The Encyclopedia of Indian Philosophies v5: Volume 5
The Encyclopedia of Indian Philosophies v5
- Language
- English
Contents
- v5.c24 Front matter — 135 parts
- v5.c25 Bibliography on Grammar (Vyākaraṇa) — 5 parts
- v5.c26 Bibliography — 3 parts
- v5.c27 1: Historical Resume — 2 parts
- v5.c28 2: Metaphysics — 47 parts
- v5.c29 3. Epistemology — 17 parts
- v5.c30 4: Word Meaning — 3 parts
- v5.c31 5. Sentence Meaning — 1 part
- v5.c32 Notes to Survey — 75 parts
- v5.c33 Cumulative Index — 79 parts
Volume 5
Front matter
v5.c24.1
The first of these, the *Gospel of the Hebrews*, was used by the Jewish Christians of Palestine, and was probably a version of the Gospel of Matthew, though it contained some traditions not found in our canonical books. The *Gospel of the Egyptians* was used by the ascetic sects of Egypt, and was marked by a strong tendency towards mysticism and asceticism. The *Gospel of Peter* was used by the Docetists, and taught that Christ did not really suffer, but only appeared to do so.
v5.c24.2
These gospels, and others like them, were not accepted by the Church as a whole, and were gradually lost or suppressed. The Church, however, felt the need for a standard collection of writings that would serve as a guide for faith and practice. This led to the formation of the New Testament canon, a process that took several centuries to complete.
v5.c24.3
The first step in this process was the collection of the letters of Paul, which were widely circulated among the churches. Then came the four Gospels, which were accepted as the authoritative accounts of the life and teachings of Jesus. The Acts of the Apostles and the other epistles were gradually added, and finally the Book of Revelation.
v5.c24.4
By the end of the second century, the core of the New Testament was generally accepted by the Church. The process of canonization was not a formal act of a council, but rather a gradual recognition by the churches of the writings that were most valuable for their life and worship. It was only in the fourth century that the canon was formally closed by the councils of the Church.
v5.c24.5
The formation of the New Testament canon was a crucial event in the history of the Church. It provided a standard of faith and practice that helped to unify the Church and to protect it from the influence of heretical sects. It also ensured that the teachings of Jesus and the apostles would be preserved for future generations.
v5.c24.6
The first of these, the *Gospel of the Hebrews*, was used by the Jewish Christians of Palestine, and was probably a version of the Gospel of Matthew, though it contained some traditions not found in our canonical books. The *Gospel of the Egyptians* was used by the ascetic sects of Egypt, and was marked by a strong tendency towards mysticism and asceticism. The *Gospel of Peter* was used by the Docetists, and taught that Christ did not really suffer, but only appeared to do so.
v5.c24.7
These gospels, and others like them, were not accepted by the Church as a whole, and were gradually lost or suppressed. The Church, however, felt the need for a standard collection of writings that would serve as a guide for faith and practice. This led to the formation of the New Testament canon, a process that took several centuries to complete.
v5.c24.8
The first step in this process was the collection of the letters of Paul, which were widely circulated among the churches. Then came the four Gospels, which were accepted as the authoritative accounts of the life and teachings of Jesus. The Acts of the Apostles and the other epistles were gradually added, and finally the Book of Revelation.
v5.c24.9
By the end of the second century, the core of the New Testament was generally accepted by the Church. The process of canonization was not a formal act of a council, but rather a gradual recognition by the churches of the writings that were most valuable for their life and worship. It was only in the fourth century that the canon was formally closed by the councils of the Church.
v5.c24.10
The formation of the New Testament canon was a crucial event in the history of the Church. It provided a standard of faith and practice that helped to unify the Church and to protect it from the influence of heretical sects. It also ensured that the teachings of Jesus and the apostles would be preserved for future generations.
v5.c24.11
The Philosophy of the Grammarians
v5.c24.12
HAROLD G. COWARD and K. KUNJUNNI RAJA
v5.c24.13
PRINCETON UNIVERSITY PRESS Princeton, New Jersey Copyright © 1990 by Princeton University Press Published by Princeton University Press, 41 William Street, Princeton, New Jersey 08540 In the United Kingdom: Princeton University Press, Oxford All Rights Reserved
v5.c24.14
Library of Congress Cataloging-in-Publication Data The Philosophy of the Grammarians / edited by Harold G. Coward and K. Kunjunni Raja. p. cm.—(Encyclopedia of Indian philosophies ; v. 5) Includes bibliographical references and index. ISBN 0-691-07331-7 (alk. paper)
1. Bhartṛhari. Vākyapadīya. 2. Pāṇini. Aṣṭadhyāyī.
3. Patañjali. Mahābhāṣya. 4. Sanskrit language—Philosophy.
5. Sanskrit language—Grammar—History. 6. Indo-Aryan languages—Grammar—History. 7. Sanskritists—India. I. Coward, Harold G. II. Kunjunni Raja, K. III. Series: Encyclopedia of Indian philosophies (Princeton, N.J.) ; vol. 5. PK475.P48 1990 181′.4—dc20 90-47434 CIP This book was composed in India by Jainendra Prakash Jain at Shri Jainendra Press, New Delhi Princeton University Press books are printed on acid-free paper, and meet the guidelines for permanence and durability of the Committee on Production Guidelines for Book Longevity of the Council on Library Resources Printed in the United States of America by Princeton University Press, Princeton, New Jersey 10 9 8 7 6 5 4 3 2 1
v5.c24.15
Contributors :
v5.c24.16
Ashok Aklujkar, University of British Columbia John G. Arapura, McMaster University S. R. Bannerjee, S. D. Joshi, University of Poona Shoryu Katsura, Hiroshima University G. B. Palsule, University of Poona Karl H. Potter, University of Washington V. K. S. N. Raghavan, University of Madras K. A. Subramania Iyer, Lucknow University The first of these, the *Gospel of the Hebrews*, was used by the Jewish Christians of Palestine, and was probably a version of the Gospel of Matthew, though it contained some traditions not found in our canonical books. The *Gospel of the Egyptians* was used by the ascetic sects of Egypt, and was marked by a strong tendency towards mysticism and asceticism. The *Gospel of Peter* was used by the Docetists, and taught that Christ did not really suffer, but only appeared to do so.
