… As space, the hare’s horns, and a barren woman’s child are non-entities except as expressed in words, so is this existence imagined. 

For this reason, Mahamati, the error-existence (or this world of illusion) is not, but as this water is manifest to other people it is not a non-existence either. When the world is understood to be nothing but Mind itself, the existence and non-existence of external objects ceases to be discriminated. Thus, Mahamati, this error being discriminated by the wise turns into tathata(suchness)… Maya is something imagined and clung to as having multitudinousness of individual forms. 

When the state of imagelessness is recognized, objects… cease to assert themselves. Citta is bound up with the objective world; the intellect’s function is to speculate; and in the excellent state of imagelessness there is the evolving of transcendental wisdom (prajna)… Where there is false imagination multitudinouness of things is recognized… As far as the duality of being and non-being extends, there is the realm of intellection; when this realm vanishes, intellection completely ceases. When the external world is not grasped (as real) there is neither causation nor reality; there is the essence of suchness (tathata), which is the (spiritual) realm of the wise. 

The continuity of thought from this Buddhist construction to the Yoga theory of cognition to Bhartrhari’s integral theory is self-evident. Bhartrhari achieves a characteristically Indian synthesis and says that the meaning of a word in relation to the phenomenal world is determinable only in its application (Ukti). When a word is used, its conceptual world constructs the external object through imposition (aropa), and in this sense all this world (Jagat) is a linguistic construct (Vikalpa). 

In the tradition of the grammarians, the word is something ‘which when articulated brings to the hearer’s mind the cognition of an object.’ So all meaning rests in the mind – is buddhi visaya. The reality we talk about is in fact constructed by us (and by the hearer) in the sense that in verbal usage, both designation and reception are mental constructs. Bhartrhari says, the word is imposed on the object: ‘that is this’ = ‘that is a cow’. It is not the object that identifies the word; the object is in fact defined/recognized in terms of the word.

The word, cow, for example, has an associated image in the speaker’s mind, which is mapped on to the object when the word is applied to the object. And the hearer also has a mental image of ‘cow’, which is matched with the object. This becomes clear when we consider the use of a familiar word for an absent object or for an unfamiliar object; in either case the object is evidently constructed in terms of a mental image. Bhartrhari has this process in mind when he says: ‘The meaning/object is something mental (Budhi visaya), but it rests on an external object and is looked upon as meaning when it is externalized (2.132).

In one of its hymnal reflections, the Rig Veda codes an ancient conception: ‘Language cuts forms in the ocean of reality.’ Reality, like water in the ocean, is sequence less, non-discrete, and continuous, and the difference (bheda) that we recognize in the phenomenal world, the multiple varied external reality that we cognize, is a construct sabdasya parinama.

This extraordinary multiplicity, our knowledge of this phenomenal world, is of the nature of vikalpa (1.119), in the precise sense that this term is used in the Yoga theory of knowledge, there being virtually no cognitive distinction between non-existent entities such as ‘rabbit-horns’ and perceptible entities such as a jar or a piece of cloth. Language designates these objects not as they are in their totality but as they are constituted at the level of appearance (pratibhasika), which is the level of bheda or differences. Complete understanding of the object may never be reached and may not, in fact be expressible in words. Just as we proceed in our understanding of reality from the Pratibhasika (apparent) through the vyavaharika (empirical) to a true understanding of the paramarthika (ultimate) reality, in the same way, in the grasp of meaning, we progress from the denotation of the particular, via the universal qualified by the particular, to the denotation of one universal existence, satta (3.1.19, 29, 33).

In Indian philosophy, the total reality of an object is a complex matrix of no less than eleven properties which can be affirmed positively or negatively about the object in question. Complete knowledge of the object, purna jnana, involves knowledge of how each of these categories is present in the given object (say, a jar, ghata) and in what manner or degree. (To be concluded)

[Source: Kapil Kapoor, “Philosophy of Language in the Vaiyakarana Tradition“, in Prabuddha Bharata, Vol. 112, No.8, August 2007, p. 463-468]

Centre For Indic studies