CC BY
37
THE DENIAL OF SEMANTIC RELATION IN INDIAN CLASSICAL CULTURE
E.A. Desnitskaya
Chair of Oriental Philosophy and Culturology Faculty of Philosophy Saint-Petersburg State University Mendeleevskaya Liniya 5, Saint-Petersburg, Russia, 199034
The existence of semantic relation was postulated in Katyayana's Varttika (3 B.C.E.), which served a key-stone for the further development of Indian language philosophy. However in different Sanskrit texts (Tantras, the Nirukta, works on ritual and poetics) the existence of semantic relation had been denied explicitly. This paper considers specific characteristics of extralinguistic kinds of activities that stipulated the genesis of these texts and elicits the reasons why their authors rejected to use ordinary language, for the sake of establishing new semantic relations.
Key words: Indian language philosophy, samdha-bhasha, Nirukta, Tantra, Indian poetic.
Indian classical culture was especially attentive to language. Nature of language, its interrelation with consciousness and the way language functions in everyday communication were central subjects of Indian language philosophy (Vyakarana), the basic principles of which go back to Panini's Grammar (~5 B.C.E.). Other Indian philosophical systems, e.g. Nyaya, Mimansa, poetics (Alankarashastra), also dealt with different linguistic problems. One of the most important subjects, investigated by almost every school of philosophy, was the meaning or — more generally — the nature of relation between language and the world.
The basis of Indian semantics is the first Varttika of Katyayana (~3 B.C.E.), which says:
siddhe shabdarthasambandhe...
"Established / permanent is the relation between the word (shabda) and its meaning (artha)..." [1. P. 7] (1).
The precise meaning of the Varttika was controversial already for Patanjali (2 B.C.E.), the author of the 'Great Commentary' on Panini's Grammar (the Mahab-hashya). First, because the compound 'shabdarthasambandha' can be analyzed in different ways, providing three different translations of the Varttika:
— Established / permanent are the word, its meaning and their relation;
— Established / permanent are the word and its relation with meaning;
— Established / permanent is the relation of the word and its meaning.
Notably, the third possible translation, which seems more plausible from our
point of view, was never mentioned by Patanjali, probably because the high respect to the word in Vedic ritual system made him to consider the word to be eternal. Nevertheless Katyayana could have in mind the third interpretation, because the subject of his interest was obviously pure grammar, not the nature of the word and its referent.
BecmHK Py^H, cepua &unoco$wi, 2014, № 1
Secondly the meaning of the word siddha, used in the Varttika, may have different explanations. The further part of the Varttika, which appeals to ordinary language, prompts to translate siddha as 'established'. Patanjali however considers siddha to be a synonym of the word nitya 'permanent' or 'eternal'. Later commentators usually followed him, providing in turn different interpretations of the notion of permanence. According to the first interpretation, Katyayana proclaimed semantic relation to be permanent from ontological point of view. The universal meaning of all words is Brahman, thus the relation of the word and its meaning is also considered to be permanent and eternal. According to the second interpretation, Katyayana spoke of the functional permanence of semantic relation. Unless semantic relation is permanent, there is no opportunity for language activity to occur. Thus the permanence of semantic relation is a pre-requisite for the use of language (2).
The first interpretation is close to Mimamsa, which considered the word and its relation with the meaning to be eternal [4. P. 37—47]. However Katyayana most probably held the second opinion, because the Varttika belonged to the 'technical', i.e. non-philosophical level of grammar, which was usually accepted by all schools as a kind of normative discipline. Different schools of Indian philosophy, notwithstanding all their disagreements in ontological matters, accepted inevitably the existence of relation between word and meaning; otherwise their own arguments would become senseless.
Having this in mind, it is more surprising to find the contrary opinion on the existence of semantic relation expressed explicitly. It is in Buddhist Tantra Prajnopayavi-nishcayasiddhi by Anangavajra [5], that Katyayana's Varttika is paraphrased in negative sense:
shabdarthayor-asambandhat (II. 5c)
...as there is no relation between the word and [its] meaning...
This statement seems to be self-contradictory, because if there were no semantic relation, the meaningfulness of all texts, including this very Tantra, would be called into doubt. One may assume, that Anangavajra denied the ontological status of semantic relation only, but not its very existence. However, the context, from which the quotation is extracted, does not support this assumption. In the whole the passage reads:
(3) Even Buddhas are unable to say 'This is that', because [that, what] everyone perceives in himself, is not grasped in external objecst.
(4) Therefore [those], awakened by three paths, bringing joy in the world, have never expounded [their knowledge] consistently in sutras, mantras, instructions etc.
(5) As there is no relation between the word and [its] meaning, it can never be learnt by hearing of the shastras, which explain [by means of] definitions (lakshana).