v5.c24.17
These gospels, and others like them, were not accepted by the Church as a whole, and were gradually lost or suppressed. The Church, however, felt the need for a standard collection of writings that would serve as a guide for faith and practice. This led to the formation of the New Testament canon, a process that took several centuries to complete.
v5.c24.18
The first step in this process was the collection of the letters of Paul, which were widely circulated among the churches. Then came the four Gospels, which were accepted as the authoritative accounts of the life and teachings of Jesus. The Acts of the Apostles and the other epistles were gradually added, and finally the Book of Revelation.
v5.c24.19
By the end of the second century, the core of the New Testament was generally accepted by the Church. The process of canonization was not a formal act of a council, but rather a gradual recognition by the churches of the writings that were most valuable for their life and worship. It was only in the fourth century that the canon was formally closed by the councils of the Church.
v5.c24.20
The formation of the New Testament canon was a crucial event in the history of the Church. It provided a standard of faith and practice that helped to unify the Church and to protect it from the influence of heretical sects. It also ensured that the teachings of Jesus and the apostles would be preserved for future generations.
v5.c24.21
PART ONE : INTRODUCTION TO THE PHILOSOPHY OF THE GRAMMARIANS
v5.c24.22
*(Harold G. Coward and K. Kunjunni Raja)*
1. Historical Résumé 3
2. Metaphysics 33
3. Epistemology 51
4. Word Meaning 63
5. Sentence Meaning 83
v5.c24.23
PART TWO : SURVEY OF THE LITERATURE OF GRAMMARIAN PHILOSOPHY
v5.c24.24
1. Philosophical Elements in Vedic Literature 101 *(John G. Arapura and K. Kunjunni Raja)*
2. Philosophical Elements in Yāska’s Nirukta 107 *(K. Kunjunni Raja)*
3. Philosophical Elements in Pāṇini’s Aṣṭādhyāyī 111 *(K. Kunjunni Raja)*
4. Philosophical Elements in Patañjali’s Mahābhāṣya 115 *(K. Kunjunni Raja)*
5. Bhartṛhari 121 *Trikāṇḍī or Vākyapadīya, with Vṛtti on Books 1 and 2* *(Ashok Aklujkar)* *Ṭīkā on Patañjali’s Mahābhāṣya (K. Kunjunni Raja)*
6. Durvinīta or Avinīta 175
7. Dharmapāla 177
8. Hari Vṛṣabha or Vṛṣabhadeva 179 *(Ashok Aklujkar)*
9. Maṇḍana Miśra 181 *Sphoṭasiddhi (G. B. Palsule, Harold G. Coward, and Karl H. Potter)*
v5.c24.25
*Commentary on Bhartṛhari’s Trikāṇḍī* *(K. Kunjunni Raja)*
11. Prameyasaṃgraha 199
12. Puṇyarāja 201
13. Kaiyaṭa 203 *Pradīpa on Patañjali’s Mahābhāṣya (S. R. Bannerjee* *and K. Kunjunni Raja)*
14. Jyeṣṭhakalāśa 205
15. Maitreya Rakṣita 207
16. Puruṣottamadeva 209
17. Dhaneśvara 211
18. (Ṛṣiputra) Parameśvara II 213 *Sphuṭasiddhigopālikā (K. A. Subramania Iyer)*
19. Śeṣa Kṛṣṇa 215 *Sphuṭatattvanirūpaṇa (G. B. Palsule)*
20. Satyānanda or Rāmacandra Sarasvatī 219
21. Śeṣa Cintāmaṇi 221
22. Śeṣa Vīreśvara or Rāmeśvara 223
23. Śeṣa Nārāyaṇa Bhaṭṭa 225
24. Viṣṇumitra 227
25. Īśvarānanda or Īśvarīdatta Sarasvatī 229
26. Bharata Miśra 231 *Sphuṭasiddhi (G. B. Palsule)*
27. Sphuṭasiddhinyāyavicāra (G. B. Palsule) 235
28. Annambhaṭṭa 237
29. Appayya Dīkṣita I 239
30. Bhaṭṭoji Dīkṣita 241 *Śabdakaustubha*
31. Śeṣa Viṣṇu 243
32. Śivarāmendra Sarasvatī 245
33. (Śeṣa) Cakrapāṇi (Datta) 247
34. Mallaya Yajvan 249
35. Nīlakaṇṭha Śukla 251
36. Nārāyaṇa (Śāstrin) 253
37. Koṇḍa (or Kauṇḍa) Bhaṭṭa 255 *Vaiyākaraṇabhūṣaṇa and Vaiyākaraṇabhūṣaṇasāra* *(S. D. Joshi)*
38. Tāraka Brahmānanda Sarasvatī 309
39. Cokkanātha or Śokanātha Dīkṣita 311
40. Tirumala Yajvan 313
41. (Rāma) Nārāyaṇa (Śarman) (Vandyopādhyāya) 315
42. Sadāśiva 317
43. Hari Dīkṣita 319
v5.c24.26
44. Rāmabhadra Dīkṣita 321 *Ṣaḍdarśinisiddhānta saṃgraha (Vyākaraṇa Section)* (K. Kunjunni Raja) *Śabdabhedanirūpaṇa* (K. Kunjunni Raja)
45. Nāgeśa (or Nāgoji) Bhaṭṭa 323 *Paramalaghumañjūṣā* (K. Kunjunni Raja) *Mahābhāṣyapradīpoddyota* (V. K. S. N. Raghavan) *Sphuṭavāda* (K. Kunjunni Raja)
46. Jñānendra Sarasvatī 351
47. Gopālakṛṣṇa Śāstrin 353
48. Dharaṇīdhara 355
49. Vaidyanātha Payaguṇḍa 357
50. Satyapriya Tīrtha Svāmin 359
51. Jayakṛṣṇa Maunin 361
52. Harivallabha 363
53. Vāsudeva Dīkṣita 365
54. Śrīkṛṣṇa Bhaṭṭa Maunin 367 *Sphuṭacandrikā* (G. B. Palsule)
55. Umāmaheśvara or Abhinava Kālidāsa 371
56. Nīlakaṇṭha Dīkṣita 373
57. Āṣādhara Bhaṭṭa 375
58. Rāmasevaka 377
59. Indradatta Upādhyāya 379
60. Kṛṣṇamitrācārya or Durbalācārya 381
61. Haribhaṭṭa 383
62. Dharaṇīdhara (II) 385
63. Mannudeva or Manyudeva or Gopāladeva 387
64. Bhairava Miśra 389
65. Kumāra Tātaya 391
66. Satārā Rāghavendrācārya (Gajendragadkar) 393
67. Gaṅgādhara Kavirāja 395
68. Tārānātha Tarkavācaspati 397
69. Khuddi Jhā (Śarman) 399
70. Nityānanda Panta Parvatīya 401
71. Dravyeśa Jhā 403
72. Sūryanārāyaṇa Śukla 405
73. Gopāla Śāstrī Nene 407
74. P. S. Anantanārāyaṇa Śāstrī 409
75. Brahmadeva 411
76. V. Kṛṣṇamācārya 413
77. Sadāśiva Śāstrī (Śarman) 415
78. Bāla Kṛṣṇa Pañcoli 417
79. Rāma Prasāda Tripāṭhī 419
80. Rudradhara Jhā Śarman 421 81. Kālikā Prasāda Śukla 423
82. Śabhāpati Śarman Upādhyāya 425
83. Raghunātha Śarman 427
84. Satyakāma Varmā 429
85. Rāmājñā Pāṇḍeya 431
v5.c24.27
BIBLIOGRAPHY ON GRAMMAR (*Vyākaraṇa*) (*Karl H. Potter*) 433
v5.c24.28
Part 1 : Authors Whose Dates Are (More or Less) Known 439
v5.c24.29
Part 2 : Authors and Works Whose Dates Are Unknown 517
v5.c24.30
Part 3 : Secondary Literature on *Vyākaraṇa*
v5.c24.31
NOTES 549
v5.c24.32
CUMULATIVE INDEX 563
v5.c24.33
This volume, the fifth in the *Encyclopedia of Indian Philosophies*, is devoted to the philosophy of the Grammarians. The introductory essay is intended to set their school in its context and to summarize the main Grammarian teachings. The summaries of primary sources that follow the introduction aim at making available the substance of the main philosophical ideas contained in these works, so that philosophers who are unable to read the original Sanskrit and who find difficulty in understanding and finding their way about in the translations (where such exist) can get an idea of the positions taken and arguments offered. The summaries, then, are intended primarily for philosophers and only secondarily for indologists. Certain sections of the works have been omitted or treated sketchily because they are repetitions or deemed less interesting for philosophers, though they may be of great interest to Sanskritists. The summaries are not likely to make interesting consecutive reading: they are provided in the spirit of a reference work. The appendix, which contains a lengthy bibliography of original and secondary writings on the philosophy of Grammar, is also presented as an aid to research.
v5.c24.34
References in the footnotes such as “G273” are to the bibliography presented in the appendix. References such as “RB10337” are to the first volume of this encyclopedia, 2nd edition (1984). Abbreviations used are listed at the beginning of the appendix.
v5.c24.35
Preparation of this volume has been made possible by grants from the American Institute of Indian Studies and the University of Calgary. These grants made possible the obtaining of the summaries and funded the travel that the editorial work required. The editors wish to thank Pradip R. Mehendiratta for his good offices. A debt of gratitude is also owed to the late Professor T. R. V. Murti, who gave generously of his time in working with Harold Coward in the volume’s planning stages. A research fellowship awarded to K. Kunjunni Raja by the Calgary Institute for the Humanities enabled the two editors to work together in completing the project. Special gratitude is due to Karl H. Potter, editor of the *Encyclopedia of Indian Philosophies*, for his many contributions, which have added greatly to the value of this volume.
v5.c24.36
1987 HAROLD G. COWARD
v5.c24.37
The first of these, the *Gospel of the Hebrews*, was used by the Jewish Christians of Palestine, and was probably a version of the Gospel of Matthew, though it contained some traditions not found in our canonical books. The *Gospel of the Egyptians* was used by the ascetic sects of Egypt, and was marked by a strong tendency towards mysticism and asceticism. The *Gospel of Peter* was used by the Docetists, and taught that Christ did not really suffer, but only appeared to do so.
v5.c24.38
These gospels, and others like them, were not accepted by the Church as a whole, and were gradually lost or suppressed. The Church, however, felt the need for a standard collection of writings that would serve as a guide for faith and practice. This led to the formation of the New Testament canon, a process that took several centuries to complete.
v5.c24.39
The first step in this process was the collection of the letters of Paul, which were widely circulated among the churches. Then came the four Gospels, which were accepted as the authoritative accounts of the life and teachings of Jesus. The Acts of the Apostles and the other epistles were gradually added, and finally the Book of Revelation.
v5.c24.40
By the end of the second century, the core of the New Testament was generally accepted by the Church. The process of canonization was not a formal act of a council, but rather a gradual recognition by the churches of the writings that were most valuable for their life and worship. It was only in the fourth century that the canon was formally closed by the councils of the Church.
v5.c24.41
The formation of the New Testament canon was a crucial event in the history of the Church. It provided a standard of faith and practice that helped to unify the Church and to protect it from the influence of heretical sects. It also ensured that the teachings of Jesus and the apostles would be preserved for future generations.