(6) Thus one should constantly and devotedly worship the teacher, because without him the realness (tattva) is not to be attained even in the course of ten million kalpas [5. P. 12].
In other words, artha, i.e. the intentional object of the discourse, is inexpressible and as such is to be grasped under the direct guidance of the teacher. Thus it is not the semantic relation, that Anangavajra denies, but the very possibility to express some
nonverbal experience in words. Similar statements are typical for many spiritual traditions in different cultures. However it was in Indian culture with its tendency to reflection on language, that the denial of semantic relation resulted in specific forms of sophisticated textual discourse. The most remarkable example of it is the Tantric 'twilight language'.
The Sanskrit term for 'twilight language' is samdhya-bhasha, in texts also referred to as samdha-bhasha, the word samdha being a corrupted form for samdhya 'twilight'. Samdha-bhasha can be also translated as 'intentional language'. This language is widely used in Buddhist and Hindu Tantras and related texts. As an example let us consider a list from the Hevajra Tantra:
madhya (wine) is madana (intoxication)
mamsa (flesh) is bala (strength)
malayaja (sandlewood) is milana (meeting)... (3)
One could suggest, we have here a kind of glossary, where words of the 'secret' language are elucidated by ordinary ones. However, Elder claims, that all the elements of this chain belong to 'twilight language', they are interchangeable and refer to one the same object, viz. some details of Tantric Yoga [6. P. 242]. If one uses 'twilight language', it means, 'that for him language has ceased to function in an ordinary way... because meanings overlap or intersect, like day and night at the time of twilight' [6. P. 250].
Twilight language is opposed to nontwilight (nasamdhibhasha), i.e. to ordinary language with all its semantic relations established. Similarly two other oppositions were elaborated in Buddhist Tantras, i. e. the opposition of hinted meaning (neyartha) vs. evident meaning (nitartha) and the opposition of standard terminology (yatharuta) vs. non-standard (coined) terminology (nayatharuta) [6. P. 236; 7]. The first opposition concerns the modes of verbal expression (shabda), the second the modes of the meaning (artha), whereas the opposition of terminology deals both with word and meaning in their interrelation (sambandha), which is an obvious throw-back to Patanjali's triad of word-meaning-relation.
'Twilight' semantic relation between word and meaning is established on the basis of samaya. The primary meaning of this word is 'coming together', but in some contexts it means 'vow' and 'opportunity' [8. P. 1630], the two latter meanings being not so far from each other as it seems, because a vow is to be made and solved in order to afford a favorable opportunity for the accomplishment of certain goals. In grammar and rhetoric samaya usually means 'conventional meaning of a word', being a synonym for samketa [9]. Thus we may conclude, that a vow, made by initiated, consists inter alia in acceptance of semantic convention concerning the use of 'twilight language', which affords a favorable opportunity for Tantric realization.
This use of language is not a unique trait of Tantras. Many other Sanskrit texts belonging to different fields — from ritual to poetics — are based on denial or violation of semantic relation. One of the earliest examples is to be found in the beginning passage of the Brihadaranyaka Upanishad, which concerns the Ashvamedha ritual:
"The head of the sacrificial horse, clearly, is the dawn; its sight is the sun; its breath is the wind; and its gaping mouth is the fire common to all men. The body (atman) of the
BecrnHK Py^H, cepua &unoco$wi, 2014, № 1
sacrificial horse is the year; its back is the sky; its abdomen is the intermediate region..." (4).
Sementsov claimed, that this passage was stipulated by its ritual function, being a part of the Ashvamedha, intended for ritual performance. The passage was not an exposition of some philosophy, but an instruction in a specific form of psychotechnique, which makes the mundane world sacral, so that for the participants of the Ashvamedha ritual the Universe becomes a gigantic sacrifice [11. P. 48—50] (5).
As a result, in the passages of this kind the relation between a word and its meaning is no more obvious and needs to be justified. This is especially relevant with regard to the words which traditional grammar considers as avyutpanna, i.e. 'nonderiv-able' according to the rules. This problem could be solved by means of Nirukta, one of the six Vedic auxiliary disciplines (Vedangas) (6).
Nirukta is often characterized as etymology, which is not true, because traditional Indian culture had no interest to history in general and to history of words in particular. According to traditional explanation, Nirukta provides derivation of a word in correspondence with its meaning [14. P. 27]. The basic principle of Nirukta — anvartha, lit. 'similarity with the object (artha)' — can be compared with European concept of iconici-ty, i.e. structural correspondence between the word and the essence of the referent. Nirukta's procedure (nirvacana) provides an answer to the question of how does a word denote, what it denotes, not in general, but in the certain context. Thus it seems more appropriate to follow Bronkhorst and Kahrs in defining Nirukta as a technique of semantic analysis [15; 16; 14].