v5.c24.42
INTRODUCTION TO THE PHILOSOPHY OF THE GRAMMARIANS
v5.c24.43
The first of these, the *Gospel of the Hebrews*, was used by the Jewish Christians of Palestine, and was probably a version of the Gospel of Matthew, though it contained some traditions not found in our canonical books. The *Gospel of the Egyptians* was used by the ascetic sects of Egypt, and was marked by a strong tendency towards mysticism and asceticism. The *Gospel of Peter* was used by the Docetists, and taught that Christ did not really suffer, but only appeared to do so.
v5.c24.44
These gospels, and others like them, were not accepted by the Church as a whole, and were gradually lost or suppressed. The Church, however, felt the need for a standard collection of writings that would serve as a guide for faith and practice. This led to the formation of the New Testament canon, a process that took several centuries to complete.
v5.c24.45
The first step in this process was the collection of the letters of Paul, which were widely circulated among the churches. Then came the four Gospels, which were accepted as the authoritative accounts of the life and teachings of Jesus. The Acts of the Apostles and the other epistles were gradually added, and finally the Book of Revelation.
v5.c24.46
By the end of the second century, the core of the New Testament was generally accepted by the Church. The process of canonization was not a formal act of a council, but rather a gradual recognition by the churches of the writings that were most valuable for their life and worship. It was only in the fourth century that the canon was formally closed by the councils of the Church.
v5.c24.47
The formation of the New Testament canon was a crucial event in the history of the Church. It provided a standard of faith and practice that helped to unify the Church and to protect it from the influence of heretical sects. It also ensured that the teachings of Jesus and the apostles would be preserved for future generations.
Historical Résumé
1. The Place of Language in Indian Philosophy
v5.c24.s1.t1.1
Language has been one of the fundamental concerns of Indian philosophy and has attracted the serious attention of all thinkers from the outset. In India the study of language has never been the monopoly of the Grammarians or the Rhetoricians. All schools of thought began their philosophical discussions from the fundamental problem of communication. The poet-philosophers of the *Rg Veda* were greatly concerned with the powers and limitations of language as a means of communicating their mystic, personal experiences of an ecstatic nature to their fellows and they tried to stretch the power of language by various means. They praised language as a powerful and benign deity (*vāc*), ever ready to bestow favors on her devotees. The entire creation of the world was attributed by some sages to divine language, and it was generally recognized that the ordinary speech of mortals was only a fraction of that language.
v5.c24.s1.t1.2
Among the six accessories to the Vedas (*Vedāngas*) two are directly concerned with language: grammar (*vyākaraṇa*), or linguistic analysis, and etymology (*nirukta*), or interpretation of the meanings of selected words in the Vedas through etymological methods. Another accessory, metrics (*chandas*), is concerned with prosody.
v5.c24.s1.t1.3
Among the systems of Indian philosophy (*darśana*), Pūrvamīmāṃsā is called *vākyaśāstra* or the science of sentence interpretation, and the Nyāya system was also intrinsically language oriented. The Buddhist and Jain schools of thought have also devoted considerable attention to the working of language. Grammar and literary criticism (*sāhitya*) are directly interested in language problems, including semantic and philosophical issues, and Grammarians have claimed the status of an independent *darśana* for themselves.
v5.c24.s1.t1.4
The Indian approach to the study of language and linguistic problems has been characterized by both analysis and synthesis. On the one hand, a systematic attempt was made to analyze speech utterance in terms of sentences and words, stems and suffixes, morphemes and phonemes. The verbal root was considered as the core element to which preverbs, primary suffixes, and secondary suffixes, as well as nominal or verbal terminations, were added to evolve the word. On the other hand, rules of coalescence (saṃdhi) between these various elements and between words in a compound word or a sentence were studied and systematized. Rules of syntax were also studied carefully and attempts made to identify the cementing factors helping to form an integral unit.
v5.c24.s1.t1.5
The analytical method was older and more popular. The Sanskrit term for grammar, vyākaraṇa, means literally "linguistic analysis". Kumārila Bhaṭṭa, in the beginning of the seventh century, said that "we cannot think of any point of time totally devoid of some work or other dealing with the grammatical rules treating of the different kinds of roots and suffixes."¹ Śakalya’s Padapāṭha of the Ṛg Veda was one of the early attempts in the direction of analysis; he broke down the saṃhitā text of the Ṛg Veda into words, identifying even the separate elements of compound words. The Bṛhaddevatā, attributed to Śaunaka, says that a sentence is made up of words, and words are made up of phonemes (varṇa).² Pāṇini, who flourished about the fifth century B.C., brought the descriptive grammar of the Sanskrit language to its highest perfection in his Aṣṭādhyāyī, which has been praised by Leonard Bloomfield, the father of modern linguistics, as "the greatest monument of human intelligence."³ Pāṇini’s primary concern was the building up of Sanskrit words, both Vedic and classical, from verbal roots, preverbs, primary and secondary suffixes, and nominal and verbal terminations; but he was also interested in syntactic problems involved in the formation of compound words and the relationship of the nouns in a sentence with the action indicated by the verb. Pāṇini did not neglect meaning, but he was aware of the fact that meaning was likely to change over time and that the final authorities regarding meaning are the people who speak the language.
v5.c24.s1.t1.6
It was the etymological school of Yāska, author of the Nirukta commentary on the Nighaṇṭu list of select words in Vedic literature, that undertook a semantic analysis of words with their components in order to explain their meanings in the contexts of their occurrence. This school generally subscribed to the view that nouns are derived from verbal roots. The Uṇādisūtras follow this view and attempt to find derivations for even apparently integral words.
v5.c24.s1.t1.7
Mīmāṃsā, called vākyaśāstra, was mainly concerned with the methodology of textual interpretation in order to give a cogent explanation of prescriptive scriptural texts. It had to deal with apparent absurdities, inconsistencies, and contradictions, besides ambiguities, and evolved rules of interpretation that were accepted generally by all schools of thought and were used freely in legal practice and in commentaries. The Mīmāṃsakas used both analysis and synthesis in their approach to textual problems. They gave a semantic definition of the sentence, evolving the concepts of mutual expectancy (*ākāṅkṣā*), consistency (*yogyatā*), and contiguity (*āsatti*) as factors necessary for the existence of a sentence. It was the Mīmāṃsā school that developed the theory of metaphor to explain the apparent absurdities and inconsistencies in Vedic texts.
v5.c24.s1.t1.8
The Nyāya school, mainly interested in the theory of knowledge and the truth or falsity of judgments, had to be concerned with the theory of meaning, because understanding the proposition was a primary requirement for making any significant study about it.
v5.c24.s1.t1.9
The literary critics who were concerned with the understanding and appreciation of literature were very much interested in the stylistic analysis of language and in finding out the deviance of literary language from ordinary language, in order to see how far poets have been able to communicate their vision of beauty and emotional experience through the medium of words.
v5.c24.s1.t1.10
It is clear that for centuries the various schools of thought in India have carried out studies that have produced insights into the working of language. The Grammarians’ interest was not confined to the description and analysis of a particular language, but extended to the true nature and potentialities of language, including its role in effecting liberation.
2. The Basic Problems of Philosophy of Language
v5.c24.s1.t2.1
A. Linguistic Elements
v5.c24.s1.t2.2
One of the fundamental problems discussed is the relation between the linguistic elements (*śabda*) and their meanings (*artha*). The term *śabda* is normally used by the Grammarians to refer to a linguistic element, a meaningful unit of speech.⁴ Patañjali’s definition is that *śabda* is that which, when articulated, is seen to convey the idea of the referent. Maṇḍana Miśra defines it in his *Sphuṭasiddhi* as the cause that produces the idea of its meaning. In any case, it is the meaning bearer. In ordinary parlance people may use the word *śabda* to mean sound, as pointed out by Patañjali himself, but for the Grammarian it is the meaning-bearing unit.
v5.c24.s1.t2.3
Is it the articulated sound, or the phoneme (*varṇa*), or the word (*pada*), or the sentence (*vākya*) that is referred to by the term *śabda*? According to the *sphoṭa* theory of Bhartṛhari it is the complete utterance of the sentence that is the unit, and it is called *vākyasphoṭa*; but at a lower analytical level the word can be considered as the unit, for which the term *padasphoṭa* is used by the Grammarian. Those who know the language very well think and speak in units of sentences and also hear whole sentences. It is only those who do not know the language properly who hear words or phonemes or bits of sounds and have to struggle with them to get the connected sentence meaning. But in grammatical texts the words are taken as the unit for the sake of easy understanding.
v5.c24.s1.t2.4
This view is not acceptable to the Mīmāṃsakas, who consider the letter (permanent articulated sound-unit) or phoneme (*varṇa*) to be the *śabda* or unit of language and the meaning bearer. They assume phonemes to be permanent and each utterance to be their realization. To the Naiyāyikas *śabda* means sound produced by the speaker and heard by the listener, and it is impermanent; *pada* means a morpheme (meaningful unit).
v5.c24.s1.t2.5
B. Meaning (*Artha*)
v5.c24.s1.t2.6
What is meant by *artha* or meaning? Is it the universal that is intended, or the particular? According to Kātyāyana and Patañjali, two different positions were held by two ancient Grammarians, Vyāḍi and Vājapyāyana, the former holding that words refer to *dravya*, "substance" or "individual", and the latter holding that words (including proper names) refer to *jāti*, "universal" or "attribute". Pāṇini seems to have left the question open, holding that words could refer to individuals or to the universals. The Mīmāṃsakas held that the primary meaning of a word is the universal and the sense of the particular in a sentence is obtained either through secondary significative power (according to Bhāṭṭa Mīmāṃsakas) or through both the universal and the particular being grasped by the same perceptive effort simultaneously (according to the Prābhākaras). The early Naiyāyikas considered the meaning of words as comprising universal (*jāti*), configuration (*ākṛti*), and particular;⁵ later Naiyāyikas held that the primary meaning of words is the individual as qualified by the universal (*jātiviśiṣṭavyakti*). The Buddhists of Dignāga's school held that the meaning is *vikalpa*, a mental construct that has no direct correspondence with the real, its nature being to exclude other things (*anyāpoha*). The function of a word or a name is the exclusion of other possibilities.
v5.c24.s1.t2.7
C. Significative Power (*Śakti*)
v5.c24.s1.t2.8
The significative power of words (*śakti*) is based on the relation that exists between a word and its meaning. The Grammarians hold that in the case of ordinary words in everyday speech it is permanent; but in the case of technical terms it is based on the convention. The Mīmāṃsakas consider the relation as "original" (*autpattika*), that is, as permanent or eternal. The Grammarians explain this permanence in two ways: *pravāhanityatā* and *yogyatānityatā*. We learn language from our elders; they in turn learned it from their forefathers; thus it could be traced back to any conceivable period of human society. This type of permanence is *pravāhanityatā*. The other view is based on the innate capacity of words to express any meaning; this capacity (*yogyatā*) is restricted by convention. Patañjali made a distinction between absolute eternity (*kūṭasthanityatā*), by which an item is not liable to any modification, and the perennial nature as used through generations of speakers (*pravāhanityatā*).
v5.c24.s1.t2.9
D. Polysemy
v5.c24.s1.t2.10
It is generally believed that in an ideal language a word must have only one meaning, and a sense must have only one word to express it. This binary relationship between a word and its meaning is accepted in principle by all schools of thought. It is also believed that this relationship, which is the basis for the significative power of words, is stable and constant because linguistic communication would be impossible without it. If there is no general understanding of the meaning of words shared by the speaker and listener there will be chaos and mutual comprehension will be jeopardized.
v5.c24.s1.t2.11
The existence of polysemy is recognized in actual practice, however. Two words may have the same form, and the same word may develop more than one meaning. The problem of homophones and homonyms has been discussed by scholars like Bhartṛhari. Yāska’s discussion about the principle of word derivation in Sanskrit also sheds considerable light on the problem of synonyms. Nouns are normally derived from verbal roots. If all nouns are so derived from verbal roots denoting action, every object will have as many names as the actions with which it is associated, and by the same token each noun could be applied to as many objects as are associated with that action indicated by that verbal root. Yāska’s answer to the problem is that there are no restrictions. Language designates things in an incomplete manner; it can choose only one of the many activities associated with an object. Hence there is some sort of permanent relation between a word and its meaning.