An example of such analysis is derivation of the word ahar 'day':
'Why ahar? [because] one carries out (upa-ahar-anti) actions in the course of it» (7).
This 'derivation' contradicts the rules of Sanskrit grammar, and is obviously absurd for everyone, who knows Sanskrit. Another example of nirvacana is the word anna 'food', derived from two verbal roots simultaneously, though only one of the derivations seems plausible [Bronkhorst 1981: 7]. Derivations of this kind are against the innate linguistic intuition, however they may help to understand the meaning of the word in some context.
European scholars usually characterize Nirukta as a folk or popular etymology that is inconsistent with modern linguistics (8). In Ancient India however there was no concept of history in European sense of the word. Indian traditional culture regarded time as cyclical, not linear. Thus one can hardly expect, that in Vedic ritual system there will emerge a specific Vedanga, concentrated on historical derivation of words. According to Bronkhorst, most probably Yaska had never considered the hypothesis of language being subject to alteration. For him Vedic language and Sanskrit did not represent different stages of language history, but were two coexistent aspects of one language [15. P. 5—7].
By means of nirvacana a new semantic relation is established, transforming an incomprehensible text into a distinct reality and at the same time modifying the way one perceives this reality. In this regard nirvacana is very similar to samaya, discussed
earlier. The problem of the affinity between Vedic culture and Tantra is too complicated to be discussed here, though certain relations as well genetic, as typological are very plausible. Anyway the denial of ordinary semantic relations and re-establishment of the new ones are common both to Vedic and Tantric ritual traditions.
The same tendency can be discovered in another cultural realm, viz. in poetry. In poetry the established semantic relation of the word to its meaning is sometimes intentionally violated, in order to make the word designate some other referent, different to its usual one, but somehow connected to it. In Indian poetics usual examples of this metaphoric transfer (lakshana) were sentences: 'The cots cry' (though in fact the babies in cots cry) and 'The cows graze on the Ganges' (on the bank of the Ganges) [17. P. 232—273].
Indian classical poetry was famous with its sophisticated means of expression, so that even educated people were unable to understand for instance Kalidasa's poem, without having recourse to commentaries. Thus we may say that even in poetry there was a kind of samaya, based on allusion and association. Anandavardhana claimed in the Dhvanyaloka, that suggested meaning is not to be grasped by means of grammar, only those understand it, who know the essence of poetry (9).
Poetical means affect the audience aesthetically, changing the mode of apprehension. Thus Indian rhetoricians had good reasons for comparing aesthetic and spiritual experience. Abhinavagupta proclaimed aesthetic experience (rasa) to be of the same nature as the direct non-duel experience of the Ultimate Reality [Cf. 19, 20].
Thus we may conclude that established semantic relation is a necessary condition for philosophy, grammar as well as for various kinds of everyday activities, based on the use of ordinary language. On the contrary some kinds of activity, i.e. ritual, religion and poetry, which are aimed to change the usual world-view, handle with semantic reality skillfully. The subject of a text of this tape is not to be grasped directly by means of established semantic relation; on the contrary it can be comprehended only after this relation has been destroyed.
FOOTNOTES
(1) According to the full text of the Varttika, introduced in further parts of the Mahabhashya, after the relation between the word (shabda) and its meaning (artha) is established from ordinary language, grammatical restrictions are applied, as in mundane and Vedic activities. [Cf. 1. P. 7] But commentators usually analyze the first part of the Varttika separately as a distinct statement.
(2) Bhartrhari in his commentary on Patanjali's Mahabhashya (the Mahabhashyadipika was probably the first to introduce the idea, that the permanence of semantic relation can be the two types of [2. P. 74, 207—208]. Later on this concept was mentioned by Kayyata (XI A.D.) in his commentary on the Paspashahnika of the Mahabhashya) [3. P. 91, 110—114].
(3) Translated by Wayman. Cited by Elder [6. P. 241].
(4) Translated by P. Olivelle [10. P.37].
(5) Probably this textual practice originates from the ancient Indo-European poetic opposition of 'language of people' and 'language of gods' (or even languages of asuras and gandharvas in India, dwarfs, jotunns, vanir etc. in Norse mythology), each of them possessing special synonyms
for designation of same realia [12. P. 43]. For example in many Vedic hymns the word 'cow' (gau) is the secret name of Speech [13. P. 23—37].
(6) Vedangas were six in number: phonetics (Shiksha), metrics (Chandas), grammar (Vyakarana), semantic analysis (Nirukta), astronomy (Jyotisha), ritual (Kalpa). The first two provide correct pronunciation of the Vedas, next two correct understanding of the Vedas and two latter correct performance of the ritual. Thus the pairs of the Vedangas regulate respectively the activity of speech, mind and body in the course of ritual.