v5.c24.s1.t2.12
It is accepted that even the primary meaning of a word is not definitely circumscribed and that the boundaries of the meaning often change on the basis of contextual factors, not only in the case of ambiguous words but even in that of ordinary words: thus “man is mortal” does not mean “woman is immortal”; but in the phrase “man and woman”, “man” does not include “woman”. When there is conflict between the correct etymological meaning and the popular usage, the meaning current in popular usage among the educated elite is to be accepted. Grammatical analysis and etymological interpretations are only means of approach; the final authority is the popular usage of the cultured.
v5.c24.s1.t2.13
E. Secondary Meaning (Lakṣaṇā)
v5.c24.s1.t2.14
Even though it is accepted that every word has a primary stable meaning core, in actual practice shifts in meaning, metaphoric transfers, and secondary usages are quite common. If there is discrepancy in sense when the primary meaning is taken, the passage will have to be explained by resorting to the secondary meaning. There are three conditions considered necessary for resorting to secondary meaning. The first, is inconsistency or incongruity of the words taken in the literal sense. A sentence like "He is an ass" or "He is a firebrand" cannot be taken in the literal sense because the human being referred to cannot be an animal or an inanimate object. A sentence like "The house is in the river" does not make sense, because a house cannot exist in the river. In such cases the primary meaning of the word has to be given up and another meaning used. The second condition is that the actual meaning and the primary meaning must be related in some way; it may be on the basis of similarity or common quality or it may be on the basis of some other relationship like proximity. The example "He is an ass" can be explained if the term "ass" is interpreted as "a fool" (as the donkey is notorious for its dullness). The example of the house on the river has to be explained by taking "river" to mean the bank of the river on the basis of proximity. The third condition for resorting to secondary significance is either sanction by popular usage, as in the case of faded metaphors, or a special purpose for which it is resorted to, as in the case of intentional metaphors. The inconsistency of primary meaning can mean impossibility of syntactic connection from the point of view of meaning, or it can mean inconsistency in the context. As an example, in "see that crows do not spoil the curd" "crows" implies all beings, including a dog, who might spoil it.
v5.c24.s1.t2.15
Literary critics like Ānandavardhana proposed the element of purpose in intentional metaphors and pointed out its importance in enriching literature's content.
v5.c24.s1.t2.16
F. Conditions for Syntactic Relation
v5.c24.s1.t2.17
How can we get a connected meaning from a sentence if each word gives only its isolated sense, which is of a universal nature? This problem has been discussed in India since ancient times, and three main factors have been pointed out as unifying of sentence meaning: expectancy (ākāṅkṣā), consistency (yogyatā), and contiguity (āsatti). Words in a sentence must have mutual expectancy. Pāṇini hinted as much when he stressed the need for sāmarthya or capacity among the meanings of words for mutual connection, mainly in compound words.⁶ This sāmarthya has been interpreted as similar to ākāṅkṣā or mutual expectancy and unity of sense. Later the Mīmāṃsakas developed this concept, and the logicians made further modifications. Mutual expectancy consists in a word being unable to convey a complete sense in the absence of another word. Literally it is the desire on the part of the listeners to know the other words in the sentence in order to complete the sense. A word is said to have expectancy for another if it cannot, without the latter, produce knowledge of its interconnection in an utterance. The Mīmāṃsakas were more interested in psychological expectancy, while the logicians and the Grammarians stressed the need for syntactic expectancy.
v5.c24.s1.t2.18
To this primary condition were added two more, yogyatā or consistency of sense and āsatti or the contiguity of the words. Grammarians did not emphasize the importance of yogyatā for to them it is enough for a sentence to give a syntactically connected meaning. Its veracity is not a condition. From the Grammarian’s point of view lakṣaṇā, secondary meaning, is also of little interest. “He is a boy” and “He is an ass” are equally valid for them. Even empty phrases like “the child of a barren woman” are linguistically valid to them, for Grammarians are not concerned with the real existence of the thing meant by an expression. Yogyatā involves a judgment on the sense or nonsense of a sentence. There is difference of opinion about whether it should be taken as a positive condition. If the lack of yogyatā—inconsistency—is only apparent and can be explained away by resorting to the metaphorical meaning of a word in the sentence, there is no difficulty in understanding the sentence’s meaning.
v5.c24.s1.t2.19
Āsatti or contiguity is the uninterrupted utterance or the unbroken apprehension of the words in a sentence. In the case of elliptical sentences, one school believes that the syntactic relation is known by supplying the necessary meaning, while another school insists that the missing words have to be supplied and the meaning obtained. Some take tātparya, the intention of the speaker known from contextual factors, as a fourth condition for understanding the meaning of a sentence.
v5.c24.s1.t2.20
G. Sentence Meaning
v5.c24.s1.t2.21
Regarding the comprehension of the sentence meaning there are two main theories, called anvitābhidhāna and abhihitānvaya. Speech is purposive in nature. People use words with the intention of conveying a connected, unified sense. Hence from the use of words in juxtaposition it is assumed that the speaker has uttered them with the intention of conveying a connected sense. Expectancy, consistency, and contiguity help in this comprehension of a unified sentence meaning. The sentence meaning is something more than the sum of the word meanings. Besides the word meanings, the syntactic connection of the word meanings has to be conveyed. The *abhihitānvaya* theory says that in a sentence each word gives out its individual isolated meaning (which is universal) and their significative power is exhausted with that. Then with the help of *lakṣaṇā* (secondary significative power) the syntactic relationship is obtained, and thus the sentence meaning is understood. According to the *anvitābhidhāna* school, by contrast, each word in a sentence conveys not only its isolated meaning but also the syntactic element. The words convey the meaning of the universal and simultaneously the meaning as referring to the particular. The words themselves also give the syntactic relationship. Thus the entire sentence meaning is conveyed by the words themselves. The Naiyāyikas, who believe that the words in a sentence denote primary meanings that are particulars as qualified by universal traits, contend that the sentence meaning is an association of the word meanings (*saṃsargamaryādā*).
v5.c24.s1.t2.22
H. *Sphota* Theory
v5.c24.s1.t2.23
Even in ancient India there were some scholars who emphasized the unreal nature of words and advocated the need for taking the sentence as a whole. In the *Nirukta* Yāska refers to Audumbarāyaṇa’s theory that it is the statement as a whole that is regularly present in the perceptive faculty of the hearer.⁷ The *sphoṭa* theory, fully promulgated by Bhartṛhari in the fifth century of the Christian era, is one of the most important contributions of India to the problem of meaning. He insisted that the fundamental linguistic fact is the complete utterance or sentence. Just as a letter or a phoneme has no parts, so also the word and the sentence are to be taken as complete integral units, not as made up of smaller elements. Bhartṛhari says that although linguistic analysis—splitting sentences into words and further into roots and suffixes and into phonemes—may be a useful means for studying language, it has no reality. In a speech situation, communication is always through complete utterances. The speaker thinks and the listener understands the utterance as a single unit. It is only those who do not know the language thoroughly who analyze it into words, and further bits, in order to get a connected meaning. Those who know the language will conceive the idea and the expression as a single unit and express it; and the listener likewise comprehends it as a whole, the understanding is as an instantaneous flash of insight (*pratibhā*). The fact that the expression has to be through the medium of phonemes, through a temporal or spatial series, does not warrant our considering it as made up of parts. When a painter conceives a picture in his mind and paints it on a canvas, he may use various colors, and make various strokes; that does not mean that the picture is not a unit. And we see the picture as a unit, not as different colors and strokes. Just as the meaning is unitary, integral, and indivisible, the symbol that signifies it must also be unitary and indivisible. This concept is called *sphoṭa*—the sentence taken as an integral symbol, in which its apparent parts are irrelevant to it as parts. It is not something hypothetically assumed to explain language behavior; it is actually experienced and known through perception. On hearing a sentence those who know the language well hear the sentence, not the phonemes or sound bits or even words. Those who do not know the language may hear only the sound bits. The *sphoṭa* theory says that hearing the whole sentence is the real experience, while the apparent experience of hearing the sound bits is only for those who do not know the language.
v5.c24.s1.t2.24
I. Componential Analysis of Word Meaning
v5.c24.s1.t2.25
It may be noted that even the so-called unity of meaning is often an illusion, for it is the language that makes the unity. Yāska in the fifth century B.C. and, following him, Bhartṛhari in the fifth century of the Christian era have pointed out that a verb conveys a series of operations or activities taking place in a particular temporal sequence. Thus the word "cooks" conveys the idea of a series of activities—preparing the fire, putting the vessel on it, pouring water in the vessel, washing the rice, putting it in the water, blowing the fire to make it burn properly, putting out the fire, removing the excess water, and so on. It is the word "cooks" that collects all of these activities into a unitary, integral action. Each of these activities can be further analyzed into a series of activities taking place in time.
v5.c24.s1.t2.26
Later philosophers of language made further componential analysis of words from the semantic point of view and declared that every verbal root (*dhātu*) involved two semantic factors, activity (*vyāpāra*) and goal or result (*phala*). The verb "he cooks" means an activity directed toward the softening of the rice, and so forth. There is a difference of opinion about whether both are primary meanings of the verbal root or one can be taken as the main meaning and the other as subsidiary. The verb was divided into the root and the suffix, and separate meaning bits assigned to them. Maṇḍana Miśra said that the meaning of the root is the result, and it is the suffix that indicates the activity. With the addition of preverbs the meaning changes considerably in Sanskrit, and there have been discussions of whether all the meanings are present in a latent form in the root, to be revealed by the preverbs, or these preverbs can be assigned specific meanings.
v5.c24.s1.t2.27
J. Suggestion (*vyañjanā*)
v5.c24.s1.t2.28
The theory of literal (primary) and metaphoric (secondary) meaning developed by the Nyāya and Mīmāṃsā schools of sentence interpretation in ancient India was extended farther by Ānandavardhana in the second half of the ninth century to include emotive and other associative meanings under linguistic meaning. He did not attack the usual division of speech into words, into stems and suffixes and the distinction between the primary and secondary meanings of words. He accepted all of these concepts, but in addition he postulated a third capability of language, which he called *vyañjanā* or the capacity to suggest meaning other than its literal or metaphoric meaning. Ānandavardhana pointed out that this suggestive function of language has a vital role to play in literature.
v5.c24.s1.t2.29
K. Time
v5.c24.s1.t2.30
The concept of time and its divisions were discussed by such Grammarians as Patañjali and Bhartṛhari. The division of time into past, present, and future has a place in grammar, but the rules of usage given are not strictly followed in actual practice. The present tense (*laṭ*) is used to indicate the entire stretch of time included from the beginning of the action till its completion. "He is cooking" can mean he has started cooking and the operations are not yet completed. The present need not be momentary. Usages like "the mountains are standing" and "the rivers are flowing" mean that they continue to stand or flow.
v5.c24.s1.t2.31
Bhartṛhari considers time to be the most important power of Śabda Brahman, relatively more independent than other powers, and to be responsible for regulating them. The ancient authority Vārṣāyani said that becoming (*bhāva*) has six stages from birth to decay.⁸
v5.c24.s1.t2.32
L. Gender
v5.c24.s1.t2.33
Grammatical gender has attracted much speculation because in Sanskrit the grammatical gender does not coincide with sex, and words referring to the same object may occur in different genders.