(7) Nirukta II. 20. Kahrs' translation slightly modified [14. P. 54].
(8) Cf. [21. P. 68—73]. A review of interpretations of Nirukta, dominated in European Indology since 19th century, see [15. P. 1—3].
(9) Dhvanyaloka I. 7: [18. P. 5].
BIBLIOGRAPHY
[1] Patanjali's Vyakarana-Mahabhashya. Edited by F. Kielhorn. Vol. I. Bombay, 1880.
[2] Iyer K.A.S. Bhartrihari. A study of Vakyapadiya in the Light of the Ancient Commentaries. Poona, 1992.
[3] Patanjali's Vyakarana-Mahabhashya. Paspashahnika. Introduction. Text. Translation and Notes by S.D. Joshi and J.A.F. Roodbergen. Pune, 1986.
[4] Devasthali G. V. Mimamsa — the Vakya-shastra of Ancient India. Bombay, 1959.
[5] Prajnopayavinishcayasiddhi of Anangavajra // Two Vajrayana Works. Gaekwad's Oriental Series, NXLIV. Baroda, 1929.
[6] Elder G.R. Problems of Language in Buddhist Tantra // History of Religions. Vol. 15/3, 1976. P. 231—250.
[7] Wayman A. Buddhist Tantra and Lexical Meaning // Current Advances in Semantic Theory. Amsterdam — Philadelphia. 1992. P. 465—478.
[8] Apte V.Sh. The Practical Sanskrit-English Dictionary. Delhi, 1998.
[9] Houben J.E.M. Bhartrihari's Samaya / Helaraja's Samketa. A Contribution to the Reconstruction of the Grammarian's Discussion with the Vaisheshikas on the Relation between Shabda and Artha // Journal of Indian Philosophy. 20. 1992. P. 219—242.
[10] The Early Upanishads. Annotated Text and Translation by Patrick Olivelle. Oxford, 1998.
[11] Семенцов В.С. Проблемы интерпретации брахманической прозы. М., 1981.
[12] Елизаренкова Т.Я., Топоров В.Н. Древнеиндийская поэтика и ее индоевропейские истоки // Литература и культура Древней и Средневековой Индии. М., 1979. С. 36—88.
[13] Гринцер П.А. Тайный язык «Ригведы» // Избранные произведения. Том I. Древнеиндийская литература. М., 2008. С. 9—74.
[14] Kahrs E. Indian Semantic Analysis: the Nirvacana Tradition. Cambridge. 1998.
[15] Bronkhorst J. Nirukta and Ashtadhyayi: their shared presuppositions // Indo-Iranian Journal. 23. 1981. P. 1—14.
[16] Bronkhorst J. Etymology and magic: Yaska's Nirukta, Plato's Cratylus, and the riddle of semantic etymologies // Numen, Vol. 48. Leiden, 2001.
[17] Raja K.K. Indian theories of meaning. Madras, 1963.
[18] Anandavardhana. Dhvanyaloka. http://sanskritdocuments.org/all_pdf/dhvanyaloka1.pdf
[19] Анандавардхана. Дхваньялока («Свет дхвани») / Пер. с санскрита, введение и комментарий Ю.М. Алихановой. М., 1974.
[20] Gnoli R. The Aesthetic Experience According to Abhinavagupta. Varanasi, 1968.
[21 ] Катенина Т.Е., Рудой В.И. Лингвистические знания в Древней Индии // История лингвистических учений. Древний мир. Л., 1980.
ОТРИЦАНИЕ СЕМАНТИЧЕСКОЙ СВЯЗИ В ИНДИЙСКОЙ КЛАССИЧЕСКОЙ КУЛЬТУРЕ
Е.А. Десницкая
Кафедра философии и культурологии Востока Философский факультет Санкт-Петербургский государственный Университет Менделеевская линия, 5, Санкт-Петербург, Россия, 199034
Существование семантической связи постулируется в варттике Катьяяны (III в. до н.э.), послужившей отправной точкой для дальнейшего развития индийской лингвофилософии. Тем удивительнее встретить отрицание существования семантической связи, эксплицитно выраженное в санскритских текстах различной тематики (тантры, «Нирукта», труды по ритуалу и поэтике). В статье рассматриваются особенности видов внеязыковой деятельности, лежащие в основе подобных текстов, и выявляются причины, побудившие авторов отказаться от использования обыденного языка и устанавливать новые семантические связи.
Ключевые слова: индийская философия языка, самдхья-бхаша, нирукта, тантра, индийская поэтика.