3. The Literature on Grammar
v5.c24.s1.t3.1
Grammar (*vyākaraṇa*) was recognized from the earliest times in India as a distinct science, a field of knowledge with its own parameters that distinguished it from other sciences such as astronomy (*jyotiṣa*), architecture (*śilpaśāstra*), agriculture (*kṛṣiśāstra*), and the like. The coverage in this encyclopedia thus presents a peculiar problem somewhat different from that faced in dealing with the literature of other philosophical systems. Whereas in the cases of those other systems the "philosophical" literature is confined to that material which relates to the overall aim of liberation and those treatises which discuss theoretical issues pertinent to that aim in a polemical context, in the case of the grammatical literature the coverage has been widened to include those works which deal with theory of meaning and the related issues discussed in the previous section. As a result, the literature that is summarized here in the subsequent pages represents a selection from the total corpus of vyākaraṇa literature classified as such in India. For that reason it seems appropriate to attempt at this point to place the writers treated in the body of this volume within the total roster of authors whose works constitute the entire corpus of grammatical literature.
v5.c24.s1.t3.2
To this end we have provided as an appendix a bibliography of vyākaraṇa works and authors that covers what is currently known of the literature on grammar in general. Within the confines of this literature the present volume provides summaries of some works that are (1) in print and (2) especially relevant to the concerns touched on in the preceding section. Many of the works that are not summarized do, without doubt, bear on these topics. The distinction drawn must seem from the standpoint of Indian tradition rather artificial. We can only remind our readers that the entire project of this encyclopedia is directed toward a readership that comprises in the main those trained in philosophy as understood in the western world. The bibliography will suggest to such readers the extent of classical writings on grammar in general. From these writings we have selected the ones that seem most informative concerning matters of general philosophical interest, as "philosophy" has been understood for the purpose of this entire encyclopedic project.
v5.c24.s1.t3.3
The origin of grammar in India, as with so much else, is unclear. What is clear is that it was recognized as a field of study from a very early time. Pāṇini and Yāska, who must represent a stage of thought several centuries prior to the Christian era, both refer to a number of grammatical authorities and their views. In some cases the names cited by both writers are identical. According to a statement in the Brāhmaṇas it was the god Indra who first analyzed a speech utterance in terms of its parts.⁹ Some scholars speak of an "Aindra" school of Grammarians in locating the origins of grammar. Patañjali refers to a tradition involving the futile attempt of Bṛhaspati to teach the language by enumerating its words and their meanings.¹⁰ The earliest historical figure who dealt with linguistic study seems to be Śākalya, author of the Padapāṭha of the Ṛg Veda, who is mentioned by Pāṇini. According to Bhartṛhari, the sage Audumbarāyaṇa (also mentioned by Yāska), together with Vārttākṣa, seems to have held views similar to the sphoṭa theory.¹¹ A late tradition makes Sphoṭāyana, mentioned by Pāṇini, the founder of the sphoṭa theory.¹² Śakaṭāyana held the view that all words must be derived from verbal roots; some people consider him to be the author of the Uṇādisūtras. Gārgya and others held the view that not all nouns can be traced to verbal roots. But no authenticated works of any of these pre-Pāṇinian writers have come down to us, and it is difficult if not impossible to say which, if any, of Pāṇini’s rules may have been taken from his predecessors.
v5.c24.s1.t3.4
Yāska and Pāṇini are the two great early writers on language. They belong to a period several centuries before Christ, possibly the fifth century. Yāska is generally considered to be earlier than Pāṇini, but Paul Thieme holds that Yāska knew Pāṇini.¹³ George Cardona thinks it wise to leave the problem open.¹⁴ The two writers are classified into different genres of literature by the Indian tradition. Yāska’s Nirukta provides the name for a discipline of etymology counted as separate from grammar (vyākaraṇa), the discipline for which Pāṇini stands as the major seminal figure. (The bibliography appended to this volume confines itself to works classed in the latter discipline.)
v5.c24.s1.t3.5
A. Pāṇini’s Aṣṭādhyāyī
v5.c24.s1.t3.6
Pāṇini’s basic work is merely titled “The Eight-Chaptered” (Aṣṭādhyāyī). But a very remarkable work it is, providing a model for recent and contemporary work in descriptive linguistics that can stand with the best efforts of modern analysts. The eight chapters constitute a complete descriptive analytical grammar of the Sanskrit language, comprising about four thousand rules called sūtras preceded by a list of sounds divided into fourteen groups, which are called the śiva-, pratyāhāra-, or maheśvara-sūtras. In these rules the language is analyzed into verbal and nominal bases, so that the bases have come to be cataloged in two lists known as dhātupāṭha—the exhaustive lists of primitive verbal roots—and gaṇapāṭha—the selective lists of nouns, verbs, and so on for application in his rules. It is not clear whether one person wrote all these different components of the fundamental Pāṇinian corpus. A further feature of Pāṇini’s method is a set of metarules of paribhāṣās, which tell us in which order to apply the rules, where exceptions are to be made, and so forth. All of these components are, in any case, made the subject of a grand commentarial tradition extending to the present.
v5.c24.s1.t3.7
Certain additional materials are ascribed to Pāṇini by the tradition, though scholarship is less agreed on the authenticity of these ascriptions. Among them, one group of sūtras, the Uṇādisūtras, provides rules for introducing affixes after verbal roots to derive nominal bases. The authorship of these Uṇādisūtras is frequently attributed to Śakaṭāyana instead of Pāṇini; and some scholars have found them to date from a later period. It is likely that there is truth in all of these views, in other words, that the sūtras represent a development of analysis over a long period.
v5.c24.s1.t3.8
A second set of ancillary sūtras are the Phiṭsūtras, which provide principles of accentuation. A feature of these rules is that accents are presupposed for nominal bases, from which the rules derive revisions of accentuation for the wholes of which those bases form a part. Because Pāṇini’s rules contain none specifying accents for nominal bases, Cardona reports that “it is clear that the *phiṭsūtras* cannot be attributed to Pāṇini.”[¹⁵] In fact, they are traditionally ascribed to Sāntanava, a rather later writer. Still, there are scholars who believe the *phiṭsūtras* date from a pre-Pāṇinian period.
First 98 of 3,704 passages. Read on